diff options
Diffstat (limited to 'drivers/net/ethernet')
642 files changed, 42555 insertions, 12635 deletions
diff --git a/drivers/net/ethernet/3com/typhoon.c b/drivers/net/ethernet/3com/typhoon.c index 049cc0158a64..05e15b6e5e2c 100644 --- a/drivers/net/ethernet/3com/typhoon.c +++ b/drivers/net/ethernet/3com/typhoon.c @@ -789,8 +789,8 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev) * it with zeros to ETH_ZLEN for us. */ if (skb_shinfo(skb)->nr_frags == 0) { - skb_dma = pci_map_single(tp->tx_pdev, skb->data, skb->len, - PCI_DMA_TODEVICE); + skb_dma = dma_map_single(&tp->tx_pdev->dev, skb->data, + skb->len, DMA_TO_DEVICE); txd->flags = TYPHOON_FRAG_DESC | TYPHOON_DESC_VALID; txd->len = cpu_to_le16(skb->len); txd->frag.addr = cpu_to_le32(skb_dma); @@ -800,8 +800,8 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev) int i, len; len = skb_headlen(skb); - skb_dma = pci_map_single(tp->tx_pdev, skb->data, len, - PCI_DMA_TODEVICE); + skb_dma = dma_map_single(&tp->tx_pdev->dev, skb->data, len, + DMA_TO_DEVICE); txd->flags = TYPHOON_FRAG_DESC | TYPHOON_DESC_VALID; txd->len = cpu_to_le16(len); txd->frag.addr = cpu_to_le32(skb_dma); @@ -818,8 +818,8 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev) len = skb_frag_size(frag); frag_addr = skb_frag_address(frag); - skb_dma = pci_map_single(tp->tx_pdev, frag_addr, len, - PCI_DMA_TODEVICE); + skb_dma = dma_map_single(&tp->tx_pdev->dev, frag_addr, + len, DMA_TO_DEVICE); txd->flags = TYPHOON_FRAG_DESC | TYPHOON_DESC_VALID; txd->len = cpu_to_le16(len); txd->frag.addr = cpu_to_le32(skb_dma); @@ -1349,12 +1349,12 @@ typhoon_download_firmware(struct typhoon *tp) image_data = typhoon_fw->data; fHdr = (struct typhoon_file_header *) image_data; - /* Cannot just map the firmware image using pci_map_single() as + /* Cannot just map the firmware image using dma_map_single() as * the firmware is vmalloc()'d and may not be physically contiguous, - * so we allocate some consistent memory to copy the sections into. + * so we allocate some coherent memory to copy the sections into. */ err = -ENOMEM; - dpage = pci_alloc_consistent(pdev, PAGE_SIZE, &dpage_dma); + dpage = dma_alloc_coherent(&pdev->dev, PAGE_SIZE, &dpage_dma, GFP_ATOMIC); if (!dpage) { netdev_err(tp->dev, "no DMA mem for firmware\n"); goto err_out; @@ -1459,7 +1459,7 @@ err_out_irq: iowrite32(irqMasked, ioaddr + TYPHOON_REG_INTR_MASK); iowrite32(irqEnabled, ioaddr + TYPHOON_REG_INTR_ENABLE); - pci_free_consistent(pdev, PAGE_SIZE, dpage, dpage_dma); + dma_free_coherent(&pdev->dev, PAGE_SIZE, dpage, dpage_dma); err_out: return err; @@ -1526,8 +1526,8 @@ typhoon_clean_tx(struct typhoon *tp, struct transmit_ring *txRing, */ skb_dma = (dma_addr_t) le32_to_cpu(tx->frag.addr); dma_len = le16_to_cpu(tx->len); - pci_unmap_single(tp->pdev, skb_dma, dma_len, - PCI_DMA_TODEVICE); + dma_unmap_single(&tp->pdev->dev, skb_dma, dma_len, + DMA_TO_DEVICE); } tx->flags = 0; @@ -1608,8 +1608,8 @@ typhoon_alloc_rx_skb(struct typhoon *tp, u32 idx) skb_reserve(skb, 2); #endif - dma_addr = pci_map_single(tp->pdev, skb->data, - PKT_BUF_SZ, PCI_DMA_FROMDEVICE); + dma_addr = dma_map_single(&tp->pdev->dev, skb->data, PKT_BUF_SZ, + DMA_FROM_DEVICE); /* Since no card does 64 bit DAC, the high bits will never * change from zero. @@ -1664,20 +1664,19 @@ typhoon_rx(struct typhoon *tp, struct basic_ring *rxRing, volatile __le32 * read if (pkt_len < rx_copybreak && (new_skb = netdev_alloc_skb(tp->dev, pkt_len + 2)) != NULL) { skb_reserve(new_skb, 2); - pci_dma_sync_single_for_cpu(tp->pdev, dma_addr, - PKT_BUF_SZ, - PCI_DMA_FROMDEVICE); + dma_sync_single_for_cpu(&tp->pdev->dev, dma_addr, + PKT_BUF_SZ, DMA_FROM_DEVICE); skb_copy_to_linear_data(new_skb, skb->data, pkt_len); - pci_dma_sync_single_for_device(tp->pdev, dma_addr, - PKT_BUF_SZ, - PCI_DMA_FROMDEVICE); + dma_sync_single_for_device(&tp->pdev->dev, dma_addr, + PKT_BUF_SZ, + DMA_FROM_DEVICE); skb_put(new_skb, pkt_len); typhoon_recycle_rx_skb(tp, idx); } else { new_skb = skb; skb_put(new_skb, pkt_len); - pci_unmap_single(tp->pdev, dma_addr, PKT_BUF_SZ, - PCI_DMA_FROMDEVICE); + dma_unmap_single(&tp->pdev->dev, dma_addr, PKT_BUF_SZ, + DMA_FROM_DEVICE); typhoon_alloc_rx_skb(tp, idx); } new_skb->protocol = eth_type_trans(new_skb, tp->dev); @@ -1791,8 +1790,8 @@ typhoon_free_rx_rings(struct typhoon *tp) for (i = 0; i < RXENT_ENTRIES; i++) { struct rxbuff_ent *rxb = &tp->rxbuffers[i]; if (rxb->skb) { - pci_unmap_single(tp->pdev, rxb->dma_addr, PKT_BUF_SZ, - PCI_DMA_FROMDEVICE); + dma_unmap_single(&tp->pdev->dev, rxb->dma_addr, + PKT_BUF_SZ, DMA_FROM_DEVICE); dev_kfree_skb(rxb->skb); rxb->skb = NULL; } @@ -2305,7 +2304,7 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) goto error_out_disable; } - err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32)); + err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)); if (err < 0) { err_msg = "No usable DMA configuration"; goto error_out_mwi; @@ -2354,8 +2353,8 @@ typhoon_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) /* allocate pci dma space for rx and tx descriptor rings */ - shared = pci_alloc_consistent(pdev, sizeof(struct typhoon_shared), - &shared_dma); + shared = dma_alloc_coherent(&pdev->dev, sizeof(struct typhoon_shared), + &shared_dma, GFP_KERNEL); if (!shared) { err_msg = "could not allocate DMA memory"; err = -ENOMEM; @@ -2508,8 +2507,8 @@ error_out_reset: typhoon_reset(ioaddr, NoWait); error_out_dma: - pci_free_consistent(pdev, sizeof(struct typhoon_shared), - shared, shared_dma); + dma_free_coherent(&pdev->dev, sizeof(struct typhoon_shared), shared, + shared_dma); error_out_remap: pci_iounmap(pdev, ioaddr); error_out_regions: @@ -2536,8 +2535,8 @@ typhoon_remove_one(struct pci_dev *pdev) pci_restore_state(pdev); typhoon_reset(tp->ioaddr, NoWait); pci_iounmap(pdev, tp->ioaddr); - pci_free_consistent(pdev, sizeof(struct typhoon_shared), - tp->shared, tp->shared_dma); + dma_free_coherent(&pdev->dev, sizeof(struct typhoon_shared), + tp->shared, tp->shared_dma); pci_release_regions(pdev); pci_clear_mwi(pdev); pci_disable_device(pdev); diff --git a/drivers/net/ethernet/8390/axnet_cs.c b/drivers/net/ethernet/8390/axnet_cs.c index a00b36f91d9f..2488bfdb9133 100644 --- a/drivers/net/ethernet/8390/axnet_cs.c +++ b/drivers/net/ethernet/8390/axnet_cs.c @@ -657,8 +657,10 @@ static void block_input(struct net_device *dev, int count, outb_p(E8390_RREAD+E8390_START, nic_base + AXNET_CMD); insw(nic_base + AXNET_DATAPORT,buf,count>>1); - if (count & 0x01) - buf[count-1] = inb(nic_base + AXNET_DATAPORT), xfer_count++; + if (count & 0x01) { + buf[count-1] = inb(nic_base + AXNET_DATAPORT); + xfer_count++; + } } @@ -1270,10 +1272,12 @@ static void ei_tx_intr(struct net_device *dev) ei_local->txing = 1; NS8390_trigger_send(dev, ei_local->tx2, ei_local->tx_start_page + 6); netif_trans_update(dev); - ei_local->tx2 = -1, + ei_local->tx2 = -1; ei_local->lasttx = 2; + } else { + ei_local->lasttx = 20; + ei_local->txing = 0; } - else ei_local->lasttx = 20, ei_local->txing = 0; } else if (ei_local->tx2 < 0) { @@ -1289,9 +1293,10 @@ static void ei_tx_intr(struct net_device *dev) netif_trans_update(dev); ei_local->tx1 = -1; ei_local->lasttx = 1; + } else { + ei_local->lasttx = 10; + ei_local->txing = 0; } - else - ei_local->lasttx = 10, ei_local->txing = 0; } // else // netdev_warn(dev, "unexpected TX-done interrupt, lasttx=%d\n", diff --git a/drivers/net/ethernet/8390/lib8390.c b/drivers/net/ethernet/8390/lib8390.c index babc92e2692e..e84021282edf 100644 --- a/drivers/net/ethernet/8390/lib8390.c +++ b/drivers/net/ethernet/8390/lib8390.c @@ -50,6 +50,7 @@ */ +#include <linux/build_bug.h> #include <linux/module.h> #include <linux/kernel.h> #include <linux/jiffies.h> @@ -112,8 +113,10 @@ static void do_set_multicast_list(struct net_device *dev); static void __NS8390_init(struct net_device *dev, int startp); static unsigned version_printed; -static u32 msg_enable; -module_param(msg_enable, uint, 0444); +static int msg_enable; +static const int default_msg_level = (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_RX_ERR | + NETIF_MSG_TX_ERR); +module_param(msg_enable, int, 0444); MODULE_PARM_DESC(msg_enable, "Debug message level (see linux/netdevice.h for bitmap)"); /* @@ -597,10 +600,12 @@ static void ei_tx_intr(struct net_device *dev) ei_local->txing = 1; NS8390_trigger_send(dev, ei_local->tx2, ei_local->tx_start_page + 6); netif_trans_update(dev); - ei_local->tx2 = -1, + ei_local->tx2 = -1; ei_local->lasttx = 2; - } else - ei_local->lasttx = 20, ei_local->txing = 0; + } else { + ei_local->lasttx = 20; + ei_local->txing = 0; + } } else if (ei_local->tx2 < 0) { if (ei_local->lasttx != 2 && ei_local->lasttx != -2) pr_err("%s: bogus last_tx_buffer %d, tx2=%d\n", @@ -612,8 +617,10 @@ static void ei_tx_intr(struct net_device *dev) netif_trans_update(dev); ei_local->tx1 = -1; ei_local->lasttx = 1; - } else - ei_local->lasttx = 10, ei_local->txing = 0; + } else { + ei_local->lasttx = 10; + ei_local->txing = 0; + } } /* else netdev_warn(dev, "unexpected TX-done interrupt, lasttx=%d\n", ei_local->lasttx); @@ -969,14 +976,14 @@ static void ethdev_setup(struct net_device *dev) { struct ei_device *ei_local = netdev_priv(dev); - if ((msg_enable & NETIF_MSG_DRV) && (version_printed++ == 0)) - pr_info("%s", version); - ether_setup(dev); spin_lock_init(&ei_local->page_lock); - ei_local->msg_enable = msg_enable; + ei_local->msg_enable = netif_msg_init(msg_enable, default_msg_level); + + if (netif_msg_drv(ei_local) && (version_printed++ == 0)) + pr_info("%s", version); } /** @@ -1014,8 +1021,7 @@ static void __NS8390_init(struct net_device *dev, int startp) ? (0x48 | ENDCFG_WTS | (ei_local->bigendian ? ENDCFG_BOS : 0)) : 0x48; - if (sizeof(struct e8390_pkt_hdr) != 4) - panic("8390.c: header struct mispacked\n"); + BUILD_BUG_ON(sizeof(struct e8390_pkt_hdr) != 4); /* Follow National Semi's recommendations for initing the DP83902. */ ei_outb_p(E8390_NODMA+E8390_PAGE0+E8390_STOP, e8390_base+E8390_CMD); /* 0x21 */ ei_outb_p(endcfg, e8390_base + EN0_DCFG); /* 0x48 or 0x49 */ diff --git a/drivers/net/ethernet/8390/pcnet_cs.c b/drivers/net/ethernet/8390/pcnet_cs.c index 164c3ed550bf..9d3b1e0e425c 100644 --- a/drivers/net/ethernet/8390/pcnet_cs.c +++ b/drivers/net/ethernet/8390/pcnet_cs.c @@ -1178,8 +1178,10 @@ static void dma_block_input(struct net_device *dev, int count, outb_p(E8390_RREAD+E8390_START, nic_base + PCNET_CMD); insw(nic_base + PCNET_DATAPORT,buf,count>>1); - if (count & 0x01) - buf[count-1] = inb(nic_base + PCNET_DATAPORT), xfer_count++; + if (count & 0x01) { + buf[count-1] = inb(nic_base + PCNET_DATAPORT); + xfer_count++; + } /* This was for the ALPHA version only, but enough people have been encountering problems that it is still here. */ diff --git a/drivers/net/ethernet/adaptec/starfire.c b/drivers/net/ethernet/adaptec/starfire.c index ba0055bb1614..555299737b51 100644 --- a/drivers/net/ethernet/adaptec/starfire.c +++ b/drivers/net/ethernet/adaptec/starfire.c @@ -886,7 +886,9 @@ static int netdev_open(struct net_device *dev) tx_ring_size = ((sizeof(starfire_tx_desc) * TX_RING_SIZE + QUEUE_ALIGN - 1) / QUEUE_ALIGN) * QUEUE_ALIGN; rx_ring_size = sizeof(struct starfire_rx_desc) * RX_RING_SIZE; np->queue_mem_size = tx_done_q_size + rx_done_q_size + tx_ring_size + rx_ring_size; - np->queue_mem = pci_alloc_consistent(np->pci_dev, np->queue_mem_size, &np->queue_mem_dma); + np->queue_mem = dma_alloc_coherent(&np->pci_dev->dev, + np->queue_mem_size, + &np->queue_mem_dma, GFP_ATOMIC); if (np->queue_mem == NULL) { free_irq(irq, dev); return -ENOMEM; @@ -1136,9 +1138,11 @@ static void init_ring(struct net_device *dev) np->rx_info[i].skb = skb; if (skb == NULL) break; - np->rx_info[i].mapping = pci_map_single(np->pci_dev, skb->data, np->rx_buf_sz, PCI_DMA_FROMDEVICE); - if (pci_dma_mapping_error(np->pci_dev, - np->rx_info[i].mapping)) { + np->rx_info[i].mapping = dma_map_single(&np->pci_dev->dev, + skb->data, + np->rx_buf_sz, + DMA_FROM_DEVICE); + if (dma_mapping_error(&np->pci_dev->dev, np->rx_info[i].mapping)) { dev_kfree_skb(skb); np->rx_info[i].skb = NULL; break; @@ -1217,18 +1221,19 @@ static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev) status |= skb_first_frag_len(skb) | (skb_num_frags(skb) << 16); np->tx_info[entry].mapping = - pci_map_single(np->pci_dev, skb->data, skb_first_frag_len(skb), PCI_DMA_TODEVICE); + dma_map_single(&np->pci_dev->dev, skb->data, + skb_first_frag_len(skb), + DMA_TO_DEVICE); } else { const skb_frag_t *this_frag = &skb_shinfo(skb)->frags[i - 1]; status |= skb_frag_size(this_frag); np->tx_info[entry].mapping = - pci_map_single(np->pci_dev, + dma_map_single(&np->pci_dev->dev, skb_frag_address(this_frag), skb_frag_size(this_frag), - PCI_DMA_TODEVICE); + DMA_TO_DEVICE); } - if (pci_dma_mapping_error(np->pci_dev, - np->tx_info[entry].mapping)) { + if (dma_mapping_error(&np->pci_dev->dev, np->tx_info[entry].mapping)) { dev->stats.tx_dropped++; goto err_out; } @@ -1271,18 +1276,16 @@ err_out: entry = prev_tx % TX_RING_SIZE; np->tx_info[entry].skb = NULL; if (i > 0) { - pci_unmap_single(np->pci_dev, + dma_unmap_single(&np->pci_dev->dev, np->tx_info[entry].mapping, - skb_first_frag_len(skb), - PCI_DMA_TODEVICE); + skb_first_frag_len(skb), DMA_TO_DEVICE); np->tx_info[entry].mapping = 0; entry = (entry + np->tx_info[entry].used_slots) % TX_RING_SIZE; for (j = 1; j < i; j++) { - pci_unmap_single(np->pci_dev, + dma_unmap_single(&np->pci_dev->dev, np->tx_info[entry].mapping, - skb_frag_size( - &skb_shinfo(skb)->frags[j-1]), - PCI_DMA_TODEVICE); + skb_frag_size(&skb_shinfo(skb)->frags[j - 1]), + DMA_TO_DEVICE); entry++; } } @@ -1356,20 +1359,20 @@ static irqreturn_t intr_handler(int irq, void *dev_instance) u16 entry = (tx_status & 0x7fff) / sizeof(starfire_tx_desc); struct sk_buff *skb = np->tx_info[entry].skb; np->tx_info[entry].skb = NULL; - pci_unmap_single(np->pci_dev, + dma_unmap_single(&np->pci_dev->dev, np->tx_info[entry].mapping, skb_first_frag_len(skb), - PCI_DMA_TODEVICE); + DMA_TO_DEVICE); np->tx_info[entry].mapping = 0; np->dirty_tx += np->tx_info[entry].used_slots; entry = (entry + np->tx_info[entry].used_slots) % TX_RING_SIZE; { int i; for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { - pci_unmap_single(np->pci_dev, + dma_unmap_single(&np->pci_dev->dev, np->tx_info[entry].mapping, skb_frag_size(&skb_shinfo(skb)->frags[i]), - PCI_DMA_TODEVICE); + DMA_TO_DEVICE); np->dirty_tx++; entry++; } @@ -1461,16 +1464,18 @@ static int __netdev_rx(struct net_device *dev, int *quota) if (pkt_len < rx_copybreak && (skb = netdev_alloc_skb(dev, pkt_len + 2)) != NULL) { skb_reserve(skb, 2); /* 16 byte align the IP header */ - pci_dma_sync_single_for_cpu(np->pci_dev, - np->rx_info[entry].mapping, - pkt_len, PCI_DMA_FROMDEVICE); + dma_sync_single_for_cpu(&np->pci_dev->dev, + np->rx_info[entry].mapping, + pkt_len, DMA_FROM_DEVICE); skb_copy_to_linear_data(skb, np->rx_info[entry].skb->data, pkt_len); - pci_dma_sync_single_for_device(np->pci_dev, - np->rx_info[entry].mapping, - pkt_len, PCI_DMA_FROMDEVICE); + dma_sync_single_for_device(&np->pci_dev->dev, + np->rx_info[entry].mapping, + pkt_len, DMA_FROM_DEVICE); skb_put(skb, pkt_len); } else { - pci_unmap_single(np->pci_dev, np->rx_info[entry].mapping, np->rx_buf_sz, PCI_DMA_FROMDEVICE); + dma_unmap_single(&np->pci_dev->dev, + np->rx_info[entry].mapping, + np->rx_buf_sz, DMA_FROM_DEVICE); skb = np->rx_info[entry].skb; skb_put(skb, pkt_len); np->rx_info[entry].skb = NULL; @@ -1588,9 +1593,9 @@ static void refill_rx_ring(struct net_device *dev) if (skb == NULL) break; /* Better luck next round. */ np->rx_info[entry].mapping = - pci_map_single(np->pci_dev, skb->data, np->rx_buf_sz, PCI_DMA_FROMDEVICE); - if (pci_dma_mapping_error(np->pci_dev, - np->rx_info[entry].mapping)) { + dma_map_single(&np->pci_dev->dev, skb->data, + np->rx_buf_sz, DMA_FROM_DEVICE); + if (dma_mapping_error(&np->pci_dev->dev, np->rx_info[entry].mapping)) { dev_kfree_skb(skb); np->rx_info[entry].skb = NULL; break; @@ -1963,7 +1968,9 @@ static int netdev_close(struct net_device *dev) for (i = 0; i < RX_RING_SIZE; i++) { np->rx_ring[i].rxaddr = cpu_to_dma(0xBADF00D0); /* An invalid address. */ if (np->rx_info[i].skb != NULL) { - pci_unmap_single(np->pci_dev, np->rx_info[i].mapping, np->rx_buf_sz, PCI_DMA_FROMDEVICE); + dma_unmap_single(&np->pci_dev->dev, + np->rx_info[i].mapping, + np->rx_buf_sz, DMA_FROM_DEVICE); dev_kfree_skb(np->rx_info[i].skb); } np->rx_info[i].skb = NULL; @@ -1973,9 +1980,8 @@ static int netdev_close(struct net_device *dev) struct sk_buff *skb = np->tx_info[i].skb; if (skb == NULL) continue; - pci_unmap_single(np->pci_dev, - np->tx_info[i].mapping, - skb_first_frag_len(skb), PCI_DMA_TODEVICE); + dma_unmap_single(&np->pci_dev->dev, np->tx_info[i].mapping, + skb_first_frag_len(skb), DMA_TO_DEVICE); np->tx_info[i].mapping = 0; dev_kfree_skb(skb); np->tx_info[i].skb = NULL; @@ -2018,7 +2024,8 @@ static void starfire_remove_one(struct pci_dev *pdev) unregister_netdev(dev); if (np->queue_mem) - pci_free_consistent(pdev, np->queue_mem_size, np->queue_mem, np->queue_mem_dma); + dma_free_coherent(&pdev->dev, np->queue_mem_size, + np->queue_mem, np->queue_mem_dma); /* XXX: add wakeup code -- requires firmware for MagicPacket */ diff --git a/drivers/net/ethernet/allwinner/sun4i-emac.c b/drivers/net/ethernet/allwinner/sun4i-emac.c index b3b8a8010142..862ea44beea7 100644 --- a/drivers/net/ethernet/allwinner/sun4i-emac.c +++ b/drivers/net/ethernet/allwinner/sun4i-emac.c @@ -640,13 +640,11 @@ static irqreturn_t emac_interrupt(int irq, void *dev_id) struct net_device *dev = dev_id; struct emac_board_info *db = netdev_priv(dev); int int_status; - unsigned long flags; unsigned int reg_val; /* A real interrupt coming */ - /* holders of db->lock must always block IRQs */ - spin_lock_irqsave(&db->lock, flags); + spin_lock(&db->lock); /* Disable all interrupts */ writel(0, db->membase + EMAC_INT_CTL_REG); @@ -680,7 +678,7 @@ static irqreturn_t emac_interrupt(int irq, void *dev_id) reg_val |= (0xf << 0) | (0x01 << 8); writel(reg_val, db->membase + EMAC_INT_CTL_REG); } - spin_unlock_irqrestore(&db->lock, flags); + spin_unlock(&db->lock); return IRQ_HANDLED; } diff --git a/drivers/net/ethernet/alteon/acenic.c b/drivers/net/ethernet/alteon/acenic.c index 8470c836fa18..1a7e4df9b3e9 100644 --- a/drivers/net/ethernet/alteon/acenic.c +++ b/drivers/net/ethernet/alteon/acenic.c @@ -465,6 +465,7 @@ static int acenic_probe_one(struct pci_dev *pdev, SET_NETDEV_DEV(dev, &pdev->dev); ap = netdev_priv(dev); + ap->ndev = dev; ap->pdev = pdev; ap->name = pci_name(pdev); @@ -1562,10 +1563,10 @@ static void ace_watchdog(struct net_device *data, unsigned int txqueue) } -static void ace_tasklet(unsigned long arg) +static void ace_tasklet(struct tasklet_struct *t) { - struct net_device *dev = (struct net_device *) arg; - struct ace_private *ap = netdev_priv(dev); + struct ace_private *ap = from_tasklet(ap, t, ace_tasklet); + struct net_device *dev = ap->ndev; int cur_size; cur_size = atomic_read(&ap->cur_rx_bufs); @@ -2269,7 +2270,7 @@ static int ace_open(struct net_device *dev) /* * Setup the bottom half rx ring refill handler */ - tasklet_init(&ap->ace_tasklet, ace_tasklet, (unsigned long)dev); + tasklet_setup(&ap->ace_tasklet, ace_tasklet); return 0; } diff --git a/drivers/net/ethernet/alteon/acenic.h b/drivers/net/ethernet/alteon/acenic.h index c670067b1541..265fa601a258 100644 --- a/drivers/net/ethernet/alteon/acenic.h +++ b/drivers/net/ethernet/alteon/acenic.h @@ -633,6 +633,7 @@ struct ace_skb */ struct ace_private { + struct net_device *ndev; /* backpointer */ struct ace_info *info; struct ace_regs __iomem *regs; /* register base */ struct ace_skb *skb; @@ -776,7 +777,7 @@ static int ace_open(struct net_device *dev); static netdev_tx_t ace_start_xmit(struct sk_buff *skb, struct net_device *dev); static int ace_close(struct net_device *dev); -static void ace_tasklet(unsigned long dev); +static void ace_tasklet(struct tasklet_struct *t); static void ace_dump_trace(struct ace_private *ap); static void ace_set_multicast_list(struct net_device *dev); static int ace_change_mtu(struct net_device *dev, int new_mtu); diff --git a/drivers/net/ethernet/amazon/ena/ena_admin_defs.h b/drivers/net/ethernet/amazon/ena/ena_admin_defs.h index b818a169c193..4164eacc5c28 100644 --- a/drivers/net/ethernet/amazon/ena/ena_admin_defs.h +++ b/drivers/net/ethernet/amazon/ena/ena_admin_defs.h @@ -1,37 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ /* - * Copyright 2015 - 2016 Amazon.com, Inc. or its affiliates. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. + * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved. */ #ifndef _ENA_ADMIN_H_ #define _ENA_ADMIN_H_ +#define ENA_ADMIN_RSS_KEY_PARTS 10 enum ena_admin_aq_opcode { ENA_ADMIN_CREATE_SQ = 1, @@ -55,6 +29,7 @@ enum ena_admin_aq_completion_status { ENA_ADMIN_RESOURCE_BUSY = 7, }; +/* subcommands for the set/get feature admin commands */ enum ena_admin_aq_feature_id { ENA_ADMIN_DEVICE_ATTRIBUTES = 1, ENA_ADMIN_MAX_QUEUES_NUM = 2, @@ -63,7 +38,7 @@ enum ena_admin_aq_feature_id { ENA_ADMIN_MAX_QUEUES_EXT = 7, ENA_ADMIN_RSS_HASH_FUNCTION = 10, ENA_ADMIN_STATELESS_OFFLOAD_CONFIG = 11, - ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG = 12, + ENA_ADMIN_RSS_INDIRECTION_TABLE_CONFIG = 12, ENA_ADMIN_MTU = 14, ENA_ADMIN_RSS_HASH_INPUT = 18, ENA_ADMIN_INTERRUPT_MODERATION = 20, @@ -117,6 +92,8 @@ enum ena_admin_completion_policy_type { enum ena_admin_get_stats_type { ENA_ADMIN_GET_STATS_TYPE_BASIC = 0, ENA_ADMIN_GET_STATS_TYPE_EXTENDED = 1, + /* extra HW stats for specific network interface */ + ENA_ADMIN_GET_STATS_TYPE_ENI = 2, }; enum ena_admin_get_stats_scope { @@ -193,7 +170,7 @@ struct ena_admin_acq_common_desc { u16 extended_status; /* indicates to the driver which AQ entry has been consumed by the - * device and could be reused + * device and could be reused */ u16 sq_head_indx; }; @@ -238,8 +215,8 @@ struct ena_admin_aq_create_sq_cmd { */ u8 sq_caps_3; - /* associated completion queue id. This CQ must be created prior to - * SQ creation + /* associated completion queue id. This CQ must be created prior to SQ + * creation */ u16 cq_idx; @@ -378,7 +355,7 @@ struct ena_admin_aq_get_stats_cmd { u16 queue_idx; /* device id, value 0xFFFF means mine. only privileged device can get - * stats of other device + * stats of other device */ u16 device_id; }; @@ -410,10 +387,43 @@ struct ena_admin_basic_stats { u32 tx_drops_high; }; +/* ENI Statistics Command. */ +struct ena_admin_eni_stats { + /* The number of packets shaped due to inbound aggregate BW + * allowance being exceeded + */ + u64 bw_in_allowance_exceeded; + + /* The number of packets shaped due to outbound aggregate BW + * allowance being exceeded + */ + u64 bw_out_allowance_exceeded; + + /* The number of packets shaped due to PPS allowance being exceeded */ + u64 pps_allowance_exceeded; + + /* The number of packets shaped due to connection tracking + * allowance being exceeded and leading to failure in establishment + * of new connections + */ + u64 conntrack_allowance_exceeded; + + /* The number of packets shaped due to linklocal packet rate + * allowance being exceeded + */ + u64 linklocal_allowance_exceeded; +}; + struct ena_admin_acq_get_stats_resp { struct ena_admin_acq_common_desc acq_common_desc; - struct ena_admin_basic_stats basic_stats; + union { + u64 raw[7]; + + struct ena_admin_basic_stats basic_stats; + + struct ena_admin_eni_stats eni_stats; + } u; }; struct ena_admin_get_set_feature_common_desc { @@ -440,7 +450,9 @@ struct ena_admin_device_attr_feature_desc { u32 device_version; - /* bitmap of ena_admin_aq_feature_id */ + /* bitmap of ena_admin_aq_feature_id, which represents supported + * subcommands for the set/get feature admin commands. + */ u32 supported_features; u32 reserved3; @@ -526,32 +538,30 @@ struct ena_admin_feature_llq_desc { u32 max_llq_depth; - /* specify the header locations the device supports. bitfield of - * enum ena_admin_llq_header_location. + /* specify the header locations the device supports. bitfield of enum + * ena_admin_llq_header_location. */ u16 header_location_ctrl_supported; /* the header location the driver selected to use. */ u16 header_location_ctrl_enabled; - /* if inline header is specified - this is the size of descriptor - * list entry. If header in a separate ring is specified - this is - * the size of header ring entry. bitfield of enum - * ena_admin_llq_ring_entry_size. specify the entry sizes the device - * supports + /* if inline header is specified - this is the size of descriptor list + * entry. If header in a separate ring is specified - this is the size + * of header ring entry. bitfield of enum ena_admin_llq_ring_entry_size. + * specify the entry sizes the device supports */ u16 entry_size_ctrl_supported; /* the entry size the driver selected to use. */ u16 entry_size_ctrl_enabled; - /* valid only if inline header is specified. First entry associated - * with the packet includes descriptors and header. Rest of the - * entries occupied by descriptors. This parameter defines the max - * number of descriptors precedding the header in the first entry. - * The field is bitfield of enum - * ena_admin_llq_num_descs_before_header and specify the values the - * device supports + /* valid only if inline header is specified. First entry associated with + * the packet includes descriptors and header. Rest of the entries + * occupied by descriptors. This parameter defines the max number of + * descriptors precedding the header in the first entry. The field is + * bitfield of enum ena_admin_llq_num_descs_before_header and specify + * the values the device supports */ u16 desc_num_before_header_supported; @@ -559,7 +569,7 @@ struct ena_admin_feature_llq_desc { u16 desc_num_before_header_enabled; /* valid only if inline was chosen. bitfield of enum - * ena_admin_llq_stride_ctrl + * ena_admin_llq_stride_ctrl */ u16 descriptors_stride_ctrl_supported; @@ -594,8 +604,8 @@ struct ena_admin_queue_ext_feature_fields { u32 max_tx_header_size; - /* Maximum Descriptors number, including meta descriptor, allowed for - * a single Tx packet + /* Maximum Descriptors number, including meta descriptor, allowed for a + * single Tx packet */ u16 max_per_packet_tx_descs; @@ -618,8 +628,8 @@ struct ena_admin_queue_feature_desc { u32 max_header_size; - /* Maximum Descriptors number, including meta descriptor, allowed for - * a single Tx packet + /* Maximum Descriptors number, including meta descriptor, allowed for a + * single Tx packet */ u16 max_packet_tx_descs; @@ -707,11 +717,11 @@ enum ena_admin_hash_functions { }; struct ena_admin_feature_rss_flow_hash_control { - u32 keys_num; + u32 key_parts; u32 reserved; - u32 key[10]; + u32 key[ENA_ADMIN_RSS_KEY_PARTS]; }; struct ena_admin_feature_rss_flow_hash_function { @@ -1007,7 +1017,7 @@ struct ena_admin_set_feat_resp { struct ena_admin_aenq_common_desc { u16 group; - u16 syndrom; + u16 syndrome; /* 0 : phase * 7:1 : reserved - MBZ @@ -1031,7 +1041,7 @@ enum ena_admin_aenq_group { ENA_ADMIN_AENQ_GROUPS_NUM = 5, }; -enum ena_admin_aenq_notification_syndrom { +enum ena_admin_aenq_notification_syndrome { ENA_ADMIN_SUSPEND = 0, ENA_ADMIN_RESUME = 1, ENA_ADMIN_UPDATE_HINTS = 2, diff --git a/drivers/net/ethernet/amazon/ena/ena_com.c b/drivers/net/ethernet/amazon/ena/ena_com.c index 435bf05a853c..5f8769aa469d 100644 --- a/drivers/net/ethernet/amazon/ena/ena_com.c +++ b/drivers/net/ethernet/amazon/ena/ena_com.c @@ -1,33 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB /* - * Copyright 2015 Amazon.com, Inc. or its affiliates. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. + * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved. */ #include "ena_com.h" @@ -98,7 +71,7 @@ static int ena_com_mem_addr_set(struct ena_com_dev *ena_dev, dma_addr_t addr) { if ((addr & GENMASK_ULL(ena_dev->dma_addr_bits - 1, 0)) != addr) { - pr_err("dma address has more bits that the device supports\n"); + pr_err("DMA address has more bits that the device supports\n"); return -EINVAL; } @@ -108,16 +81,16 @@ static int ena_com_mem_addr_set(struct ena_com_dev *ena_dev, return 0; } -static int ena_com_admin_init_sq(struct ena_com_admin_queue *queue) +static int ena_com_admin_init_sq(struct ena_com_admin_queue *admin_queue) { - struct ena_com_admin_sq *sq = &queue->sq; - u16 size = ADMIN_SQ_SIZE(queue->q_depth); + struct ena_com_admin_sq *sq = &admin_queue->sq; + u16 size = ADMIN_SQ_SIZE(admin_queue->q_depth); - sq->entries = dma_alloc_coherent(queue->q_dmadev, size, &sq->dma_addr, - GFP_KERNEL); + sq->entries = dma_alloc_coherent(admin_queue->q_dmadev, size, + &sq->dma_addr, GFP_KERNEL); if (!sq->entries) { - pr_err("memory allocation failed\n"); + pr_err("Memory allocation failed\n"); return -ENOMEM; } @@ -130,16 +103,16 @@ static int ena_com_admin_init_sq(struct ena_com_admin_queue *queue) return 0; } -static int ena_com_admin_init_cq(struct ena_com_admin_queue *queue) +static int ena_com_admin_init_cq(struct ena_com_admin_queue *admin_queue) { - struct ena_com_admin_cq *cq = &queue->cq; - u16 size = ADMIN_CQ_SIZE(queue->q_depth); + struct ena_com_admin_cq *cq = &admin_queue->cq; + u16 size = ADMIN_CQ_SIZE(admin_queue->q_depth); - cq->entries = dma_alloc_coherent(queue->q_dmadev, size, &cq->dma_addr, - GFP_KERNEL); + cq->entries = dma_alloc_coherent(admin_queue->q_dmadev, size, + &cq->dma_addr, GFP_KERNEL); if (!cq->entries) { - pr_err("memory allocation failed\n"); + pr_err("Memory allocation failed\n"); return -ENOMEM; } @@ -149,20 +122,20 @@ static int ena_com_admin_init_cq(struct ena_com_admin_queue *queue) return 0; } -static int ena_com_admin_init_aenq(struct ena_com_dev *dev, +static int ena_com_admin_init_aenq(struct ena_com_dev *ena_dev, struct ena_aenq_handlers *aenq_handlers) { - struct ena_com_aenq *aenq = &dev->aenq; + struct ena_com_aenq *aenq = &ena_dev->aenq; u32 addr_low, addr_high, aenq_caps; u16 size; - dev->aenq.q_depth = ENA_ASYNC_QUEUE_DEPTH; + ena_dev->aenq.q_depth = ENA_ASYNC_QUEUE_DEPTH; size = ADMIN_AENQ_SIZE(ENA_ASYNC_QUEUE_DEPTH); - aenq->entries = dma_alloc_coherent(dev->dmadev, size, &aenq->dma_addr, - GFP_KERNEL); + aenq->entries = dma_alloc_coherent(ena_dev->dmadev, size, + &aenq->dma_addr, GFP_KERNEL); if (!aenq->entries) { - pr_err("memory allocation failed\n"); + pr_err("Memory allocation failed\n"); return -ENOMEM; } @@ -172,18 +145,18 @@ static int ena_com_admin_init_aenq(struct ena_com_dev *dev, addr_low = ENA_DMA_ADDR_TO_UINT32_LOW(aenq->dma_addr); addr_high = ENA_DMA_ADDR_TO_UINT32_HIGH(aenq->dma_addr); - writel(addr_low, dev->reg_bar + ENA_REGS_AENQ_BASE_LO_OFF); - writel(addr_high, dev->reg_bar + ENA_REGS_AENQ_BASE_HI_OFF); + writel(addr_low, ena_dev->reg_bar + ENA_REGS_AENQ_BASE_LO_OFF); + writel(addr_high, ena_dev->reg_bar + ENA_REGS_AENQ_BASE_HI_OFF); aenq_caps = 0; - aenq_caps |= dev->aenq.q_depth & ENA_REGS_AENQ_CAPS_AENQ_DEPTH_MASK; + aenq_caps |= ena_dev->aenq.q_depth & ENA_REGS_AENQ_CAPS_AENQ_DEPTH_MASK; aenq_caps |= (sizeof(struct ena_admin_aenq_entry) << ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_SHIFT) & ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_MASK; - writel(aenq_caps, dev->reg_bar + ENA_REGS_AENQ_CAPS_OFF); + writel(aenq_caps, ena_dev->reg_bar + ENA_REGS_AENQ_CAPS_OFF); if (unlikely(!aenq_handlers)) { - pr_err("aenq handlers pointer is NULL\n"); + pr_err("AENQ handlers pointer is NULL\n"); return -EINVAL; } @@ -199,31 +172,31 @@ static void comp_ctxt_release(struct ena_com_admin_queue *queue, atomic_dec(&queue->outstanding_cmds); } -static struct ena_comp_ctx *get_comp_ctxt(struct ena_com_admin_queue *queue, +static struct ena_comp_ctx *get_comp_ctxt(struct ena_com_admin_queue *admin_queue, u16 command_id, bool capture) { - if (unlikely(command_id >= queue->q_depth)) { - pr_err("command id is larger than the queue size. cmd_id: %u queue size %d\n", - command_id, queue->q_depth); + if (unlikely(command_id >= admin_queue->q_depth)) { + pr_err("Command id is larger than the queue size. cmd_id: %u queue size %d\n", + command_id, admin_queue->q_depth); return NULL; } - if (unlikely(!queue->comp_ctx)) { + if (unlikely(!admin_queue->comp_ctx)) { pr_err("Completion context is NULL\n"); return NULL; } - if (unlikely(queue->comp_ctx[command_id].occupied && capture)) { + if (unlikely(admin_queue->comp_ctx[command_id].occupied && capture)) { pr_err("Completion context is occupied\n"); return NULL; } if (capture) { - atomic_inc(&queue->outstanding_cmds); - queue->comp_ctx[command_id].occupied = true; + atomic_inc(&admin_queue->outstanding_cmds); + admin_queue->comp_ctx[command_id].occupied = true; } - return &queue->comp_ctx[command_id]; + return &admin_queue->comp_ctx[command_id]; } static struct ena_comp_ctx *__ena_com_submit_admin_cmd(struct ena_com_admin_queue *admin_queue, @@ -244,7 +217,7 @@ static struct ena_comp_ctx *__ena_com_submit_admin_cmd(struct ena_com_admin_queu /* In case of queue FULL */ cnt = (u16)atomic_read(&admin_queue->outstanding_cmds); if (cnt >= admin_queue->q_depth) { - pr_debug("admin queue is full.\n"); + pr_debug("Admin queue is full.\n"); admin_queue->stats.out_of_space++; return ERR_PTR(-ENOSPC); } @@ -284,20 +257,21 @@ static struct ena_comp_ctx *__ena_com_submit_admin_cmd(struct ena_com_admin_queu return comp_ctx; } -static int ena_com_init_comp_ctxt(struct ena_com_admin_queue *queue) +static int ena_com_init_comp_ctxt(struct ena_com_admin_queue *admin_queue) { - size_t size = queue->q_depth * sizeof(struct ena_comp_ctx); + size_t size = admin_queue->q_depth * sizeof(struct ena_comp_ctx); struct ena_comp_ctx *comp_ctx; u16 i; - queue->comp_ctx = devm_kzalloc(queue->q_dmadev, size, GFP_KERNEL); - if (unlikely(!queue->comp_ctx)) { - pr_err("memory allocation failed\n"); + admin_queue->comp_ctx = + devm_kzalloc(admin_queue->q_dmadev, size, GFP_KERNEL); + if (unlikely(!admin_queue->comp_ctx)) { + pr_err("Memory allocation failed\n"); return -ENOMEM; } - for (i = 0; i < queue->q_depth; i++) { - comp_ctx = get_comp_ctxt(queue, i, false); + for (i = 0; i < admin_queue->q_depth; i++) { + comp_ctx = get_comp_ctxt(admin_queue, i, false); if (comp_ctx) init_completion(&comp_ctx->wait_event); } @@ -363,7 +337,7 @@ static int ena_com_init_io_sq(struct ena_com_dev *ena_dev, } if (!io_sq->desc_addr.virt_addr) { - pr_err("memory allocation failed\n"); + pr_err("Memory allocation failed\n"); return -ENOMEM; } } @@ -389,7 +363,7 @@ static int ena_com_init_io_sq(struct ena_com_dev *ena_dev, devm_kzalloc(ena_dev->dmadev, size, GFP_KERNEL); if (!io_sq->bounce_buf_ctrl.base_buffer) { - pr_err("bounce buffer memory allocation failed\n"); + pr_err("Bounce buffer memory allocation failed\n"); return -ENOMEM; } @@ -449,7 +423,7 @@ static int ena_com_init_io_cq(struct ena_com_dev *ena_dev, } if (!io_cq->cdesc_addr.virt_addr) { - pr_err("memory allocation failed\n"); + pr_err("Memory allocation failed\n"); return -ENOMEM; } @@ -525,7 +499,7 @@ static void ena_com_handle_admin_completion(struct ena_com_admin_queue *admin_qu static int ena_com_comp_status_to_errno(u8 comp_status) { if (unlikely(comp_status != 0)) - pr_err("admin command failed[%u]\n", comp_status); + pr_err("Admin command failed[%u]\n", comp_status); switch (comp_status) { case ENA_ADMIN_SUCCESS: @@ -539,6 +513,8 @@ static int ena_com_comp_status_to_errno(u8 comp_status) case ENA_ADMIN_ILLEGAL_PARAMETER: case ENA_ADMIN_UNKNOWN_ERROR: return -EINVAL; + case ENA_ADMIN_RESOURCE_BUSY: + return -EAGAIN; } return -EINVAL; @@ -603,7 +579,7 @@ err: return ret; } -/** +/* * Set the LLQ configurations of the firmware * * The driver provides only the enabled feature values to the device, @@ -717,7 +693,7 @@ static int ena_com_config_llq_info(struct ena_com_dev *ena_dev, /* The desc list entry size should be whole multiply of 8 * This requirement comes from __iowrite64_copy() */ - pr_err("illegal entry size %d\n", llq_info->desc_list_entry_size); + pr_err("Illegal entry size %d\n", llq_info->desc_list_entry_size); return -EINVAL; } @@ -858,7 +834,7 @@ static u32 ena_com_reg_bar_read32(struct ena_com_dev *ena_dev, u16 offset) } if (unlikely(i == timeout)) { - pr_err("reading reg failed for timeout. expected: req id[%hu] offset[%hu] actual: req id[%hu] offset[%hu]\n", + pr_err("Reading reg failed for timeout. expected: req id[%hu] offset[%hu] actual: req id[%hu] offset[%hu]\n", mmio_read->seq_num, offset, read_resp->req_id, read_resp->reg_off); ret = ENA_MMIO_READ_TIMEOUT; @@ -925,7 +901,7 @@ static int ena_com_destroy_io_sq(struct ena_com_dev *ena_dev, sizeof(destroy_resp)); if (unlikely(ret && (ret != -ENODEV))) - pr_err("failed to destroy io sq error: %d\n", ret); + pr_err("Failed to destroy io sq error: %d\n", ret); return ret; } @@ -1034,7 +1010,7 @@ static int ena_com_get_feature_ex(struct ena_com_dev *ena_dev, &get_cmd.control_buffer.address, control_buf_dma_addr); if (unlikely(ret)) { - pr_err("memory address set failed\n"); + pr_err("Memory address set failed\n"); return ret; } @@ -1081,11 +1057,10 @@ static void ena_com_hash_key_fill_default_key(struct ena_com_dev *ena_dev) (ena_dev->rss).hash_key; netdev_rss_key_fill(&hash_key->key, sizeof(hash_key->key)); - /* The key is stored in the device in u32 array - * as well as the API requires the key to be passed in this - * format. Thus the size of our array should be divided by 4 + /* The key buffer is stored in the device in an array of + * uint32 elements. */ - hash_key->keys_num = sizeof(hash_key->key) / sizeof(u32); + hash_key->key_parts = ENA_ADMIN_RSS_KEY_PARTS; } static int ena_com_hash_key_allocate(struct ena_com_dev *ena_dev) @@ -1149,13 +1124,13 @@ static int ena_com_indirect_table_allocate(struct ena_com_dev *ena_dev, int ret; ret = ena_com_get_feature(ena_dev, &get_resp, - ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG, 0); + ENA_ADMIN_RSS_INDIRECTION_TABLE_CONFIG, 0); if (unlikely(ret)) return ret; if ((get_resp.u.ind_table.min_size > log_size) || (get_resp.u.ind_table.max_size < log_size)) { - pr_err("indirect table size doesn't fit. requested size: %d while min is:%d and max %d\n", + pr_err("Indirect table size doesn't fit. requested size: %d while min is:%d and max %d\n", 1 << log_size, 1 << get_resp.u.ind_table.min_size, 1 << get_resp.u.ind_table.max_size); return -EINVAL; @@ -1248,7 +1223,7 @@ static int ena_com_create_io_sq(struct ena_com_dev *ena_dev, &create_cmd.sq_ba, io_sq->desc_addr.phys_addr); if (unlikely(ret)) { - pr_err("memory address set failed\n"); + pr_err("Memory address set failed\n"); return ret; } } @@ -1277,7 +1252,7 @@ static int ena_com_create_io_sq(struct ena_com_dev *ena_dev, cmd_completion.llq_descriptors_offset); } - pr_debug("created sq[%u], depth[%u]\n", io_sq->idx, io_sq->q_depth); + pr_debug("Created sq[%u], depth[%u]\n", io_sq->idx, io_sq->q_depth); return ret; } @@ -1390,7 +1365,7 @@ int ena_com_create_io_cq(struct ena_com_dev *ena_dev, &create_cmd.cq_ba, io_cq->cdesc_addr.phys_addr); if (unlikely(ret)) { - pr_err("memory address set failed\n"); + pr_err("Memory address set failed\n"); return ret; } @@ -1419,7 +1394,7 @@ int ena_com_create_io_cq(struct ena_com_dev *ena_dev, (u32 __iomem *)((uintptr_t)ena_dev->reg_bar + cmd_completion.numa_node_register_offset); - pr_debug("created cq[%u], depth[%u]\n", io_cq->idx, io_cq->q_depth); + pr_debug("Created cq[%u], depth[%u]\n", io_cq->idx, io_cq->q_depth); return ret; } @@ -1612,12 +1587,12 @@ int ena_com_validate_version(struct ena_com_dev *ena_dev) return -ETIME; } - pr_info("ena device version: %d.%d\n", + pr_info("ENA device version: %d.%d\n", (ver & ENA_REGS_VERSION_MAJOR_VERSION_MASK) >> ENA_REGS_VERSION_MAJOR_VERSION_SHIFT, ver & ENA_REGS_VERSION_MINOR_VERSION_MASK); - pr_info("ena controller version: %d.%d.%d implementation version %d\n", + pr_info("ENA controller version: %d.%d.%d implementation version %d\n", (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_MASK) >> ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_SHIFT, (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_MASK) >> @@ -1640,6 +1615,19 @@ int ena_com_validate_version(struct ena_com_dev *ena_dev) return 0; } +static void +ena_com_free_ena_admin_queue_comp_ctx(struct ena_com_dev *ena_dev, + struct ena_com_admin_queue *admin_queue) + +{ + if (!admin_queue->comp_ctx) + return; + + devm_kfree(ena_dev->dmadev, admin_queue->comp_ctx); + + admin_queue->comp_ctx = NULL; +} + void ena_com_admin_destroy(struct ena_com_dev *ena_dev) { struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue; @@ -1648,9 +1636,8 @@ void ena_com_admin_destroy(struct ena_com_dev *ena_dev) struct ena_com_aenq *aenq = &ena_dev->aenq; u16 size; - if (admin_queue->comp_ctx) - devm_kfree(ena_dev->dmadev, admin_queue->comp_ctx); - admin_queue->comp_ctx = NULL; + ena_com_free_ena_admin_queue_comp_ctx(ena_dev, admin_queue); + size = ADMIN_SQ_SIZE(admin_queue->q_depth); if (sq->entries) dma_free_coherent(ena_dev->dmadev, size, sq->entries, @@ -1928,6 +1915,7 @@ int ena_com_get_dev_attr_feat(struct ena_com_dev *ena_dev, memcpy(&get_feat_ctx->dev_attr, &get_resp.u.dev_attr, sizeof(get_resp.u.dev_attr)); + ena_dev->supported_features = get_resp.u.dev_attr.supported_features; if (ena_dev->supported_features & BIT(ENA_ADMIN_MAX_QUEUES_EXT)) { @@ -2006,10 +1994,10 @@ void ena_com_admin_q_comp_intr_handler(struct ena_com_dev *ena_dev) /* ena_handle_specific_aenq_event: * return the handler that is relevant to the specific event group */ -static ena_aenq_handler ena_com_get_specific_aenq_cb(struct ena_com_dev *dev, +static ena_aenq_handler ena_com_get_specific_aenq_cb(struct ena_com_dev *ena_dev, u16 group) { - struct ena_aenq_handlers *aenq_handlers = dev->aenq.aenq_handlers; + struct ena_aenq_handlers *aenq_handlers = ena_dev->aenq.aenq_handlers; if ((group < ENA_MAX_HANDLERS) && aenq_handlers->handlers[group]) return aenq_handlers->handlers[group]; @@ -2021,11 +2009,11 @@ static ena_aenq_handler ena_com_get_specific_aenq_cb(struct ena_com_dev *dev, * handles the aenq incoming events. * pop events from the queue and apply the specific handler */ -void ena_com_aenq_intr_handler(struct ena_com_dev *dev, void *data) +void ena_com_aenq_intr_handler(struct ena_com_dev *ena_dev, void *data) { struct ena_admin_aenq_entry *aenq_e; struct ena_admin_aenq_common_desc *aenq_common; - struct ena_com_aenq *aenq = &dev->aenq; + struct ena_com_aenq *aenq = &ena_dev->aenq; u64 timestamp; ena_aenq_handler handler_cb; u16 masked_head, processed = 0; @@ -2045,12 +2033,13 @@ void ena_com_aenq_intr_handler(struct ena_com_dev *dev, void *data) dma_rmb(); timestamp = (u64)aenq_common->timestamp_low | - ((u64)aenq_common->timestamp_high << 32); - pr_debug("AENQ! Group[%x] Syndrom[%x] timestamp: [%llus]\n", - aenq_common->group, aenq_common->syndrom, timestamp); + ((u64)aenq_common->timestamp_high << 32); + + pr_debug("AENQ! Group[%x] Syndrome[%x] timestamp: [%llus]\n", + aenq_common->group, aenq_common->syndrome, timestamp); /* Handle specific event*/ - handler_cb = ena_com_get_specific_aenq_cb(dev, + handler_cb = ena_com_get_specific_aenq_cb(ena_dev, aenq_common->group); handler_cb(data, aenq_e); /* call the actual event handler*/ @@ -2075,7 +2064,8 @@ void ena_com_aenq_intr_handler(struct ena_com_dev *dev, void *data) /* write the aenq doorbell after all AENQ descriptors were read */ mb(); - writel_relaxed((u32)aenq->head, dev->reg_bar + ENA_REGS_AENQ_HEAD_DB_OFF); + writel_relaxed((u32)aenq->head, + ena_dev->reg_bar + ENA_REGS_AENQ_HEAD_DB_OFF); } int ena_com_dev_reset(struct ena_com_dev *ena_dev, @@ -2167,6 +2157,21 @@ static int ena_get_dev_stats(struct ena_com_dev *ena_dev, return ret; } +int ena_com_get_eni_stats(struct ena_com_dev *ena_dev, + struct ena_admin_eni_stats *stats) +{ + struct ena_com_stats_ctx ctx; + int ret; + + memset(&ctx, 0x0, sizeof(ctx)); + ret = ena_get_dev_stats(ena_dev, &ctx, ENA_ADMIN_GET_STATS_TYPE_ENI); + if (likely(ret == 0)) + memcpy(stats, &ctx.get_resp.u.eni_stats, + sizeof(ctx.get_resp.u.eni_stats)); + + return ret; +} + int ena_com_get_dev_basic_stats(struct ena_com_dev *ena_dev, struct ena_admin_basic_stats *stats) { @@ -2176,8 +2181,8 @@ int ena_com_get_dev_basic_stats(struct ena_com_dev *ena_dev, memset(&ctx, 0x0, sizeof(ctx)); ret = ena_get_dev_stats(ena_dev, &ctx, ENA_ADMIN_GET_STATS_TYPE_BASIC); if (likely(ret == 0)) - memcpy(stats, &ctx.get_resp.basic_stats, - sizeof(ctx.get_resp.basic_stats)); + memcpy(stats, &ctx.get_resp.u.basic_stats, + sizeof(ctx.get_resp.u.basic_stats)); return ret; } @@ -2273,7 +2278,7 @@ int ena_com_set_hash_function(struct ena_com_dev *ena_dev) &cmd.control_buffer.address, rss->hash_key_dma_addr); if (unlikely(ret)) { - pr_err("memory address set failed\n"); + pr_err("Memory address set failed\n"); return ret; } @@ -2331,7 +2336,7 @@ int ena_com_fill_hash_function(struct ena_com_dev *ena_dev, } memcpy(hash_key->key, key, key_len); rss->hash_init_val = init_val; - hash_key->keys_num = key_len >> 2; + hash_key->key_parts = key_len / sizeof(hash_key->key[0]); } break; case ENA_ADMIN_CRC32: @@ -2386,7 +2391,8 @@ int ena_com_get_hash_key(struct ena_com_dev *ena_dev, u8 *key) ena_dev->rss.hash_key; if (key) - memcpy(key, hash_key->key, (size_t)(hash_key->keys_num) << 2); + memcpy(key, hash_key->key, + (size_t)(hash_key->key_parts) * sizeof(hash_key->key[0])); return 0; } @@ -2442,7 +2448,7 @@ int ena_com_set_hash_ctrl(struct ena_com_dev *ena_dev) &cmd.control_buffer.address, rss->hash_ctrl_dma_addr); if (unlikely(ret)) { - pr_err("memory address set failed\n"); + pr_err("Memory address set failed\n"); return ret; } cmd.control_buffer.length = sizeof(*hash_ctrl); @@ -2503,7 +2509,7 @@ int ena_com_set_default_hash_ctrl(struct ena_com_dev *ena_dev) available_fields = hash_ctrl->selected_fields[i].fields & hash_ctrl->supported_fields[i].fields; if (available_fields != hash_ctrl->selected_fields[i].fields) { - pr_err("hash control doesn't support all the desire configuration. proto %x supported %x selected %x\n", + pr_err("Hash control doesn't support all the desire configuration. proto %x supported %x selected %x\n", i, hash_ctrl->supported_fields[i].fields, hash_ctrl->selected_fields[i].fields); return -EOPNOTSUPP; @@ -2541,7 +2547,7 @@ int ena_com_fill_hash_ctrl(struct ena_com_dev *ena_dev, /* Make sure all the fields are supported */ supported_fields = hash_ctrl->supported_fields[proto].fields; if ((hash_fields & supported_fields) != hash_fields) { - pr_err("proto %d doesn't support the required fields %x. supports only: %x\n", + pr_err("Proto %d doesn't support the required fields %x. supports only: %x\n", proto, hash_fields, supported_fields); } @@ -2581,9 +2587,9 @@ int ena_com_indirect_table_set(struct ena_com_dev *ena_dev) int ret; if (!ena_com_check_supported_feature_id( - ena_dev, ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG)) { + ena_dev, ENA_ADMIN_RSS_INDIRECTION_TABLE_CONFIG)) { pr_debug("Feature %d isn't supported\n", - ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG); + ENA_ADMIN_RSS_INDIRECTION_TABLE_CONFIG); return -EOPNOTSUPP; } @@ -2598,7 +2604,7 @@ int ena_com_indirect_table_set(struct ena_com_dev *ena_dev) cmd.aq_common_descriptor.opcode = ENA_ADMIN_SET_FEATURE; cmd.aq_common_descriptor.flags = ENA_ADMIN_AQ_COMMON_DESC_CTRL_DATA_INDIRECT_MASK; - cmd.feat_common.feature_id = ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG; + cmd.feat_common.feature_id = ENA_ADMIN_RSS_INDIRECTION_TABLE_CONFIG; cmd.u.ind_table.size = rss->tbl_log_size; cmd.u.ind_table.inline_index = 0xFFFFFFFF; @@ -2606,7 +2612,7 @@ int ena_com_indirect_table_set(struct ena_com_dev *ena_dev) &cmd.control_buffer.address, rss->rss_ind_tbl_dma_addr); if (unlikely(ret)) { - pr_err("memory address set failed\n"); + pr_err("Memory address set failed\n"); return ret; } @@ -2636,7 +2642,7 @@ int ena_com_indirect_table_get(struct ena_com_dev *ena_dev, u32 *ind_tbl) sizeof(struct ena_admin_rss_ind_table_entry); rc = ena_com_get_feature_ex(ena_dev, &get_resp, - ENA_ADMIN_RSS_REDIRECTION_TABLE_CONFIG, + ENA_ADMIN_RSS_INDIRECTION_TABLE_CONFIG, rss->rss_ind_tbl_dma_addr, tbl_size, 0); if (unlikely(rc)) @@ -2719,8 +2725,7 @@ int ena_com_allocate_debug_area(struct ena_com_dev *ena_dev, host_attr->debug_area_virt_addr = dma_alloc_coherent(ena_dev->dmadev, debug_area_size, - &host_attr->debug_area_dma_addr, - GFP_KERNEL); + &host_attr->debug_area_dma_addr, GFP_KERNEL); if (unlikely(!host_attr->debug_area_virt_addr)) { host_attr->debug_area_size = 0; return -ENOMEM; @@ -2777,7 +2782,7 @@ int ena_com_set_host_attributes(struct ena_com_dev *ena_dev) &cmd.u.host_attr.debug_ba, host_attr->debug_area_dma_addr); if (unlikely(ret)) { - pr_err("memory address set failed\n"); + pr_err("Memory address set failed\n"); return ret; } @@ -2785,7 +2790,7 @@ int ena_com_set_host_attributes(struct ena_com_dev *ena_dev) &cmd.u.host_attr.os_info_ba, host_attr->host_info_dma_addr); if (unlikely(ret)) { - pr_err("memory address set failed\n"); + pr_err("Memory address set failed\n"); return ret; } @@ -2904,7 +2909,7 @@ int ena_com_config_dev_mode(struct ena_com_dev *ena_dev, (llq_info->descs_num_before_header * sizeof(struct ena_eth_io_tx_desc)); if (unlikely(ena_dev->tx_max_header_size == 0)) { - pr_err("the size of the LLQ entry is smaller than needed\n"); + pr_err("The size of the LLQ entry is smaller than needed\n"); return -EINVAL; } diff --git a/drivers/net/ethernet/amazon/ena/ena_com.h b/drivers/net/ethernet/amazon/ena/ena_com.h index 4287d47b2b0b..55097750d062 100644 --- a/drivers/net/ethernet/amazon/ena/ena_com.h +++ b/drivers/net/ethernet/amazon/ena/ena_com.h @@ -1,33 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ /* - * Copyright 2015 Amazon.com, Inc. or its affiliates. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. + * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved. */ #ifndef ENA_COM @@ -536,7 +509,7 @@ void ena_com_admin_q_comp_intr_handler(struct ena_com_dev *ena_dev); * This method goes over the async event notification queue and calls the proper * aenq handler. */ -void ena_com_aenq_intr_handler(struct ena_com_dev *dev, void *data); +void ena_com_aenq_intr_handler(struct ena_com_dev *ena_dev, void *data); /* ena_com_abort_admin_commands - Abort all the outstanding admin commands. * @ena_dev: ENA communication layer struct @@ -616,6 +589,15 @@ int ena_com_get_dev_attr_feat(struct ena_com_dev *ena_dev, int ena_com_get_dev_basic_stats(struct ena_com_dev *ena_dev, struct ena_admin_basic_stats *stats); +/* ena_com_get_eni_stats - Get extended network interface statistics + * @ena_dev: ENA communication layer struct + * @stats: stats return value + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_get_eni_stats(struct ena_com_dev *ena_dev, + struct ena_admin_eni_stats *stats); + /* ena_com_set_dev_mtu - Configure the device mtu. * @ena_dev: ENA communication layer struct * @mtu: mtu value diff --git a/drivers/net/ethernet/amazon/ena/ena_common_defs.h b/drivers/net/ethernet/amazon/ena/ena_common_defs.h index 8a8ded0de9ac..e210c8a81fc0 100644 --- a/drivers/net/ethernet/amazon/ena/ena_common_defs.h +++ b/drivers/net/ethernet/amazon/ena/ena_common_defs.h @@ -1,33 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ /* - * Copyright 2015 - 2016 Amazon.com, Inc. or its affiliates. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. + * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved. */ #ifndef _ENA_COMMON_H_ #define _ENA_COMMON_H_ diff --git a/drivers/net/ethernet/amazon/ena/ena_eth_com.c b/drivers/net/ethernet/amazon/ena/ena_eth_com.c index ccd440589565..ad30cacc1622 100644 --- a/drivers/net/ethernet/amazon/ena/ena_eth_com.c +++ b/drivers/net/ethernet/amazon/ena/ena_eth_com.c @@ -1,33 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB /* - * Copyright 2015 Amazon.com, Inc. or its affiliates. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. + * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved. */ #include "ena_eth_com.h" @@ -45,8 +18,9 @@ static struct ena_eth_io_rx_cdesc_base *ena_com_get_next_rx_cdesc( cdesc = (struct ena_eth_io_rx_cdesc_base *)(io_cq->cdesc_addr.virt_addr + (head_masked * io_cq->cdesc_entry_size_in_bytes)); - desc_phase = (READ_ONCE(cdesc->status) & ENA_ETH_IO_RX_CDESC_BASE_PHASE_MASK) >> - ENA_ETH_IO_RX_CDESC_BASE_PHASE_SHIFT; + desc_phase = (READ_ONCE(cdesc->status) & + ENA_ETH_IO_RX_CDESC_BASE_PHASE_MASK) >> + ENA_ETH_IO_RX_CDESC_BASE_PHASE_SHIFT; if (desc_phase != expected_phase) return NULL; @@ -89,7 +63,7 @@ static int ena_com_write_bounce_buffer_to_dev(struct ena_com_io_sq *io_sq, } io_sq->entries_in_tx_burst_left--; - pr_debug("decreasing entries_in_tx_burst_left of queue %d to %d\n", + pr_debug("Decreasing entries_in_tx_burst_left of queue %d to %d\n", io_sq->qid, io_sq->entries_in_tx_burst_left); } @@ -128,12 +102,12 @@ static int ena_com_write_header_to_bounce(struct ena_com_io_sq *io_sq, if (unlikely((header_offset + header_len) > llq_info->desc_list_entry_size)) { - pr_err("trying to write header larger than llq entry can accommodate\n"); + pr_err("Trying to write header larger than llq entry can accommodate\n"); return -EFAULT; } if (unlikely(!bounce_buffer)) { - pr_err("bounce buffer is NULL\n"); + pr_err("Bounce buffer is NULL\n"); return -EFAULT; } @@ -151,7 +125,7 @@ static void *get_sq_desc_llq(struct ena_com_io_sq *io_sq) bounce_buffer = pkt_ctrl->curr_bounce_buf; if (unlikely(!bounce_buffer)) { - pr_err("bounce buffer is NULL\n"); + pr_err("Bounce buffer is NULL\n"); return NULL; } @@ -262,8 +236,9 @@ static u16 ena_com_cdesc_rx_pkt_get(struct ena_com_io_cq *io_cq, ena_com_cq_inc_head(io_cq); count++; - last = (READ_ONCE(cdesc->status) & ENA_ETH_IO_RX_CDESC_BASE_LAST_MASK) >> - ENA_ETH_IO_RX_CDESC_BASE_LAST_SHIFT; + last = (READ_ONCE(cdesc->status) & + ENA_ETH_IO_RX_CDESC_BASE_LAST_MASK) >> + ENA_ETH_IO_RX_CDESC_BASE_LAST_SHIFT; } while (!last); if (last) { @@ -275,7 +250,7 @@ static u16 ena_com_cdesc_rx_pkt_get(struct ena_com_io_cq *io_cq, io_cq->cur_rx_pkt_cdesc_count = 0; io_cq->cur_rx_pkt_cdesc_start_idx = head_masked; - pr_debug("ena q_id: %d packets were completed. first desc idx %u descs# %d\n", + pr_debug("ENA q_id: %d packets were completed. first desc idx %u descs# %d\n", io_cq->qid, *first_cdesc_idx, count); } else { io_cq->cur_rx_pkt_cdesc_count += count; @@ -291,6 +266,9 @@ static int ena_com_create_meta(struct ena_com_io_sq *io_sq, struct ena_eth_io_tx_meta_desc *meta_desc = NULL; meta_desc = get_sq_desc(io_sq); + if (unlikely(!meta_desc)) + return -EFAULT; + memset(meta_desc, 0x0, sizeof(struct ena_eth_io_tx_meta_desc)); meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_META_DESC_MASK; @@ -298,7 +276,7 @@ static int ena_com_create_meta(struct ena_com_io_sq *io_sq, meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_EXT_VALID_MASK; /* bits 0-9 of the mss */ - meta_desc->word2 |= (ena_meta->mss << + meta_desc->word2 |= ((u32)ena_meta->mss << ENA_ETH_IO_TX_META_DESC_MSS_LO_SHIFT) & ENA_ETH_IO_TX_META_DESC_MSS_LO_MASK; /* bits 10-13 of the mss */ @@ -308,7 +286,7 @@ static int ena_com_create_meta(struct ena_com_io_sq *io_sq, /* Extended meta desc */ meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_ETH_META_TYPE_MASK; - meta_desc->len_ctrl |= (io_sq->phase << + meta_desc->len_ctrl |= ((u32)io_sq->phase << ENA_ETH_IO_TX_META_DESC_PHASE_SHIFT) & ENA_ETH_IO_TX_META_DESC_PHASE_MASK; @@ -321,7 +299,7 @@ static int ena_com_create_meta(struct ena_com_io_sq *io_sq, ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_SHIFT) & ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_MASK; - meta_desc->word2 |= (ena_meta->l4_hdr_len << + meta_desc->word2 |= ((u32)ena_meta->l4_hdr_len << ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_SHIFT) & ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_MASK; @@ -358,7 +336,7 @@ static int ena_com_create_and_store_tx_meta_desc(struct ena_com_io_sq *io_sq, } static void ena_com_rx_set_flags(struct ena_com_rx_ctx *ena_rx_ctx, - struct ena_eth_io_rx_cdesc_base *cdesc) + struct ena_eth_io_rx_cdesc_base *cdesc) { ena_rx_ctx->l3_proto = cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_L3_PROTO_IDX_MASK; @@ -379,7 +357,7 @@ static void ena_com_rx_set_flags(struct ena_com_rx_ctx *ena_rx_ctx, (cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_MASK) >> ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_SHIFT; - pr_debug("ena_rx_ctx->l3_proto %d ena_rx_ctx->l4_proto %d\nena_rx_ctx->l3_csum_err %d ena_rx_ctx->l4_csum_err %d\nhash frag %d frag: %d cdesc_status: %x\n", + pr_debug("l3_proto %d l4_proto %d l3_csum_err %d l4_csum_err %d hash %d frag %d cdesc_status %x\n", ena_rx_ctx->l3_proto, ena_rx_ctx->l4_proto, ena_rx_ctx->l3_csum_err, ena_rx_ctx->l4_csum_err, ena_rx_ctx->hash, ena_rx_ctx->frag, cdesc->status); @@ -412,7 +390,7 @@ int ena_com_prepare_tx(struct ena_com_io_sq *io_sq, } if (unlikely(header_len > io_sq->tx_max_header_size)) { - pr_err("header size is too large %d max header: %d\n", + pr_err("Header size is too large %d max header: %d\n", header_len, io_sq->tx_max_header_size); return -EINVAL; } @@ -427,7 +405,7 @@ int ena_com_prepare_tx(struct ena_com_io_sq *io_sq, rc = ena_com_create_and_store_tx_meta_desc(io_sq, ena_tx_ctx, &have_meta); if (unlikely(rc)) { - pr_err("failed to create and store tx meta desc\n"); + pr_err("Failed to create and store tx meta desc\n"); return rc; } @@ -447,16 +425,16 @@ int ena_com_prepare_tx(struct ena_com_io_sq *io_sq, if (!have_meta) desc->len_ctrl |= ENA_ETH_IO_TX_DESC_FIRST_MASK; - desc->buff_addr_hi_hdr_sz |= (header_len << + desc->buff_addr_hi_hdr_sz |= ((u32)header_len << ENA_ETH_IO_TX_DESC_HEADER_LENGTH_SHIFT) & ENA_ETH_IO_TX_DESC_HEADER_LENGTH_MASK; - desc->len_ctrl |= (io_sq->phase << ENA_ETH_IO_TX_DESC_PHASE_SHIFT) & + desc->len_ctrl |= ((u32)io_sq->phase << ENA_ETH_IO_TX_DESC_PHASE_SHIFT) & ENA_ETH_IO_TX_DESC_PHASE_MASK; desc->len_ctrl |= ENA_ETH_IO_TX_DESC_COMP_REQ_MASK; /* Bits 0-9 */ - desc->meta_ctrl |= (ena_tx_ctx->req_id << + desc->meta_ctrl |= ((u32)ena_tx_ctx->req_id << ENA_ETH_IO_TX_DESC_REQ_ID_LO_SHIFT) & ENA_ETH_IO_TX_DESC_REQ_ID_LO_MASK; @@ -502,7 +480,7 @@ int ena_com_prepare_tx(struct ena_com_io_sq *io_sq, memset(desc, 0x0, sizeof(struct ena_eth_io_tx_desc)); - desc->len_ctrl |= (io_sq->phase << + desc->len_ctrl |= ((u32)io_sq->phase << ENA_ETH_IO_TX_DESC_PHASE_SHIFT) & ENA_ETH_IO_TX_DESC_PHASE_MASK; } @@ -550,7 +528,7 @@ int ena_com_rx_pkt(struct ena_com_io_cq *io_cq, return 0; } - pr_debug("fetch rx packet: queue %d completed desc: %d\n", io_cq->qid, + pr_debug("Fetch rx packet: queue %d completed desc: %d\n", io_cq->qid, nb_hw_desc); if (unlikely(nb_hw_desc > ena_rx_ctx->max_bufs)) { @@ -606,9 +584,9 @@ int ena_com_add_single_rx_desc(struct ena_com_io_sq *io_sq, desc->length = ena_buf->len; desc->ctrl = ENA_ETH_IO_RX_DESC_FIRST_MASK | - ENA_ETH_IO_RX_DESC_LAST_MASK | - (io_sq->phase & ENA_ETH_IO_RX_DESC_PHASE_MASK) | - ENA_ETH_IO_RX_DESC_COMP_REQ_MASK; + ENA_ETH_IO_RX_DESC_LAST_MASK | + (io_sq->phase & ENA_ETH_IO_RX_DESC_PHASE_MASK) | + ENA_ETH_IO_RX_DESC_COMP_REQ_MASK; desc->req_id = req_id; diff --git a/drivers/net/ethernet/amazon/ena/ena_eth_com.h b/drivers/net/ethernet/amazon/ena/ena_eth_com.h index b6592cb93b04..2c16c218818a 100644 --- a/drivers/net/ethernet/amazon/ena/ena_eth_com.h +++ b/drivers/net/ethernet/amazon/ena/ena_eth_com.h @@ -1,33 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ /* - * Copyright 2015 Amazon.com, Inc. or its affiliates. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. + * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved. */ #ifndef ENA_ETH_COM_H_ @@ -167,7 +140,7 @@ static inline bool ena_com_is_doorbell_needed(struct ena_com_io_sq *io_sq, llq_info->descs_per_entry); } - pr_debug("queue: %d num_descs: %d num_entries_needed: %d\n", io_sq->qid, + pr_debug("Queue: %d num_descs: %d num_entries_needed: %d\n", io_sq->qid, num_descs, num_entries_needed); return num_entries_needed > io_sq->entries_in_tx_burst_left; @@ -178,13 +151,13 @@ static inline int ena_com_write_sq_doorbell(struct ena_com_io_sq *io_sq) u16 max_entries_in_tx_burst = io_sq->llq_info.max_entries_in_tx_burst; u16 tail = io_sq->tail; - pr_debug("write submission queue doorbell for queue: %d tail: %d\n", + pr_debug("Write submission queue doorbell for queue: %d tail: %d\n", io_sq->qid, tail); writel(tail, io_sq->db_addr); if (is_llq_max_tx_burst_exists(io_sq)) { - pr_debug("reset available entries in tx burst for queue %d to %d\n", + pr_debug("Reset available entries in tx burst for queue %d to %d\n", io_sq->qid, max_entries_in_tx_burst); io_sq->entries_in_tx_burst_left = max_entries_in_tx_burst; } diff --git a/drivers/net/ethernet/amazon/ena/ena_eth_io_defs.h b/drivers/net/ethernet/amazon/ena/ena_eth_io_defs.h index d105c9c56192..332ac0d28ac7 100644 --- a/drivers/net/ethernet/amazon/ena/ena_eth_io_defs.h +++ b/drivers/net/ethernet/amazon/ena/ena_eth_io_defs.h @@ -1,33 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ /* - * Copyright 2015 - 2016 Amazon.com, Inc. or its affiliates. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. + * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved. */ #ifndef _ENA_ETH_IO_H_ #define _ENA_ETH_IO_H_ diff --git a/drivers/net/ethernet/amazon/ena/ena_ethtool.c b/drivers/net/ethernet/amazon/ena/ena_ethtool.c index 430275bc0d04..3b2cd28f962d 100644 --- a/drivers/net/ethernet/amazon/ena/ena_ethtool.c +++ b/drivers/net/ethernet/amazon/ena/ena_ethtool.c @@ -1,33 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB /* - * Copyright 2015 Amazon.com, Inc. or its affiliates. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. + * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved. */ #include <linux/pci.h> @@ -41,12 +14,17 @@ struct ena_stats { #define ENA_STAT_ENA_COM_ENTRY(stat) { \ .name = #stat, \ - .stat_offset = offsetof(struct ena_com_stats_admin, stat) \ + .stat_offset = offsetof(struct ena_com_stats_admin, stat) / sizeof(u64) \ } #define ENA_STAT_ENTRY(stat, stat_type) { \ .name = #stat, \ - .stat_offset = offsetof(struct ena_stats_##stat_type, stat) \ + .stat_offset = offsetof(struct ena_stats_##stat_type, stat) / sizeof(u64) \ +} + +#define ENA_STAT_HW_ENTRY(stat, stat_type) { \ + .name = #stat, \ + .stat_offset = offsetof(struct ena_admin_##stat_type, stat) / sizeof(u64) \ } #define ENA_STAT_RX_ENTRY(stat) \ @@ -58,6 +36,9 @@ struct ena_stats { #define ENA_STAT_GLOBAL_ENTRY(stat) \ ENA_STAT_ENTRY(stat, dev) +#define ENA_STAT_ENI_ENTRY(stat) \ + ENA_STAT_HW_ENTRY(stat, eni_stats) + static const struct ena_stats ena_stats_global_strings[] = { ENA_STAT_GLOBAL_ENTRY(tx_timeout), ENA_STAT_GLOBAL_ENTRY(suspend), @@ -68,6 +49,14 @@ static const struct ena_stats ena_stats_global_strings[] = { ENA_STAT_GLOBAL_ENTRY(admin_q_pause), }; +static const struct ena_stats ena_stats_eni_strings[] = { + ENA_STAT_ENI_ENTRY(bw_in_allowance_exceeded), + ENA_STAT_ENI_ENTRY(bw_out_allowance_exceeded), + ENA_STAT_ENI_ENTRY(pps_allowance_exceeded), + ENA_STAT_ENI_ENTRY(conntrack_allowance_exceeded), + ENA_STAT_ENI_ENTRY(linklocal_allowance_exceeded), +}; + static const struct ena_stats ena_stats_tx_strings[] = { ENA_STAT_TX_ENTRY(cnt), ENA_STAT_TX_ENTRY(bytes), @@ -100,6 +89,11 @@ static const struct ena_stats ena_stats_rx_strings[] = { ENA_STAT_RX_ENTRY(bad_req_id), ENA_STAT_RX_ENTRY(empty_rx_ring), ENA_STAT_RX_ENTRY(csum_unchecked), + ENA_STAT_RX_ENTRY(xdp_aborted), + ENA_STAT_RX_ENTRY(xdp_drop), + ENA_STAT_RX_ENTRY(xdp_pass), + ENA_STAT_RX_ENTRY(xdp_tx), + ENA_STAT_RX_ENTRY(xdp_invalid), }; static const struct ena_stats ena_stats_ena_com_strings[] = { @@ -110,10 +104,12 @@ static const struct ena_stats ena_stats_ena_com_strings[] = { ENA_STAT_ENA_COM_ENTRY(no_completion), }; -#define ENA_STATS_ARRAY_GLOBAL ARRAY_SIZE(ena_stats_global_strings) -#define ENA_STATS_ARRAY_TX ARRAY_SIZE(ena_stats_tx_strings) -#define ENA_STATS_ARRAY_RX ARRAY_SIZE(ena_stats_rx_strings) -#define ENA_STATS_ARRAY_ENA_COM ARRAY_SIZE(ena_stats_ena_com_strings) +#define ENA_STATS_ARRAY_GLOBAL ARRAY_SIZE(ena_stats_global_strings) +#define ENA_STATS_ARRAY_TX ARRAY_SIZE(ena_stats_tx_strings) +#define ENA_STATS_ARRAY_RX ARRAY_SIZE(ena_stats_rx_strings) +#define ENA_STATS_ARRAY_ENA_COM ARRAY_SIZE(ena_stats_ena_com_strings) +#define ENA_STATS_ARRAY_ENI(adapter) \ + (ARRAY_SIZE(ena_stats_eni_strings) * (adapter)->eni_stats_supported) static void ena_safe_update_stat(u64 *src, u64 *dst, struct u64_stats_sync *syncp) @@ -134,29 +130,30 @@ static void ena_queue_stats(struct ena_adapter *adapter, u64 **data) u64 *ptr; int i, j; - for (i = 0; i < adapter->num_io_queues; i++) { + for (i = 0; i < adapter->num_io_queues + adapter->xdp_num_queues; i++) { /* Tx stats */ ring = &adapter->tx_ring[i]; for (j = 0; j < ENA_STATS_ARRAY_TX; j++) { ena_stats = &ena_stats_tx_strings[j]; - ptr = (u64 *)((uintptr_t)&ring->tx_stats + - (uintptr_t)ena_stats->stat_offset); + ptr = (u64 *)&ring->tx_stats + ena_stats->stat_offset; ena_safe_update_stat(ptr, (*data)++, &ring->syncp); } + /* XDP TX queues don't have a RX queue counterpart */ + if (!ENA_IS_XDP_INDEX(adapter, i)) { + /* Rx stats */ + ring = &adapter->rx_ring[i]; - /* Rx stats */ - ring = &adapter->rx_ring[i]; - - for (j = 0; j < ENA_STATS_ARRAY_RX; j++) { - ena_stats = &ena_stats_rx_strings[j]; + for (j = 0; j < ENA_STATS_ARRAY_RX; j++) { + ena_stats = &ena_stats_rx_strings[j]; - ptr = (u64 *)((uintptr_t)&ring->rx_stats + - (uintptr_t)ena_stats->stat_offset); + ptr = (u64 *)&ring->rx_stats + + ena_stats->stat_offset; - ena_safe_update_stat(ptr, (*data)++, &ring->syncp); + ena_safe_update_stat(ptr, (*data)++, &ring->syncp); + } } } } @@ -170,18 +167,17 @@ static void ena_dev_admin_queue_stats(struct ena_adapter *adapter, u64 **data) for (i = 0; i < ENA_STATS_ARRAY_ENA_COM; i++) { ena_stats = &ena_stats_ena_com_strings[i]; - ptr = (u64 *)((uintptr_t)&adapter->ena_dev->admin_queue.stats + - (uintptr_t)ena_stats->stat_offset); + ptr = (u64 *)&adapter->ena_dev->admin_queue.stats + + ena_stats->stat_offset; *(*data)++ = *ptr; } } -static void ena_get_ethtool_stats(struct net_device *netdev, - struct ethtool_stats *stats, - u64 *data) +static void ena_get_stats(struct ena_adapter *adapter, + u64 *data, + bool eni_stats_needed) { - struct ena_adapter *adapter = netdev_priv(netdev); const struct ena_stats *ena_stats; u64 *ptr; int i; @@ -189,16 +185,48 @@ static void ena_get_ethtool_stats(struct net_device *netdev, for (i = 0; i < ENA_STATS_ARRAY_GLOBAL; i++) { ena_stats = &ena_stats_global_strings[i]; - ptr = (u64 *)((uintptr_t)&adapter->dev_stats + - (uintptr_t)ena_stats->stat_offset); + ptr = (u64 *)&adapter->dev_stats + ena_stats->stat_offset; ena_safe_update_stat(ptr, data++, &adapter->syncp); } + if (eni_stats_needed) { + ena_update_hw_stats(adapter); + for (i = 0; i < ENA_STATS_ARRAY_ENI(adapter); i++) { + ena_stats = &ena_stats_eni_strings[i]; + + ptr = (u64 *)&adapter->eni_stats + + ena_stats->stat_offset; + + ena_safe_update_stat(ptr, data++, &adapter->syncp); + } + } + ena_queue_stats(adapter, &data); ena_dev_admin_queue_stats(adapter, &data); } +static void ena_get_ethtool_stats(struct net_device *netdev, + struct ethtool_stats *stats, + u64 *data) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + + ena_get_stats(adapter, data, adapter->eni_stats_supported); +} + +static int ena_get_sw_stats_count(struct ena_adapter *adapter) +{ + return adapter->num_io_queues * (ENA_STATS_ARRAY_TX + ENA_STATS_ARRAY_RX) + + adapter->xdp_num_queues * ENA_STATS_ARRAY_TX + + ENA_STATS_ARRAY_GLOBAL + ENA_STATS_ARRAY_ENA_COM; +} + +static int ena_get_hw_stats_count(struct ena_adapter *adapter) +{ + return ENA_STATS_ARRAY_ENI(adapter); +} + int ena_get_sset_count(struct net_device *netdev, int sset) { struct ena_adapter *adapter = netdev_priv(netdev); @@ -206,31 +234,38 @@ int ena_get_sset_count(struct net_device *netdev, int sset) if (sset != ETH_SS_STATS) return -EOPNOTSUPP; - return adapter->num_io_queues * (ENA_STATS_ARRAY_TX + ENA_STATS_ARRAY_RX) - + ENA_STATS_ARRAY_GLOBAL + ENA_STATS_ARRAY_ENA_COM; + return ena_get_sw_stats_count(adapter) + ena_get_hw_stats_count(adapter); } static void ena_queue_strings(struct ena_adapter *adapter, u8 **data) { const struct ena_stats *ena_stats; + bool is_xdp; int i, j; - for (i = 0; i < adapter->num_io_queues; i++) { + for (i = 0; i < adapter->num_io_queues + adapter->xdp_num_queues; i++) { + is_xdp = ENA_IS_XDP_INDEX(adapter, i); /* Tx stats */ for (j = 0; j < ENA_STATS_ARRAY_TX; j++) { ena_stats = &ena_stats_tx_strings[j]; snprintf(*data, ETH_GSTRING_LEN, - "queue_%u_tx_%s", i, ena_stats->name); + "queue_%u_%s_%s", i, + is_xdp ? "xdp_tx" : "tx", ena_stats->name); (*data) += ETH_GSTRING_LEN; } - /* Rx stats */ - for (j = 0; j < ENA_STATS_ARRAY_RX; j++) { - ena_stats = &ena_stats_rx_strings[j]; - snprintf(*data, ETH_GSTRING_LEN, - "queue_%u_rx_%s", i, ena_stats->name); - (*data) += ETH_GSTRING_LEN; + if (!is_xdp) { + /* RX stats, in XDP there isn't a RX queue + * counterpart + */ + for (j = 0; j < ENA_STATS_ARRAY_RX; j++) { + ena_stats = &ena_stats_rx_strings[j]; + + snprintf(*data, ETH_GSTRING_LEN, + "queue_%u_rx_%s", i, ena_stats->name); + (*data) += ETH_GSTRING_LEN; + } } } } @@ -249,25 +284,43 @@ static void ena_com_dev_strings(u8 **data) } } -static void ena_get_strings(struct net_device *netdev, u32 sset, u8 *data) +static void ena_get_strings(struct ena_adapter *adapter, + u8 *data, + bool eni_stats_needed) { - struct ena_adapter *adapter = netdev_priv(netdev); const struct ena_stats *ena_stats; int i; - if (sset != ETH_SS_STATS) - return; - for (i = 0; i < ENA_STATS_ARRAY_GLOBAL; i++) { ena_stats = &ena_stats_global_strings[i]; memcpy(data, ena_stats->name, ETH_GSTRING_LEN); data += ETH_GSTRING_LEN; } + if (eni_stats_needed) { + for (i = 0; i < ENA_STATS_ARRAY_ENI(adapter); i++) { + ena_stats = &ena_stats_eni_strings[i]; + memcpy(data, ena_stats->name, ETH_GSTRING_LEN); + data += ETH_GSTRING_LEN; + } + } + ena_queue_strings(adapter, &data); ena_com_dev_strings(&data); } +static void ena_get_ethtool_strings(struct net_device *netdev, + u32 sset, + u8 *data) +{ + struct ena_adapter *adapter = netdev_priv(netdev); + + if (sset != ETH_SS_STATS) + return; + + ena_get_strings(adapter, data, adapter->eni_stats_supported); +} + static int ena_get_link_ksettings(struct net_device *netdev, struct ethtool_link_ksettings *link_ksettings) { @@ -847,7 +900,7 @@ static const struct ethtool_ops ena_ethtool_ops = { .get_ringparam = ena_get_ringparam, .set_ringparam = ena_set_ringparam, .get_sset_count = ena_get_sset_count, - .get_strings = ena_get_strings, + .get_strings = ena_get_ethtool_strings, .get_ethtool_stats = ena_get_ethtool_stats, .get_rxnfc = ena_get_rxnfc, .set_rxnfc = ena_set_rxnfc, @@ -875,7 +928,7 @@ static void ena_dump_stats_ex(struct ena_adapter *adapter, u8 *buf) int strings_num; int i, rc; - strings_num = ena_get_sset_count(netdev, ETH_SS_STATS); + strings_num = ena_get_sw_stats_count(adapter); if (strings_num <= 0) { netif_err(adapter, drv, netdev, "Can't get stats num\n"); return; @@ -886,7 +939,7 @@ static void ena_dump_stats_ex(struct ena_adapter *adapter, u8 *buf) GFP_ATOMIC); if (!strings_buf) { netif_err(adapter, drv, netdev, - "failed to alloc strings_buf\n"); + "Failed to allocate strings_buf\n"); return; } @@ -895,13 +948,13 @@ static void ena_dump_stats_ex(struct ena_adapter *adapter, u8 *buf) GFP_ATOMIC); if (!data_buf) { netif_err(adapter, drv, netdev, - "failed to allocate data buf\n"); + "Failed to allocate data buf\n"); devm_kfree(&adapter->pdev->dev, strings_buf); return; } - ena_get_strings(netdev, ETH_SS_STATS, strings_buf); - ena_get_ethtool_stats(netdev, NULL, data_buf); + ena_get_strings(adapter, strings_buf, false); + ena_get_stats(adapter, data_buf, false); /* If there is a buffer, dump stats, otherwise print them to dmesg */ if (buf) diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index a3a8edf9a734..e8131dadc22c 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -1,33 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB /* - * Copyright 2015 Amazon.com, Inc. or its affiliates. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. + * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved. */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt @@ -139,7 +112,7 @@ static int ena_change_mtu(struct net_device *dev, int new_mtu) ret = ena_com_set_dev_mtu(adapter->ena_dev, new_mtu); if (!ret) { - netif_dbg(adapter, drv, dev, "set MTU to %d\n", new_mtu); + netif_dbg(adapter, drv, dev, "Set MTU to %d\n", new_mtu); update_rx_ring_mtu(adapter, new_mtu); dev->mtu = new_mtu; } else { @@ -178,7 +151,7 @@ static int ena_xmit_common(struct net_device *dev, */ if (unlikely(rc)) { netif_err(adapter, tx_queued, dev, - "failed to prepare tx bufs\n"); + "Failed to prepare tx bufs\n"); u64_stats_update_begin(&ring->syncp); ring->tx_stats.prepare_ctx_err++; u64_stats_update_end(&ring->syncp); @@ -292,7 +265,7 @@ error_report_dma_error: u64_stats_update_begin(&xdp_ring->syncp); xdp_ring->tx_stats.dma_mapping_err++; u64_stats_update_end(&xdp_ring->syncp); - netdev_warn(adapter->netdev, "failed to map xdp buff\n"); + netif_warn(adapter, tx_queued, adapter->netdev, "Failed to map xdp buff\n"); xdp_return_frame_rx_napi(tx_info->xdpf); tx_info->xdpf = NULL; @@ -365,6 +338,7 @@ static int ena_xdp_execute(struct ena_ring *rx_ring, { struct bpf_prog *xdp_prog; u32 verdict = XDP_PASS; + u64 *xdp_stat; rcu_read_lock(); xdp_prog = READ_ONCE(rx_ring->xdp_bpf_prog); @@ -374,17 +348,31 @@ static int ena_xdp_execute(struct ena_ring *rx_ring, verdict = bpf_prog_run_xdp(xdp_prog, xdp); - if (verdict == XDP_TX) + if (verdict == XDP_TX) { ena_xdp_xmit_buff(rx_ring->netdev, xdp, rx_ring->qid + rx_ring->adapter->num_io_queues, rx_info); - else if (unlikely(verdict == XDP_ABORTED)) + + xdp_stat = &rx_ring->rx_stats.xdp_tx; + } else if (unlikely(verdict == XDP_ABORTED)) { trace_xdp_exception(rx_ring->netdev, xdp_prog, verdict); - else if (unlikely(verdict > XDP_TX)) + xdp_stat = &rx_ring->rx_stats.xdp_aborted; + } else if (unlikely(verdict == XDP_DROP)) { + xdp_stat = &rx_ring->rx_stats.xdp_drop; + } else if (unlikely(verdict == XDP_PASS)) { + xdp_stat = &rx_ring->rx_stats.xdp_pass; + } else { bpf_warn_invalid_xdp_action(verdict); + xdp_stat = &rx_ring->rx_stats.xdp_invalid; + } + + u64_stats_update_begin(&rx_ring->syncp); + (*xdp_stat)++; + u64_stats_update_end(&rx_ring->syncp); out: rcu_read_unlock(); + return verdict; } @@ -549,7 +537,7 @@ static int ena_xdp_set(struct net_device *netdev, struct netdev_bpf *bpf) if (!old_bpf_prog) netif_info(adapter, drv, adapter->netdev, - "xdp program set, changing the max_mtu from %d to %d", + "XDP program is set, changing the max_mtu from %d to %d", prev_mtu, netdev->max_mtu); } else if (rc == ENA_XDP_CURRENT_MTU_TOO_LARGE) { @@ -968,7 +956,7 @@ static int ena_alloc_rx_page(struct ena_ring *rx_ring, return -EIO; } netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev, - "alloc page %p, rx_info %p\n", page, rx_info); + "Allocate page %p, rx_info %p\n", page, rx_info); rx_info->page = page; rx_info->page_offset = 0; @@ -1018,7 +1006,7 @@ static int ena_refill_rx_bufs(struct ena_ring *rx_ring, u32 num) GFP_ATOMIC | __GFP_COMP); if (unlikely(rc < 0)) { netif_warn(rx_ring->adapter, rx_err, rx_ring->netdev, - "failed to alloc buffer for rx queue %d\n", + "Failed to allocate buffer for rx queue %d\n", rx_ring->qid); break; } @@ -1027,7 +1015,7 @@ static int ena_refill_rx_bufs(struct ena_ring *rx_ring, u32 num) req_id); if (unlikely(rc)) { netif_warn(rx_ring->adapter, rx_status, rx_ring->netdev, - "failed to add buffer for rx queue %d\n", + "Failed to add buffer for rx queue %d\n", rx_ring->qid); break; } @@ -1039,9 +1027,9 @@ static int ena_refill_rx_bufs(struct ena_ring *rx_ring, u32 num) u64_stats_update_begin(&rx_ring->syncp); rx_ring->rx_stats.refil_partial++; u64_stats_update_end(&rx_ring->syncp); - netdev_warn(rx_ring->netdev, - "refilled rx qid %d with only %d buffers (from %d)\n", - rx_ring->qid, i, num); + netif_warn(rx_ring->adapter, rx_err, rx_ring->netdev, + "Refilled rx qid %d with only %d buffers (from %d)\n", + rx_ring->qid, i, num); } /* ena_com_write_sq_doorbell issues a wmb() */ @@ -1082,7 +1070,7 @@ static void ena_refill_all_rx_bufs(struct ena_adapter *adapter) if (unlikely(rc != bufs_num)) netif_warn(rx_ring->adapter, rx_status, rx_ring->netdev, - "refilling Queue %d failed. allocated %d buffers from: %d\n", + "Refilling Queue %d failed. allocated %d buffers from: %d\n", i, rc, bufs_num); } } @@ -1140,14 +1128,14 @@ static void ena_free_tx_bufs(struct ena_ring *tx_ring) continue; if (print_once) { - netdev_notice(tx_ring->netdev, - "free uncompleted tx skb qid %d idx 0x%x\n", - tx_ring->qid, i); + netif_notice(tx_ring->adapter, ifdown, tx_ring->netdev, + "Free uncompleted tx skb qid %d idx 0x%x\n", + tx_ring->qid, i); print_once = false; } else { - netdev_dbg(tx_ring->netdev, - "free uncompleted tx skb qid %d idx 0x%x\n", - tx_ring->qid, i); + netif_dbg(tx_ring->adapter, ifdown, tx_ring->netdev, + "Free uncompleted tx skb qid %d idx 0x%x\n", + tx_ring->qid, i); } ena_unmap_tx_buff(tx_ring, tx_info); @@ -1399,7 +1387,7 @@ static struct sk_buff *ena_rx_skb(struct ena_ring *rx_ring, return NULL; netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev, - "rx allocated small packet. len %d. data_len %d\n", + "RX allocated small packet. len %d. data_len %d\n", skb->len, skb->data_len); /* sync this buffer for CPU use */ @@ -1436,7 +1424,7 @@ static struct sk_buff *ena_rx_skb(struct ena_ring *rx_ring, rx_info->page_offset = 0; netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev, - "rx skb updated. len %d. data_len %d\n", + "RX skb updated. len %d. data_len %d\n", skb->len, skb->data_len); rx_info->page = NULL; @@ -1643,6 +1631,11 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi, &next_to_clean); if (unlikely(!skb)) { + /* The page might not actually be freed here since the + * page reference count is incremented in + * ena_xdp_xmit_buff(), and it will be decreased only + * when send completion was received from the device + */ if (xdp_verdict == XDP_TX) ena_free_rx_page(rx_ring, &rx_ring->rx_buffer_info[rx_ring->ena_bufs[0].req_id]); @@ -1770,6 +1763,7 @@ static void ena_unmask_interrupt(struct ena_ring *tx_ring, u64_stats_update_begin(&tx_ring->syncp); tx_ring->tx_stats.unmask_interrupt++; u64_stats_update_end(&tx_ring->syncp); + /* It is a shared MSI-X. * Tx and Rx CQ have pointer to it. * So we use one of them to reach the intr reg @@ -1987,7 +1981,7 @@ static int ena_enable_msix(struct ena_adapter *adapter) /* Reserved the max msix vectors we might need */ msix_vecs = ENA_MAX_MSIX_VEC(adapter->max_num_io_queues); netif_dbg(adapter, probe, adapter->netdev, - "trying to enable MSI-X, vectors %d\n", msix_vecs); + "Trying to enable MSI-X, vectors %d\n", msix_vecs); irq_cnt = pci_alloc_irq_vectors(adapter->pdev, ENA_MIN_MSIX_VEC, msix_vecs, PCI_IRQ_MSIX); @@ -2000,7 +1994,7 @@ static int ena_enable_msix(struct ena_adapter *adapter) if (irq_cnt != msix_vecs) { netif_notice(adapter, probe, adapter->netdev, - "enable only %d MSI-X (out of %d), reduce the number of queues\n", + "Enable only %d MSI-X (out of %d), reduce the number of queues\n", irq_cnt, msix_vecs); adapter->num_io_queues = irq_cnt - ENA_ADMIN_MSIX_VEC; } @@ -2070,12 +2064,12 @@ static int ena_request_mgmnt_irq(struct ena_adapter *adapter) irq->data); if (rc) { netif_err(adapter, probe, adapter->netdev, - "failed to request admin irq\n"); + "Failed to request admin irq\n"); return rc; } netif_dbg(adapter, probe, adapter->netdev, - "set affinity hint of mgmnt irq.to 0x%lx (irq vector: %d)\n", + "Set affinity hint of mgmnt irq.to 0x%lx (irq vector: %d)\n", irq->affinity_hint_mask.bits[0], irq->vector); irq_set_affinity_hint(irq->vector, &irq->affinity_hint_mask); @@ -2108,7 +2102,7 @@ static int ena_request_io_irq(struct ena_adapter *adapter) } netif_dbg(adapter, ifup, adapter->netdev, - "set affinity hint of irq. index %d to 0x%lx (irq vector: %d)\n", + "Set affinity hint of irq. index %d to 0x%lx (irq vector: %d)\n", i, irq->affinity_hint_mask.bits[0], irq->vector); irq_set_affinity_hint(irq->vector, &irq->affinity_hint_mask); @@ -2548,7 +2542,7 @@ static int ena_up(struct ena_adapter *adapter) { int io_queue_count, rc, i; - netdev_dbg(adapter->netdev, "%s\n", __func__); + netif_dbg(adapter, ifup, adapter->netdev, "%s\n", __func__); io_queue_count = adapter->num_io_queues + adapter->xdp_num_queues; ena_setup_io_intr(adapter); @@ -2632,7 +2626,8 @@ static void ena_down(struct ena_adapter *adapter) rc = ena_com_dev_reset(adapter->ena_dev, adapter->reset_reason); if (rc) - dev_err(&adapter->pdev->dev, "Device reset failed\n"); + netif_err(adapter, ifdown, adapter->netdev, + "Device reset failed\n"); /* stop submitting admin commands on a device that was reset */ ena_com_set_admin_running_state(adapter->ena_dev, false); } @@ -2954,7 +2949,7 @@ error_report_dma_error: u64_stats_update_begin(&tx_ring->syncp); tx_ring->tx_stats.dma_mapping_err++; u64_stats_update_end(&tx_ring->syncp); - netdev_warn(adapter->netdev, "failed to map skb\n"); + netif_warn(adapter, tx_queued, adapter->netdev, "Failed to map skb\n"); tx_info->skb = NULL; @@ -3092,13 +3087,14 @@ static u16 ena_select_queue(struct net_device *dev, struct sk_buff *skb, static void ena_config_host_info(struct ena_com_dev *ena_dev, struct pci_dev *pdev) { + struct device *dev = &pdev->dev; struct ena_admin_host_info *host_info; int rc; /* Allocate only the host info */ rc = ena_com_allocate_host_info(ena_dev); if (rc) { - pr_err("Cannot allocate host info\n"); + dev_err(dev, "Cannot allocate host info\n"); return; } @@ -3128,9 +3124,9 @@ static void ena_config_host_info(struct ena_com_dev *ena_dev, struct pci_dev *pd rc = ena_com_set_host_attributes(ena_dev); if (rc) { if (rc == -EOPNOTSUPP) - pr_warn("Cannot set host attributes\n"); + dev_warn(dev, "Cannot set host attributes\n"); else - pr_err("Cannot set host attributes\n"); + dev_err(dev, "Cannot set host attributes\n"); goto err; } @@ -3158,7 +3154,8 @@ static void ena_config_debug_area(struct ena_adapter *adapter) rc = ena_com_allocate_debug_area(adapter->ena_dev, debug_area_size); if (rc) { - pr_err("Cannot allocate debug area\n"); + netif_err(adapter, drv, adapter->netdev, + "Cannot allocate debug area\n"); return; } @@ -3178,6 +3175,19 @@ err: ena_com_delete_debug_area(adapter->ena_dev); } +int ena_update_hw_stats(struct ena_adapter *adapter) +{ + int rc = 0; + + rc = ena_com_get_eni_stats(adapter->ena_dev, &adapter->eni_stats); + if (rc) { + dev_info_once(&adapter->pdev->dev, "Failed to get ENI stats\n"); + return rc; + } + + return 0; +} + static void ena_get_stats64(struct net_device *netdev, struct rtnl_link_stats64 *stats) { @@ -3349,7 +3359,7 @@ static int ena_device_init(struct ena_com_dev *ena_dev, struct pci_dev *pdev, rc = ena_com_mmio_reg_read_request_init(ena_dev); if (rc) { - dev_err(dev, "failed to init mmio read less\n"); + dev_err(dev, "Failed to init mmio read less\n"); return rc; } @@ -3367,7 +3377,7 @@ static int ena_device_init(struct ena_com_dev *ena_dev, struct pci_dev *pdev, rc = ena_com_validate_version(ena_dev); if (rc) { - dev_err(dev, "device version is too low\n"); + dev_err(dev, "Device version is too low\n"); goto err_mmio_read_less; } @@ -3436,7 +3446,7 @@ static int ena_device_init(struct ena_com_dev *ena_dev, struct pci_dev *pdev, rc = ena_set_queues_placement_policy(pdev, ena_dev, &get_feat_ctx->llq, &llq_config); if (rc) { - dev_err(&pdev->dev, "ena device init failed\n"); + dev_err(dev, "ENA device init failed\n"); goto err_admin_init; } @@ -3572,9 +3582,10 @@ static int ena_restore_device(struct ena_adapter *adapter) netif_carrier_on(adapter->netdev); mod_timer(&adapter->timer_service, round_jiffies(jiffies + HZ)); - dev_err(&pdev->dev, "Device reset completed successfully\n"); adapter->last_keep_alive_jiffies = jiffies; + dev_err(&pdev->dev, "Device reset completed successfully\n"); + return rc; err_disable_msix: ena_free_mgmnt_irq(adapter); @@ -3776,7 +3787,7 @@ static void check_for_empty_rx_ring(struct ena_adapter *adapter) u64_stats_update_end(&rx_ring->syncp); netif_err(adapter, drv, adapter->netdev, - "trigger refill for ring %d\n", i); + "Trigger refill for ring %d\n", i); napi_schedule(rx_ring->napi); rx_ring->empty_rx_queue = 0; @@ -4138,14 +4149,13 @@ static int ena_calc_io_queue_size(struct ena_calc_queue_size_ctx *ctx) */ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { - struct ena_calc_queue_size_ctx calc_queue_ctx = { 0 }; + struct ena_calc_queue_size_ctx calc_queue_ctx = {}; struct ena_com_dev_get_features_ctx get_feat_ctx; struct ena_com_dev *ena_dev = NULL; struct ena_adapter *adapter; struct net_device *netdev; static int adapters_found; u32 max_num_io_queues; - char *queue_type_str; bool wd_state; int bars, rc; @@ -4177,7 +4187,7 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent) pci_resource_start(pdev, ENA_REG_BAR), pci_resource_len(pdev, ENA_REG_BAR)); if (!ena_dev->reg_bar) { - dev_err(&pdev->dev, "failed to remap regs bar\n"); + dev_err(&pdev->dev, "Failed to remap regs bar\n"); rc = -EFAULT; goto err_free_region; } @@ -4188,7 +4198,7 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent) rc = ena_device_init(ena_dev, pdev, &get_feat_ctx, &wd_state); if (rc) { - dev_err(&pdev->dev, "ena device init failed\n"); + dev_err(&pdev->dev, "ENA device init failed\n"); if (rc == -ETIME) rc = -EPROBE_DEFER; goto err_free_region; @@ -4196,7 +4206,7 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent) rc = ena_map_llq_mem_bar(pdev, ena_dev, bars); if (rc) { - dev_err(&pdev->dev, "ena llq bar mapping failed\n"); + dev_err(&pdev->dev, "ENA llq bar mapping failed\n"); goto err_free_ena_dev; } @@ -4296,6 +4306,11 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent) ena_config_debug_area(adapter); + if (!ena_update_hw_stats(adapter)) + adapter->eni_stats_supported = true; + else + adapter->eni_stats_supported = false; + memcpy(adapter->netdev->perm_addr, adapter->mac_addr, netdev->addr_len); netif_carrier_off(netdev); @@ -4318,15 +4333,10 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent) timer_setup(&adapter->timer_service, ena_timer_service, 0); mod_timer(&adapter->timer_service, round_jiffies(jiffies + HZ)); - if (ena_dev->tx_mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_HOST) - queue_type_str = "Regular"; - else - queue_type_str = "Low Latency"; - dev_info(&pdev->dev, - "%s found at mem %lx, mac addr %pM, Placement policy: %s\n", + "%s found at mem %lx, mac addr %pM\n", DEVICE_NAME, (long)pci_resource_start(pdev, 0), - netdev->dev_addr, queue_type_str); + netdev->dev_addr); set_bit(ENA_FLAG_DEVICE_RUNNING, &adapter->flags); @@ -4456,7 +4466,7 @@ static int __maybe_unused ena_suspend(struct device *dev_d) rtnl_lock(); if (unlikely(test_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags))) { dev_err(&pdev->dev, - "ignoring device reset request as the device is being suspended\n"); + "Ignoring device reset request as the device is being suspended\n"); clear_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags); } ena_destroy_device(adapter, true); @@ -4531,7 +4541,7 @@ static void ena_update_on_link_change(void *adapter_data, ENA_ADMIN_AENQ_LINK_CHANGE_DESC_LINK_STATUS_MASK; if (status) { - netdev_dbg(adapter->netdev, "%s\n", __func__); + netif_dbg(adapter, ifup, adapter->netdev, "%s\n", __func__); set_bit(ENA_FLAG_LINK_UP, &adapter->flags); if (!test_bit(ENA_FLAG_ONGOING_RESET, &adapter->flags)) netif_carrier_on(adapter->netdev); @@ -4575,7 +4585,7 @@ static void ena_notification(void *adapter_data, aenq_e->aenq_common_desc.group, ENA_ADMIN_NOTIFICATION); - switch (aenq_e->aenq_common_desc.syndrom) { + switch (aenq_e->aenq_common_desc.syndrome) { case ENA_ADMIN_UPDATE_HINTS: hints = (struct ena_admin_ena_hw_hints *) (&aenq_e->inline_data_w4); @@ -4584,7 +4594,7 @@ static void ena_notification(void *adapter_data, default: netif_err(adapter, drv, adapter->netdev, "Invalid aenq notification link state %d\n", - aenq_e->aenq_common_desc.syndrom); + aenq_e->aenq_common_desc.syndrome); } } diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h index 0c8504006247..30eb686749dc 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.h +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h @@ -1,33 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ /* - * Copyright 2015 Amazon.com, Inc. or its affiliates. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. + * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved. */ #ifndef ENA_H @@ -261,6 +234,11 @@ struct ena_stats_rx { u64 bad_req_id; u64 empty_rx_ring; u64 csum_unchecked; + u64 xdp_aborted; + u64 xdp_drop; + u64 xdp_pass; + u64 xdp_tx; + u64 xdp_invalid; }; struct ena_ring { @@ -405,6 +383,8 @@ struct ena_adapter { struct u64_stats_sync syncp; struct ena_stats_dev dev_stats; + struct ena_admin_eni_stats eni_stats; + bool eni_stats_supported; /* last queue index that was checked for uncompleted tx packets */ u32 last_monitored_tx_qid; @@ -422,6 +402,8 @@ void ena_dump_stats_to_dmesg(struct ena_adapter *adapter); void ena_dump_stats_to_buf(struct ena_adapter *adapter, u8 *buf); +int ena_update_hw_stats(struct ena_adapter *adapter); + int ena_update_queue_sizes(struct ena_adapter *adapter, u32 new_tx_size, u32 new_rx_size); diff --git a/drivers/net/ethernet/amazon/ena/ena_pci_id_tbl.h b/drivers/net/ethernet/amazon/ena/ena_pci_id_tbl.h index 426e57e10a7f..3ecdf29160ca 100644 --- a/drivers/net/ethernet/amazon/ena/ena_pci_id_tbl.h +++ b/drivers/net/ethernet/amazon/ena/ena_pci_id_tbl.h @@ -1,33 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ /* - * Copyright 2015 Amazon.com, Inc. or its affiliates. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. + * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved. */ #ifndef ENA_PCI_ID_TBL_H_ diff --git a/drivers/net/ethernet/amazon/ena/ena_regs_defs.h b/drivers/net/ethernet/amazon/ena/ena_regs_defs.h index b514bb1b855d..1e007a41a525 100644 --- a/drivers/net/ethernet/amazon/ena/ena_regs_defs.h +++ b/drivers/net/ethernet/amazon/ena/ena_regs_defs.h @@ -1,33 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ /* - * Copyright 2015 - 2016 Amazon.com, Inc. or its affiliates. - * - * This software is available to you under a choice of one of two - * licenses. You may choose to be licensed under the terms of the GNU - * General Public License (GPL) Version 2, available from the file - * COPYING in the main directory of this source tree, or the - * BSD license below: - * - * Redistribution and use in source and binary forms, with or - * without modification, are permitted provided that the following - * conditions are met: - * - * - Redistributions of source code must retain the above - * copyright notice, this list of conditions and the following - * disclaimer. - * - * - Redistributions in binary form must reproduce the above - * copyright notice, this list of conditions and the following - * disclaimer in the documentation and/or other materials - * provided with the distribution. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, - * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF - * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND - * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS - * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN - * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN - * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - * SOFTWARE. + * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved. */ #ifndef _ENA_REGS_H_ #define _ENA_REGS_H_ diff --git a/drivers/net/ethernet/amd/sun3lance.c b/drivers/net/ethernet/amd/sun3lance.c index e1fde585fd0d..00ae1081254d 100644 --- a/drivers/net/ethernet/amd/sun3lance.c +++ b/drivers/net/ethernet/amd/sun3lance.c @@ -657,16 +657,6 @@ static irqreturn_t lance_interrupt( int irq, void *dev_id) struct net_device *dev = dev_id; struct lance_private *lp = netdev_priv(dev); int csr0; - static int in_interrupt; - - if (dev == NULL) { - DPRINTK( 1, ( "lance_interrupt(): invalid dev_id\n" )); - return IRQ_NONE; - } - - if (in_interrupt) - DPRINTK( 2, ( "%s: Re-entering the interrupt handler.\n", dev->name )); - in_interrupt = 1; still_more: flush_cache_all(); @@ -774,7 +764,6 @@ static irqreturn_t lance_interrupt( int irq, void *dev_id) DPRINTK( 2, ( "%s: exiting interrupt, csr0=%#04x.\n", dev->name, DREG )); - in_interrupt = 0; return IRQ_HANDLED; } diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c index 4ba75551cb17..2709a2db5657 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c @@ -403,9 +403,9 @@ static bool xgbe_ecc_ded(struct xgbe_prv_data *pdata, unsigned long *period, return false; } -static void xgbe_ecc_isr_task(unsigned long data) +static void xgbe_ecc_isr_task(struct tasklet_struct *t) { - struct xgbe_prv_data *pdata = (struct xgbe_prv_data *)data; + struct xgbe_prv_data *pdata = from_tasklet(pdata, t, tasklet_ecc); unsigned int ecc_isr; bool stop = false; @@ -468,14 +468,14 @@ static irqreturn_t xgbe_ecc_isr(int irq, void *data) if (pdata->isr_as_tasklet) tasklet_schedule(&pdata->tasklet_ecc); else - xgbe_ecc_isr_task((unsigned long)pdata); + xgbe_ecc_isr_task(&pdata->tasklet_ecc); return IRQ_HANDLED; } -static void xgbe_isr_task(unsigned long data) +static void xgbe_isr_task(struct tasklet_struct *t) { - struct xgbe_prv_data *pdata = (struct xgbe_prv_data *)data; + struct xgbe_prv_data *pdata = from_tasklet(pdata, t, tasklet_dev); struct xgbe_hw_if *hw_if = &pdata->hw_if; struct xgbe_channel *channel; unsigned int dma_isr, dma_ch_isr; @@ -582,7 +582,7 @@ isr_done: /* If there is not a separate ECC irq, handle it here */ if (pdata->vdata->ecc_support && (pdata->dev_irq == pdata->ecc_irq)) - xgbe_ecc_isr_task((unsigned long)pdata); + xgbe_ecc_isr_task(&pdata->tasklet_ecc); /* If there is not a separate I2C irq, handle it here */ if (pdata->vdata->i2c_support && (pdata->dev_irq == pdata->i2c_irq)) @@ -607,7 +607,7 @@ static irqreturn_t xgbe_isr(int irq, void *data) if (pdata->isr_as_tasklet) tasklet_schedule(&pdata->tasklet_dev); else - xgbe_isr_task((unsigned long)pdata); + xgbe_isr_task(&pdata->tasklet_dev); return IRQ_HANDLED; } @@ -991,9 +991,8 @@ static int xgbe_request_irqs(struct xgbe_prv_data *pdata) unsigned int i; int ret; - tasklet_init(&pdata->tasklet_dev, xgbe_isr_task, (unsigned long)pdata); - tasklet_init(&pdata->tasklet_ecc, xgbe_ecc_isr_task, - (unsigned long)pdata); + tasklet_setup(&pdata->tasklet_dev, xgbe_isr_task); + tasklet_setup(&pdata->tasklet_ecc, xgbe_ecc_isr_task); ret = devm_request_irq(pdata->dev, pdata->dev_irq, xgbe_isr, 0, netdev_name(netdev), pdata); diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c b/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c index 4d9062d35930..22d4fc547a0a 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c @@ -274,9 +274,9 @@ static void xgbe_i2c_clear_isr_interrupts(struct xgbe_prv_data *pdata, XI2C_IOREAD(pdata, IC_CLR_STOP_DET); } -static void xgbe_i2c_isr_task(unsigned long data) +static void xgbe_i2c_isr_task(struct tasklet_struct *t) { - struct xgbe_prv_data *pdata = (struct xgbe_prv_data *)data; + struct xgbe_prv_data *pdata = from_tasklet(pdata, t, tasklet_i2c); struct xgbe_i2c_op_state *state = &pdata->i2c.op_state; unsigned int isr; @@ -324,7 +324,7 @@ static irqreturn_t xgbe_i2c_isr(int irq, void *data) if (pdata->isr_as_tasklet) tasklet_schedule(&pdata->tasklet_i2c); else - xgbe_i2c_isr_task((unsigned long)pdata); + xgbe_i2c_isr_task(&pdata->tasklet_i2c); return IRQ_HANDLED; } @@ -369,7 +369,7 @@ static void xgbe_i2c_set_target(struct xgbe_prv_data *pdata, unsigned int addr) static irqreturn_t xgbe_i2c_combined_isr(struct xgbe_prv_data *pdata) { - xgbe_i2c_isr_task((unsigned long)pdata); + xgbe_i2c_isr_task(&pdata->tasklet_i2c); return IRQ_HANDLED; } @@ -462,8 +462,7 @@ static int xgbe_i2c_start(struct xgbe_prv_data *pdata) /* If we have a separate I2C irq, enable it */ if (pdata->dev_irq != pdata->i2c_irq) { - tasklet_init(&pdata->tasklet_i2c, xgbe_i2c_isr_task, - (unsigned long)pdata); + tasklet_setup(&pdata->tasklet_i2c, xgbe_i2c_isr_task); ret = devm_request_irq(pdata->dev, pdata->i2c_irq, xgbe_i2c_isr, 0, pdata->i2c_name, diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c index 8a3a60bb2688..93ef5a30cb8d 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c @@ -688,9 +688,9 @@ static void xgbe_an73_isr(struct xgbe_prv_data *pdata) } } -static void xgbe_an_isr_task(unsigned long data) +static void xgbe_an_isr_task(struct tasklet_struct *t) { - struct xgbe_prv_data *pdata = (struct xgbe_prv_data *)data; + struct xgbe_prv_data *pdata = from_tasklet(pdata, t, tasklet_an); netif_dbg(pdata, intr, pdata->netdev, "AN interrupt received\n"); @@ -715,14 +715,14 @@ static irqreturn_t xgbe_an_isr(int irq, void *data) if (pdata->isr_as_tasklet) tasklet_schedule(&pdata->tasklet_an); else - xgbe_an_isr_task((unsigned long)pdata); + xgbe_an_isr_task(&pdata->tasklet_an); return IRQ_HANDLED; } static irqreturn_t xgbe_an_combined_isr(struct xgbe_prv_data *pdata) { - xgbe_an_isr_task((unsigned long)pdata); + xgbe_an_isr_task(&pdata->tasklet_an); return IRQ_HANDLED; } @@ -1414,8 +1414,7 @@ static int xgbe_phy_start(struct xgbe_prv_data *pdata) /* If we have a separate AN irq, enable it */ if (pdata->dev_irq != pdata->an_irq) { - tasklet_init(&pdata->tasklet_an, xgbe_an_isr_task, - (unsigned long)pdata); + tasklet_setup(&pdata->tasklet_an, xgbe_an_isr_task); ret = devm_request_irq(pdata->dev, pdata->an_irq, xgbe_an_isr, 0, pdata->an_name, diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h index d35a338120cf..643f5e646740 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h @@ -18,6 +18,7 @@ #include <linux/of_platform.h> #include <linux/of_net.h> #include <linux/of_mdio.h> +#include <linux/mdio/mdio-xgene.h> #include <linux/module.h> #include <net/ip.h> #include <linux/prefetch.h> @@ -26,7 +27,6 @@ #include "xgene_enet_hw.h" #include "xgene_enet_cle.h" #include "xgene_enet_ring2.h" -#include "../../../phy/mdio-xgene.h" #define ETHER_MIN_PACKET 64 #define ETHER_STD_PACKET 1518 diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c index 1ab5314c4c1b..de2a9348bc3f 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c @@ -917,6 +917,57 @@ static int aq_ethtool_set_priv_flags(struct net_device *ndev, u32 flags) return ret; } +static int aq_ethtool_get_phy_tunable(struct net_device *ndev, + const struct ethtool_tunable *tuna, void *data) +{ + struct aq_nic_s *aq_nic = netdev_priv(ndev); + + switch (tuna->id) { + case ETHTOOL_PHY_EDPD: { + u16 *val = data; + + *val = aq_nic->aq_nic_cfg.is_media_detect ? AQ_HW_MEDIA_DETECT_CNT : 0; + break; + } + case ETHTOOL_PHY_DOWNSHIFT: { + u8 *val = data; + + *val = (u8)aq_nic->aq_nic_cfg.downshift_counter; + break; + } + default: + return -EOPNOTSUPP; + } + + return 0; +} + +static int aq_ethtool_set_phy_tunable(struct net_device *ndev, + const struct ethtool_tunable *tuna, const void *data) +{ + int err = -EOPNOTSUPP; + struct aq_nic_s *aq_nic = netdev_priv(ndev); + + switch (tuna->id) { + case ETHTOOL_PHY_EDPD: { + const u16 *val = data; + + err = aq_nic_set_media_detect(aq_nic, *val); + break; + } + case ETHTOOL_PHY_DOWNSHIFT: { + const u8 *val = data; + + err = aq_nic_set_downshift(aq_nic, *val); + break; + } + default: + break; + } + + return err; +} + const struct ethtool_ops aq_ethtool_ops = { .supported_coalesce_params = ETHTOOL_COALESCE_USECS | ETHTOOL_COALESCE_MAX_FRAMES, @@ -952,4 +1003,6 @@ const struct ethtool_ops aq_ethtool_ops = { .get_coalesce = aq_ethtool_get_coalesce, .set_coalesce = aq_ethtool_set_coalesce, .get_ts_info = aq_ethtool_get_ts_info, + .get_phy_tunable = aq_ethtool_get_phy_tunable, + .set_phy_tunable = aq_ethtool_set_phy_tunable, }; diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h index 7df74015fbc9..bed481816ea3 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h @@ -143,6 +143,8 @@ struct aq_stats_s { #define AQ_HW_LED_BLINK 0x2U #define AQ_HW_LED_DEFAULT 0x0U +#define AQ_HW_MEDIA_DETECT_CNT 6000 + enum aq_priv_flags { AQ_HW_LOOPBACK_DMA_SYS, AQ_HW_LOOPBACK_PKT_SYS, @@ -386,6 +388,10 @@ struct aq_fw_ops { int (*get_eee_rate)(struct aq_hw_s *self, u32 *rate, u32 *supported_rates); + int (*set_downshift)(struct aq_hw_s *self, u32 counter); + + int (*set_media_detect)(struct aq_hw_s *self, bool enable); + u32 (*get_link_capabilities)(struct aq_hw_s *self); int (*send_macsec_req)(struct aq_hw_s *self, diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c index c6bdf1d677d1..0f865daeb36d 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c @@ -405,6 +405,10 @@ int aq_nic_init(struct aq_nic_s *self) mutex_unlock(&self->fwreq_mutex); if (err < 0) goto err_exit; + /* Restore default settings */ + aq_nic_set_downshift(self, self->aq_nic_cfg.downshift_counter); + aq_nic_set_media_detect(self, self->aq_nic_cfg.is_media_detect ? + AQ_HW_MEDIA_DETECT_CNT : 0); err = self->aq_hw_ops->hw_init(self->aq_hw, aq_nic_get_ndev(self)->dev_addr); @@ -1398,6 +1402,52 @@ void aq_nic_release_filter(struct aq_nic_s *self, enum aq_rx_filter_type type, } } +int aq_nic_set_downshift(struct aq_nic_s *self, int val) +{ + int err = 0; + struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg; + + if (!self->aq_fw_ops->set_downshift) + return -EOPNOTSUPP; + + if (val > 15) { + netdev_err(self->ndev, "downshift counter should be <= 15\n"); + return -EINVAL; + } + cfg->downshift_counter = val; + + mutex_lock(&self->fwreq_mutex); + err = self->aq_fw_ops->set_downshift(self->aq_hw, cfg->downshift_counter); + mutex_unlock(&self->fwreq_mutex); + + return err; +} + +int aq_nic_set_media_detect(struct aq_nic_s *self, int val) +{ + struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg; + int err = 0; + + if (!self->aq_fw_ops->set_media_detect) + return -EOPNOTSUPP; + + if (val > 0 && val != AQ_HW_MEDIA_DETECT_CNT) { + netdev_err(self->ndev, "EDPD on this device could have only fixed value of %d\n", + AQ_HW_MEDIA_DETECT_CNT); + return -EINVAL; + } + + mutex_lock(&self->fwreq_mutex); + err = self->aq_fw_ops->set_media_detect(self->aq_hw, !!val); + mutex_unlock(&self->fwreq_mutex); + + /* msecs plays no role - configuration is always fixed in PHY */ + if (!err) + cfg->is_media_detect = !!val; + + return err; +} + int aq_nic_setup_tc_mqprio(struct aq_nic_s *self, u32 tcs, u8 *prio_tc_map) { struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg; diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.h b/drivers/net/ethernet/aquantia/atlantic/aq_nic.h index eb7d8430f2f5..926cca9a0c83 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.h @@ -62,6 +62,8 @@ struct aq_nic_cfg_s { bool is_lro; bool is_qos; bool is_ptp; + bool is_media_detect; + int downshift_counter; enum aq_tc_mode tc_mode; u32 priv_flags; u8 tcs; @@ -195,6 +197,8 @@ int aq_nic_set_link_ksettings(struct aq_nic_s *self, struct aq_nic_cfg_s *aq_nic_get_cfg(struct aq_nic_s *self); u32 aq_nic_get_fw_version(struct aq_nic_s *self); int aq_nic_set_loopback(struct aq_nic_s *self); +int aq_nic_set_downshift(struct aq_nic_s *self, int val); +int aq_nic_set_media_detect(struct aq_nic_s *self, int val); int aq_nic_update_interrupt_moderation_settings(struct aq_nic_s *self); void aq_nic_shutdown(struct aq_nic_s *self); u8 aq_nic_reserve_filter(struct aq_nic_s *self, enum aq_rx_filter_type type); diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c index 8941ac4df9e3..9f1b15077e7d 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c @@ -1536,7 +1536,7 @@ static int hw_atl_b0_hw_fl2_clear(struct aq_hw_s *self, return aq_hw_err_from_flags(self); } -/** +/* * @brief Set VLAN filter table * @details Configure VLAN filter table to accept (and assign the queue) traffic * for the particular vlan ids. diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c index 93c06dfa6c55..ee0c22d04935 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c @@ -612,6 +612,41 @@ static u32 aq_fw2x_state2_get(struct aq_hw_s *self) return aq_hw_read_reg(self, HW_ATL_FW2X_MPI_STATE2_ADDR); } +static int aq_fw2x_set_downshift(struct aq_hw_s *self, u32 counter) +{ + int err = 0; + u32 mpi_opts; + u32 offset; + + offset = offsetof(struct hw_atl_utils_settings, downshift_retry_count); + err = hw_atl_write_fwsettings_dwords(self, offset, &counter, 1); + if (err) + return err; + + mpi_opts = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR); + if (counter) + mpi_opts |= HW_ATL_FW2X_CTRL_DOWNSHIFT; + else + mpi_opts &= ~HW_ATL_FW2X_CTRL_DOWNSHIFT; + aq_hw_write_reg(self, HW_ATL_FW2X_MPI_CONTROL2_ADDR, mpi_opts); + + return err; +} + +static int aq_fw2x_set_media_detect(struct aq_hw_s *self, bool on) +{ + u32 enable; + u32 offset; + + if (self->fw_ver_actual < HW_ATL_FW_VER_MEDIA_CONTROL) + return -EOPNOTSUPP; + + offset = offsetof(struct hw_atl_utils_settings, media_detect); + enable = on; + + return hw_atl_write_fwsettings_dwords(self, offset, &enable, 1); +} + static u32 aq_fw2x_get_link_capabilities(struct aq_hw_s *self) { int err = 0; @@ -692,6 +727,8 @@ const struct aq_fw_ops aq_fw_2x_ops = { .enable_ptp = aq_fw3x_enable_ptp, .led_control = aq_fw2x_led_control, .set_phyloopback = aq_fw2x_set_phyloopback, + .set_downshift = aq_fw2x_set_downshift, + .set_media_detect = aq_fw2x_set_media_detect, .adjust_ptp = aq_fw3x_adjust_ptp, .get_link_capabilities = aq_fw2x_get_link_capabilities, .send_macsec_req = aq_fw2x_send_macsec_req, diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils_fw.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils_fw.c index 85628acbcc1d..dd259c8f2f4f 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils_fw.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils_fw.c @@ -519,6 +519,18 @@ int hw_atl2_utils_get_action_resolve_table_caps(struct aq_hw_s *self, return 0; } +static int aq_a2_fw_set_downshift(struct aq_hw_s *self, u32 counter) +{ + struct link_options_s link_options; + + hw_atl2_shared_buffer_get(self, link_options, link_options); + link_options.downshift = !!counter; + link_options.downshift_retry = counter; + hw_atl2_shared_buffer_write(self, link_options, link_options); + + return hw_atl2_shared_buffer_finish_ack(self); +} + const struct aq_fw_ops aq_a2_fw_ops = { .init = aq_a2_fw_init, .deinit = aq_a2_fw_deinit, @@ -536,4 +548,5 @@ const struct aq_fw_ops aq_a2_fw_ops = { .set_flow_control = aq_a2_fw_set_flow_control, .get_flow_control = aq_a2_fw_get_flow_control, .set_phyloopback = aq_a2_fw_set_phyloopback, + .set_downshift = aq_a2_fw_set_downshift, }; diff --git a/drivers/net/ethernet/arc/emac_arc.c b/drivers/net/ethernet/arc/emac_arc.c index 1c7736b7eaf7..800620b8f10d 100644 --- a/drivers/net/ethernet/arc/emac_arc.c +++ b/drivers/net/ethernet/arc/emac_arc.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0-or-later /** - * emac_arc.c - ARC EMAC specific glue layer + * DOC: emac_arc.c - ARC EMAC specific glue layer * * Copyright (C) 2014 Romain Perier * diff --git a/drivers/net/ethernet/atheros/ag71xx.c b/drivers/net/ethernet/atheros/ag71xx.c index 38cce66ef212..dd5c8a9038bb 100644 --- a/drivers/net/ethernet/atheros/ag71xx.c +++ b/drivers/net/ethernet/atheros/ag71xx.c @@ -235,6 +235,59 @@ | NETIF_MSG_RX_ERR \ | NETIF_MSG_TX_ERR) +struct ag71xx_statistic { + unsigned short offset; + u32 mask; + const char name[ETH_GSTRING_LEN]; +}; + +static const struct ag71xx_statistic ag71xx_statistics[] = { + { 0x0080, GENMASK(17, 0), "Tx/Rx 64 Byte", }, + { 0x0084, GENMASK(17, 0), "Tx/Rx 65-127 Byte", }, + { 0x0088, GENMASK(17, 0), "Tx/Rx 128-255 Byte", }, + { 0x008C, GENMASK(17, 0), "Tx/Rx 256-511 Byte", }, + { 0x0090, GENMASK(17, 0), "Tx/Rx 512-1023 Byte", }, + { 0x0094, GENMASK(17, 0), "Tx/Rx 1024-1518 Byte", }, + { 0x0098, GENMASK(17, 0), "Tx/Rx 1519-1522 Byte VLAN", }, + { 0x009C, GENMASK(23, 0), "Rx Byte", }, + { 0x00A0, GENMASK(17, 0), "Rx Packet", }, + { 0x00A4, GENMASK(11, 0), "Rx FCS Error", }, + { 0x00A8, GENMASK(17, 0), "Rx Multicast Packet", }, + { 0x00AC, GENMASK(21, 0), "Rx Broadcast Packet", }, + { 0x00B0, GENMASK(17, 0), "Rx Control Frame Packet", }, + { 0x00B4, GENMASK(11, 0), "Rx Pause Frame Packet", }, + { 0x00B8, GENMASK(11, 0), "Rx Unknown OPCode Packet", }, + { 0x00BC, GENMASK(11, 0), "Rx Alignment Error", }, + { 0x00C0, GENMASK(15, 0), "Rx Frame Length Error", }, + { 0x00C4, GENMASK(11, 0), "Rx Code Error", }, + { 0x00C8, GENMASK(11, 0), "Rx Carrier Sense Error", }, + { 0x00CC, GENMASK(11, 0), "Rx Undersize Packet", }, + { 0x00D0, GENMASK(11, 0), "Rx Oversize Packet", }, + { 0x00D4, GENMASK(11, 0), "Rx Fragments", }, + { 0x00D8, GENMASK(11, 0), "Rx Jabber", }, + { 0x00DC, GENMASK(11, 0), "Rx Dropped Packet", }, + { 0x00E0, GENMASK(23, 0), "Tx Byte", }, + { 0x00E4, GENMASK(17, 0), "Tx Packet", }, + { 0x00E8, GENMASK(17, 0), "Tx Multicast Packet", }, + { 0x00EC, GENMASK(17, 0), "Tx Broadcast Packet", }, + { 0x00F0, GENMASK(11, 0), "Tx Pause Control Frame", }, + { 0x00F4, GENMASK(11, 0), "Tx Deferral Packet", }, + { 0x00F8, GENMASK(11, 0), "Tx Excessive Deferral Packet", }, + { 0x00FC, GENMASK(11, 0), "Tx Single Collision Packet", }, + { 0x0100, GENMASK(11, 0), "Tx Multiple Collision", }, + { 0x0104, GENMASK(11, 0), "Tx Late Collision Packet", }, + { 0x0108, GENMASK(11, 0), "Tx Excessive Collision Packet", }, + { 0x010C, GENMASK(12, 0), "Tx Total Collision", }, + { 0x0110, GENMASK(11, 0), "Tx Pause Frames Honored", }, + { 0x0114, GENMASK(11, 0), "Tx Drop Frame", }, + { 0x0118, GENMASK(11, 0), "Tx Jabber Frame", }, + { 0x011C, GENMASK(11, 0), "Tx FCS Error", }, + { 0x0120, GENMASK(11, 0), "Tx Control Frame", }, + { 0x0124, GENMASK(11, 0), "Tx Oversize Frame", }, + { 0x0128, GENMASK(11, 0), "Tx Undersize Frame", }, + { 0x012C, GENMASK(11, 0), "Tx Fragment", }, +}; + #define DESC_EMPTY BIT(31) #define DESC_MORE BIT(24) #define DESC_PKTLEN_M 0xfff @@ -394,6 +447,99 @@ static void ag71xx_int_disable(struct ag71xx *ag, u32 ints) ag71xx_cb(ag, AG71XX_REG_INT_ENABLE, ints); } +static void ag71xx_get_drvinfo(struct net_device *ndev, + struct ethtool_drvinfo *info) +{ + struct ag71xx *ag = netdev_priv(ndev); + + strlcpy(info->driver, "ag71xx", sizeof(info->driver)); + strlcpy(info->bus_info, of_node_full_name(ag->pdev->dev.of_node), + sizeof(info->bus_info)); +} + +static int ag71xx_get_link_ksettings(struct net_device *ndev, + struct ethtool_link_ksettings *kset) +{ + struct ag71xx *ag = netdev_priv(ndev); + + return phylink_ethtool_ksettings_get(ag->phylink, kset); +} + +static int ag71xx_set_link_ksettings(struct net_device *ndev, + const struct ethtool_link_ksettings *kset) +{ + struct ag71xx *ag = netdev_priv(ndev); + + return phylink_ethtool_ksettings_set(ag->phylink, kset); +} + +static int ag71xx_ethtool_nway_reset(struct net_device *ndev) +{ + struct ag71xx *ag = netdev_priv(ndev); + + return phylink_ethtool_nway_reset(ag->phylink); +} + +static void ag71xx_ethtool_get_pauseparam(struct net_device *ndev, + struct ethtool_pauseparam *pause) +{ + struct ag71xx *ag = netdev_priv(ndev); + + phylink_ethtool_get_pauseparam(ag->phylink, pause); +} + +static int ag71xx_ethtool_set_pauseparam(struct net_device *ndev, + struct ethtool_pauseparam *pause) +{ + struct ag71xx *ag = netdev_priv(ndev); + + return phylink_ethtool_set_pauseparam(ag->phylink, pause); +} + +static void ag71xx_ethtool_get_strings(struct net_device *netdev, u32 sset, + u8 *data) +{ + if (sset == ETH_SS_STATS) { + int i; + + for (i = 0; i < ARRAY_SIZE(ag71xx_statistics); i++) + memcpy(data + i * ETH_GSTRING_LEN, + ag71xx_statistics[i].name, ETH_GSTRING_LEN); + } +} + +static void ag71xx_ethtool_get_stats(struct net_device *ndev, + struct ethtool_stats *stats, u64 *data) +{ + struct ag71xx *ag = netdev_priv(ndev); + int i; + + for (i = 0; i < ARRAY_SIZE(ag71xx_statistics); i++) + *data++ = ag71xx_rr(ag, ag71xx_statistics[i].offset) + & ag71xx_statistics[i].mask; +} + +static int ag71xx_ethtool_get_sset_count(struct net_device *ndev, int sset) +{ + if (sset == ETH_SS_STATS) + return ARRAY_SIZE(ag71xx_statistics); + return -EOPNOTSUPP; +} + +static const struct ethtool_ops ag71xx_ethtool_ops = { + .get_drvinfo = ag71xx_get_drvinfo, + .get_link = ethtool_op_get_link, + .get_ts_info = ethtool_op_get_ts_info, + .get_link_ksettings = ag71xx_get_link_ksettings, + .set_link_ksettings = ag71xx_set_link_ksettings, + .nway_reset = ag71xx_ethtool_nway_reset, + .get_pauseparam = ag71xx_ethtool_get_pauseparam, + .set_pauseparam = ag71xx_ethtool_set_pauseparam, + .get_strings = ag71xx_ethtool_get_strings, + .get_ethtool_stats = ag71xx_ethtool_get_stats, + .get_sset_count = ag71xx_ethtool_get_sset_count, +}; + static int ag71xx_mdio_wait_busy(struct ag71xx *ag) { struct net_device *ndev = ag->ndev; @@ -910,6 +1056,8 @@ static void ag71xx_mac_validate(struct phylink_config *config, phylink_set(mask, MII); + phylink_set(mask, Pause); + phylink_set(mask, Asym_Pause); phylink_set(mask, Autoneg); phylink_set(mask, 10baseT_Half); phylink_set(mask, 10baseT_Full); @@ -960,7 +1108,7 @@ static void ag71xx_mac_link_up(struct phylink_config *config, bool tx_pause, bool rx_pause) { struct ag71xx *ag = netdev_priv(to_net_dev(config->dev)); - u32 cfg2; + u32 cfg1, cfg2; u32 ifctl; u32 fifo5; @@ -994,6 +1142,15 @@ static void ag71xx_mac_link_up(struct phylink_config *config, ag71xx_wr(ag, AG71XX_REG_FIFO_CFG5, fifo5); ag71xx_wr(ag, AG71XX_REG_MAC_IFCTL, ifctl); + cfg1 = ag71xx_rr(ag, AG71XX_REG_MAC_CFG1); + cfg1 &= ~(MAC_CFG1_TFC | MAC_CFG1_RFC); + if (tx_pause) + cfg1 |= MAC_CFG1_TFC; + + if (rx_pause) + cfg1 |= MAC_CFG1_RFC; + ag71xx_wr(ag, AG71XX_REG_MAC_CFG1, cfg1); + ag71xx_hw_start(ag); } @@ -1769,6 +1926,7 @@ static int ag71xx_probe(struct platform_device *pdev) } ndev->netdev_ops = &ag71xx_netdev_ops; + ndev->ethtool_ops = &ag71xx_ethtool_ops; INIT_DELAYED_WORK(&ag->restart_work, ag71xx_restart_work_func); timer_setup(&ag->oom_timer, ag71xx_oom_timer_handler, 0); diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c index decab9a8e4a8..0c12cf7bda50 100644 --- a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c +++ b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c @@ -204,7 +204,7 @@ static u32 atl1c_wait_until_idle(struct atl1c_hw *hw, u32 modu_ctrl) /** * atl1c_phy_config - Timer Call-back - * @data: pointer to netdev cast into an unsigned long + * @t: timer list containing pointer to netdev cast into an unsigned long */ static void atl1c_phy_config(struct timer_list *t) { @@ -220,7 +220,6 @@ static void atl1c_phy_config(struct timer_list *t) void atl1c_reinit_locked(struct atl1c_adapter *adapter) { - WARN_ON(in_interrupt()); atl1c_down(adapter); atl1c_up(adapter); clear_bit(__AT_RESETTING, &adapter->flags); @@ -346,6 +345,7 @@ static void atl1c_del_timer(struct atl1c_adapter *adapter) /** * atl1c_tx_timeout - Respond to a Tx Hang * @netdev: network interface device structure + * @txqueue: index of hanging tx queue */ static void atl1c_tx_timeout(struct net_device *netdev, unsigned int txqueue) { @@ -826,16 +826,16 @@ static inline void atl1c_clean_buffer(struct pci_dev *pdev, return; if (buffer_info->dma) { if (buffer_info->flags & ATL1C_PCIMAP_FROMDEVICE) - pci_driection = PCI_DMA_FROMDEVICE; + pci_driection = DMA_FROM_DEVICE; else - pci_driection = PCI_DMA_TODEVICE; + pci_driection = DMA_TO_DEVICE; if (buffer_info->flags & ATL1C_PCIMAP_SINGLE) - pci_unmap_single(pdev, buffer_info->dma, - buffer_info->length, pci_driection); + dma_unmap_single(&pdev->dev, buffer_info->dma, + buffer_info->length, pci_driection); else if (buffer_info->flags & ATL1C_PCIMAP_PAGE) - pci_unmap_page(pdev, buffer_info->dma, - buffer_info->length, pci_driection); + dma_unmap_page(&pdev->dev, buffer_info->dma, + buffer_info->length, pci_driection); } if (buffer_info->skb) dev_consume_skb_any(buffer_info->skb); @@ -846,6 +846,7 @@ static inline void atl1c_clean_buffer(struct pci_dev *pdev, /** * atl1c_clean_tx_ring - Free Tx-skb * @adapter: board private structure + * @type: type of transmit queue */ static void atl1c_clean_tx_ring(struct atl1c_adapter *adapter, enum atl1c_trans_queue type) @@ -933,9 +934,8 @@ static void atl1c_free_ring_resources(struct atl1c_adapter *adapter) { struct pci_dev *pdev = adapter->pdev; - pci_free_consistent(pdev, adapter->ring_header.size, - adapter->ring_header.desc, - adapter->ring_header.dma); + dma_free_coherent(&pdev->dev, adapter->ring_header.size, + adapter->ring_header.desc, adapter->ring_header.dma); adapter->ring_header.desc = NULL; /* Note: just free tdp_ring.buffer_info, @@ -1717,10 +1717,9 @@ static int atl1c_alloc_rx_buffer(struct atl1c_adapter *adapter) ATL1C_SET_BUFFER_STATE(buffer_info, ATL1C_BUFFER_BUSY); buffer_info->skb = skb; buffer_info->length = adapter->rx_buffer_len; - mapping = pci_map_single(pdev, vir_addr, - buffer_info->length, - PCI_DMA_FROMDEVICE); - if (unlikely(pci_dma_mapping_error(pdev, mapping))) { + mapping = dma_map_single(&pdev->dev, vir_addr, + buffer_info->length, DMA_FROM_DEVICE); + if (unlikely(dma_mapping_error(&pdev->dev, mapping))) { dev_kfree_skb(skb); buffer_info->skb = NULL; buffer_info->length = 0; @@ -1831,8 +1830,8 @@ rrs_checked: rfd_index = (rrs->word0 >> RRS_RX_RFD_INDEX_SHIFT) & RRS_RX_RFD_INDEX_MASK; buffer_info = &rfd_ring->buffer_info[rfd_index]; - pci_unmap_single(pdev, buffer_info->dma, - buffer_info->length, PCI_DMA_FROMDEVICE); + dma_unmap_single(&pdev->dev, buffer_info->dma, + buffer_info->length, DMA_FROM_DEVICE); skb = buffer_info->skb; } else { /* TODO */ @@ -1863,6 +1862,8 @@ rrs_checked: /** * atl1c_clean - NAPI Rx polling callback + * @napi: napi info + * @budget: limit of packets to clean */ static int atl1c_clean(struct napi_struct *napi, int budget) { @@ -2106,10 +2107,10 @@ static int atl1c_tx_map(struct atl1c_adapter *adapter, buffer_info = atl1c_get_tx_buffer(adapter, use_tpd); buffer_info->length = map_len; - buffer_info->dma = pci_map_single(adapter->pdev, - skb->data, hdr_len, PCI_DMA_TODEVICE); - if (unlikely(pci_dma_mapping_error(adapter->pdev, - buffer_info->dma))) + buffer_info->dma = dma_map_single(&adapter->pdev->dev, + skb->data, hdr_len, + DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(&adapter->pdev->dev, buffer_info->dma))) goto err_dma; ATL1C_SET_BUFFER_STATE(buffer_info, ATL1C_BUFFER_BUSY); ATL1C_SET_PCIMAP_TYPE(buffer_info, ATL1C_PCIMAP_SINGLE, @@ -2131,10 +2132,10 @@ static int atl1c_tx_map(struct atl1c_adapter *adapter, buffer_info = atl1c_get_tx_buffer(adapter, use_tpd); buffer_info->length = buf_len - mapped_len; buffer_info->dma = - pci_map_single(adapter->pdev, skb->data + mapped_len, - buffer_info->length, PCI_DMA_TODEVICE); - if (unlikely(pci_dma_mapping_error(adapter->pdev, - buffer_info->dma))) + dma_map_single(&adapter->pdev->dev, + skb->data + mapped_len, + buffer_info->length, DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(&adapter->pdev->dev, buffer_info->dma))) goto err_dma; ATL1C_SET_BUFFER_STATE(buffer_info, ATL1C_BUFFER_BUSY); @@ -2542,8 +2543,8 @@ static int atl1c_probe(struct pci_dev *pdev, const struct pci_device_id *ent) * various kernel subsystems to support the mechanics required by a * fixed-high-32-bit system. */ - if ((pci_set_dma_mask(pdev, DMA_BIT_MASK(32)) != 0) || - (pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32)) != 0)) { + if ((dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)) != 0) || + (dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32)) != 0)) { dev_err(&pdev->dev, "No usable DMA configuration,aborting\n"); goto err_dma; } diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c index 223ef846123e..098b0328e3cb 100644 --- a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c +++ b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c @@ -111,7 +111,7 @@ static inline void atl1e_irq_reset(struct atl1e_adapter *adapter) /** * atl1e_phy_config - Timer Call-back - * @data: pointer to netdev cast into an unsigned long + * @t: timer list containing pointer to netdev cast into an unsigned long */ static void atl1e_phy_config(struct timer_list *t) { @@ -127,8 +127,6 @@ static void atl1e_phy_config(struct timer_list *t) void atl1e_reinit_locked(struct atl1e_adapter *adapter) { - - WARN_ON(in_interrupt()); while (test_and_set_bit(__AT_RESETTING, &adapter->flags)) msleep(1); atl1e_down(adapter); @@ -196,7 +194,7 @@ static int atl1e_check_link(struct atl1e_adapter *adapter) /** * atl1e_link_chg_task - deal with link change event Out of interrupt context - * @netdev: network interface device structure + * @work: work struct with driver info */ static void atl1e_link_chg_task(struct work_struct *work) { @@ -246,6 +244,7 @@ static void atl1e_cancel_work(struct atl1e_adapter *adapter) /** * atl1e_tx_timeout - Respond to a Tx Hang * @netdev: network interface device structure + * @txqueue: the index of the hanging queue */ static void atl1e_tx_timeout(struct net_device *netdev, unsigned int txqueue) { @@ -654,11 +653,13 @@ static void atl1e_clean_tx_ring(struct atl1e_adapter *adapter) tx_buffer = &tx_ring->tx_buffer[index]; if (tx_buffer->dma) { if (tx_buffer->flags & ATL1E_TX_PCIMAP_SINGLE) - pci_unmap_single(pdev, tx_buffer->dma, - tx_buffer->length, PCI_DMA_TODEVICE); + dma_unmap_single(&pdev->dev, tx_buffer->dma, + tx_buffer->length, + DMA_TO_DEVICE); else if (tx_buffer->flags & ATL1E_TX_PCIMAP_PAGE) - pci_unmap_page(pdev, tx_buffer->dma, - tx_buffer->length, PCI_DMA_TODEVICE); + dma_unmap_page(&pdev->dev, tx_buffer->dma, + tx_buffer->length, + DMA_TO_DEVICE); tx_buffer->dma = 0; } } @@ -774,8 +775,8 @@ static void atl1e_free_ring_resources(struct atl1e_adapter *adapter) atl1e_clean_rx_ring(adapter); if (adapter->ring_vir_addr) { - pci_free_consistent(pdev, adapter->ring_size, - adapter->ring_vir_addr, adapter->ring_dma); + dma_free_coherent(&pdev->dev, adapter->ring_size, + adapter->ring_vir_addr, adapter->ring_dma); adapter->ring_vir_addr = NULL; } @@ -810,11 +811,12 @@ static int atl1e_setup_ring_resources(struct atl1e_adapter *adapter) /* real ring DMA buffer */ size = adapter->ring_size; - adapter->ring_vir_addr = pci_zalloc_consistent(pdev, adapter->ring_size, - &adapter->ring_dma); + adapter->ring_vir_addr = dma_alloc_coherent(&pdev->dev, + adapter->ring_size, + &adapter->ring_dma, GFP_KERNEL); if (adapter->ring_vir_addr == NULL) { netdev_err(adapter->netdev, - "pci_alloc_consistent failed, size = D%d\n", size); + "dma_alloc_coherent failed, size = D%d\n", size); return -ENOMEM; } @@ -870,8 +872,8 @@ static int atl1e_setup_ring_resources(struct atl1e_adapter *adapter) return 0; failed: if (adapter->ring_vir_addr != NULL) { - pci_free_consistent(pdev, adapter->ring_size, - adapter->ring_vir_addr, adapter->ring_dma); + dma_free_coherent(&pdev->dev, adapter->ring_size, + adapter->ring_vir_addr, adapter->ring_dma); adapter->ring_vir_addr = NULL; } return err; @@ -1233,11 +1235,15 @@ static bool atl1e_clean_tx_irq(struct atl1e_adapter *adapter) tx_buffer = &tx_ring->tx_buffer[next_to_clean]; if (tx_buffer->dma) { if (tx_buffer->flags & ATL1E_TX_PCIMAP_SINGLE) - pci_unmap_single(adapter->pdev, tx_buffer->dma, - tx_buffer->length, PCI_DMA_TODEVICE); + dma_unmap_single(&adapter->pdev->dev, + tx_buffer->dma, + tx_buffer->length, + DMA_TO_DEVICE); else if (tx_buffer->flags & ATL1E_TX_PCIMAP_PAGE) - pci_unmap_page(adapter->pdev, tx_buffer->dma, - tx_buffer->length, PCI_DMA_TODEVICE); + dma_unmap_page(&adapter->pdev->dev, + tx_buffer->dma, + tx_buffer->length, + DMA_TO_DEVICE); tx_buffer->dma = 0; } @@ -1495,6 +1501,8 @@ fatal_err: /** * atl1e_clean - NAPI Rx polling callback + * @napi: napi info + * @budget: number of packets to clean */ static int atl1e_clean(struct napi_struct *napi, int budget) { @@ -1710,8 +1718,9 @@ static int atl1e_tx_map(struct atl1e_adapter *adapter, tx_buffer = atl1e_get_tx_buffer(adapter, use_tpd); tx_buffer->length = map_len; - tx_buffer->dma = pci_map_single(adapter->pdev, - skb->data, hdr_len, PCI_DMA_TODEVICE); + tx_buffer->dma = dma_map_single(&adapter->pdev->dev, + skb->data, hdr_len, + DMA_TO_DEVICE); if (dma_mapping_error(&adapter->pdev->dev, tx_buffer->dma)) return -ENOSPC; @@ -1739,8 +1748,9 @@ static int atl1e_tx_map(struct atl1e_adapter *adapter, ((buf_len - mapped_len) >= MAX_TX_BUF_LEN) ? MAX_TX_BUF_LEN : (buf_len - mapped_len); tx_buffer->dma = - pci_map_single(adapter->pdev, skb->data + mapped_len, - map_len, PCI_DMA_TODEVICE); + dma_map_single(&adapter->pdev->dev, + skb->data + mapped_len, map_len, + DMA_TO_DEVICE); if (dma_mapping_error(&adapter->pdev->dev, tx_buffer->dma)) { /* We need to unwind the mappings we've done */ @@ -1749,8 +1759,10 @@ static int atl1e_tx_map(struct atl1e_adapter *adapter, while (adapter->tx_ring.next_to_use != ring_end) { tpd = atl1e_get_tpd(adapter); tx_buffer = atl1e_get_tx_buffer(adapter, tpd); - pci_unmap_single(adapter->pdev, tx_buffer->dma, - tx_buffer->length, PCI_DMA_TODEVICE); + dma_unmap_single(&adapter->pdev->dev, + tx_buffer->dma, + tx_buffer->length, + DMA_TO_DEVICE); } /* Reset the tx rings next pointer */ adapter->tx_ring.next_to_use = ring_start; @@ -2300,8 +2312,8 @@ static int atl1e_probe(struct pci_dev *pdev, const struct pci_device_id *ent) * various kernel subsystems to support the mechanics required by a * fixed-high-32-bit system. */ - if ((pci_set_dma_mask(pdev, DMA_BIT_MASK(32)) != 0) || - (pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32)) != 0)) { + if ((dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)) != 0) || + (dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32)) != 0)) { dev_err(&pdev->dev, "No usable DMA configuration,aborting\n"); goto err_dma; } diff --git a/drivers/net/ethernet/atheros/atlx/atl1.c b/drivers/net/ethernet/atheros/atlx/atl1.c index b35fcfcd692d..eaf96d002fa5 100644 --- a/drivers/net/ethernet/atheros/atlx/atl1.c +++ b/drivers/net/ethernet/atheros/atlx/atl1.c @@ -1050,11 +1050,11 @@ static s32 atl1_setup_ring_resources(struct atl1_adapter *adapter) + sizeof(struct stats_msg_block) + 40; - ring_header->desc = pci_alloc_consistent(pdev, ring_header->size, - &ring_header->dma); + ring_header->desc = dma_alloc_coherent(&pdev->dev, ring_header->size, + &ring_header->dma, GFP_KERNEL); if (unlikely(!ring_header->desc)) { if (netif_msg_drv(adapter)) - dev_err(&pdev->dev, "pci_alloc_consistent failed\n"); + dev_err(&pdev->dev, "dma_alloc_coherent failed\n"); goto err_nomem; } @@ -1136,8 +1136,8 @@ static void atl1_clean_rx_ring(struct atl1_adapter *adapter) for (i = 0; i < rfd_ring->count; i++) { buffer_info = &rfd_ring->buffer_info[i]; if (buffer_info->dma) { - pci_unmap_page(pdev, buffer_info->dma, - buffer_info->length, PCI_DMA_FROMDEVICE); + dma_unmap_page(&pdev->dev, buffer_info->dma, + buffer_info->length, DMA_FROM_DEVICE); buffer_info->dma = 0; } if (buffer_info->skb) { @@ -1175,8 +1175,8 @@ static void atl1_clean_tx_ring(struct atl1_adapter *adapter) for (i = 0; i < tpd_ring->count; i++) { buffer_info = &tpd_ring->buffer_info[i]; if (buffer_info->dma) { - pci_unmap_page(pdev, buffer_info->dma, - buffer_info->length, PCI_DMA_TODEVICE); + dma_unmap_page(&pdev->dev, buffer_info->dma, + buffer_info->length, DMA_TO_DEVICE); buffer_info->dma = 0; } } @@ -1217,8 +1217,8 @@ static void atl1_free_ring_resources(struct atl1_adapter *adapter) atl1_clean_rx_ring(adapter); kfree(tpd_ring->buffer_info); - pci_free_consistent(pdev, ring_header->size, ring_header->desc, - ring_header->dma); + dma_free_coherent(&pdev->dev, ring_header->size, ring_header->desc, + ring_header->dma); tpd_ring->buffer_info = NULL; tpd_ring->desc = NULL; @@ -1866,9 +1866,9 @@ static u16 atl1_alloc_rx_buffers(struct atl1_adapter *adapter) buffer_info->length = (u16) adapter->rx_buffer_len; page = virt_to_page(skb->data); offset = offset_in_page(skb->data); - buffer_info->dma = pci_map_page(pdev, page, offset, + buffer_info->dma = dma_map_page(&pdev->dev, page, offset, adapter->rx_buffer_len, - PCI_DMA_FROMDEVICE); + DMA_FROM_DEVICE); rfd_desc->buffer_addr = cpu_to_le64(buffer_info->dma); rfd_desc->buf_len = cpu_to_le16(adapter->rx_buffer_len); rfd_desc->coalese = 0; @@ -1992,8 +1992,8 @@ rrd_ok: } /* Good Receive */ - pci_unmap_page(adapter->pdev, buffer_info->dma, - buffer_info->length, PCI_DMA_FROMDEVICE); + dma_unmap_page(&adapter->pdev->dev, buffer_info->dma, + buffer_info->length, DMA_FROM_DEVICE); buffer_info->dma = 0; skb = buffer_info->skb; length = le16_to_cpu(rrd->xsz.xsum_sz.pkt_size); @@ -2062,8 +2062,8 @@ static int atl1_intr_tx(struct atl1_adapter *adapter) while (cmb_tpd_next_to_clean != sw_tpd_next_to_clean) { buffer_info = &tpd_ring->buffer_info[sw_tpd_next_to_clean]; if (buffer_info->dma) { - pci_unmap_page(adapter->pdev, buffer_info->dma, - buffer_info->length, PCI_DMA_TODEVICE); + dma_unmap_page(&adapter->pdev->dev, buffer_info->dma, + buffer_info->length, DMA_TO_DEVICE); buffer_info->dma = 0; } @@ -2210,9 +2210,9 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb, buffer_info->length = hdr_len; page = virt_to_page(skb->data); offset = offset_in_page(skb->data); - buffer_info->dma = pci_map_page(adapter->pdev, page, + buffer_info->dma = dma_map_page(&adapter->pdev->dev, page, offset, hdr_len, - PCI_DMA_TODEVICE); + DMA_TO_DEVICE); if (++next_to_use == tpd_ring->count) next_to_use = 0; @@ -2235,9 +2235,10 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb, (hdr_len + i * ATL1_MAX_TX_BUF_LEN)); offset = offset_in_page(skb->data + (hdr_len + i * ATL1_MAX_TX_BUF_LEN)); - buffer_info->dma = pci_map_page(adapter->pdev, - page, offset, buffer_info->length, - PCI_DMA_TODEVICE); + buffer_info->dma = dma_map_page(&adapter->pdev->dev, + page, offset, + buffer_info->length, + DMA_TO_DEVICE); if (++next_to_use == tpd_ring->count) next_to_use = 0; } @@ -2247,8 +2248,9 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb, buffer_info->length = buf_len; page = virt_to_page(skb->data); offset = offset_in_page(skb->data); - buffer_info->dma = pci_map_page(adapter->pdev, page, - offset, buf_len, PCI_DMA_TODEVICE); + buffer_info->dma = dma_map_page(&adapter->pdev->dev, page, + offset, buf_len, + DMA_TO_DEVICE); if (++next_to_use == tpd_ring->count) next_to_use = 0; } @@ -2550,7 +2552,7 @@ static irqreturn_t atl1_intr(int irq, void *data) /** * atl1_phy_config - Timer Call-back - * @data: pointer to netdev cast into an unsigned long + * @t: timer_list containing pointer to netdev cast into an unsigned long */ static void atl1_phy_config(struct timer_list *t) { @@ -2922,7 +2924,7 @@ static int atl1_probe(struct pci_dev *pdev, const struct pci_device_id *ent) * various kernel subsystems to support the mechanics required by a * fixed-high-32-bit system. */ - err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32)); + err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)); if (err) { dev_err(&pdev->dev, "no usable DMA configuration\n"); goto err_dma; diff --git a/drivers/net/ethernet/atheros/atlx/atl2.c b/drivers/net/ethernet/atheros/atlx/atl2.c index c915852b8892..7b80d924632a 100644 --- a/drivers/net/ethernet/atheros/atlx/atl2.c +++ b/drivers/net/ethernet/atheros/atlx/atl2.c @@ -281,8 +281,8 @@ static s32 atl2_setup_ring_resources(struct atl2_adapter *adapter) adapter->txs_ring_size * 4 + 7 + /* dword align */ adapter->rxd_ring_size * 1536 + 127; /* 128bytes align */ - adapter->ring_vir_addr = pci_alloc_consistent(pdev, size, - &adapter->ring_dma); + adapter->ring_vir_addr = dma_alloc_coherent(&pdev->dev, size, + &adapter->ring_dma, GFP_KERNEL); if (!adapter->ring_vir_addr) return -ENOMEM; @@ -663,8 +663,8 @@ static int atl2_request_irq(struct atl2_adapter *adapter) static void atl2_free_ring_resources(struct atl2_adapter *adapter) { struct pci_dev *pdev = adapter->pdev; - pci_free_consistent(pdev, adapter->ring_size, adapter->ring_vir_addr, - adapter->ring_dma); + dma_free_coherent(&pdev->dev, adapter->ring_size, + adapter->ring_vir_addr, adapter->ring_dma); } /** @@ -994,6 +994,7 @@ static int atl2_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) /** * atl2_tx_timeout - Respond to a Tx Hang * @netdev: network interface device structure + * @txqueue: index of the hanging transmit queue */ static void atl2_tx_timeout(struct net_device *netdev, unsigned int txqueue) { @@ -1005,7 +1006,7 @@ static void atl2_tx_timeout(struct net_device *netdev, unsigned int txqueue) /** * atl2_watchdog - Timer Call-back - * @data: pointer to netdev cast into an unsigned long + * @t: timer list containing a pointer to netdev cast into an unsigned long */ static void atl2_watchdog(struct timer_list *t) { @@ -1030,7 +1031,7 @@ static void atl2_watchdog(struct timer_list *t) /** * atl2_phy_config - Timer Call-back - * @data: pointer to netdev cast into an unsigned long + * @t: timer list containing a pointer to netdev cast into an unsigned long */ static void atl2_phy_config(struct timer_list *t) { @@ -1085,7 +1086,6 @@ err_up: static void atl2_reinit_locked(struct atl2_adapter *adapter) { - WARN_ON(in_interrupt()); while (test_and_set_bit(__ATL2_RESETTING, &adapter->flags)) msleep(1); atl2_down(adapter); @@ -1235,6 +1235,7 @@ static int atl2_check_link(struct atl2_adapter *adapter) /** * atl2_link_chg_task - deal with link change event Out of interrupt context + * @work: pointer to work struct with private info */ static void atl2_link_chg_task(struct work_struct *work) { @@ -1328,8 +1329,8 @@ static int atl2_probe(struct pci_dev *pdev, const struct pci_device_id *ent) * until the kernel has the proper infrastructure to support 64-bit DMA * on these devices. */ - if (pci_set_dma_mask(pdev, DMA_BIT_MASK(32)) && - pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32))) { + if (dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)) && + dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32))) { printk(KERN_ERR "atl2: No usable DMA configuration, aborting\n"); err = -EIO; goto err_dma; diff --git a/drivers/net/ethernet/broadcom/b44.c b/drivers/net/ethernet/broadcom/b44.c index 6fb620e25208..74c1778d841e 100644 --- a/drivers/net/ethernet/broadcom/b44.c +++ b/drivers/net/ethernet/broadcom/b44.c @@ -2210,12 +2210,12 @@ static void b44_adjust_link(struct net_device *dev) { struct b44 *bp = netdev_priv(dev); struct phy_device *phydev = dev->phydev; - bool status_changed = 0; + bool status_changed = false; BUG_ON(!phydev); if (bp->old_link != phydev->link) { - status_changed = 1; + status_changed = true; bp->old_link = phydev->link; } @@ -2223,11 +2223,11 @@ static void b44_adjust_link(struct net_device *dev) if (phydev->link) { if ((phydev->duplex == DUPLEX_HALF) && (bp->flags & B44_FLAG_FULL_DUPLEX)) { - status_changed = 1; + status_changed = true; bp->flags &= ~B44_FLAG_FULL_DUPLEX; } else if ((phydev->duplex == DUPLEX_FULL) && !(bp->flags & B44_FLAG_FULL_DUPLEX)) { - status_changed = 1; + status_changed = true; bp->flags |= B44_FLAG_FULL_DUPLEX; } } diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index 0762d5d1a810..0fdd19d99d99 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -20,6 +20,7 @@ #include <linux/phy.h> #include <linux/phy_fixed.h> #include <net/dsa.h> +#include <linux/clk.h> #include <net/ip.h> #include <net/ipv6.h> @@ -186,6 +187,11 @@ static int bcm_sysport_set_features(struct net_device *dev, netdev_features_t features) { struct bcm_sysport_priv *priv = netdev_priv(dev); + int ret; + + ret = clk_prepare_enable(priv->clk); + if (ret) + return ret; /* Read CRC forward */ if (!priv->is_lite) @@ -197,6 +203,8 @@ static int bcm_sysport_set_features(struct net_device *dev, bcm_sysport_set_rx_csum(dev, features); bcm_sysport_set_tx_csum(dev, features); + clk_disable_unprepare(priv->clk); + return 0; } @@ -1940,6 +1948,8 @@ static int bcm_sysport_open(struct net_device *dev) unsigned int i; int ret; + clk_prepare_enable(priv->clk); + /* Reset UniMAC */ umac_reset(priv); @@ -1970,7 +1980,8 @@ static int bcm_sysport_open(struct net_device *dev) 0, priv->phy_interface); if (!phydev) { netdev_err(dev, "could not attach to PHY\n"); - return -ENODEV; + ret = -ENODEV; + goto out_clk_disable; } /* Reset house keeping link status */ @@ -2048,6 +2059,8 @@ out_free_irq0: free_irq(priv->irq0, dev); out_phy_disconnect: phy_disconnect(phydev); +out_clk_disable: + clk_disable_unprepare(priv->clk); return ret; } @@ -2106,6 +2119,8 @@ static int bcm_sysport_stop(struct net_device *dev) /* Disconnect from PHY */ phy_disconnect(dev->phydev); + clk_disable_unprepare(priv->clk); + return 0; } @@ -2487,6 +2502,10 @@ static int bcm_sysport_probe(struct platform_device *pdev) /* Initialize private members */ priv = netdev_priv(dev); + priv->clk = devm_clk_get_optional(&pdev->dev, "sw_sysport"); + if (IS_ERR(priv->clk)) + return PTR_ERR(priv->clk); + /* Allocate number of TX rings */ priv->tx_rings = devm_kcalloc(&pdev->dev, txq, sizeof(struct bcm_sysport_tx_ring), @@ -2566,6 +2585,10 @@ static int bcm_sysport_probe(struct platform_device *pdev) if (!ret) device_set_wakeup_capable(&pdev->dev, 1); + priv->wol_clk = devm_clk_get_optional(&pdev->dev, "sw_sysportwol"); + if (IS_ERR(priv->wol_clk)) + return PTR_ERR(priv->wol_clk); + /* Set the needed headroom once and for all */ BUILD_BUG_ON(sizeof(struct bcm_tsb) != 8); dev->needed_headroom += sizeof(struct bcm_tsb); @@ -2590,6 +2613,8 @@ static int bcm_sysport_probe(struct platform_device *pdev) goto err_deregister_notifier; } + clk_prepare_enable(priv->clk); + priv->rev = topctrl_readl(priv, REV_CNTL) & REV_MASK; dev_info(&pdev->dev, "Broadcom SYSTEMPORT%s " REV_FMT @@ -2598,6 +2623,8 @@ static int bcm_sysport_probe(struct platform_device *pdev) (priv->rev >> 8) & 0xff, priv->rev & 0xff, priv->irq0, priv->irq1, txq, rxq); + clk_disable_unprepare(priv->clk); + return 0; err_deregister_notifier: @@ -2751,8 +2778,12 @@ static int __maybe_unused bcm_sysport_suspend(struct device *d) bcm_sysport_fini_rx_ring(priv); /* Get prepared for Wake-on-LAN */ - if (device_may_wakeup(d) && priv->wolopts) + if (device_may_wakeup(d) && priv->wolopts) { + clk_prepare_enable(priv->wol_clk); ret = bcm_sysport_suspend_to_wol(priv); + } + + clk_disable_unprepare(priv->clk); return ret; } @@ -2767,6 +2798,10 @@ static int __maybe_unused bcm_sysport_resume(struct device *d) if (!netif_running(dev)) return 0; + clk_prepare_enable(priv->clk); + if (priv->wolopts) + clk_disable_unprepare(priv->wol_clk); + umac_reset(priv); /* Disable the UniMAC RX/TX */ @@ -2846,6 +2881,7 @@ out_free_rx_ring: out_free_tx_rings: for (i = 0; i < dev->num_tx_queues; i++) bcm_sysport_fini_tx_ring(priv, i); + clk_disable_unprepare(priv->clk); return ret; } diff --git a/drivers/net/ethernet/broadcom/bcmsysport.h b/drivers/net/ethernet/broadcom/bcmsysport.h index 6d80735fbc7f..3a5cb6f128f5 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.h +++ b/drivers/net/ethernet/broadcom/bcmsysport.h @@ -770,6 +770,8 @@ struct bcm_sysport_priv { u32 wolopts; u8 sopass[SOPASS_MAX]; unsigned int wol_irq_disabled:1; + struct clk *clk; + struct clk *wol_clk; /* MIB related fields */ struct bcm_sysport_mib mib; diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index e3d92e4f2193..1a6ec1a12d53 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -504,6 +504,7 @@ static void bnx2x_tpa_start(struct bnx2x_fastpath *fp, u16 queue, * @len_on_bd: total length of the first packet for the * aggregation. * @pkt_len: length of all segments + * @num_of_coalesced_segs: count of segments * * Approximate value of the MSS for this aggregation calculated using * the first packet of it. @@ -1958,6 +1959,7 @@ void bnx2x_set_num_queues(struct bnx2x *bp) * bnx2x_set_real_num_queues - configure netdev->real_num_[tx,rx]_queues * * @bp: Driver handle + * @include_cnic: handle cnic case * * We currently support for at most 16 Tx queues for each CoS thus we will * allocate a multiple of 16 for ETH L2 rings according to the value of the @@ -4229,8 +4231,8 @@ void bnx2x_get_c2s_mapping(struct bnx2x *bp, u8 *c2s_map, u8 *c2s_default) /** * bnx2x_setup_tc - routine to configure net_device for multi tc * - * @netdev: net device to configure - * @tc: number of traffic classes to enable + * @dev: net device to configure + * @num_tc: number of traffic classes to enable * * callback connected to the ndo_setup_tc function pointer */ diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h index 7e4c93be4451..d8b1824c334d 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h @@ -825,9 +825,9 @@ static inline void bnx2x_del_all_napi_cnic(struct bnx2x *bp) int i; for_each_rx_queue_cnic(bp, i) { - napi_hash_del(&bnx2x_fp(bp, i, napi)); - netif_napi_del(&bnx2x_fp(bp, i, napi)); + __netif_napi_del(&bnx2x_fp(bp, i, napi)); } + synchronize_net(); } static inline void bnx2x_del_all_napi(struct bnx2x *bp) @@ -835,9 +835,9 @@ static inline void bnx2x_del_all_napi(struct bnx2x *bp) int i; for_each_eth_queue(bp, i) { - napi_hash_del(&bnx2x_fp(bp, i, napi)); - netif_napi_del(&bnx2x_fp(bp, i, napi)); + __netif_napi_del(&bnx2x_fp(bp, i, napi)); } + synchronize_net(); } int bnx2x_set_int_mode(struct bnx2x *bp); diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c index 7cea33803f7f..32245bbe88a8 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c @@ -839,8 +839,9 @@ static bool bnx2x_is_wreg_in_chip(struct bnx2x *bp, /** * bnx2x_read_pages_regs - read "paged" registers * - * @bp device handle - * @p output buffer + * @bp: device handle + * @p: output buffer + * @preset: the preset value * * Reads "paged" memories: memories that may only be read by first writing to a * specific address ("write address") and then reading from a specific address @@ -3561,6 +3562,7 @@ static void bnx2x_get_channels(struct net_device *dev, * bnx2x_change_num_queues - change the number of RSS queues. * * @bp: bnx2x private structure + * @num_rss: rss count * * Re-configure interrupt mode to get the new number of MSI-X * vectors and re-add NAPI objects. diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 3c543dd7a8f3..28069b290862 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -3086,9 +3086,9 @@ void bnx2x_func_init(struct bnx2x *bp, struct bnx2x_func_init_params *p) /** * bnx2x_get_common_flags - Return common flags * - * @bp device handle - * @fp queue handle - * @zero_stats TRUE if statistics zeroing is needed + * @bp: device handle + * @fp: queue handle + * @zero_stats: TRUE if statistics zeroing is needed * * Return the flags that are common for the Tx-only and not normal connections. */ @@ -6313,11 +6313,11 @@ static void bnx2x_init_internal(struct bnx2x *bp, u32 load_code) case FW_MSG_CODE_DRV_LOAD_COMMON: case FW_MSG_CODE_DRV_LOAD_COMMON_CHIP: bnx2x_init_internal_common(bp); - /* no break */ + fallthrough; case FW_MSG_CODE_DRV_LOAD_PORT: /* nothing to do */ - /* no break */ + fallthrough; case FW_MSG_CODE_DRV_LOAD_FUNCTION: /* internal memory per function is @@ -12390,7 +12390,7 @@ static int bnx2x_init_bp(struct bnx2x *bp) } if (CHIP_IS_E1(bp)) - bp->dropless_fc = 0; + bp->dropless_fc = false; else bp->dropless_fc = dropless_fc | bnx2x_get_dropless_info(bp); @@ -13591,8 +13591,8 @@ static int bnx2x_set_qm_cid_count(struct bnx2x *bp) /** * bnx2x_get_num_none_def_sbs - return the number of none default SBs - * - * @dev: pci device + * @pdev: pci device + * @cnic_cnt: count * */ static int bnx2x_get_num_non_def_sbs(struct pci_dev *pdev, int cnic_cnt) @@ -14451,9 +14451,7 @@ module_exit(bnx2x_cleanup); /** * bnx2x_set_iscsi_eth_mac_addr - set iSCSI MAC(s). - * * @bp: driver handle - * @set: set or clear the CAM entry * * This function will wait until the ramrod completion returns. * Return 0 if success, -ENODEV if ramrod doesn't return. @@ -15412,7 +15410,7 @@ static int bnx2x_hwtstamp_ioctl(struct bnx2x *bp, struct ifreq *ifr) return -EINVAL; } - bp->hwtstamp_ioctl_called = 1; + bp->hwtstamp_ioctl_called = true; bp->tx_type = config.tx_type; bp->rx_filter = config.rx_filter; @@ -15494,7 +15492,7 @@ void bnx2x_init_ptp(struct bnx2x *bp) bnx2x_init_cyclecounter(bp); timecounter_init(&bp->timecounter, &bp->cyclecounter, ktime_to_ns(ktime_get_real())); - bp->timecounter_init_done = 1; + bp->timecounter_init_done = true; } DP(BNX2X_MSG_PTP, "PTP initialization ended successfully\n"); diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c index e26f4da5a6d7..6cd1523ad9e5 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c @@ -37,10 +37,12 @@ /** * bnx2x_exe_queue_init - init the Exe Queue object * + * @bp: driver handle * @o: pointer to the object * @exe_len: length * @owner: pointer to the owner * @validate: validate function pointer + * @remove: remove function pointer * @optimize: optimize function pointer * @exec: execute function pointer * @get: get function pointer @@ -103,7 +105,7 @@ static inline int bnx2x_exe_queue_length(struct bnx2x_exe_queue_obj *o) * * @bp: driver handle * @o: queue - * @cmd: new command to add + * @elem: new command to add * @restore: true - do not optimize the command * * If the element is optimized or is illegal, frees it. @@ -277,7 +279,7 @@ static void bnx2x_raw_set_pending(struct bnx2x_raw_obj *o) * * @bp: device handle * @state: state which is to be cleared - * @state_p: state buffer + * @pstate: state buffer * */ static inline int bnx2x_state_wait(struct bnx2x *bp, int state, @@ -424,8 +426,8 @@ static bool bnx2x_put_credit_vlan_mac(struct bnx2x_vlan_mac_obj *o) * @bp: device handle * @o: vlan_mac object * - * @details: Non-blocking implementation; should be called under execution - * queue lock. + * Context: Non-blocking implementation; should be called under execution + * queue lock. */ static int __bnx2x_vlan_mac_h_write_trylock(struct bnx2x *bp, struct bnx2x_vlan_mac_obj *o) @@ -445,7 +447,7 @@ static int __bnx2x_vlan_mac_h_write_trylock(struct bnx2x *bp, * @bp: device handle * @o: vlan_mac object * - * @details Should be called under execution queue lock; notice it might release + * details Should be called under execution queue lock; notice it might release * and reclaim it during its run. */ static void __bnx2x_vlan_mac_h_exec_pending(struct bnx2x *bp, @@ -475,7 +477,7 @@ static void __bnx2x_vlan_mac_h_exec_pending(struct bnx2x *bp, * @o: vlan_mac object * @ramrod_flags: ramrod flags of missed execution * - * @details Should be called under execution queue lock. + * Context: Should be called under execution queue lock. */ static void __bnx2x_vlan_mac_h_pend(struct bnx2x *bp, struct bnx2x_vlan_mac_obj *o, @@ -493,7 +495,7 @@ static void __bnx2x_vlan_mac_h_pend(struct bnx2x *bp, * @bp: device handle * @o: vlan_mac object * - * @details Should be called under execution queue lock. Notice if a pending + * Context: Should be called under execution queue lock. Notice if a pending * execution exists, it would perform it - possibly releasing and * reclaiming the execution queue lock. */ @@ -516,7 +518,7 @@ static void __bnx2x_vlan_mac_h_write_unlock(struct bnx2x *bp, * @bp: device handle * @o: vlan_mac object * - * @details Should be called under the execution queue lock. May sleep. May + * Context: Should be called under the execution queue lock. May sleep. May * release and reclaim execution queue lock during its run. */ static int __bnx2x_vlan_mac_h_read_lock(struct bnx2x *bp, @@ -536,7 +538,7 @@ static int __bnx2x_vlan_mac_h_read_lock(struct bnx2x *bp, * @bp: device handle * @o: vlan_mac object * - * @details May sleep. Claims and releases execution queue lock during its run. + * Context: May sleep. Claims and releases execution queue lock during its run. */ int bnx2x_vlan_mac_h_read_lock(struct bnx2x *bp, struct bnx2x_vlan_mac_obj *o) @@ -556,7 +558,7 @@ int bnx2x_vlan_mac_h_read_lock(struct bnx2x *bp, * @bp: device handle * @o: vlan_mac object * - * @details Should be called under execution queue lock. Notice if a pending + * Context: Should be called under execution queue lock. Notice if a pending * execution exists, it would be performed if this was the last * reader. possibly releasing and reclaiming the execution queue lock. */ @@ -591,7 +593,7 @@ static void __bnx2x_vlan_mac_h_read_unlock(struct bnx2x *bp, * @bp: device handle * @o: vlan_mac object * - * @details Notice if a pending execution exists, it would be performed if this + * Context: Notice if a pending execution exists, it would be performed if this * was the last reader. Claims and releases the execution queue lock * during its run. */ @@ -968,7 +970,7 @@ static void bnx2x_set_one_mac_e2(struct bnx2x *bp, * * @bp: device handle * @o: queue - * @type: + * @type: the type of echo * @cam_offset: offset in cam memory * @hdr: pointer to a header to setup * @@ -1608,8 +1610,8 @@ static int __bnx2x_vlan_mac_execute_step(struct bnx2x *bp, * * @bp: device handle * @o: bnx2x_vlan_mac_obj - * @cqe: - * @cont: if true schedule next execution chunk + * @cqe: completion element + * @ramrod_flags: if set schedule next execution chunk * */ static int bnx2x_complete_vlan_mac(struct bnx2x *bp, @@ -1656,7 +1658,7 @@ static int bnx2x_complete_vlan_mac(struct bnx2x *bp, * bnx2x_optimize_vlan_mac - optimize ADD and DEL commands. * * @bp: device handle - * @o: bnx2x_qable_obj + * @qo: bnx2x_qable_obj * @elem: bnx2x_exeq_elem */ static int bnx2x_optimize_vlan_mac(struct bnx2x *bp, @@ -1714,10 +1716,10 @@ static int bnx2x_optimize_vlan_mac(struct bnx2x *bp, * bnx2x_vlan_mac_get_registry_elem - prepare a registry element * * @bp: device handle - * @o: - * @elem: - * @restore: - * @re: + * @o: vlan object + * @elem: element + * @restore: to restore or not + * @re: registry * * prepare a registry element according to the current command request. */ @@ -1768,9 +1770,9 @@ static inline int bnx2x_vlan_mac_get_registry_elem( * bnx2x_execute_vlan_mac - execute vlan mac command * * @bp: device handle - * @qo: - * @exe_chunk: - * @ramrod_flags: + * @qo: bnx2x_qable_obj pointer + * @exe_chunk: chunk + * @ramrod_flags: flags * * go and send a ramrod! */ @@ -2006,8 +2008,8 @@ int bnx2x_config_vlan_mac(struct bnx2x *bp, * bnx2x_vlan_mac_del_all - delete elements with given vlan_mac_flags spec * * @bp: device handle - * @o: - * @vlan_mac_flags: + * @o: vlan object info + * @vlan_mac_flags: vlan flags * @ramrod_flags: execution flags to be used for this deletion * * if the last operation has completed successfully and there are no @@ -2767,7 +2769,7 @@ static int bnx2x_mcast_enqueue_cmd(struct bnx2x *bp, /** * bnx2x_mcast_get_next_bin - get the next set bin (index) * - * @o: + * @o: multicast object info * @last: index to start looking from (including) * * returns the next found (set) bin or a negative value if none is found. @@ -2892,7 +2894,7 @@ static void bnx2x_mcast_set_one_rule_e2(struct bnx2x *bp, * bnx2x_mcast_handle_restore_cmd_e2 - restore configuration from the registry * * @bp: device handle - * @o: + * @o: multicast object info * @start_bin: index in the registry to start from (including) * @rdata_idx: index in the ramrod data to start from * @@ -3202,11 +3204,11 @@ static inline void bnx2x_mcast_hdl_del(struct bnx2x *bp, } /** - * bnx2x_mcast_handle_current_cmd - + * bnx2x_mcast_handle_current_cmd - send command if room * * @bp: device handle - * @p: - * @cmd: + * @p: ramrod mcast info + * @cmd: command * @start_cnt: first line in the ramrod data that may be used * * This function is called iff there is enough place for the current command in @@ -3323,7 +3325,7 @@ static void bnx2x_mcast_revert_e2(struct bnx2x *bp, * bnx2x_mcast_set_rdata_hdr_e2 - sets a header values * * @bp: device handle - * @p: + * @p: ramrod parameters * @len: number of rules to handle */ static inline void bnx2x_mcast_set_rdata_hdr_e2(struct bnx2x *bp, @@ -3684,7 +3686,7 @@ static void bnx2x_mcast_set_one_rule_e1(struct bnx2x *bp, * bnx2x_mcast_set_rdata_hdr_e1 - set header values in mac_configuration_cmd * * @bp: device handle - * @p: + * @p: ramrod parameters * @len: number of rules to handle */ static inline void bnx2x_mcast_set_rdata_hdr_e1(struct bnx2x *bp, @@ -3711,7 +3713,7 @@ static inline void bnx2x_mcast_set_rdata_hdr_e1(struct bnx2x *bp, * bnx2x_mcast_handle_restore_cmd_e1 - restore command for 57710 * * @bp: device handle - * @o: + * @o: multicast info * @start_idx: index in the registry to start from * @rdata_idx: index in the ramrod data to start from * @@ -3798,10 +3800,10 @@ static inline int bnx2x_mcast_handle_pending_cmds_e1( /** * bnx2x_get_fw_mac_addr - revert the bnx2x_set_fw_mac_addr(). * - * @fw_hi: - * @fw_mid: - * @fw_lo: - * @mac: + * @fw_hi: address + * @fw_mid: address + * @fw_lo: address + * @mac: mac address */ static inline void bnx2x_get_fw_mac_addr(__le16 *fw_hi, __le16 *fw_mid, __le16 *fw_lo, u8 *mac) @@ -3818,7 +3820,7 @@ static inline void bnx2x_get_fw_mac_addr(__le16 *fw_hi, __le16 *fw_mid, * bnx2x_mcast_refresh_registry_e1 - * * @bp: device handle - * @cnt: + * @o: multicast info * * Check the ramrod data first entry flag to see if it's a DELETE or ADD command * and update the registry correspondingly: if ADD - allocate a memory and add @@ -4311,7 +4313,7 @@ static bool bnx2x_credit_pool_get_entry_always_true( /** * bnx2x_init_credit_pool - initialize credit pool internals. * - * @p: + * @p: credit pool * @base: Base entry in the CAM to use. * @credit: pool size. * @@ -4725,8 +4727,8 @@ static int bnx2x_queue_wait_comp(struct bnx2x *bp, * bnx2x_queue_comp_cmd - complete the state change command. * * @bp: device handle - * @o: - * @cmd: + * @o: queue info + * @cmd: command to exec * * Checks that the arrived completion is expected. */ @@ -5477,8 +5479,8 @@ static int bnx2x_queue_send_cmd_e2(struct bnx2x *bp, * bnx2x_queue_chk_transition - check state machine of a regular Queue * * @bp: device handle - * @o: - * @params: + * @o: queue info + * @params: queue state * * (not Forwarding) * It both checks if the requested command is legal in a current @@ -5735,8 +5737,8 @@ static int bnx2x_func_wait_comp(struct bnx2x *bp, * bnx2x_func_state_change_comp - complete the state machine transition * * @bp: device handle - * @o: - * @cmd: + * @o: function info + * @cmd: more info * * Called on state change transition. Completes the state * machine transition only - no HW interaction. @@ -5776,8 +5778,8 @@ static inline int bnx2x_func_state_change_comp(struct bnx2x *bp, * bnx2x_func_comp_cmd - complete the state change command * * @bp: device handle - * @o: - * @cmd: + * @o: function info + * @cmd: more info * * Checks that the arrived completion is expected. */ @@ -5796,8 +5798,8 @@ static int bnx2x_func_comp_cmd(struct bnx2x *bp, * bnx2x_func_chk_transition - perform function state machine transition * * @bp: device handle - * @o: - * @params: + * @o: function info + * @params: state parameters * * It both checks if the requested command is legal in a current * state and, if it's legal, sets a `next_state' in the object diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 7b7e8b7883c8..fa147865e33f 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -69,6 +69,7 @@ #include "bnxt_debugfs.h" #define BNXT_TX_TIMEOUT (5 * HZ) +#define BNXT_DEF_MSG_ENABLE (NETIF_MSG_DRV | NETIF_MSG_HW) MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("Broadcom BCM573xx network driver"); @@ -254,6 +255,7 @@ static const u16 bnxt_async_events_arr[] = { ASYNC_EVENT_CMPL_EVENT_ID_PORT_PHY_CFG_CHANGE, ASYNC_EVENT_CMPL_EVENT_ID_RESET_NOTIFY, ASYNC_EVENT_CMPL_EVENT_ID_ERROR_RECOVERY, + ASYNC_EVENT_CMPL_EVENT_ID_RING_MONITOR_MSG, }; static struct workqueue_struct *bnxt_pf_wq; @@ -1172,7 +1174,10 @@ static void bnxt_sched_reset(struct bnxt *bp, struct bnxt_rx_ring_info *rxr) { if (!rxr->bnapi->in_reset) { rxr->bnapi->in_reset = true; - set_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event); + if (bp->flags & BNXT_FLAG_CHIP_P5) + set_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event); + else + set_bit(BNXT_RST_RING_SP_EVENT, &bp->sp_event); bnxt_queue_sp_work(bp); } rxr->rx_next_cons = 0xffff; @@ -1738,8 +1743,10 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr, if (unlikely(cons != rxr->rx_next_cons)) { int rc1 = bnxt_discard_rx(bp, cpr, raw_cons, rxcmp); - netdev_warn(bp->dev, "RX cons %x != expected cons %x\n", - cons, rxr->rx_next_cons); + /* 0xffff is forced error, don't print it */ + if (rxr->rx_next_cons != 0xffff) + netdev_warn(bp->dev, "RX cons %x != expected cons %x\n", + cons, rxr->rx_next_cons); bnxt_sched_reset(bp, rxr); return rc1; } @@ -1772,9 +1779,10 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr, rc = -EIO; if (rx_err & RX_CMPL_ERRORS_BUFFER_ERROR_MASK) { bnapi->cp_ring.sw_stats.rx.rx_buf_errors++; - if (!(bp->flags & BNXT_FLAG_CHIP_P5)) { - netdev_warn(bp->dev, "RX buffer error %x\n", - rx_err); + if (!(bp->flags & BNXT_FLAG_CHIP_P5) && + !(bp->fw_cap & BNXT_FW_CAP_RING_MONITOR)) { + netdev_warn_once(bp->dev, "RX buffer error %x\n", + rx_err); bnxt_sched_reset(bp, rxr); } } @@ -1941,19 +1949,43 @@ u32 bnxt_fw_health_readl(struct bnxt *bp, int reg_idx) return val; } +static u16 bnxt_agg_ring_id_to_grp_idx(struct bnxt *bp, u16 ring_id) +{ + int i; + + for (i = 0; i < bp->rx_nr_rings; i++) { + u16 grp_idx = bp->rx_ring[i].bnapi->index; + struct bnxt_ring_grp_info *grp_info; + + grp_info = &bp->grp_info[grp_idx]; + if (grp_info->agg_fw_ring_id == ring_id) + return grp_idx; + } + return INVALID_HW_RING_ID; +} + #define BNXT_GET_EVENT_PORT(data) \ ((data) & \ ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_PORT_ID_MASK) +#define BNXT_EVENT_RING_TYPE(data2) \ + ((data2) & \ + ASYNC_EVENT_CMPL_RING_MONITOR_MSG_EVENT_DATA2_DISABLE_RING_TYPE_MASK) + +#define BNXT_EVENT_RING_TYPE_RX(data2) \ + (BNXT_EVENT_RING_TYPE(data2) == \ + ASYNC_EVENT_CMPL_RING_MONITOR_MSG_EVENT_DATA2_DISABLE_RING_TYPE_RX) + static int bnxt_async_event_process(struct bnxt *bp, struct hwrm_async_event_cmpl *cmpl) { u16 event_id = le16_to_cpu(cmpl->event_id); + u32 data1 = le32_to_cpu(cmpl->event_data1); + u32 data2 = le32_to_cpu(cmpl->event_data2); /* TODO CHIMP_FW: Define event id's for link change, error etc */ switch (event_id) { case ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CFG_CHANGE: { - u32 data1 = le32_to_cpu(cmpl->event_data1); struct bnxt_link_info *link_info = &bp->link_info; if (BNXT_VF(bp)) @@ -1983,7 +2015,6 @@ static int bnxt_async_event_process(struct bnxt *bp, set_bit(BNXT_HWRM_PF_UNLOAD_SP_EVENT, &bp->sp_event); break; case ASYNC_EVENT_CMPL_EVENT_ID_PORT_CONN_NOT_ALLOWED: { - u32 data1 = le32_to_cpu(cmpl->event_data1); u16 port_id = BNXT_GET_EVENT_PORT(data1); if (BNXT_VF(bp)) @@ -2000,9 +2031,10 @@ static int bnxt_async_event_process(struct bnxt *bp, goto async_event_process_exit; set_bit(BNXT_RESET_TASK_SILENT_SP_EVENT, &bp->sp_event); break; - case ASYNC_EVENT_CMPL_EVENT_ID_RESET_NOTIFY: { - u32 data1 = le32_to_cpu(cmpl->event_data1); - + case ASYNC_EVENT_CMPL_EVENT_ID_RESET_NOTIFY: + if (netif_msg_hw(bp)) + netdev_warn(bp->dev, "Received RESET_NOTIFY event, data1: 0x%x, data2: 0x%x\n", + data1, data2); if (!bp->fw_health) goto async_event_process_exit; @@ -2022,10 +2054,8 @@ static int bnxt_async_event_process(struct bnxt *bp, } set_bit(BNXT_FW_RESET_NOTIFY_SP_EVENT, &bp->sp_event); break; - } case ASYNC_EVENT_CMPL_EVENT_ID_ERROR_RECOVERY: { struct bnxt_fw_health *fw_health = bp->fw_health; - u32 data1 = le32_to_cpu(cmpl->event_data1); if (!fw_health) goto async_event_process_exit; @@ -2052,6 +2082,28 @@ static int bnxt_async_event_process(struct bnxt *bp, bnxt_fw_health_readl(bp, BNXT_FW_RESET_CNT_REG); goto async_event_process_exit; } + case ASYNC_EVENT_CMPL_EVENT_ID_RING_MONITOR_MSG: { + struct bnxt_rx_ring_info *rxr; + u16 grp_idx; + + if (bp->flags & BNXT_FLAG_CHIP_P5) + goto async_event_process_exit; + + netdev_warn(bp->dev, "Ring monitor event, ring type %lu id 0x%x\n", + BNXT_EVENT_RING_TYPE(data2), data1); + if (!BNXT_EVENT_RING_TYPE_RX(data2)) + goto async_event_process_exit; + + grp_idx = bnxt_agg_ring_id_to_grp_idx(bp, data1); + if (grp_idx == INVALID_HW_RING_ID) { + netdev_warn(bp->dev, "Unknown RX agg ring id 0x%x\n", + data1); + goto async_event_process_exit; + } + rxr = bp->bnapi[grp_idx]->rx_ring; + bnxt_sched_reset(bp, rxr); + goto async_event_process_exit; + } default: goto async_event_process_exit; } @@ -2250,7 +2302,7 @@ static void __bnxt_poll_work_done(struct bnxt *bp, struct bnxt_napi *bnapi) bnapi->tx_pkts = 0; } - if (bnapi->events & BNXT_RX_EVENT) { + if ((bnapi->events & BNXT_RX_EVENT) && !(bnapi->in_reset)) { struct bnxt_rx_ring_info *rxr = bnapi->rx_ring; if (bnapi->events & BNXT_AGG_EVENT) @@ -2540,93 +2592,91 @@ static void bnxt_free_tx_skbs(struct bnxt *bp) } } -static void bnxt_free_rx_skbs(struct bnxt *bp) +static void bnxt_free_one_rx_ring_skbs(struct bnxt *bp, int ring_nr) { - int i, max_idx, max_agg_idx; + struct bnxt_rx_ring_info *rxr = &bp->rx_ring[ring_nr]; struct pci_dev *pdev = bp->pdev; - - if (!bp->rx_ring) - return; + struct bnxt_tpa_idx_map *map; + int i, max_idx, max_agg_idx; max_idx = bp->rx_nr_pages * RX_DESC_CNT; max_agg_idx = bp->rx_agg_nr_pages * RX_DESC_CNT; - for (i = 0; i < bp->rx_nr_rings; i++) { - struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i]; - struct bnxt_tpa_idx_map *map; - int j; - - if (rxr->rx_tpa) { - for (j = 0; j < bp->max_tpa; j++) { - struct bnxt_tpa_info *tpa_info = - &rxr->rx_tpa[j]; - u8 *data = tpa_info->data; + if (!rxr->rx_tpa) + goto skip_rx_tpa_free; - if (!data) - continue; + for (i = 0; i < bp->max_tpa; i++) { + struct bnxt_tpa_info *tpa_info = &rxr->rx_tpa[i]; + u8 *data = tpa_info->data; - dma_unmap_single_attrs(&pdev->dev, - tpa_info->mapping, - bp->rx_buf_use_size, - bp->rx_dir, - DMA_ATTR_WEAK_ORDERING); + if (!data) + continue; - tpa_info->data = NULL; + dma_unmap_single_attrs(&pdev->dev, tpa_info->mapping, + bp->rx_buf_use_size, bp->rx_dir, + DMA_ATTR_WEAK_ORDERING); - kfree(data); - } - } + tpa_info->data = NULL; - for (j = 0; j < max_idx; j++) { - struct bnxt_sw_rx_bd *rx_buf = &rxr->rx_buf_ring[j]; - dma_addr_t mapping = rx_buf->mapping; - void *data = rx_buf->data; + kfree(data); + } - if (!data) - continue; +skip_rx_tpa_free: + for (i = 0; i < max_idx; i++) { + struct bnxt_sw_rx_bd *rx_buf = &rxr->rx_buf_ring[i]; + dma_addr_t mapping = rx_buf->mapping; + void *data = rx_buf->data; - rx_buf->data = NULL; + if (!data) + continue; - if (BNXT_RX_PAGE_MODE(bp)) { - mapping -= bp->rx_dma_offset; - dma_unmap_page_attrs(&pdev->dev, mapping, - PAGE_SIZE, bp->rx_dir, - DMA_ATTR_WEAK_ORDERING); - page_pool_recycle_direct(rxr->page_pool, data); - } else { - dma_unmap_single_attrs(&pdev->dev, mapping, - bp->rx_buf_use_size, - bp->rx_dir, - DMA_ATTR_WEAK_ORDERING); - kfree(data); - } + rx_buf->data = NULL; + if (BNXT_RX_PAGE_MODE(bp)) { + mapping -= bp->rx_dma_offset; + dma_unmap_page_attrs(&pdev->dev, mapping, PAGE_SIZE, + bp->rx_dir, + DMA_ATTR_WEAK_ORDERING); + page_pool_recycle_direct(rxr->page_pool, data); + } else { + dma_unmap_single_attrs(&pdev->dev, mapping, + bp->rx_buf_use_size, bp->rx_dir, + DMA_ATTR_WEAK_ORDERING); + kfree(data); } + } + for (i = 0; i < max_agg_idx; i++) { + struct bnxt_sw_rx_agg_bd *rx_agg_buf = &rxr->rx_agg_ring[i]; + struct page *page = rx_agg_buf->page; - for (j = 0; j < max_agg_idx; j++) { - struct bnxt_sw_rx_agg_bd *rx_agg_buf = - &rxr->rx_agg_ring[j]; - struct page *page = rx_agg_buf->page; - - if (!page) - continue; + if (!page) + continue; - dma_unmap_page_attrs(&pdev->dev, rx_agg_buf->mapping, - BNXT_RX_PAGE_SIZE, - PCI_DMA_FROMDEVICE, - DMA_ATTR_WEAK_ORDERING); + dma_unmap_page_attrs(&pdev->dev, rx_agg_buf->mapping, + BNXT_RX_PAGE_SIZE, PCI_DMA_FROMDEVICE, + DMA_ATTR_WEAK_ORDERING); - rx_agg_buf->page = NULL; - __clear_bit(j, rxr->rx_agg_bmap); + rx_agg_buf->page = NULL; + __clear_bit(i, rxr->rx_agg_bmap); - __free_page(page); - } - if (rxr->rx_page) { - __free_page(rxr->rx_page); - rxr->rx_page = NULL; - } - map = rxr->rx_tpa_idx_map; - if (map) - memset(map->agg_idx_bmap, 0, sizeof(map->agg_idx_bmap)); + __free_page(page); + } + if (rxr->rx_page) { + __free_page(rxr->rx_page); + rxr->rx_page = NULL; } + map = rxr->rx_tpa_idx_map; + if (map) + memset(map->agg_idx_bmap, 0, sizeof(map->agg_idx_bmap)); +} + +static void bnxt_free_rx_skbs(struct bnxt *bp) +{ + int i; + + if (!bp->rx_ring) + return; + + for (i = 0; i < bp->rx_nr_rings; i++) + bnxt_free_one_rx_ring_skbs(bp, i); } static void bnxt_free_skbs(struct bnxt *bp) @@ -3165,31 +3215,16 @@ static void bnxt_init_rxbd_pages(struct bnxt_ring_struct *ring, u32 type) } } -static int bnxt_init_one_rx_ring(struct bnxt *bp, int ring_nr) +static int bnxt_alloc_one_rx_ring(struct bnxt *bp, int ring_nr) { + struct bnxt_rx_ring_info *rxr = &bp->rx_ring[ring_nr]; struct net_device *dev = bp->dev; - struct bnxt_rx_ring_info *rxr; - struct bnxt_ring_struct *ring; - u32 prod, type; + u32 prod; int i; - type = (bp->rx_buf_use_size << RX_BD_LEN_SHIFT) | - RX_BD_TYPE_RX_PACKET_BD | RX_BD_FLAGS_EOP; - - if (NET_IP_ALIGN == 2) - type |= RX_BD_FLAGS_SOP; - - rxr = &bp->rx_ring[ring_nr]; - ring = &rxr->rx_ring_struct; - bnxt_init_rxbd_pages(ring, type); - - if (BNXT_RX_PAGE_MODE(bp) && bp->xdp_prog) { - bpf_prog_add(bp->xdp_prog, 1); - rxr->xdp_prog = bp->xdp_prog; - } prod = rxr->rx_prod; for (i = 0; i < bp->rx_ring_size; i++) { - if (bnxt_alloc_rx_data(bp, rxr, prod, GFP_KERNEL) != 0) { + if (bnxt_alloc_rx_data(bp, rxr, prod, GFP_KERNEL)) { netdev_warn(dev, "init'ed rx ring %d with %d/%d skbs only\n", ring_nr, i, bp->rx_ring_size); break; @@ -3197,22 +3232,13 @@ static int bnxt_init_one_rx_ring(struct bnxt *bp, int ring_nr) prod = NEXT_RX(prod); } rxr->rx_prod = prod; - ring->fw_ring_id = INVALID_HW_RING_ID; - - ring = &rxr->rx_agg_ring_struct; - ring->fw_ring_id = INVALID_HW_RING_ID; if (!(bp->flags & BNXT_FLAG_AGG_RINGS)) return 0; - type = ((u32)BNXT_RX_PAGE_SIZE << RX_BD_LEN_SHIFT) | - RX_BD_TYPE_RX_AGG_BD | RX_BD_FLAGS_SOP; - - bnxt_init_rxbd_pages(ring, type); - prod = rxr->rx_agg_prod; for (i = 0; i < bp->rx_agg_ring_size; i++) { - if (bnxt_alloc_rx_page(bp, rxr, prod, GFP_KERNEL) != 0) { + if (bnxt_alloc_rx_page(bp, rxr, prod, GFP_KERNEL)) { netdev_warn(dev, "init'ed rx ring %d with %d/%d pages only\n", ring_nr, i, bp->rx_ring_size); break; @@ -3221,30 +3247,58 @@ static int bnxt_init_one_rx_ring(struct bnxt *bp, int ring_nr) } rxr->rx_agg_prod = prod; - if (bp->flags & BNXT_FLAG_TPA) { - if (rxr->rx_tpa) { - u8 *data; - dma_addr_t mapping; + if (rxr->rx_tpa) { + dma_addr_t mapping; + u8 *data; - for (i = 0; i < bp->max_tpa; i++) { - data = __bnxt_alloc_rx_data(bp, &mapping, - GFP_KERNEL); - if (!data) - return -ENOMEM; + for (i = 0; i < bp->max_tpa; i++) { + data = __bnxt_alloc_rx_data(bp, &mapping, GFP_KERNEL); + if (!data) + return -ENOMEM; - rxr->rx_tpa[i].data = data; - rxr->rx_tpa[i].data_ptr = data + bp->rx_offset; - rxr->rx_tpa[i].mapping = mapping; - } - } else { - netdev_err(bp->dev, "No resource allocated for LRO/GRO\n"); - return -ENOMEM; + rxr->rx_tpa[i].data = data; + rxr->rx_tpa[i].data_ptr = data + bp->rx_offset; + rxr->rx_tpa[i].mapping = mapping; } } - return 0; } +static int bnxt_init_one_rx_ring(struct bnxt *bp, int ring_nr) +{ + struct bnxt_rx_ring_info *rxr; + struct bnxt_ring_struct *ring; + u32 type; + + type = (bp->rx_buf_use_size << RX_BD_LEN_SHIFT) | + RX_BD_TYPE_RX_PACKET_BD | RX_BD_FLAGS_EOP; + + if (NET_IP_ALIGN == 2) + type |= RX_BD_FLAGS_SOP; + + rxr = &bp->rx_ring[ring_nr]; + ring = &rxr->rx_ring_struct; + bnxt_init_rxbd_pages(ring, type); + + if (BNXT_RX_PAGE_MODE(bp) && bp->xdp_prog) { + bpf_prog_add(bp->xdp_prog, 1); + rxr->xdp_prog = bp->xdp_prog; + } + ring->fw_ring_id = INVALID_HW_RING_ID; + + ring = &rxr->rx_agg_ring_struct; + ring->fw_ring_id = INVALID_HW_RING_ID; + + if ((bp->flags & BNXT_FLAG_AGG_RINGS)) { + type = ((u32)BNXT_RX_PAGE_SIZE << RX_BD_LEN_SHIFT) | + RX_BD_TYPE_RX_AGG_BD | RX_BD_FLAGS_SOP; + + bnxt_init_rxbd_pages(ring, type); + } + + return bnxt_alloc_one_rx_ring(bp, ring_nr); +} + static void bnxt_init_cp_rings(struct bnxt *bp) { int i, j; @@ -4269,6 +4323,8 @@ static int bnxt_hwrm_to_stderr(u32 hwrm_err) switch (hwrm_err) { case HWRM_ERR_CODE_SUCCESS: return 0; + case HWRM_ERR_CODE_RESOURCE_LOCKED: + return -EROFS; case HWRM_ERR_CODE_RESOURCE_ACCESS_DENIED: return -EACCES; case HWRM_ERR_CODE_RESOURCE_ALLOC_ERROR: @@ -5343,13 +5399,16 @@ static int bnxt_hwrm_vnic_qcaps(struct bnxt *bp) * VLAN_STRIP_CAP properly. */ if ((flags & VNIC_QCAPS_RESP_FLAGS_VLAN_STRIP_CAP) || - ((bp->flags & BNXT_FLAG_CHIP_P5) && + (BNXT_CHIP_P5_THOR(bp) && !(bp->fw_cap & BNXT_FW_CAP_EXT_HW_STATS_SUPPORTED))) bp->fw_cap |= BNXT_FW_CAP_VLAN_RX_STRIP; bp->max_tpa_v2 = le16_to_cpu(resp->max_aggs_supported); - if (bp->max_tpa_v2) - bp->hw_ring_stats_size = - sizeof(struct ctx_hw_stats_ext); + if (bp->max_tpa_v2) { + if (BNXT_CHIP_P5_THOR(bp)) + bp->hw_ring_stats_size = BNXT_RING_STATS_SIZE_P5; + else + bp->hw_ring_stats_size = BNXT_RING_STATS_SIZE_P5_SR2; + } } mutex_unlock(&bp->hwrm_cmd_lock); return rc; @@ -6639,6 +6698,8 @@ static int bnxt_hwrm_func_qcfg(struct bnxt *bp) } if (BNXT_PF(bp) && (flags & FUNC_QCFG_RESP_FLAGS_MULTI_HOST)) bp->flags |= BNXT_FLAG_MULTI_HOST; + if (flags & FUNC_QCFG_RESP_FLAGS_RING_MONITOR_ENABLED) + bp->fw_cap |= BNXT_FW_CAP_RING_MONITOR; switch (resp->port_partition_type) { case FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR1_0: @@ -7333,6 +7394,77 @@ hwrm_cfa_adv_qcaps_exit: return rc; } +static int __bnxt_alloc_fw_health(struct bnxt *bp) +{ + if (bp->fw_health) + return 0; + + bp->fw_health = kzalloc(sizeof(*bp->fw_health), GFP_KERNEL); + if (!bp->fw_health) + return -ENOMEM; + + return 0; +} + +static int bnxt_alloc_fw_health(struct bnxt *bp) +{ + int rc; + + if (!(bp->fw_cap & BNXT_FW_CAP_HOT_RESET) && + !(bp->fw_cap & BNXT_FW_CAP_ERROR_RECOVERY)) + return 0; + + rc = __bnxt_alloc_fw_health(bp); + if (rc) { + bp->fw_cap &= ~BNXT_FW_CAP_HOT_RESET; + bp->fw_cap &= ~BNXT_FW_CAP_ERROR_RECOVERY; + return rc; + } + + return 0; +} + +static void __bnxt_map_fw_health_reg(struct bnxt *bp, u32 reg) +{ + writel(reg & BNXT_GRC_BASE_MASK, bp->bar0 + + BNXT_GRCPF_REG_WINDOW_BASE_OUT + + BNXT_FW_HEALTH_WIN_MAP_OFF); +} + +static void bnxt_try_map_fw_health_reg(struct bnxt *bp) +{ + void __iomem *hs; + u32 status_loc; + u32 reg_type; + u32 sig; + + __bnxt_map_fw_health_reg(bp, HCOMM_STATUS_STRUCT_LOC); + hs = bp->bar0 + BNXT_FW_HEALTH_WIN_OFF(HCOMM_STATUS_STRUCT_LOC); + + sig = readl(hs + offsetof(struct hcomm_status, sig_ver)); + if ((sig & HCOMM_STATUS_SIGNATURE_MASK) != HCOMM_STATUS_SIGNATURE_VAL) { + if (bp->fw_health) + bp->fw_health->status_reliable = false; + return; + } + + if (__bnxt_alloc_fw_health(bp)) { + netdev_warn(bp->dev, "no memory for firmware status checks\n"); + return; + } + + status_loc = readl(hs + offsetof(struct hcomm_status, fw_status_loc)); + bp->fw_health->regs[BNXT_FW_HEALTH_REG] = status_loc; + reg_type = BNXT_FW_HEALTH_REG_TYPE(status_loc); + if (reg_type == BNXT_FW_HEALTH_REG_TYPE_GRC) { + __bnxt_map_fw_health_reg(bp, status_loc); + bp->fw_health->mapped_regs[BNXT_FW_HEALTH_REG] = + BNXT_FW_HEALTH_WIN_OFF(status_loc); + } + + bp->fw_health->status_reliable = true; +} + static int bnxt_map_fw_health_regs(struct bnxt *bp) { struct bnxt_fw_health *fw_health = bp->fw_health; @@ -7349,14 +7481,12 @@ static int bnxt_map_fw_health_regs(struct bnxt *bp) reg_base = reg & BNXT_GRC_BASE_MASK; if ((reg & BNXT_GRC_BASE_MASK) != reg_base) return -ERANGE; - fw_health->mapped_regs[i] = BNXT_FW_HEALTH_WIN_BASE + - (reg & BNXT_GRC_OFFSET_MASK); + fw_health->mapped_regs[i] = BNXT_FW_HEALTH_WIN_OFF(reg); } if (reg_base == 0xffffffff) return 0; - writel(reg_base, bp->bar0 + BNXT_GRCPF_REG_WINDOW_BASE_OUT + - BNXT_FW_HEALTH_WIN_MAP_OFF); + __bnxt_map_fw_health_reg(bp, reg_base); return 0; } @@ -7432,6 +7562,16 @@ static int bnxt_hwrm_func_reset(struct bnxt *bp) return hwrm_send_message(bp, &req, sizeof(req), HWRM_RESET_TIMEOUT); } +static void bnxt_nvm_cfg_ver_get(struct bnxt *bp) +{ + struct hwrm_nvm_get_dev_info_output nvm_info; + + if (!bnxt_hwrm_nvm_get_dev_info(bp, &nvm_info)) + snprintf(bp->nvm_cfg_ver, FW_VER_STR_LEN, "%d.%d.%d", + nvm_info.nvm_cfg_ver_maj, nvm_info.nvm_cfg_ver_min, + nvm_info.nvm_cfg_ver_upd); +} + static int bnxt_hwrm_queue_qportcfg(struct bnxt *bp) { int rc = 0; @@ -8635,10 +8775,9 @@ static void bnxt_del_napi(struct bnxt *bp) for (i = 0; i < bp->cp_nr_rings; i++) { struct bnxt_napi *bnapi = bp->bnapi[i]; - napi_hash_del(&bnapi->napi); - netif_napi_del(&bnapi->napi); + __netif_napi_del(&bnapi->napi); } - /* We called napi_hash_del() before netif_napi_del(), we need + /* We called __netif_napi_del(), we need * to respect an RCU grace period before freeing napi structures. */ synchronize_net(); @@ -8694,14 +8833,19 @@ static void bnxt_enable_napi(struct bnxt *bp) int i; for (i = 0; i < bp->cp_nr_rings; i++) { - struct bnxt_cp_ring_info *cpr = &bp->bnapi[i]->cp_ring; - bp->bnapi[i]->in_reset = false; + struct bnxt_napi *bnapi = bp->bnapi[i]; + struct bnxt_cp_ring_info *cpr; + + cpr = &bnapi->cp_ring; + if (bnapi->in_reset) + cpr->sw_stats.rx.rx_resets++; + bnapi->in_reset = false; - if (bp->bnapi[i]->rx_ring) { + if (bnapi->rx_ring) { INIT_WORK(&cpr->dim.work, bnxt_dim_work); cpr->dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE; } - napi_enable(&bp->bnapi[i]->napi); + napi_enable(&bnapi->napi); } } @@ -8735,6 +8879,30 @@ void bnxt_tx_enable(struct bnxt *bp) netif_carrier_on(bp->dev); } +static char *bnxt_report_fec(struct bnxt_link_info *link_info) +{ + u8 active_fec = link_info->active_fec_sig_mode & + PORT_PHY_QCFG_RESP_ACTIVE_FEC_MASK; + + switch (active_fec) { + default: + case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_NONE_ACTIVE: + return "None"; + case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_CLAUSE74_ACTIVE: + return "Clause 74 BaseR"; + case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_CLAUSE91_ACTIVE: + return "Clause 91 RS(528,514)"; + case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS544_1XN_ACTIVE: + return "Clause 91 RS544_1XN"; + case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS544_IEEE_ACTIVE: + return "Clause 91 RS(544,514)"; + case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS272_1XN_ACTIVE: + return "Clause 91 RS272_1XN"; + case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS272_IEEE_ACTIVE: + return "Clause 91 RS(272,257)"; + } +} + static void bnxt_report_link(struct bnxt *bp) { if (bp->link_info.link_up) { @@ -8744,6 +8912,11 @@ static void bnxt_report_link(struct bnxt *bp) u16 fec; netif_carrier_on(bp->dev); + speed = bnxt_fw_to_ethtool_speed(bp->link_info.link_speed); + if (speed == SPEED_UNKNOWN) { + netdev_info(bp->dev, "NIC Link is Up, speed unknown\n"); + return; + } if (bp->link_info.duplex == BNXT_LINK_DUPLEX_FULL) duplex = "full"; else @@ -8756,7 +8929,6 @@ static void bnxt_report_link(struct bnxt *bp) flow_ctrl = "ON - receive"; else flow_ctrl = "none"; - speed = bnxt_fw_to_ethtool_speed(bp->link_info.link_speed); netdev_info(bp->dev, "NIC Link is Up, %u Mbps %s duplex, Flow control: %s\n", speed, duplex, flow_ctrl); if (bp->flags & BNXT_FLAG_EEE_CAP) @@ -8765,16 +8937,25 @@ static void bnxt_report_link(struct bnxt *bp) "not active"); fec = bp->link_info.fec_cfg; if (!(fec & PORT_PHY_QCFG_RESP_FEC_CFG_FEC_NONE_SUPPORTED)) - netdev_info(bp->dev, "FEC autoneg %s encodings: %s\n", + netdev_info(bp->dev, "FEC autoneg %s encoding: %s\n", (fec & BNXT_FEC_AUTONEG) ? "on" : "off", - (fec & BNXT_FEC_ENC_BASE_R) ? "BaseR" : - (fec & BNXT_FEC_ENC_RS) ? "RS" : "None"); + bnxt_report_fec(&bp->link_info)); } else { netif_carrier_off(bp->dev); netdev_err(bp->dev, "NIC Link is Down\n"); } } +static bool bnxt_phy_qcaps_no_speed(struct hwrm_port_phy_qcaps_output *resp) +{ + if (!resp->supported_speeds_auto_mode && + !resp->supported_speeds_force_mode && + !resp->supported_pam4_speeds_auto_mode && + !resp->supported_pam4_speeds_force_mode) + return true; + return false; +} + static int bnxt_hwrm_phy_qcaps(struct bnxt *bp) { int rc = 0; @@ -8822,9 +9003,24 @@ static int bnxt_hwrm_phy_qcaps(struct bnxt *bp) if (resp->flags & PORT_PHY_QCAPS_RESP_FLAGS_CUMULATIVE_COUNTERS_ON_RESET) bp->fw_cap |= BNXT_FW_CAP_PORT_STATS_NO_RESET; + if (bp->hwrm_spec_code >= 0x10a01) { + if (bnxt_phy_qcaps_no_speed(resp)) { + link_info->phy_state = BNXT_PHY_STATE_DISABLED; + netdev_warn(bp->dev, "Ethernet link disabled\n"); + } else if (link_info->phy_state == BNXT_PHY_STATE_DISABLED) { + link_info->phy_state = BNXT_PHY_STATE_ENABLED; + netdev_info(bp->dev, "Ethernet link enabled\n"); + /* Phy re-enabled, reprobe the speeds */ + link_info->support_auto_speeds = 0; + link_info->support_pam4_auto_speeds = 0; + } + } if (resp->supported_speeds_auto_mode) link_info->support_auto_speeds = le16_to_cpu(resp->supported_speeds_auto_mode); + if (resp->supported_pam4_speeds_auto_mode) + link_info->support_pam4_auto_speeds = + le16_to_cpu(resp->supported_pam4_speeds_auto_mode); bp->port_count = resp->port_cnt; @@ -8833,14 +9029,21 @@ hwrm_phy_qcaps_exit: return rc; } -static int bnxt_update_link(struct bnxt *bp, bool chng_link_state) +static bool bnxt_support_dropped(u16 advertising, u16 supported) +{ + u16 diff = advertising ^ supported; + + return ((supported | diff) != supported); +} + +int bnxt_update_link(struct bnxt *bp, bool chng_link_state) { int rc = 0; struct bnxt_link_info *link_info = &bp->link_info; struct hwrm_port_phy_qcfg_input req = {0}; struct hwrm_port_phy_qcfg_output *resp = bp->hwrm_cmd_resp_addr; u8 link_up = link_info->link_up; - u16 diff; + bool support_changed = false; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_PHY_QCFG, -1, -1); @@ -8867,10 +9070,17 @@ static int bnxt_update_link(struct bnxt *bp, bool chng_link_state) else link_info->link_speed = 0; link_info->force_link_speed = le16_to_cpu(resp->force_link_speed); + link_info->force_pam4_link_speed = + le16_to_cpu(resp->force_pam4_link_speed); link_info->support_speeds = le16_to_cpu(resp->support_speeds); + link_info->support_pam4_speeds = le16_to_cpu(resp->support_pam4_speeds); link_info->auto_link_speeds = le16_to_cpu(resp->auto_link_speed_mask); + link_info->auto_pam4_link_speeds = + le16_to_cpu(resp->auto_pam4_link_speed_mask); link_info->lp_auto_link_speeds = le16_to_cpu(resp->link_partner_adv_speeds); + link_info->lp_auto_pam4_link_speeds = + resp->link_partner_pam4_adv_speeds; link_info->preemphasis = le32_to_cpu(resp->preemphasis); link_info->phy_ver[0] = resp->phy_maj; link_info->phy_ver[1] = resp->phy_min; @@ -8919,9 +9129,10 @@ static int bnxt_update_link(struct bnxt *bp, bool chng_link_state) } link_info->fec_cfg = PORT_PHY_QCFG_RESP_FEC_CFG_FEC_NONE_SUPPORTED; - if (bp->hwrm_spec_code >= 0x10504) + if (bp->hwrm_spec_code >= 0x10504) { link_info->fec_cfg = le16_to_cpu(resp->fec_cfg); - + link_info->active_fec_sig_mode = resp->active_fec_signal_mode; + } /* TODO: need to add more logic to report VF link */ if (chng_link_state) { if (link_info->phy_link_status == BNXT_LINK_LINK) @@ -8939,17 +9150,21 @@ static int bnxt_update_link(struct bnxt *bp, bool chng_link_state) if (!BNXT_PHY_CFG_ABLE(bp)) return 0; - diff = link_info->support_auto_speeds ^ link_info->advertising; - if ((link_info->support_auto_speeds | diff) != - link_info->support_auto_speeds) { - /* An advertised speed is no longer supported, so we need to - * update the advertisement settings. Caller holds RTNL - * so we can modify link settings. - */ + /* Check if any advertised speeds are no longer supported. The caller + * holds the link_lock mutex, so we can modify link_info settings. + */ + if (bnxt_support_dropped(link_info->advertising, + link_info->support_auto_speeds)) { link_info->advertising = link_info->support_auto_speeds; - if (link_info->autoneg & BNXT_AUTONEG_SPEED) - bnxt_hwrm_set_link_setting(bp, true, false); + support_changed = true; + } + if (bnxt_support_dropped(link_info->advertising_pam4, + link_info->support_pam4_auto_speeds)) { + link_info->advertising_pam4 = link_info->support_pam4_auto_speeds; + support_changed = true; } + if (support_changed && (link_info->autoneg & BNXT_AUTONEG_SPEED)) + bnxt_hwrm_set_link_setting(bp, true, false); return 0; } @@ -9008,27 +9223,30 @@ bnxt_hwrm_set_pause_common(struct bnxt *bp, struct hwrm_port_phy_cfg_input *req) } } -static void bnxt_hwrm_set_link_common(struct bnxt *bp, - struct hwrm_port_phy_cfg_input *req) +static void bnxt_hwrm_set_link_common(struct bnxt *bp, struct hwrm_port_phy_cfg_input *req) { - u8 autoneg = bp->link_info.autoneg; - u16 fw_link_speed = bp->link_info.req_link_speed; - u16 advertising = bp->link_info.advertising; - - if (autoneg & BNXT_AUTONEG_SPEED) { - req->auto_mode |= - PORT_PHY_CFG_REQ_AUTO_MODE_SPEED_MASK; - - req->enables |= cpu_to_le32( - PORT_PHY_CFG_REQ_ENABLES_AUTO_LINK_SPEED_MASK); - req->auto_link_speed_mask = cpu_to_le16(advertising); - + if (bp->link_info.autoneg & BNXT_AUTONEG_SPEED) { + req->auto_mode |= PORT_PHY_CFG_REQ_AUTO_MODE_SPEED_MASK; + if (bp->link_info.advertising) { + req->enables |= cpu_to_le32(PORT_PHY_CFG_REQ_ENABLES_AUTO_LINK_SPEED_MASK); + req->auto_link_speed_mask = cpu_to_le16(bp->link_info.advertising); + } + if (bp->link_info.advertising_pam4) { + req->enables |= + cpu_to_le32(PORT_PHY_CFG_REQ_ENABLES_AUTO_PAM4_LINK_SPEED_MASK); + req->auto_link_pam4_speed_mask = + cpu_to_le16(bp->link_info.advertising_pam4); + } req->enables |= cpu_to_le32(PORT_PHY_CFG_REQ_ENABLES_AUTO_MODE); - req->flags |= - cpu_to_le32(PORT_PHY_CFG_REQ_FLAGS_RESTART_AUTONEG); + req->flags |= cpu_to_le32(PORT_PHY_CFG_REQ_FLAGS_RESTART_AUTONEG); } else { - req->force_link_speed = cpu_to_le16(fw_link_speed); req->flags |= cpu_to_le32(PORT_PHY_CFG_REQ_FLAGS_FORCE); + if (bp->link_info.req_signal_mode == BNXT_SIG_MODE_PAM4) { + req->force_pam4_link_speed = cpu_to_le16(bp->link_info.req_link_speed); + req->enables |= cpu_to_le32(PORT_PHY_CFG_REQ_ENABLES_FORCE_PAM4_LINK_SPEED); + } else { + req->force_link_speed = cpu_to_le16(bp->link_info.req_link_speed); + } } /* tell chimp that the setting takes effect immediately */ @@ -9424,14 +9642,19 @@ static int bnxt_update_phy_setting(struct bnxt *bp) if (!(link_info->autoneg & BNXT_AUTONEG_SPEED)) { if (BNXT_AUTO_MODE(link_info->auto_mode)) update_link = true; - if (link_info->req_link_speed != link_info->force_link_speed) + if (link_info->req_signal_mode == BNXT_SIG_MODE_NRZ && + link_info->req_link_speed != link_info->force_link_speed) + update_link = true; + else if (link_info->req_signal_mode == BNXT_SIG_MODE_PAM4 && + link_info->req_link_speed != link_info->force_pam4_link_speed) update_link = true; if (link_info->req_duplex != link_info->duplex_setting) update_link = true; } else { if (link_info->auto_mode == BNXT_LINK_AUTO_NONE) update_link = true; - if (link_info->advertising != link_info->auto_link_speeds) + if (link_info->advertising != link_info->auto_link_speeds || + link_info->advertising_pam4 != link_info->auto_pam4_link_speeds) update_link = true; } @@ -10362,6 +10585,23 @@ static void bnxt_dbg_dump_states(struct bnxt *bp) } } +static int bnxt_hwrm_rx_ring_reset(struct bnxt *bp, int ring_nr) +{ + struct bnxt_rx_ring_info *rxr = &bp->rx_ring[ring_nr]; + struct hwrm_ring_reset_input req = {0}; + struct bnxt_napi *bnapi = rxr->bnapi; + struct bnxt_cp_ring_info *cpr; + u16 cp_ring_id; + + cpr = &bnapi->cp_ring; + cp_ring_id = cpr->cp_ring_struct.fw_ring_id; + bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_RING_RESET, cp_ring_id, -1); + req.ring_type = RING_RESET_REQ_RING_TYPE_RX_RING_GRP; + req.ring_id = cpu_to_le16(bp->grp_info[bnapi->index].fw_grp_id); + return hwrm_send_message_silent(bp, &req, sizeof(req), + HWRM_CMD_TIMEOUT); +} + static void bnxt_reset_task(struct bnxt *bp, bool silent) { if (!silent) @@ -10497,6 +10737,55 @@ static void bnxt_reset(struct bnxt *bp, bool silent) bnxt_rtnl_unlock_sp(bp); } +/* Only called from bnxt_sp_task() */ +static void bnxt_rx_ring_reset(struct bnxt *bp) +{ + int i; + + bnxt_rtnl_lock_sp(bp); + if (!test_bit(BNXT_STATE_OPEN, &bp->state)) { + bnxt_rtnl_unlock_sp(bp); + return; + } + /* Disable and flush TPA before resetting the RX ring */ + if (bp->flags & BNXT_FLAG_TPA) + bnxt_set_tpa(bp, false); + for (i = 0; i < bp->rx_nr_rings; i++) { + struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i]; + struct bnxt_cp_ring_info *cpr; + int rc; + + if (!rxr->bnapi->in_reset) + continue; + + rc = bnxt_hwrm_rx_ring_reset(bp, i); + if (rc) { + if (rc == -EINVAL || rc == -EOPNOTSUPP) + netdev_info_once(bp->dev, "RX ring reset not supported by firmware, falling back to global reset\n"); + else + netdev_warn(bp->dev, "RX ring reset failed, rc = %d, falling back to global reset\n", + rc); + bnxt_reset_task(bp, true); + break; + } + bnxt_free_one_rx_ring_skbs(bp, i); + rxr->rx_prod = 0; + rxr->rx_agg_prod = 0; + rxr->rx_sw_agg_prod = 0; + rxr->rx_next_cons = 0; + rxr->bnapi->in_reset = false; + bnxt_alloc_one_rx_ring(bp, i); + cpr = &rxr->bnapi->cp_ring; + cpr->sw_stats.rx.rx_resets++; + if (bp->flags & BNXT_FLAG_AGG_RINGS) + bnxt_db_write(bp, &rxr->rx_agg_db, rxr->rx_agg_prod); + bnxt_db_write(bp, &rxr->rx_db, rxr->rx_prod); + } + if (bp->flags & BNXT_FLAG_TPA) + bnxt_set_tpa(bp, true); + bnxt_rtnl_unlock_sp(bp); +} + static void bnxt_fw_reset_close(struct bnxt *bp) { bnxt_ulp_stop(bp); @@ -10691,8 +10980,15 @@ static void bnxt_init_ethtool_link_settings(struct bnxt *bp) link_info->autoneg |= BNXT_AUTONEG_FLOW_CTRL; } link_info->advertising = link_info->auto_link_speeds; + link_info->advertising_pam4 = link_info->auto_pam4_link_speeds; } else { link_info->req_link_speed = link_info->force_link_speed; + link_info->req_signal_mode = BNXT_SIG_MODE_NRZ; + if (link_info->force_pam4_link_speed) { + link_info->req_link_speed = + link_info->force_pam4_link_speed; + link_info->req_signal_mode = BNXT_SIG_MODE_PAM4; + } link_info->req_duplex = link_info->duplex_setting; } if (link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL) @@ -10778,6 +11074,9 @@ static void bnxt_sp_task(struct work_struct *work) if (test_and_clear_bit(BNXT_RESET_TASK_SILENT_SP_EVENT, &bp->sp_event)) bnxt_reset(bp, true); + if (test_and_clear_bit(BNXT_RST_RING_SP_EVENT, &bp->sp_event)) + bnxt_rx_ring_reset(bp); + if (test_and_clear_bit(BNXT_FW_RESET_NOTIFY_SP_EVENT, &bp->sp_event)) bnxt_devlink_health_report(bp, BNXT_FW_RESET_NOTIFY_SP_EVENT); @@ -10882,21 +11181,19 @@ static void bnxt_init_dflt_coal(struct bnxt *bp) bp->stats_coal_ticks = BNXT_DEF_STATS_COAL_TICKS; } -static void bnxt_alloc_fw_health(struct bnxt *bp) +static int bnxt_fw_reset_via_optee(struct bnxt *bp) { - if (bp->fw_health) - return; +#ifdef CONFIG_TEE_BNXT_FW + int rc = tee_bnxt_fw_load(); - if (!(bp->fw_cap & BNXT_FW_CAP_HOT_RESET) && - !(bp->fw_cap & BNXT_FW_CAP_ERROR_RECOVERY)) - return; + if (rc) + netdev_err(bp->dev, "Failed FW reset via OP-TEE, rc=%d\n", rc); - bp->fw_health = kzalloc(sizeof(*bp->fw_health), GFP_KERNEL); - if (!bp->fw_health) { - netdev_warn(bp->dev, "Failed to allocate fw_health\n"); - bp->fw_cap &= ~BNXT_FW_CAP_HOT_RESET; - bp->fw_cap &= ~BNXT_FW_CAP_ERROR_RECOVERY; - } + return rc; +#else + netdev_err(bp->dev, "OP-TEE not supported\n"); + return -ENODEV; +#endif } static int bnxt_fw_init_one_p1(struct bnxt *bp) @@ -10905,8 +11202,24 @@ static int bnxt_fw_init_one_p1(struct bnxt *bp) bp->fw_cap = 0; rc = bnxt_hwrm_ver_get(bp); - if (rc) - return rc; + bnxt_try_map_fw_health_reg(bp); + if (rc) { + if (bp->fw_health && bp->fw_health->status_reliable) { + u32 sts = bnxt_fw_health_readl(bp, BNXT_FW_HEALTH_REG); + + netdev_err(bp->dev, + "Firmware not responding, status: 0x%x\n", + sts); + if (sts & FW_STATUS_REG_CRASHED_NO_MASTER) { + netdev_warn(bp->dev, "Firmware recover via OP-TEE requested\n"); + rc = bnxt_fw_reset_via_optee(bp); + if (!rc) + rc = bnxt_hwrm_ver_get(bp); + } + } + if (rc) + return rc; + } if (bp->fw_cap & BNXT_FW_CAP_KONG_MB_CHNL) { rc = bnxt_alloc_kong_hwrm_resources(bp); @@ -10920,6 +11233,8 @@ static int bnxt_fw_init_one_p1(struct bnxt *bp) if (rc) return rc; } + bnxt_nvm_cfg_ver_get(bp); + rc = bnxt_hwrm_func_reset(bp); if (rc) return -ENODEV; @@ -10945,11 +11260,14 @@ static int bnxt_fw_init_one_p2(struct bnxt *bp) netdev_warn(bp->dev, "hwrm query adv flow mgnt failure rc: %d\n", rc); - bnxt_alloc_fw_health(bp); - rc = bnxt_hwrm_error_recovery_qcfg(bp); - if (rc) - netdev_warn(bp->dev, "hwrm query error recovery failure rc: %d\n", - rc); + if (bnxt_alloc_fw_health(bp)) { + netdev_warn(bp->dev, "no memory for firmware error recovery\n"); + } else { + rc = bnxt_hwrm_error_recovery_qcfg(bp); + if (rc) + netdev_warn(bp->dev, "hwrm query error recovery failure rc: %d\n", + rc); + } rc = bnxt_hwrm_func_drv_rgtr(bp, NULL, 0, false); if (rc) @@ -11075,12 +11393,8 @@ static void bnxt_reset_all(struct bnxt *bp) int i, rc; if (bp->fw_cap & BNXT_FW_CAP_ERR_RECOVER_RELOAD) { -#ifdef CONFIG_TEE_BNXT_FW - rc = tee_bnxt_fw_load(); - if (rc) - netdev_err(bp->dev, "Unable to reset FW rc=%d\n", rc); + bnxt_fw_reset_via_optee(bp); bp->fw_reset_timestamp = jiffies; -#endif return; } @@ -11199,7 +11513,7 @@ static void bnxt_fw_reset_task(struct work_struct *work) if (time_after(jiffies, bp->fw_reset_timestamp + (bp->fw_reset_max_dsecs * HZ / 10))) { netdev_err(bp->dev, "Firmware reset aborted\n"); - goto fw_reset_abort; + goto fw_reset_abort_status; } bnxt_queue_fw_reset_work(bp, HZ / 5); return; @@ -11233,6 +11547,13 @@ static void bnxt_fw_reset_task(struct work_struct *work) } return; +fw_reset_abort_status: + if (bp->fw_health->status_reliable || + (bp->fw_cap & BNXT_FW_CAP_ERROR_RECOVERY)) { + u32 sts = bnxt_fw_health_readl(bp, BNXT_FW_HEALTH_REG); + + netdev_err(bp->dev, "fw_health_status 0x%x\n", sts); + } fw_reset_abort: clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state); if (bp->fw_reset_state != BNXT_FW_RESET_STATE_POLL_VF) @@ -12203,6 +12524,7 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) return -ENOMEM; bp = netdev_priv(dev); + bp->msg_enable = BNXT_DEF_MSG_ENABLE; bnxt_set_max_func_irqs(bp, max_irqs); if (bnxt_vf_pciid(ent->driver_data)) @@ -12234,8 +12556,11 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (rc) goto init_err_pci_clean; - if (BNXT_CHIP_P5(bp)) + if (BNXT_CHIP_P5(bp)) { bp->flags |= BNXT_FLAG_CHIP_P5; + if (BNXT_CHIP_SR2(bp)) + bp->flags |= BNXT_FLAG_CHIP_SR2; + } rc = bnxt_alloc_rss_indir_tbl(bp); if (rc) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 0ef89dabfd61..21ef1c21f602 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -907,6 +907,7 @@ struct bnxt_rx_ring_info { struct bnxt_rx_sw_stats { u64 rx_l4_csum_errors; + u64 rx_resets; u64 rx_buf_errors; }; @@ -1142,50 +1143,6 @@ struct bnxt_ntuple_filter { #define BNXT_FLTR_UPDATE 1 }; -struct hwrm_port_phy_qcfg_output_compat { - __le16 error_code; - __le16 req_type; - __le16 seq_id; - __le16 resp_len; - u8 link; - u8 link_signal_mode; - __le16 link_speed; - u8 duplex_cfg; - u8 pause; - __le16 support_speeds; - __le16 force_link_speed; - u8 auto_mode; - u8 auto_pause; - __le16 auto_link_speed; - __le16 auto_link_speed_mask; - u8 wirespeed; - u8 lpbk; - u8 force_pause; - u8 module_status; - __le32 preemphasis; - u8 phy_maj; - u8 phy_min; - u8 phy_bld; - u8 phy_type; - u8 media_type; - u8 xcvr_pkg_type; - u8 eee_config_phy_addr; - u8 parallel_detect; - __le16 link_partner_adv_speeds; - u8 link_partner_adv_auto_mode; - u8 link_partner_adv_pause; - __le16 adv_eee_link_speed_mask; - __le16 link_partner_adv_eee_link_speed_mask; - __le32 xcvr_identifier_type_tx_lpi_timer; - __le16 fec_cfg; - u8 duplex_state; - u8 option_flags; - char phy_vendor_name[16]; - char phy_vendor_partnumber[16]; - u8 unused_0[7]; - u8 valid; -}; - struct bnxt_link_info { u8 phy_type; u8 media_type; @@ -1196,7 +1153,10 @@ struct bnxt_link_info { #define BNXT_LINK_SIGNAL PORT_PHY_QCFG_RESP_LINK_SIGNAL #define BNXT_LINK_LINK PORT_PHY_QCFG_RESP_LINK_LINK u8 wire_speed; - u8 loop_back; + u8 phy_state; +#define BNXT_PHY_STATE_ENABLED 0 +#define BNXT_PHY_STATE_DISABLED 1 + u8 link_up; u8 duplex; #define BNXT_LINK_DUPLEX_HALF PORT_PHY_QCFG_RESP_DUPLEX_STATE_HALF @@ -1232,6 +1192,7 @@ struct bnxt_link_info { #define BNXT_LINK_SPEED_50GB PORT_PHY_QCFG_RESP_LINK_SPEED_50GB #define BNXT_LINK_SPEED_100GB PORT_PHY_QCFG_RESP_LINK_SPEED_100GB u16 support_speeds; + u16 support_pam4_speeds; u16 auto_link_speeds; /* fw adv setting */ #define BNXT_LINK_SPEED_MSK_100MB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_100MB #define BNXT_LINK_SPEED_MSK_1GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_1GB @@ -1243,24 +1204,51 @@ struct bnxt_link_info { #define BNXT_LINK_SPEED_MSK_40GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_40GB #define BNXT_LINK_SPEED_MSK_50GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_50GB #define BNXT_LINK_SPEED_MSK_100GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_100GB + u16 auto_pam4_link_speeds; +#define BNXT_LINK_PAM4_SPEED_MSK_50GB PORT_PHY_QCFG_RESP_SUPPORT_PAM4_SPEEDS_50G +#define BNXT_LINK_PAM4_SPEED_MSK_100GB PORT_PHY_QCFG_RESP_SUPPORT_PAM4_SPEEDS_100G +#define BNXT_LINK_PAM4_SPEED_MSK_200GB PORT_PHY_QCFG_RESP_SUPPORT_PAM4_SPEEDS_200G u16 support_auto_speeds; + u16 support_pam4_auto_speeds; u16 lp_auto_link_speeds; + u16 lp_auto_pam4_link_speeds; u16 force_link_speed; + u16 force_pam4_link_speed; u32 preemphasis; u8 module_status; + u8 active_fec_sig_mode; u16 fec_cfg; +#define BNXT_FEC_NONE PORT_PHY_QCFG_RESP_FEC_CFG_FEC_NONE_SUPPORTED +#define BNXT_FEC_AUTONEG_CAP PORT_PHY_QCFG_RESP_FEC_CFG_FEC_AUTONEG_SUPPORTED #define BNXT_FEC_AUTONEG PORT_PHY_QCFG_RESP_FEC_CFG_FEC_AUTONEG_ENABLED +#define BNXT_FEC_ENC_BASE_R_CAP \ + PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE74_SUPPORTED #define BNXT_FEC_ENC_BASE_R PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE74_ENABLED -#define BNXT_FEC_ENC_RS PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_ENABLED +#define BNXT_FEC_ENC_RS_CAP \ + PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_SUPPORTED +#define BNXT_FEC_ENC_LLRS_CAP \ + (PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS272_1XN_SUPPORTED | \ + PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS272_IEEE_SUPPORTED) +#define BNXT_FEC_ENC_RS \ + (PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_ENABLED | \ + PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS544_1XN_ENABLED | \ + PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS544_IEEE_ENABLED) +#define BNXT_FEC_ENC_LLRS \ + (PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS272_1XN_ENABLED | \ + PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS272_IEEE_ENABLED) /* copy of requested setting from ethtool cmd */ u8 autoneg; #define BNXT_AUTONEG_SPEED 1 #define BNXT_AUTONEG_FLOW_CTRL 2 + u8 req_signal_mode; +#define BNXT_SIG_MODE_NRZ PORT_PHY_QCFG_RESP_SIGNAL_MODE_NRZ +#define BNXT_SIG_MODE_PAM4 PORT_PHY_QCFG_RESP_SIGNAL_MODE_PAM4 u8 req_duplex; u8 req_flow_ctrl; u16 req_link_speed; u16 advertising; /* user adv setting */ + u16 advertising_pam4; bool force_link_chng; bool phy_retry; @@ -1272,6 +1260,49 @@ struct bnxt_link_info { struct hwrm_port_phy_qcfg_output phy_qcfg_resp; }; +#define BNXT_FEC_RS544_ON \ + (PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_1XN_ENABLE | \ + PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_IEEE_ENABLE) + +#define BNXT_FEC_RS544_OFF \ + (PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_1XN_DISABLE | \ + PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_IEEE_DISABLE) + +#define BNXT_FEC_RS272_ON \ + (PORT_PHY_CFG_REQ_FLAGS_FEC_RS272_1XN_ENABLE | \ + PORT_PHY_CFG_REQ_FLAGS_FEC_RS272_IEEE_ENABLE) + +#define BNXT_FEC_RS272_OFF \ + (PORT_PHY_CFG_REQ_FLAGS_FEC_RS272_1XN_DISABLE | \ + PORT_PHY_CFG_REQ_FLAGS_FEC_RS272_IEEE_DISABLE) + +#define BNXT_PAM4_SUPPORTED(link_info) \ + ((link_info)->support_pam4_speeds) + +#define BNXT_FEC_RS_ON(link_info) \ + (PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_ENABLE | \ + PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_DISABLE | \ + (BNXT_PAM4_SUPPORTED(link_info) ? \ + (BNXT_FEC_RS544_ON | BNXT_FEC_RS272_OFF) : 0)) + +#define BNXT_FEC_LLRS_ON \ + (PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_ENABLE | \ + PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_DISABLE | \ + BNXT_FEC_RS272_ON | BNXT_FEC_RS544_OFF) + +#define BNXT_FEC_RS_OFF(link_info) \ + (PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_DISABLE | \ + (BNXT_PAM4_SUPPORTED(link_info) ? \ + (BNXT_FEC_RS544_OFF | BNXT_FEC_RS272_OFF) : 0)) + +#define BNXT_FEC_BASE_R_ON(link_info) \ + (PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_ENABLE | \ + BNXT_FEC_RS_OFF(link_info)) + +#define BNXT_FEC_ALL_OFF(link_info) \ + (PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_DISABLE | \ + BNXT_FEC_RS_OFF(link_info)) + #define BNXT_MAX_QUEUE 8 struct bnxt_queue_info { @@ -1464,6 +1495,7 @@ struct bnxt_fw_health { u8 enabled:1; u8 master:1; u8 fatal:1; + u8 status_reliable:1; u8 tmr_multiplier; u8 tmr_counter; u8 fw_reset_seq_cnt; @@ -1491,6 +1523,9 @@ struct bnxt_fw_reporter_ctx { #define BNXT_FW_HEALTH_WIN_BASE 0x3000 #define BNXT_FW_HEALTH_WIN_MAP_OFF 8 +#define BNXT_FW_HEALTH_WIN_OFF(reg) (BNXT_FW_HEALTH_WIN_BASE + \ + ((reg) & BNXT_GRC_OFFSET_MASK)) + #define BNXT_FW_STATUS_HEALTHY 0x8000 #define BNXT_FW_STATUS_SHUTDOWN 0x100000 @@ -1535,6 +1570,8 @@ struct bnxt { u8 chip_rev; +#define CHIP_NUM_58818 0xd818 + #define BNXT_CHIP_NUM_5730X(chip_num) \ ((chip_num) >= CHIP_NUM_57301 && \ (chip_num) <= CHIP_NUM_57304) @@ -1613,6 +1650,7 @@ struct bnxt { BNXT_FLAG_ROCEV2_CAP) #define BNXT_FLAG_NO_AGG_RINGS 0x20000 #define BNXT_FLAG_RX_PAGE_MODE 0x40000 + #define BNXT_FLAG_CHIP_SR2 0x80000 #define BNXT_FLAG_MULTI_HOST 0x100000 #define BNXT_FLAG_DSN_VALID 0x200000 #define BNXT_FLAG_DOUBLE_DB 0x400000 @@ -1630,20 +1668,27 @@ struct bnxt { #define BNXT_NPAR(bp) ((bp)->port_partition_type) #define BNXT_MH(bp) ((bp)->flags & BNXT_FLAG_MULTI_HOST) #define BNXT_SINGLE_PF(bp) (BNXT_PF(bp) && !BNXT_NPAR(bp) && !BNXT_MH(bp)) -#define BNXT_PHY_CFG_ABLE(bp) (BNXT_SINGLE_PF(bp) || \ - ((bp)->fw_cap & BNXT_FW_CAP_SHARED_PORT_CFG)) +#define BNXT_PHY_CFG_ABLE(bp) ((BNXT_SINGLE_PF(bp) || \ + ((bp)->fw_cap & BNXT_FW_CAP_SHARED_PORT_CFG)) && \ + (bp)->link_info.phy_state == BNXT_PHY_STATE_ENABLED) #define BNXT_CHIP_TYPE_NITRO_A0(bp) ((bp)->flags & BNXT_FLAG_CHIP_NITRO_A0) #define BNXT_RX_PAGE_MODE(bp) ((bp)->flags & BNXT_FLAG_RX_PAGE_MODE) #define BNXT_SUPPORTS_TPA(bp) (!BNXT_CHIP_TYPE_NITRO_A0(bp) && \ (!((bp)->flags & BNXT_FLAG_CHIP_P5) || \ (bp)->max_tpa_v2) && !is_kdump_kernel()) -/* Chip class phase 5 */ -#define BNXT_CHIP_P5(bp) \ +#define BNXT_CHIP_SR2(bp) \ + ((bp)->chip_num == CHIP_NUM_58818) + +#define BNXT_CHIP_P5_THOR(bp) \ ((bp)->chip_num == CHIP_NUM_57508 || \ (bp)->chip_num == CHIP_NUM_57504 || \ (bp)->chip_num == CHIP_NUM_57502) +/* Chip class phase 5 */ +#define BNXT_CHIP_P5(bp) \ + (BNXT_CHIP_P5_THOR(bp) || BNXT_CHIP_SR2(bp)) + /* Chip class phase 4.x */ #define BNXT_CHIP_P4(bp) \ (BNXT_CHIP_NUM_57X1X((bp)->chip_num) || \ @@ -1777,6 +1822,7 @@ struct bnxt { #define BNXT_FW_CAP_VLAN_TX_INSERT 0x02000000 #define BNXT_FW_CAP_EXT_HW_STATS_SUPPORTED 0x04000000 #define BNXT_FW_CAP_PORT_STATS_NO_RESET 0x10000000 + #define BNXT_FW_CAP_RING_MONITOR 0x40000000 #define BNXT_NEW_RM(bp) ((bp)->fw_cap & BNXT_FW_CAP_NEW_RM) u32 hwrm_spec_code; @@ -1810,6 +1856,7 @@ struct bnxt { #define PHY_VER_STR_LEN (FW_VER_STR_LEN - BC_HWRM_STR_LEN) char fw_ver_str[FW_VER_STR_LEN]; char hwrm_ver_supp[FW_VER_STR_LEN]; + char nvm_cfg_ver[FW_VER_STR_LEN]; u64 fw_ver_code; #define BNXT_FW_VER_CODE(maj, min, bld, rsv) \ ((u64)(maj) << 48 | (u64)(min) << 32 | (u64)(bld) << 16 | (rsv)) @@ -1935,6 +1982,20 @@ struct bnxt { struct device *hwmon_dev; }; +#define BNXT_NUM_RX_RING_STATS 8 +#define BNXT_NUM_TX_RING_STATS 8 +#define BNXT_NUM_TPA_RING_STATS 4 +#define BNXT_NUM_TPA_RING_STATS_P5 5 +#define BNXT_NUM_TPA_RING_STATS_P5_SR2 6 + +#define BNXT_RING_STATS_SIZE_P5 \ + ((BNXT_NUM_RX_RING_STATS + BNXT_NUM_TX_RING_STATS + \ + BNXT_NUM_TPA_RING_STATS_P5) * 8) + +#define BNXT_RING_STATS_SIZE_P5_SR2 \ + ((BNXT_NUM_RX_RING_STATS + BNXT_NUM_TX_RING_STATS + \ + BNXT_NUM_TPA_RING_STATS_P5_SR2) * 8) + #define BNXT_GET_RING_STATS64(sw, counter) \ (*((sw) + offsetof(struct ctx_hw_stats, counter) / 8)) @@ -2114,6 +2175,7 @@ int bnxt_get_avail_msix(struct bnxt *bp, int num); int bnxt_reserve_rings(struct bnxt *bp, bool irq_re_init); void bnxt_tx_disable(struct bnxt *bp); void bnxt_tx_enable(struct bnxt *bp); +int bnxt_update_link(struct bnxt *bp, bool chng_link_state); int bnxt_hwrm_set_pause(struct bnxt *); int bnxt_hwrm_set_link_setting(struct bnxt *, bool, bool); int bnxt_hwrm_alloc_wol_fltr(struct bnxt *bp); diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c index 3a854195d5b0..184b6d0513b2 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c @@ -17,15 +17,13 @@ #include "bnxt_ethtool.h" static int -bnxt_dl_flash_update(struct devlink *dl, const char *filename, - const char *region, struct netlink_ext_ack *extack) +bnxt_dl_flash_update(struct devlink *dl, + struct devlink_flash_update_params *params, + struct netlink_ext_ack *extack) { struct bnxt *bp = bnxt_get_bp_from_dl(dl); int rc; - if (region) - return -EOPNOTSUPP; - if (!BNXT_PF(bp)) { NL_SET_ERR_MSG_MOD(extack, "flash update not supported from a VF"); @@ -33,15 +31,12 @@ bnxt_dl_flash_update(struct devlink *dl, const char *filename, } devlink_flash_update_begin_notify(dl); - devlink_flash_update_status_notify(dl, "Preparing to flash", region, 0, - 0); - rc = bnxt_flash_package_from_file(bp->dev, filename, 0); + devlink_flash_update_status_notify(dl, "Preparing to flash", NULL, 0, 0); + rc = bnxt_flash_package_from_file(bp->dev, params->file_name, 0); if (!rc) - devlink_flash_update_status_notify(dl, "Flashing done", region, - 0, 0); + devlink_flash_update_status_notify(dl, "Flashing done", NULL, 0, 0); else - devlink_flash_update_status_notify(dl, "Flashing failed", - region, 0, 0); + devlink_flash_update_status_notify(dl, "Flashing failed", NULL, 0, 0); devlink_flash_update_end_notify(dl); return rc; } @@ -387,15 +382,41 @@ static int bnxt_hwrm_get_nvm_cfg_ver(struct bnxt *bp, return rc; } +static int bnxt_dl_info_put(struct bnxt *bp, struct devlink_info_req *req, + enum bnxt_dl_version_type type, const char *key, + char *buf) +{ + if (!strlen(buf)) + return 0; + + if ((bp->flags & BNXT_FLAG_CHIP_P5) && + (!strcmp(key, DEVLINK_INFO_VERSION_GENERIC_FW_NCSI) || + !strcmp(key, DEVLINK_INFO_VERSION_GENERIC_FW_ROCE))) + return 0; + + switch (type) { + case BNXT_VERSION_FIXED: + return devlink_info_version_fixed_put(req, key, buf); + case BNXT_VERSION_RUNNING: + return devlink_info_version_running_put(req, key, buf); + case BNXT_VERSION_STORED: + return devlink_info_version_stored_put(req, key, buf); + } + return 0; +} + +#define HWRM_FW_VER_STR_LEN 16 + static int bnxt_dl_info_get(struct devlink *dl, struct devlink_info_req *req, struct netlink_ext_ack *extack) { + struct hwrm_nvm_get_dev_info_output nvm_dev_info; struct bnxt *bp = bnxt_get_bp_from_dl(dl); union devlink_param_value nvm_cfg_ver; struct hwrm_ver_get_output *ver_resp; char mgmt_ver[FW_VER_STR_LEN]; char roce_ver[FW_VER_STR_LEN]; - char fw_ver[FW_VER_STR_LEN]; + char ncsi_ver[FW_VER_STR_LEN]; char buf[32]; int rc; @@ -403,10 +424,11 @@ static int bnxt_dl_info_get(struct devlink *dl, struct devlink_info_req *req, if (rc) return rc; - if (strlen(bp->board_partno)) { - rc = devlink_info_version_fixed_put(req, - DEVLINK_INFO_VERSION_GENERIC_BOARD_ID, - bp->board_partno); + if (BNXT_PF(bp) && (bp->flags & BNXT_FLAG_DSN_VALID)) { + sprintf(buf, "%02X-%02X-%02X-%02X-%02X-%02X-%02X-%02X", + bp->dsn[7], bp->dsn[6], bp->dsn[5], bp->dsn[4], + bp->dsn[3], bp->dsn[2], bp->dsn[1], bp->dsn[0]); + rc = devlink_info_serial_number_put(req, buf); if (rc) return rc; } @@ -417,54 +439,56 @@ static int bnxt_dl_info_get(struct devlink *dl, struct devlink_info_req *req, return rc; } + rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_FIXED, + DEVLINK_INFO_VERSION_GENERIC_BOARD_ID, + bp->board_partno); + if (rc) + return rc; + sprintf(buf, "%X", bp->chip_num); - rc = devlink_info_version_fixed_put(req, - DEVLINK_INFO_VERSION_GENERIC_ASIC_ID, buf); + rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_FIXED, + DEVLINK_INFO_VERSION_GENERIC_ASIC_ID, buf); if (rc) return rc; ver_resp = &bp->ver_resp; sprintf(buf, "%X", ver_resp->chip_rev); - rc = devlink_info_version_fixed_put(req, - DEVLINK_INFO_VERSION_GENERIC_ASIC_REV, buf); + rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_FIXED, + DEVLINK_INFO_VERSION_GENERIC_ASIC_REV, buf); if (rc) return rc; - if (BNXT_PF(bp)) { - sprintf(buf, "%02X-%02X-%02X-%02X-%02X-%02X-%02X-%02X", - bp->dsn[7], bp->dsn[6], bp->dsn[5], bp->dsn[4], - bp->dsn[3], bp->dsn[2], bp->dsn[1], bp->dsn[0]); - rc = devlink_info_serial_number_put(req, buf); - if (rc) - return rc; - } + rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_RUNNING, + DEVLINK_INFO_VERSION_GENERIC_FW_PSID, + bp->nvm_cfg_ver); + if (rc) + return rc; - if (strlen(ver_resp->active_pkg_name)) { - rc = - devlink_info_version_running_put(req, - DEVLINK_INFO_VERSION_GENERIC_FW, - ver_resp->active_pkg_name); - if (rc) - return rc; - } + buf[0] = 0; + strncat(buf, ver_resp->active_pkg_name, HWRM_FW_VER_STR_LEN); + rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_RUNNING, + DEVLINK_INFO_VERSION_GENERIC_FW, buf); + if (rc) + return rc; if (BNXT_PF(bp) && !bnxt_hwrm_get_nvm_cfg_ver(bp, &nvm_cfg_ver)) { u32 ver = nvm_cfg_ver.vu32; sprintf(buf, "%X.%X.%X", (ver >> 16) & 0xF, (ver >> 8) & 0xF, ver & 0xF); - rc = devlink_info_version_running_put(req, - DEVLINK_INFO_VERSION_GENERIC_FW_PSID, buf); + rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_STORED, + DEVLINK_INFO_VERSION_GENERIC_FW_PSID, + buf); if (rc) return rc; } if (ver_resp->flags & VER_GET_RESP_FLAGS_EXT_VER_AVAIL) { - snprintf(fw_ver, FW_VER_STR_LEN, "%d.%d.%d.%d", + snprintf(mgmt_ver, FW_VER_STR_LEN, "%d.%d.%d.%d", ver_resp->hwrm_fw_major, ver_resp->hwrm_fw_minor, ver_resp->hwrm_fw_build, ver_resp->hwrm_fw_patch); - snprintf(mgmt_ver, FW_VER_STR_LEN, "%d.%d.%d.%d", + snprintf(ncsi_ver, FW_VER_STR_LEN, "%d.%d.%d.%d", ver_resp->mgmt_fw_major, ver_resp->mgmt_fw_minor, ver_resp->mgmt_fw_build, ver_resp->mgmt_fw_patch); @@ -472,11 +496,11 @@ static int bnxt_dl_info_get(struct devlink *dl, struct devlink_info_req *req, ver_resp->roce_fw_major, ver_resp->roce_fw_minor, ver_resp->roce_fw_build, ver_resp->roce_fw_patch); } else { - snprintf(fw_ver, FW_VER_STR_LEN, "%d.%d.%d.%d", + snprintf(mgmt_ver, FW_VER_STR_LEN, "%d.%d.%d.%d", ver_resp->hwrm_fw_maj_8b, ver_resp->hwrm_fw_min_8b, ver_resp->hwrm_fw_bld_8b, ver_resp->hwrm_fw_rsvd_8b); - snprintf(mgmt_ver, FW_VER_STR_LEN, "%d.%d.%d.%d", + snprintf(ncsi_ver, FW_VER_STR_LEN, "%d.%d.%d.%d", ver_resp->mgmt_fw_maj_8b, ver_resp->mgmt_fw_min_8b, ver_resp->mgmt_fw_bld_8b, ver_resp->mgmt_fw_rsvd_8b); @@ -484,29 +508,60 @@ static int bnxt_dl_info_get(struct devlink *dl, struct devlink_info_req *req, ver_resp->roce_fw_maj_8b, ver_resp->roce_fw_min_8b, ver_resp->roce_fw_bld_8b, ver_resp->roce_fw_rsvd_8b); } - rc = devlink_info_version_running_put(req, - DEVLINK_INFO_VERSION_GENERIC_FW_MGMT, fw_ver); + rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_RUNNING, + DEVLINK_INFO_VERSION_GENERIC_FW_MGMT, mgmt_ver); if (rc) return rc; - rc = devlink_info_version_running_put(req, - DEVLINK_INFO_VERSION_GENERIC_FW_MGMT_API, - bp->hwrm_ver_supp); + rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_RUNNING, + DEVLINK_INFO_VERSION_GENERIC_FW_MGMT_API, + bp->hwrm_ver_supp); if (rc) return rc; - if (!(bp->flags & BNXT_FLAG_CHIP_P5)) { - rc = devlink_info_version_running_put(req, - DEVLINK_INFO_VERSION_GENERIC_FW_NCSI, mgmt_ver); - if (rc) - return rc; + rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_RUNNING, + DEVLINK_INFO_VERSION_GENERIC_FW_NCSI, ncsi_ver); + if (rc) + return rc; - rc = devlink_info_version_running_put(req, - DEVLINK_INFO_VERSION_GENERIC_FW_ROCE, roce_ver); - if (rc) - return rc; - } - return 0; + rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_RUNNING, + DEVLINK_INFO_VERSION_GENERIC_FW_ROCE, roce_ver); + if (rc) + return rc; + + rc = bnxt_hwrm_nvm_get_dev_info(bp, &nvm_dev_info); + if (rc || + !(nvm_dev_info.flags & NVM_GET_DEV_INFO_RESP_FLAGS_FW_VER_VALID)) + return 0; + + buf[0] = 0; + strncat(buf, nvm_dev_info.pkg_name, HWRM_FW_VER_STR_LEN); + rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_STORED, + DEVLINK_INFO_VERSION_GENERIC_FW, buf); + if (rc) + return rc; + + snprintf(mgmt_ver, FW_VER_STR_LEN, "%d.%d.%d.%d", + nvm_dev_info.hwrm_fw_major, nvm_dev_info.hwrm_fw_minor, + nvm_dev_info.hwrm_fw_build, nvm_dev_info.hwrm_fw_patch); + rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_STORED, + DEVLINK_INFO_VERSION_GENERIC_FW_MGMT, mgmt_ver); + if (rc) + return rc; + + snprintf(ncsi_ver, FW_VER_STR_LEN, "%d.%d.%d.%d", + nvm_dev_info.mgmt_fw_major, nvm_dev_info.mgmt_fw_minor, + nvm_dev_info.mgmt_fw_build, nvm_dev_info.mgmt_fw_patch); + rc = bnxt_dl_info_put(bp, req, BNXT_VERSION_STORED, + DEVLINK_INFO_VERSION_GENERIC_FW_NCSI, ncsi_ver); + if (rc) + return rc; + + snprintf(roce_ver, FW_VER_STR_LEN, "%d.%d.%d.%d", + nvm_dev_info.roce_fw_major, nvm_dev_info.roce_fw_minor, + nvm_dev_info.roce_fw_build, nvm_dev_info.roce_fw_patch); + return bnxt_dl_info_put(bp, req, BNXT_VERSION_STORED, + DEVLINK_INFO_VERSION_GENERIC_FW_ROCE, roce_ver); } static int bnxt_hwrm_nvm_req(struct bnxt *bp, u32 param_id, void *msg, diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h index d5c8bd49383a..d22cab5d6856 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.h @@ -60,6 +60,12 @@ struct bnxt_dl_nvm_param { u8 dl_num_bytes; }; +enum bnxt_dl_version_type { + BNXT_VERSION_FIXED, + BNXT_VERSION_RUNNING, + BNXT_VERSION_STORED, +}; + void bnxt_devlink_health_report(struct bnxt *bp, unsigned long event); void bnxt_dl_health_status_update(struct bnxt *bp, bool healthy); void bnxt_dl_health_recovery_done(struct bnxt *bp); diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index fecdfd875af1..53687bc7fcf5 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -11,6 +11,7 @@ #include <linux/ctype.h> #include <linux/stringify.h> #include <linux/ethtool.h> +#include <linux/linkmode.h> #include <linux/interrupt.h> #include <linux/pci.h> #include <linux/etherdevice.h> @@ -172,10 +173,12 @@ static const char * const bnxt_ring_tpa2_stats_str[] = { "rx_tpa_pkt", "rx_tpa_bytes", "rx_tpa_errors", + "rx_tpa_events", }; static const char * const bnxt_rx_sw_stats_str[] = { "rx_l4_csum_errors", + "rx_resets", "rx_buf_errors", }; @@ -462,9 +465,12 @@ static const struct { static int bnxt_get_num_tpa_ring_stats(struct bnxt *bp) { if (BNXT_SUPPORTS_TPA(bp)) { - if (bp->max_tpa_v2) - return ARRAY_SIZE(bnxt_ring_tpa2_stats_str); - return ARRAY_SIZE(bnxt_ring_tpa_stats_str); + if (bp->max_tpa_v2) { + if (BNXT_CHIP_P5_THOR(bp)) + return BNXT_NUM_TPA_RING_STATS_P5; + return BNXT_NUM_TPA_RING_STATS_P5_SR2; + } + return BNXT_NUM_TPA_RING_STATS; } return 0; } @@ -796,7 +802,7 @@ static void bnxt_get_channels(struct net_device *dev, struct bnxt *bp = netdev_priv(dev); struct bnxt_hw_resc *hw_resc = &bp->hw_resc; int max_rx_rings, max_tx_rings, tcs; - int max_tx_sch_inputs; + int max_tx_sch_inputs, tx_grps; /* Get the most up-to-date max_tx_sch_inputs. */ if (netif_running(dev) && BNXT_NEW_RM(bp)) @@ -806,6 +812,12 @@ static void bnxt_get_channels(struct net_device *dev, bnxt_get_max_rings(bp, &max_rx_rings, &max_tx_rings, true); if (max_tx_sch_inputs) max_tx_rings = min_t(int, max_tx_rings, max_tx_sch_inputs); + + tcs = netdev_get_num_tc(dev); + tx_grps = max(tcs, 1); + if (bp->tx_nr_rings_xdp) + tx_grps++; + max_tx_rings /= tx_grps; channel->max_combined = min_t(int, max_rx_rings, max_tx_rings); if (bnxt_get_max_rings(bp, &max_rx_rings, &max_tx_rings, false)) { @@ -815,7 +827,6 @@ static void bnxt_get_channels(struct net_device *dev, if (max_tx_sch_inputs) max_tx_rings = min_t(int, max_tx_rings, max_tx_sch_inputs); - tcs = netdev_get_num_tc(dev); if (tcs > 1) max_tx_rings /= tcs; @@ -1503,6 +1514,53 @@ u32 _bnxt_fw_to_ethtool_adv_spds(u16 fw_speeds, u8 fw_pause) (fw_speeds) |= BNXT_LINK_SPEED_MSK_100GB; \ } +#define BNXT_FW_TO_ETHTOOL_PAM4_SPDS(fw_speeds, lk_ksettings, name) \ +{ \ + if ((fw_speeds) & BNXT_LINK_PAM4_SPEED_MSK_50GB) \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + 50000baseCR_Full); \ + if ((fw_speeds) & BNXT_LINK_PAM4_SPEED_MSK_100GB) \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + 100000baseCR2_Full);\ + if ((fw_speeds) & BNXT_LINK_PAM4_SPEED_MSK_200GB) \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + 200000baseCR4_Full);\ +} + +#define BNXT_ETHTOOL_TO_FW_PAM4_SPDS(fw_speeds, lk_ksettings, name) \ +{ \ + if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 50000baseCR_Full)) \ + (fw_speeds) |= BNXT_LINK_PAM4_SPEED_MSK_50GB; \ + if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 100000baseCR2_Full)) \ + (fw_speeds) |= BNXT_LINK_PAM4_SPEED_MSK_100GB; \ + if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 200000baseCR4_Full)) \ + (fw_speeds) |= BNXT_LINK_PAM4_SPEED_MSK_200GB; \ +} + +static void bnxt_fw_to_ethtool_advertised_fec(struct bnxt_link_info *link_info, + struct ethtool_link_ksettings *lk_ksettings) +{ + u16 fec_cfg = link_info->fec_cfg; + + if ((fec_cfg & BNXT_FEC_NONE) || !(fec_cfg & BNXT_FEC_AUTONEG)) { + linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_NONE_BIT, + lk_ksettings->link_modes.advertising); + return; + } + if (fec_cfg & BNXT_FEC_ENC_BASE_R) + linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_BASER_BIT, + lk_ksettings->link_modes.advertising); + if (fec_cfg & BNXT_FEC_ENC_RS) + linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_RS_BIT, + lk_ksettings->link_modes.advertising); + if (fec_cfg & BNXT_FEC_ENC_LLRS) + linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_LLRS_BIT, + lk_ksettings->link_modes.advertising); +} + static void bnxt_fw_to_ethtool_advertised_spds(struct bnxt_link_info *link_info, struct ethtool_link_ksettings *lk_ksettings) { @@ -1513,6 +1571,9 @@ static void bnxt_fw_to_ethtool_advertised_spds(struct bnxt_link_info *link_info, fw_pause = link_info->auto_pause_setting; BNXT_FW_TO_ETHTOOL_SPDS(fw_speeds, fw_pause, lk_ksettings, advertising); + fw_speeds = link_info->advertising_pam4; + BNXT_FW_TO_ETHTOOL_PAM4_SPDS(fw_speeds, lk_ksettings, advertising); + bnxt_fw_to_ethtool_advertised_fec(link_info, lk_ksettings); } static void bnxt_fw_to_ethtool_lp_adv(struct bnxt_link_info *link_info, @@ -1526,6 +1587,29 @@ static void bnxt_fw_to_ethtool_lp_adv(struct bnxt_link_info *link_info, BNXT_FW_TO_ETHTOOL_SPDS(fw_speeds, fw_pause, lk_ksettings, lp_advertising); + fw_speeds = link_info->lp_auto_pam4_link_speeds; + BNXT_FW_TO_ETHTOOL_PAM4_SPDS(fw_speeds, lk_ksettings, lp_advertising); +} + +static void bnxt_fw_to_ethtool_support_fec(struct bnxt_link_info *link_info, + struct ethtool_link_ksettings *lk_ksettings) +{ + u16 fec_cfg = link_info->fec_cfg; + + if (fec_cfg & BNXT_FEC_NONE) { + linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_NONE_BIT, + lk_ksettings->link_modes.supported); + return; + } + if (fec_cfg & BNXT_FEC_ENC_BASE_R_CAP) + linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_BASER_BIT, + lk_ksettings->link_modes.supported); + if (fec_cfg & BNXT_FEC_ENC_RS_CAP) + linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_RS_BIT, + lk_ksettings->link_modes.supported); + if (fec_cfg & BNXT_FEC_ENC_LLRS_CAP) + linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_LLRS_BIT, + lk_ksettings->link_modes.supported); } static void bnxt_fw_to_ethtool_support_spds(struct bnxt_link_info *link_info, @@ -1534,14 +1618,18 @@ static void bnxt_fw_to_ethtool_support_spds(struct bnxt_link_info *link_info, u16 fw_speeds = link_info->support_speeds; BNXT_FW_TO_ETHTOOL_SPDS(fw_speeds, 0, lk_ksettings, supported); + fw_speeds = link_info->support_pam4_speeds; + BNXT_FW_TO_ETHTOOL_PAM4_SPDS(fw_speeds, lk_ksettings, supported); ethtool_link_ksettings_add_link_mode(lk_ksettings, supported, Pause); ethtool_link_ksettings_add_link_mode(lk_ksettings, supported, Asym_Pause); - if (link_info->support_auto_speeds) + if (link_info->support_auto_speeds || + link_info->support_pam4_auto_speeds) ethtool_link_ksettings_add_link_mode(lk_ksettings, supported, Autoneg); + bnxt_fw_to_ethtool_support_fec(link_info, lk_ksettings); } u32 bnxt_fw_to_ethtool_speed(u16 fw_link_speed) @@ -1632,55 +1720,86 @@ static int bnxt_get_link_ksettings(struct net_device *dev, return 0; } -static u32 bnxt_get_fw_speed(struct net_device *dev, u32 ethtool_speed) +static int bnxt_force_link_speed(struct net_device *dev, u32 ethtool_speed) { struct bnxt *bp = netdev_priv(dev); struct bnxt_link_info *link_info = &bp->link_info; + u16 support_pam4_spds = link_info->support_pam4_speeds; u16 support_spds = link_info->support_speeds; - u32 fw_speed = 0; + u8 sig_mode = BNXT_SIG_MODE_NRZ; + u16 fw_speed = 0; switch (ethtool_speed) { case SPEED_100: if (support_spds & BNXT_LINK_SPEED_MSK_100MB) - fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_100MB; + fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_100MB; break; case SPEED_1000: if (support_spds & BNXT_LINK_SPEED_MSK_1GB) - fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_1GB; + fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_1GB; break; case SPEED_2500: if (support_spds & BNXT_LINK_SPEED_MSK_2_5GB) - fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_2_5GB; + fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_2_5GB; break; case SPEED_10000: if (support_spds & BNXT_LINK_SPEED_MSK_10GB) - fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_10GB; + fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_10GB; break; case SPEED_20000: if (support_spds & BNXT_LINK_SPEED_MSK_20GB) - fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_20GB; + fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_20GB; break; case SPEED_25000: if (support_spds & BNXT_LINK_SPEED_MSK_25GB) - fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_25GB; + fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_25GB; break; case SPEED_40000: if (support_spds & BNXT_LINK_SPEED_MSK_40GB) - fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_40GB; + fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_40GB; break; case SPEED_50000: - if (support_spds & BNXT_LINK_SPEED_MSK_50GB) - fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_50GB; + if (support_spds & BNXT_LINK_SPEED_MSK_50GB) { + fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_50GB; + } else if (support_pam4_spds & BNXT_LINK_PAM4_SPEED_MSK_50GB) { + fw_speed = PORT_PHY_CFG_REQ_FORCE_PAM4_LINK_SPEED_50GB; + sig_mode = BNXT_SIG_MODE_PAM4; + } break; case SPEED_100000: - if (support_spds & BNXT_LINK_SPEED_MSK_100GB) - fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_100GB; + if (support_spds & BNXT_LINK_SPEED_MSK_100GB) { + fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_100GB; + } else if (support_pam4_spds & BNXT_LINK_PAM4_SPEED_MSK_100GB) { + fw_speed = PORT_PHY_CFG_REQ_FORCE_PAM4_LINK_SPEED_100GB; + sig_mode = BNXT_SIG_MODE_PAM4; + } break; - default: - netdev_err(dev, "unsupported speed!\n"); + case SPEED_200000: + if (support_pam4_spds & BNXT_LINK_PAM4_SPEED_MSK_200GB) { + fw_speed = PORT_PHY_CFG_REQ_FORCE_PAM4_LINK_SPEED_200GB; + sig_mode = BNXT_SIG_MODE_PAM4; + } break; } - return fw_speed; + + if (!fw_speed) { + netdev_err(dev, "unsupported speed!\n"); + return -EINVAL; + } + + if (link_info->req_link_speed == fw_speed && + link_info->req_signal_mode == sig_mode && + link_info->autoneg == 0) + return -EALREADY; + + link_info->req_link_speed = fw_speed; + link_info->req_signal_mode = sig_mode; + link_info->req_duplex = BNXT_LINK_DUPLEX_FULL; + link_info->autoneg = 0; + link_info->advertising = 0; + link_info->advertising_pam4 = 0; + + return 0; } u16 bnxt_get_fw_auto_link_speeds(u32 advertising) @@ -1712,7 +1831,6 @@ static int bnxt_set_link_ksettings(struct net_device *dev, struct bnxt_link_info *link_info = &bp->link_info; const struct ethtool_link_settings *base = &lk_ksettings->base; bool set_pause = false; - u16 fw_advertising = 0; u32 speed; int rc = 0; @@ -1721,19 +1839,23 @@ static int bnxt_set_link_ksettings(struct net_device *dev, mutex_lock(&bp->link_lock); if (base->autoneg == AUTONEG_ENABLE) { - BNXT_ETHTOOL_TO_FW_SPDS(fw_advertising, lk_ksettings, + link_info->advertising = 0; + link_info->advertising_pam4 = 0; + BNXT_ETHTOOL_TO_FW_SPDS(link_info->advertising, lk_ksettings, advertising); + BNXT_ETHTOOL_TO_FW_PAM4_SPDS(link_info->advertising_pam4, + lk_ksettings, advertising); link_info->autoneg |= BNXT_AUTONEG_SPEED; - if (!fw_advertising) + if (!link_info->advertising && !link_info->advertising_pam4) { link_info->advertising = link_info->support_auto_speeds; - else - link_info->advertising = fw_advertising; + link_info->advertising_pam4 = + link_info->support_pam4_auto_speeds; + } /* any change to autoneg will cause link change, therefore the * driver should put back the original pause setting in autoneg */ set_pause = true; } else { - u16 fw_speed; u8 phy_type = link_info->phy_type; if (phy_type == PORT_PHY_QCFG_RESP_PHY_TYPE_BASET || @@ -1749,15 +1871,12 @@ static int bnxt_set_link_ksettings(struct net_device *dev, goto set_setting_exit; } speed = base->speed; - fw_speed = bnxt_get_fw_speed(dev, speed); - if (!fw_speed) { - rc = -EINVAL; + rc = bnxt_force_link_speed(dev, speed); + if (rc) { + if (rc == -EALREADY) + rc = 0; goto set_setting_exit; } - link_info->req_link_speed = fw_speed; - link_info->req_duplex = BNXT_LINK_DUPLEX_FULL; - link_info->autoneg = 0; - link_info->advertising = 0; } if (netif_running(dev)) @@ -1768,6 +1887,110 @@ set_setting_exit: return rc; } +static int bnxt_get_fecparam(struct net_device *dev, + struct ethtool_fecparam *fec) +{ + struct bnxt *bp = netdev_priv(dev); + struct bnxt_link_info *link_info; + u8 active_fec; + u16 fec_cfg; + + link_info = &bp->link_info; + fec_cfg = link_info->fec_cfg; + active_fec = link_info->active_fec_sig_mode & + PORT_PHY_QCFG_RESP_ACTIVE_FEC_MASK; + if (fec_cfg & BNXT_FEC_NONE) { + fec->fec = ETHTOOL_FEC_NONE; + fec->active_fec = ETHTOOL_FEC_NONE; + return 0; + } + if (fec_cfg & BNXT_FEC_AUTONEG) + fec->fec |= ETHTOOL_FEC_AUTO; + if (fec_cfg & BNXT_FEC_ENC_BASE_R) + fec->fec |= ETHTOOL_FEC_BASER; + if (fec_cfg & BNXT_FEC_ENC_RS) + fec->fec |= ETHTOOL_FEC_RS; + if (fec_cfg & BNXT_FEC_ENC_LLRS) + fec->fec |= ETHTOOL_FEC_LLRS; + + switch (active_fec) { + case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_CLAUSE74_ACTIVE: + fec->active_fec |= ETHTOOL_FEC_BASER; + break; + case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_CLAUSE91_ACTIVE: + case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS544_1XN_ACTIVE: + case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS544_IEEE_ACTIVE: + fec->active_fec |= ETHTOOL_FEC_RS; + break; + case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS272_1XN_ACTIVE: + case PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS272_IEEE_ACTIVE: + fec->active_fec |= ETHTOOL_FEC_LLRS; + break; + } + return 0; +} + +static u32 bnxt_ethtool_forced_fec_to_fw(struct bnxt_link_info *link_info, + u32 fec) +{ + u32 fw_fec = PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_DISABLE; + + if (fec & ETHTOOL_FEC_BASER) + fw_fec |= BNXT_FEC_BASE_R_ON(link_info); + else if (fec & ETHTOOL_FEC_RS) + fw_fec |= BNXT_FEC_RS_ON(link_info); + else if (fec & ETHTOOL_FEC_LLRS) + fw_fec |= BNXT_FEC_LLRS_ON; + return fw_fec; +} + +static int bnxt_set_fecparam(struct net_device *dev, + struct ethtool_fecparam *fecparam) +{ + struct hwrm_port_phy_cfg_input req = {0}; + struct bnxt *bp = netdev_priv(dev); + struct bnxt_link_info *link_info; + u32 new_cfg, fec = fecparam->fec; + u16 fec_cfg; + int rc; + + link_info = &bp->link_info; + fec_cfg = link_info->fec_cfg; + if (fec_cfg & BNXT_FEC_NONE) + return -EOPNOTSUPP; + + if (fec & ETHTOOL_FEC_OFF) { + new_cfg = PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_DISABLE | + BNXT_FEC_ALL_OFF(link_info); + goto apply_fec; + } + if (((fec & ETHTOOL_FEC_AUTO) && !(fec_cfg & BNXT_FEC_AUTONEG_CAP)) || + ((fec & ETHTOOL_FEC_RS) && !(fec_cfg & BNXT_FEC_ENC_RS_CAP)) || + ((fec & ETHTOOL_FEC_LLRS) && !(fec_cfg & BNXT_FEC_ENC_LLRS_CAP)) || + ((fec & ETHTOOL_FEC_BASER) && !(fec_cfg & BNXT_FEC_ENC_BASE_R_CAP))) + return -EINVAL; + + if (fec & ETHTOOL_FEC_AUTO) { + if (!link_info->autoneg) + return -EINVAL; + new_cfg = PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_ENABLE; + } else { + new_cfg = bnxt_ethtool_forced_fec_to_fw(link_info, fec); + } + +apply_fec: + bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_PHY_CFG, -1, -1); + req.flags = cpu_to_le32(new_cfg | PORT_PHY_CFG_REQ_FLAGS_RESET_PHY); + rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); + /* update current settings */ + if (!rc) { + mutex_lock(&bp->link_lock); + bnxt_update_link(bp, false); + mutex_unlock(&bp->link_lock); + } + return rc; +} + static void bnxt_get_pauseparam(struct net_device *dev, struct ethtool_pauseparam *epause) { @@ -1781,6 +2004,22 @@ static void bnxt_get_pauseparam(struct net_device *dev, epause->tx_pause = !!(link_info->req_flow_ctrl & BNXT_LINK_PAUSE_TX); } +static void bnxt_get_pause_stats(struct net_device *dev, + struct ethtool_pause_stats *epstat) +{ + struct bnxt *bp = netdev_priv(dev); + u64 *rx, *tx; + + if (BNXT_VF(bp) || !(bp->flags & BNXT_FLAG_PORT_STATS)) + return; + + rx = bp->port_stats.sw_stats; + tx = bp->port_stats.sw_stats + BNXT_TX_PORT_STATS_BYTE_OFFSET / 8; + + epstat->rx_pause_frames = BNXT_GET_RX_PORT_STATS64(rx, rx_pause_frames); + epstat->tx_pause_frames = BNXT_GET_TX_PORT_STATS64(tx, tx_pause_frames); +} + static int bnxt_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam *epause) { @@ -1833,6 +2072,22 @@ static u32 bnxt_get_link(struct net_device *dev) return bp->link_info.link_up; } +int bnxt_hwrm_nvm_get_dev_info(struct bnxt *bp, + struct hwrm_nvm_get_dev_info_output *nvm_dev_info) +{ + struct hwrm_nvm_get_dev_info_output *resp = bp->hwrm_cmd_resp_addr; + struct hwrm_nvm_get_dev_info_input req = {0}; + int rc; + + bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_NVM_GET_DEV_INFO, -1, -1); + mutex_lock(&bp->hwrm_cmd_lock); + rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); + if (!rc) + memcpy(nvm_dev_info, resp, sizeof(*resp)); + mutex_unlock(&bp->hwrm_cmd_lock); + return rc; +} + static void bnxt_print_admin_err(struct bnxt *bp) { netdev_info(bp->dev, "PF does not have admin privileges to flash or reset the device\n"); @@ -3059,7 +3314,7 @@ static void bnxt_self_test(struct net_device *dev, struct ethtool_test *etest, u8 test_mask = 0; int rc = 0, i; - if (!bp->num_tests || !BNXT_SINGLE_PF(bp)) + if (!bp->num_tests || !BNXT_PF(bp)) return; memset(buf, 0, sizeof(u64) * bp->num_tests); if (!netif_running(dev)) { @@ -3072,9 +3327,9 @@ static void bnxt_self_test(struct net_device *dev, struct ethtool_test *etest, do_ext_lpbk = true; if (etest->flags & ETH_TEST_FL_OFFLINE) { - if (bp->pf.active_vfs) { + if (bp->pf.active_vfs || !BNXT_SINGLE_PF(bp)) { etest->flags |= ETH_TEST_FL_FAILED; - netdev_warn(dev, "Offline tests cannot be run with active VFs\n"); + netdev_warn(dev, "Offline tests cannot be run with active VFs or on shared PF\n"); return; } offline = true; @@ -3590,7 +3845,7 @@ void bnxt_ethtool_init(struct bnxt *bp) bnxt_get_pkgver(dev); bp->num_tests = 0; - if (bp->hwrm_spec_code < 0x10704 || !BNXT_SINGLE_PF(bp)) + if (bp->hwrm_spec_code < 0x10704 || !BNXT_PF(bp)) return; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_SELFTEST_QLIST, -1, -1); @@ -3657,6 +3912,9 @@ const struct ethtool_ops bnxt_ethtool_ops = { ETHTOOL_COALESCE_USE_ADAPTIVE_RX, .get_link_ksettings = bnxt_get_link_ksettings, .set_link_ksettings = bnxt_set_link_ksettings, + .get_fecparam = bnxt_get_fecparam, + .set_fecparam = bnxt_set_fecparam, + .get_pause_stats = bnxt_get_pause_stats, .get_pauseparam = bnxt_get_pauseparam, .set_pauseparam = bnxt_set_pauseparam, .get_drvinfo = bnxt_get_drvinfo, diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h index 34f44ddfad79..fa6fbde52bea 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h @@ -92,6 +92,8 @@ u32 bnxt_get_rxfh_indir_size(struct net_device *dev); u32 _bnxt_fw_to_ethtool_adv_spds(u16, u8); u32 bnxt_fw_to_ethtool_speed(u16); u16 bnxt_get_fw_auto_link_speeds(u32); +int bnxt_hwrm_nvm_get_dev_info(struct bnxt *bp, + struct hwrm_nvm_get_dev_info_output *nvm_dev_info); int bnxt_flash_package_from_file(struct net_device *dev, const char *filename, u32 install_type); void bnxt_ethtool_init(struct bnxt *bp); diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h index c4af6bf15e36..2d3e962bdac3 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h @@ -213,7 +213,10 @@ struct cmd_nums { #define HWRM_PORT_PHY_MDIO_BUS_ACQUIRE 0xb7UL #define HWRM_PORT_PHY_MDIO_BUS_RELEASE 0xb8UL #define HWRM_PORT_QSTATS_EXT_PFC_WD 0xb9UL - #define HWRM_PORT_ECN_QSTATS 0xbaUL + #define HWRM_RESERVED7 0xbaUL + #define HWRM_PORT_TX_FIR_CFG 0xbbUL + #define HWRM_PORT_TX_FIR_QCFG 0xbcUL + #define HWRM_PORT_ECN_QSTATS 0xbdUL #define HWRM_FW_RESET 0xc0UL #define HWRM_FW_QSTATUS 0xc1UL #define HWRM_FW_HEALTH_CHECK 0xc2UL @@ -370,6 +373,9 @@ struct cmd_nums { #define HWRM_TF_SESSION_RESC_FLUSH 0x2cfUL #define HWRM_TF_TBL_TYPE_GET 0x2daUL #define HWRM_TF_TBL_TYPE_SET 0x2dbUL + #define HWRM_TF_TBL_TYPE_BULK_GET 0x2dcUL + #define HWRM_TF_CTXT_MEM_ALLOC 0x2e2UL + #define HWRM_TF_CTXT_MEM_FREE 0x2e3UL #define HWRM_TF_CTXT_MEM_RGTR 0x2e4UL #define HWRM_TF_CTXT_MEM_UNRGTR 0x2e5UL #define HWRM_TF_EXT_EM_QCAPS 0x2e6UL @@ -384,6 +390,8 @@ struct cmd_nums { #define HWRM_TF_TCAM_FREE 0x2fbUL #define HWRM_TF_GLOBAL_CFG_SET 0x2fcUL #define HWRM_TF_GLOBAL_CFG_GET 0x2fdUL + #define HWRM_TF_IF_TBL_SET 0x2feUL + #define HWRM_TF_IF_TBL_GET 0x2ffUL #define HWRM_SV 0x400UL #define HWRM_DBG_READ_DIRECT 0xff10UL #define HWRM_DBG_READ_INDIRECT 0xff11UL @@ -447,6 +455,7 @@ struct ret_codes { #define HWRM_ERR_CODE_KEY_ALREADY_EXISTS 0xeUL #define HWRM_ERR_CODE_HWRM_ERROR 0xfUL #define HWRM_ERR_CODE_BUSY 0x10UL + #define HWRM_ERR_CODE_RESOURCE_LOCKED 0x11UL #define HWRM_ERR_CODE_TLV_ENCAPSULATED_RESPONSE 0x8000UL #define HWRM_ERR_CODE_UNKNOWN_ERR 0xfffeUL #define HWRM_ERR_CODE_CMD_NOT_SUPPORTED 0xffffUL @@ -478,8 +487,8 @@ struct hwrm_err_output { #define HWRM_VERSION_MAJOR 1 #define HWRM_VERSION_MINOR 10 #define HWRM_VERSION_UPDATE 1 -#define HWRM_VERSION_RSVD 54 -#define HWRM_VERSION_STR "1.10.1.54" +#define HWRM_VERSION_RSVD 68 +#define HWRM_VERSION_STR "1.10.1.68" /* hwrm_ver_get_input (size:192b/24B) */ struct hwrm_ver_get_input { @@ -675,6 +684,7 @@ struct hwrm_async_event_cmpl { #define ASYNC_EVENT_CMPL_EVENT_ID_PORT_PHY_CFG_CHANGE 0x7UL #define ASYNC_EVENT_CMPL_EVENT_ID_RESET_NOTIFY 0x8UL #define ASYNC_EVENT_CMPL_EVENT_ID_ERROR_RECOVERY 0x9UL + #define ASYNC_EVENT_CMPL_EVENT_ID_RING_MONITOR_MSG 0xaUL #define ASYNC_EVENT_CMPL_EVENT_ID_FUNC_DRVR_UNLOAD 0x10UL #define ASYNC_EVENT_CMPL_EVENT_ID_FUNC_DRVR_LOAD 0x11UL #define ASYNC_EVENT_CMPL_EVENT_ID_FUNC_FLR_PROC_CMPLT 0x12UL @@ -851,6 +861,32 @@ struct hwrm_async_event_cmpl_error_recovery { #define ASYNC_EVENT_CMPL_ERROR_RECOVERY_EVENT_DATA1_FLAGS_RECOVERY_ENABLED 0x2UL }; +/* hwrm_async_event_cmpl_ring_monitor_msg (size:128b/16B) */ +struct hwrm_async_event_cmpl_ring_monitor_msg { + __le16 type; + #define ASYNC_EVENT_CMPL_RING_MONITOR_MSG_TYPE_MASK 0x3fUL + #define ASYNC_EVENT_CMPL_RING_MONITOR_MSG_TYPE_SFT 0 + #define ASYNC_EVENT_CMPL_RING_MONITOR_MSG_TYPE_HWRM_ASYNC_EVENT 0x2eUL + #define ASYNC_EVENT_CMPL_RING_MONITOR_MSG_TYPE_LAST ASYNC_EVENT_CMPL_RING_MONITOR_MSG_TYPE_HWRM_ASYNC_EVENT + __le16 event_id; + #define ASYNC_EVENT_CMPL_RING_MONITOR_MSG_EVENT_ID_RING_MONITOR_MSG 0xaUL + #define ASYNC_EVENT_CMPL_RING_MONITOR_MSG_EVENT_ID_LAST ASYNC_EVENT_CMPL_RING_MONITOR_MSG_EVENT_ID_RING_MONITOR_MSG + __le32 event_data2; + #define ASYNC_EVENT_CMPL_RING_MONITOR_MSG_EVENT_DATA2_DISABLE_RING_TYPE_MASK 0xffUL + #define ASYNC_EVENT_CMPL_RING_MONITOR_MSG_EVENT_DATA2_DISABLE_RING_TYPE_SFT 0 + #define ASYNC_EVENT_CMPL_RING_MONITOR_MSG_EVENT_DATA2_DISABLE_RING_TYPE_TX 0x0UL + #define ASYNC_EVENT_CMPL_RING_MONITOR_MSG_EVENT_DATA2_DISABLE_RING_TYPE_RX 0x1UL + #define ASYNC_EVENT_CMPL_RING_MONITOR_MSG_EVENT_DATA2_DISABLE_RING_TYPE_CMPL 0x2UL + #define ASYNC_EVENT_CMPL_RING_MONITOR_MSG_EVENT_DATA2_DISABLE_RING_TYPE_LAST ASYNC_EVENT_CMPL_RING_MONITOR_MSG_EVENT_DATA2_DISABLE_RING_TYPE_CMPL + u8 opaque_v; + #define ASYNC_EVENT_CMPL_RING_MONITOR_MSG_V 0x1UL + #define ASYNC_EVENT_CMPL_RING_MONITOR_MSG_OPAQUE_MASK 0xfeUL + #define ASYNC_EVENT_CMPL_RING_MONITOR_MSG_OPAQUE_SFT 1 + u8 timestamp_lo; + __le16 timestamp_hi; + __le32 event_data1; +}; + /* hwrm_async_event_cmpl_vf_cfg_change (size:128b/16B) */ struct hwrm_async_event_cmpl_vf_cfg_change { __le16 type; @@ -975,6 +1011,28 @@ struct hwrm_async_event_cmpl_eem_cache_flush_done { #define ASYNC_EVENT_CMPL_EEM_CACHE_FLUSH_DONE_EVENT_DATA1_FID_SFT 0 }; +/* hwrm_async_event_cmpl_deferred_response (size:128b/16B) */ +struct hwrm_async_event_cmpl_deferred_response { + __le16 type; + #define ASYNC_EVENT_CMPL_DEFERRED_RESPONSE_TYPE_MASK 0x3fUL + #define ASYNC_EVENT_CMPL_DEFERRED_RESPONSE_TYPE_SFT 0 + #define ASYNC_EVENT_CMPL_DEFERRED_RESPONSE_TYPE_HWRM_ASYNC_EVENT 0x2eUL + #define ASYNC_EVENT_CMPL_DEFERRED_RESPONSE_TYPE_LAST ASYNC_EVENT_CMPL_DEFERRED_RESPONSE_TYPE_HWRM_ASYNC_EVENT + __le16 event_id; + #define ASYNC_EVENT_CMPL_DEFERRED_RESPONSE_EVENT_ID_DEFERRED_RESPONSE 0x40UL + #define ASYNC_EVENT_CMPL_DEFERRED_RESPONSE_EVENT_ID_LAST ASYNC_EVENT_CMPL_DEFERRED_RESPONSE_EVENT_ID_DEFERRED_RESPONSE + __le32 event_data2; + #define ASYNC_EVENT_CMPL_DEFERRED_RESPONSE_EVENT_DATA2_SEQ_ID_MASK 0xffffUL + #define ASYNC_EVENT_CMPL_DEFERRED_RESPONSE_EVENT_DATA2_SEQ_ID_SFT 0 + u8 opaque_v; + #define ASYNC_EVENT_CMPL_DEFERRED_RESPONSE_V 0x1UL + #define ASYNC_EVENT_CMPL_DEFERRED_RESPONSE_OPAQUE_MASK 0xfeUL + #define ASYNC_EVENT_CMPL_DEFERRED_RESPONSE_OPAQUE_SFT 1 + u8 timestamp_lo; + __le16 timestamp_hi; + __le32 event_data1; +}; + /* hwrm_func_reset_input (size:192b/24B) */ struct hwrm_func_reset_input { __le16 req_type; @@ -1214,7 +1272,13 @@ struct hwrm_func_qcaps_output { #define FUNC_QCAPS_RESP_FLAGS_EXT_SCHQ_SUPPORTED 0x40UL #define FUNC_QCAPS_RESP_FLAGS_EXT_PPP_PUSH_MODE_SUPPORTED 0x80UL u8 max_schqs; - u8 unused_1[2]; + u8 mpc_chnls_cap; + #define FUNC_QCAPS_RESP_MPC_CHNLS_CAP_TCE 0x1UL + #define FUNC_QCAPS_RESP_MPC_CHNLS_CAP_RCE 0x2UL + #define FUNC_QCAPS_RESP_MPC_CHNLS_CAP_TE_CFA 0x4UL + #define FUNC_QCAPS_RESP_MPC_CHNLS_CAP_RE_CFA 0x8UL + #define FUNC_QCAPS_RESP_MPC_CHNLS_CAP_PRIMATE 0x10UL + u8 unused_1; u8 valid; }; @@ -1250,6 +1314,7 @@ struct hwrm_func_qcfg_output { #define FUNC_QCFG_RESP_FLAGS_PREBOOT_LEGACY_L2_RINGS 0x100UL #define FUNC_QCFG_RESP_FLAGS_HOT_RESET_ALLOWED 0x200UL #define FUNC_QCFG_RESP_FLAGS_PPP_PUSH_MODE_ENABLED 0x400UL + #define FUNC_QCFG_RESP_FLAGS_RING_MONITOR_ENABLED 0x800UL u8 mac_address[6]; __le16 pci_id; __le16 alloc_rsscos_ctx; @@ -1341,7 +1406,13 @@ struct hwrm_func_qcfg_output { #define FUNC_QCFG_RESP_SVIF_INFO_SVIF_MASK 0x7fffUL #define FUNC_QCFG_RESP_SVIF_INFO_SVIF_SFT 0 #define FUNC_QCFG_RESP_SVIF_INFO_SVIF_VALID 0x8000UL - u8 unused_2[7]; + u8 mpc_chnls; + #define FUNC_QCFG_RESP_MPC_CHNLS_TCE_ENABLED 0x1UL + #define FUNC_QCFG_RESP_MPC_CHNLS_RCE_ENABLED 0x2UL + #define FUNC_QCFG_RESP_MPC_CHNLS_TE_CFA_ENABLED 0x4UL + #define FUNC_QCFG_RESP_MPC_CHNLS_RE_CFA_ENABLED 0x8UL + #define FUNC_QCFG_RESP_MPC_CHNLS_PRIMATE_ENABLED 0x10UL + u8 unused_2[6]; u8 valid; }; @@ -1405,6 +1476,7 @@ struct hwrm_func_cfg_input { #define FUNC_CFG_REQ_ENABLES_ADMIN_LINK_STATE 0x400000UL #define FUNC_CFG_REQ_ENABLES_HOT_RESET_IF_SUPPORT 0x800000UL #define FUNC_CFG_REQ_ENABLES_SCHQ_ID 0x1000000UL + #define FUNC_CFG_REQ_ENABLES_MPC_CHNLS 0x2000000UL __le16 mtu; __le16 mru; __le16 num_rsscos_ctxs; @@ -1479,7 +1551,18 @@ struct hwrm_func_cfg_input { #define FUNC_CFG_REQ_OPTIONS_RSVD_SFT 4 __le16 num_mcast_filters; __le16 schq_id; - u8 unused_0[6]; + __le16 mpc_chnls; + #define FUNC_CFG_REQ_MPC_CHNLS_TCE_ENABLE 0x1UL + #define FUNC_CFG_REQ_MPC_CHNLS_TCE_DISABLE 0x2UL + #define FUNC_CFG_REQ_MPC_CHNLS_RCE_ENABLE 0x4UL + #define FUNC_CFG_REQ_MPC_CHNLS_RCE_DISABLE 0x8UL + #define FUNC_CFG_REQ_MPC_CHNLS_TE_CFA_ENABLE 0x10UL + #define FUNC_CFG_REQ_MPC_CHNLS_TE_CFA_DISABLE 0x20UL + #define FUNC_CFG_REQ_MPC_CHNLS_RE_CFA_ENABLE 0x40UL + #define FUNC_CFG_REQ_MPC_CHNLS_RE_CFA_DISABLE 0x80UL + #define FUNC_CFG_REQ_MPC_CHNLS_PRIMATE_ENABLE 0x100UL + #define FUNC_CFG_REQ_MPC_CHNLS_PRIMATE_DISABLE 0x200UL + u8 unused_0[4]; }; /* hwrm_func_cfg_output (size:128b/16B) */ @@ -1559,7 +1642,7 @@ struct hwrm_func_qstats_ext_input { u8 unused_1[4]; }; -/* hwrm_func_qstats_ext_output (size:1472b/184B) */ +/* hwrm_func_qstats_ext_output (size:1536b/192B) */ struct hwrm_func_qstats_ext_output { __le16 error_code; __le16 req_type; @@ -1586,6 +1669,7 @@ struct hwrm_func_qstats_ext_output { __le64 rx_tpa_pkt; __le64 rx_tpa_bytes; __le64 rx_tpa_errors; + __le64 rx_tpa_events; u8 unused_0[7]; u8 valid; }; @@ -2412,25 +2496,29 @@ struct hwrm_port_phy_cfg_input { __le16 target_id; __le64 resp_addr; __le32 flags; - #define PORT_PHY_CFG_REQ_FLAGS_RESET_PHY 0x1UL - #define PORT_PHY_CFG_REQ_FLAGS_DEPRECATED 0x2UL - #define PORT_PHY_CFG_REQ_FLAGS_FORCE 0x4UL - #define PORT_PHY_CFG_REQ_FLAGS_RESTART_AUTONEG 0x8UL - #define PORT_PHY_CFG_REQ_FLAGS_EEE_ENABLE 0x10UL - #define PORT_PHY_CFG_REQ_FLAGS_EEE_DISABLE 0x20UL - #define PORT_PHY_CFG_REQ_FLAGS_EEE_TX_LPI_ENABLE 0x40UL - #define PORT_PHY_CFG_REQ_FLAGS_EEE_TX_LPI_DISABLE 0x80UL - #define PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_ENABLE 0x100UL - #define PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_DISABLE 0x200UL - #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_ENABLE 0x400UL - #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_DISABLE 0x800UL - #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_ENABLE 0x1000UL - #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_DISABLE 0x2000UL - #define PORT_PHY_CFG_REQ_FLAGS_FORCE_LINK_DWN 0x4000UL - #define PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_1XN_ENABLE 0x8000UL - #define PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_1XN_DISABLE 0x10000UL - #define PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_2XN_ENABLE 0x20000UL - #define PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_2XN_DISABLE 0x40000UL + #define PORT_PHY_CFG_REQ_FLAGS_RESET_PHY 0x1UL + #define PORT_PHY_CFG_REQ_FLAGS_DEPRECATED 0x2UL + #define PORT_PHY_CFG_REQ_FLAGS_FORCE 0x4UL + #define PORT_PHY_CFG_REQ_FLAGS_RESTART_AUTONEG 0x8UL + #define PORT_PHY_CFG_REQ_FLAGS_EEE_ENABLE 0x10UL + #define PORT_PHY_CFG_REQ_FLAGS_EEE_DISABLE 0x20UL + #define PORT_PHY_CFG_REQ_FLAGS_EEE_TX_LPI_ENABLE 0x40UL + #define PORT_PHY_CFG_REQ_FLAGS_EEE_TX_LPI_DISABLE 0x80UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_ENABLE 0x100UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_DISABLE 0x200UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_ENABLE 0x400UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_DISABLE 0x800UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_ENABLE 0x1000UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_DISABLE 0x2000UL + #define PORT_PHY_CFG_REQ_FLAGS_FORCE_LINK_DWN 0x4000UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_1XN_ENABLE 0x8000UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_1XN_DISABLE 0x10000UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_IEEE_ENABLE 0x20000UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_RS544_IEEE_DISABLE 0x40000UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_RS272_1XN_ENABLE 0x80000UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_RS272_1XN_DISABLE 0x100000UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_RS272_IEEE_ENABLE 0x200000UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_RS272_IEEE_DISABLE 0x400000UL __le32 enables; #define PORT_PHY_CFG_REQ_ENABLES_AUTO_MODE 0x1UL #define PORT_PHY_CFG_REQ_ENABLES_AUTO_DUPLEX 0x2UL @@ -2573,7 +2661,7 @@ struct hwrm_port_phy_qcfg_input { u8 unused_0[6]; }; -/* hwrm_port_phy_qcfg_output (size:832b/104B) */ +/* hwrm_port_phy_qcfg_output (size:768b/96B) */ struct hwrm_port_phy_qcfg_output { __le16 error_code; __le16 req_type; @@ -2584,10 +2672,22 @@ struct hwrm_port_phy_qcfg_output { #define PORT_PHY_QCFG_RESP_LINK_SIGNAL 0x1UL #define PORT_PHY_QCFG_RESP_LINK_LINK 0x2UL #define PORT_PHY_QCFG_RESP_LINK_LAST PORT_PHY_QCFG_RESP_LINK_LINK - u8 link_signal_mode; - #define PORT_PHY_QCFG_RESP_LINK_SIGNAL_MODE_NRZ 0x0UL - #define PORT_PHY_QCFG_RESP_LINK_SIGNAL_MODE_PAM4 0x1UL - #define PORT_PHY_QCFG_RESP_LINK_SIGNAL_MODE_LAST PORT_PHY_QCFG_RESP_LINK_SIGNAL_MODE_PAM4 + u8 active_fec_signal_mode; + #define PORT_PHY_QCFG_RESP_SIGNAL_MODE_MASK 0xfUL + #define PORT_PHY_QCFG_RESP_SIGNAL_MODE_SFT 0 + #define PORT_PHY_QCFG_RESP_SIGNAL_MODE_NRZ 0x0UL + #define PORT_PHY_QCFG_RESP_SIGNAL_MODE_PAM4 0x1UL + #define PORT_PHY_QCFG_RESP_SIGNAL_MODE_LAST PORT_PHY_QCFG_RESP_SIGNAL_MODE_PAM4 + #define PORT_PHY_QCFG_RESP_ACTIVE_FEC_MASK 0xf0UL + #define PORT_PHY_QCFG_RESP_ACTIVE_FEC_SFT 4 + #define PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_NONE_ACTIVE (0x0UL << 4) + #define PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_CLAUSE74_ACTIVE (0x1UL << 4) + #define PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_CLAUSE91_ACTIVE (0x2UL << 4) + #define PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS544_1XN_ACTIVE (0x3UL << 4) + #define PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS544_IEEE_ACTIVE (0x4UL << 4) + #define PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS272_1XN_ACTIVE (0x5UL << 4) + #define PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS272_IEEE_ACTIVE (0x6UL << 4) + #define PORT_PHY_QCFG_RESP_ACTIVE_FEC_LAST PORT_PHY_QCFG_RESP_ACTIVE_FEC_FEC_RS272_IEEE_ACTIVE __le16 link_speed; #define PORT_PHY_QCFG_RESP_LINK_SPEED_100MB 0x1UL #define PORT_PHY_QCFG_RESP_LINK_SPEED_1GB 0xaUL @@ -2809,21 +2909,21 @@ struct hwrm_port_phy_qcfg_output { #define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_QSFP28 (0x11UL << 24) #define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_LAST PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_QSFP28 __le16 fec_cfg; - #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_NONE_SUPPORTED 0x1UL - #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_AUTONEG_SUPPORTED 0x2UL - #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_AUTONEG_ENABLED 0x4UL - #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE74_SUPPORTED 0x8UL - #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE74_ENABLED 0x10UL - #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_SUPPORTED 0x20UL - #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_ENABLED 0x40UL - #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS544_1XN_SUPPORTED 0x80UL - #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS544_1XN_ENABLED 0x100UL - #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS544_2XN_SUPPORTED 0x200UL - #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS544_2XN_ENABLED 0x400UL - #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE74_ACTIVE 0x800UL - #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_ACTIVE 0x1000UL - #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS544_1XN_ACTIVE 0x2000UL - #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS544_2XN_ACTIVE 0x4000UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_NONE_SUPPORTED 0x1UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_AUTONEG_SUPPORTED 0x2UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_AUTONEG_ENABLED 0x4UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE74_SUPPORTED 0x8UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE74_ENABLED 0x10UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_SUPPORTED 0x20UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_ENABLED 0x40UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS544_1XN_SUPPORTED 0x80UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS544_1XN_ENABLED 0x100UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS544_IEEE_SUPPORTED 0x200UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS544_IEEE_ENABLED 0x400UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS272_1XN_SUPPORTED 0x800UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS272_1XN_ENABLED 0x1000UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS272_IEEE_SUPPORTED 0x2000UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_RS272_IEEE_ENABLED 0x4000UL u8 duplex_state; #define PORT_PHY_QCFG_RESP_DUPLEX_STATE_HALF 0x0UL #define PORT_PHY_QCFG_RESP_DUPLEX_STATE_FULL 0x1UL @@ -2845,11 +2945,10 @@ struct hwrm_port_phy_qcfg_output { #define PORT_PHY_QCFG_RESP_AUTO_PAM4_LINK_SPEED_MASK_50G 0x1UL #define PORT_PHY_QCFG_RESP_AUTO_PAM4_LINK_SPEED_MASK_100G 0x2UL #define PORT_PHY_QCFG_RESP_AUTO_PAM4_LINK_SPEED_MASK_200G 0x4UL - __le16 link_partner_pam4_adv_speeds; + u8 link_partner_pam4_adv_speeds; #define PORT_PHY_QCFG_RESP_LINK_PARTNER_PAM4_ADV_SPEEDS_50GB 0x1UL #define PORT_PHY_QCFG_RESP_LINK_PARTNER_PAM4_ADV_SPEEDS_100GB 0x2UL #define PORT_PHY_QCFG_RESP_LINK_PARTNER_PAM4_ADV_SPEEDS_200GB 0x4UL - u8 unused_0[7]; u8 valid; }; @@ -3293,6 +3392,47 @@ struct hwrm_port_lpbk_qstats_output { u8 valid; }; +/* hwrm_port_ecn_qstats_input (size:256b/32B) */ +struct hwrm_port_ecn_qstats_input { + __le16 req_type; + __le16 cmpl_ring; + __le16 seq_id; + __le16 target_id; + __le64 resp_addr; + __le16 port_id; + __le16 ecn_stat_buf_size; + u8 flags; + #define PORT_ECN_QSTATS_REQ_FLAGS_UNUSED 0x0UL + #define PORT_ECN_QSTATS_REQ_FLAGS_COUNTER_MASK 0x1UL + #define PORT_ECN_QSTATS_REQ_FLAGS_LAST PORT_ECN_QSTATS_REQ_FLAGS_COUNTER_MASK + u8 unused_0[3]; + __le64 ecn_stat_host_addr; +}; + +/* hwrm_port_ecn_qstats_output (size:128b/16B) */ +struct hwrm_port_ecn_qstats_output { + __le16 error_code; + __le16 req_type; + __le16 seq_id; + __le16 resp_len; + __le16 ecn_stat_buf_size; + u8 mark_en; + u8 unused_0[4]; + u8 valid; +}; + +/* port_stats_ecn (size:512b/64B) */ +struct port_stats_ecn { + __le64 mark_cnt_cos0; + __le64 mark_cnt_cos1; + __le64 mark_cnt_cos2; + __le64 mark_cnt_cos3; + __le64 mark_cnt_cos4; + __le64 mark_cnt_cos5; + __le64 mark_cnt_cos6; + __le64 mark_cnt_cos7; +}; + /* hwrm_port_clr_stats_input (size:192b/24B) */ struct hwrm_port_clr_stats_input { __le16 req_type; @@ -3387,8 +3527,9 @@ struct hwrm_port_phy_qcaps_output { #define PORT_PHY_QCAPS_RESP_FLAGS_AUTONEG_LPBK_SUPPORTED 0x4UL #define PORT_PHY_QCAPS_RESP_FLAGS_SHARED_PHY_CFG_SUPPORTED 0x8UL #define PORT_PHY_QCAPS_RESP_FLAGS_CUMULATIVE_COUNTERS_ON_RESET 0x10UL - #define PORT_PHY_QCAPS_RESP_FLAGS_RSVD1_MASK 0xe0UL - #define PORT_PHY_QCAPS_RESP_FLAGS_RSVD1_SFT 5 + #define PORT_PHY_QCAPS_RESP_FLAGS_LOCAL_LPBK_NOT_SUPPORTED 0x20UL + #define PORT_PHY_QCAPS_RESP_FLAGS_RSVD1_MASK 0xc0UL + #define PORT_PHY_QCAPS_RESP_FLAGS_RSVD1_SFT 6 u8 port_cnt; #define PORT_PHY_QCAPS_RESP_PORT_CNT_UNKNOWN 0x0UL #define PORT_PHY_QCAPS_RESP_PORT_CNT_1 0x1UL @@ -5365,6 +5506,7 @@ struct hwrm_ring_alloc_input { #define RING_ALLOC_REQ_ENABLES_NQ_RING_ID_VALID 0x80UL #define RING_ALLOC_REQ_ENABLES_RX_BUF_SIZE_VALID 0x100UL #define RING_ALLOC_REQ_ENABLES_SCHQ_ID 0x200UL + #define RING_ALLOC_REQ_ENABLES_MPC_CHNLS_TYPE 0x400UL u8 ring_type; #define RING_ALLOC_REQ_RING_TYPE_L2_CMPL 0x0UL #define RING_ALLOC_REQ_RING_TYPE_TX 0x1UL @@ -5424,7 +5566,14 @@ struct hwrm_ring_alloc_input { #define RING_ALLOC_REQ_INT_MODE_MSIX 0x2UL #define RING_ALLOC_REQ_INT_MODE_POLL 0x3UL #define RING_ALLOC_REQ_INT_MODE_LAST RING_ALLOC_REQ_INT_MODE_POLL - u8 unused_4[3]; + u8 mpc_chnls_type; + #define RING_ALLOC_REQ_MPC_CHNLS_TYPE_TCE 0x0UL + #define RING_ALLOC_REQ_MPC_CHNLS_TYPE_RCE 0x1UL + #define RING_ALLOC_REQ_MPC_CHNLS_TYPE_TE_CFA 0x2UL + #define RING_ALLOC_REQ_MPC_CHNLS_TYPE_RE_CFA 0x3UL + #define RING_ALLOC_REQ_MPC_CHNLS_TYPE_PRIMATE 0x4UL + #define RING_ALLOC_REQ_MPC_CHNLS_TYPE_LAST RING_ALLOC_REQ_MPC_CHNLS_TYPE_PRIMATE + u8 unused_4[2]; __le64 cq_handle; }; @@ -6661,7 +6810,7 @@ struct hwrm_cfa_vfr_alloc_output { u8 valid; }; -/* hwrm_cfa_vfr_free_input (size:384b/48B) */ +/* hwrm_cfa_vfr_free_input (size:448b/56B) */ struct hwrm_cfa_vfr_free_input { __le16 req_type; __le16 cmpl_ring; @@ -6669,6 +6818,9 @@ struct hwrm_cfa_vfr_free_input { __le16 target_id; __le64 resp_addr; char vfr_name[32]; + __le16 vf_id; + __le16 reserved; + u8 unused_0[4]; }; /* hwrm_cfa_vfr_free_output (size:128b/16B) */ @@ -6970,7 +7122,7 @@ struct ctx_hw_stats { __le64 tpa_aborts; }; -/* ctx_hw_stats_ext (size:1344b/168B) */ +/* ctx_hw_stats_ext (size:1408b/176B) */ struct ctx_hw_stats_ext { __le64 rx_ucast_pkts; __le64 rx_mcast_pkts; @@ -6993,6 +7145,7 @@ struct ctx_hw_stats_ext { __le64 rx_tpa_pkt; __le64 rx_tpa_bytes; __le64 rx_tpa_errors; + __le64 rx_tpa_events; }; /* hwrm_stat_ctx_alloc_input (size:256b/32B) */ @@ -7065,16 +7218,16 @@ struct hwrm_stat_ctx_query_output { __le64 tx_ucast_pkts; __le64 tx_mcast_pkts; __le64 tx_bcast_pkts; - __le64 tx_err_pkts; - __le64 tx_drop_pkts; + __le64 tx_discard_pkts; + __le64 tx_error_pkts; __le64 tx_ucast_bytes; __le64 tx_mcast_bytes; __le64 tx_bcast_bytes; __le64 rx_ucast_pkts; __le64 rx_mcast_pkts; __le64 rx_bcast_pkts; - __le64 rx_err_pkts; - __le64 rx_drop_pkts; + __le64 rx_discard_pkts; + __le64 rx_error_pkts; __le64 rx_ucast_bytes; __le64 rx_mcast_bytes; __le64 rx_bcast_bytes; @@ -7099,7 +7252,7 @@ struct hwrm_stat_ext_ctx_query_input { u8 unused_0[3]; }; -/* hwrm_stat_ext_ctx_query_output (size:1472b/184B) */ +/* hwrm_stat_ext_ctx_query_output (size:1536b/192B) */ struct hwrm_stat_ext_ctx_query_output { __le16 error_code; __le16 req_type; @@ -7126,6 +7279,7 @@ struct hwrm_stat_ext_ctx_query_output { __le64 rx_tpa_pkt; __le64 rx_tpa_bytes; __le64 rx_tpa_errors; + __le64 rx_tpa_events; u8 unused_0[7]; u8 valid; }; @@ -7702,6 +7856,77 @@ struct hwrm_dbg_read_direct_output { u8 valid; }; +/* hwrm_dbg_qcaps_input (size:192b/24B) */ +struct hwrm_dbg_qcaps_input { + __le16 req_type; + __le16 cmpl_ring; + __le16 seq_id; + __le16 target_id; + __le64 resp_addr; + __le16 fid; + u8 unused_0[6]; +}; + +/* hwrm_dbg_qcaps_output (size:192b/24B) */ +struct hwrm_dbg_qcaps_output { + __le16 error_code; + __le16 req_type; + __le16 seq_id; + __le16 resp_len; + __le16 fid; + u8 unused_0[2]; + __le32 coredump_component_disable_caps; + #define DBG_QCAPS_RESP_COREDUMP_COMPONENT_DISABLE_CAPS_NVRAM 0x1UL + __le32 flags; + #define DBG_QCAPS_RESP_FLAGS_CRASHDUMP_NVM 0x1UL + #define DBG_QCAPS_RESP_FLAGS_CRASHDUMP_HOST_DDR 0x2UL + #define DBG_QCAPS_RESP_FLAGS_CRASHDUMP_SOC_DDR 0x4UL + u8 unused_1[3]; + u8 valid; +}; + +/* hwrm_dbg_qcfg_input (size:192b/24B) */ +struct hwrm_dbg_qcfg_input { + __le16 req_type; + __le16 cmpl_ring; + __le16 seq_id; + __le16 target_id; + __le64 resp_addr; + __le16 fid; + __le16 flags; + #define DBG_QCFG_REQ_FLAGS_CRASHDUMP_SIZE_FOR_DEST_MASK 0x3UL + #define DBG_QCFG_REQ_FLAGS_CRASHDUMP_SIZE_FOR_DEST_SFT 0 + #define DBG_QCFG_REQ_FLAGS_CRASHDUMP_SIZE_FOR_DEST_DEST_NVM 0x0UL + #define DBG_QCFG_REQ_FLAGS_CRASHDUMP_SIZE_FOR_DEST_DEST_HOST_DDR 0x1UL + #define DBG_QCFG_REQ_FLAGS_CRASHDUMP_SIZE_FOR_DEST_DEST_SOC_DDR 0x2UL + #define DBG_QCFG_REQ_FLAGS_CRASHDUMP_SIZE_FOR_DEST_LAST DBG_QCFG_REQ_FLAGS_CRASHDUMP_SIZE_FOR_DEST_DEST_SOC_DDR + __le32 coredump_component_disable_flags; + #define DBG_QCFG_REQ_COREDUMP_COMPONENT_DISABLE_FLAGS_NVRAM 0x1UL +}; + +/* hwrm_dbg_qcfg_output (size:256b/32B) */ +struct hwrm_dbg_qcfg_output { + __le16 error_code; + __le16 req_type; + __le16 seq_id; + __le16 resp_len; + __le16 fid; + u8 unused_0[2]; + __le32 coredump_size; + __le32 flags; + #define DBG_QCFG_RESP_FLAGS_UART_LOG 0x1UL + #define DBG_QCFG_RESP_FLAGS_UART_LOG_SECONDARY 0x2UL + #define DBG_QCFG_RESP_FLAGS_FW_TRACE 0x4UL + #define DBG_QCFG_RESP_FLAGS_FW_TRACE_SECONDARY 0x8UL + #define DBG_QCFG_RESP_FLAGS_DEBUG_NOTIFY 0x10UL + #define DBG_QCFG_RESP_FLAGS_JTAG_DEBUG 0x20UL + __le16 async_cmpl_ring; + u8 unused_2[2]; + __le32 crashdump_size; + u8 unused_3[3]; + u8 valid; +}; + /* coredump_segment_record (size:128b/16B) */ struct coredump_segment_record { __le16 component_id; @@ -8048,7 +8273,7 @@ struct hwrm_nvm_get_dev_info_input { __le64 resp_addr; }; -/* hwrm_nvm_get_dev_info_output (size:256b/32B) */ +/* hwrm_nvm_get_dev_info_output (size:640b/80B) */ struct hwrm_nvm_get_dev_info_output { __le16 error_code; __le16 req_type; @@ -8063,6 +8288,22 @@ struct hwrm_nvm_get_dev_info_output { u8 nvm_cfg_ver_maj; u8 nvm_cfg_ver_min; u8 nvm_cfg_ver_upd; + u8 flags; + #define NVM_GET_DEV_INFO_RESP_FLAGS_FW_VER_VALID 0x1UL + char pkg_name[16]; + __le16 hwrm_fw_major; + __le16 hwrm_fw_minor; + __le16 hwrm_fw_build; + __le16 hwrm_fw_patch; + __le16 mgmt_fw_major; + __le16 mgmt_fw_minor; + __le16 mgmt_fw_build; + __le16 mgmt_fw_patch; + __le16 roce_fw_major; + __le16 roce_fw_minor; + __le16 roce_fw_build; + __le16 roce_fw_patch; + u8 unused_0[7]; u8 valid; }; @@ -8381,6 +8622,16 @@ struct hwrm_selftest_irq_output { u8 valid; }; +/* db_push_info (size:64b/8B) */ +struct db_push_info { + u32 push_size_push_index; + #define DB_PUSH_INFO_PUSH_INDEX_MASK 0xffffffUL + #define DB_PUSH_INFO_PUSH_INDEX_SFT 0 + #define DB_PUSH_INFO_PUSH_SIZE_MASK 0x1f000000UL + #define DB_PUSH_INFO_PUSH_SIZE_SFT 24 + u32 reserved32; +}; + /* fw_status_reg (size:32b/4B) */ struct fw_status_reg { u32 fw_status; @@ -8393,6 +8644,32 @@ struct fw_status_reg { #define FW_STATUS_REG_CRASHDUMP_ONGOING 0x40000UL #define FW_STATUS_REG_CRASHDUMP_COMPLETE 0x80000UL #define FW_STATUS_REG_SHUTDOWN 0x100000UL -}; + #define FW_STATUS_REG_CRASHED_NO_MASTER 0x200000UL +}; + +/* hcomm_status (size:64b/8B) */ +struct hcomm_status { + u32 sig_ver; + #define HCOMM_STATUS_VER_MASK 0xffUL + #define HCOMM_STATUS_VER_SFT 0 + #define HCOMM_STATUS_VER_LATEST 0x1UL + #define HCOMM_STATUS_VER_LAST HCOMM_STATUS_VER_LATEST + #define HCOMM_STATUS_SIGNATURE_MASK 0xffffff00UL + #define HCOMM_STATUS_SIGNATURE_SFT 8 + #define HCOMM_STATUS_SIGNATURE_VAL (0x484353UL << 8) + #define HCOMM_STATUS_SIGNATURE_LAST HCOMM_STATUS_SIGNATURE_VAL + u32 fw_status_loc; + #define HCOMM_STATUS_TRUE_ADDR_SPACE_MASK 0x3UL + #define HCOMM_STATUS_TRUE_ADDR_SPACE_SFT 0 + #define HCOMM_STATUS_FW_STATUS_LOC_ADDR_SPACE_PCIE_CFG 0x0UL + #define HCOMM_STATUS_FW_STATUS_LOC_ADDR_SPACE_GRC 0x1UL + #define HCOMM_STATUS_FW_STATUS_LOC_ADDR_SPACE_BAR0 0x2UL + #define HCOMM_STATUS_FW_STATUS_LOC_ADDR_SPACE_BAR1 0x3UL + #define HCOMM_STATUS_FW_STATUS_LOC_ADDR_SPACE_LAST HCOMM_STATUS_FW_STATUS_LOC_ADDR_SPACE_BAR1 + #define HCOMM_STATUS_TRUE_OFFSET_MASK 0xfffffffcUL + #define HCOMM_STATUS_TRUE_OFFSET_SFT 2 +}; + +#define HCOMM_STATUS_STRUCT_LOC 0x31001F0UL #endif /* _BNXT_HSI_H_ */ diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c index cc2ee4d0bd18..23b80aa171dd 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c @@ -1029,7 +1029,7 @@ static int bnxt_vf_set_link(struct bnxt *bp, struct bnxt_vf_info *vf) rc = bnxt_hwrm_exec_fwd_resp( bp, vf, sizeof(struct hwrm_port_phy_qcfg_input)); } else { - struct hwrm_port_phy_qcfg_output_compat phy_qcfg_resp = {0}; + struct hwrm_port_phy_qcfg_output phy_qcfg_resp = {0}; struct hwrm_port_phy_qcfg_input *phy_qcfg_req; phy_qcfg_req = diff --git a/drivers/net/ethernet/broadcom/cnic.c b/drivers/net/ethernet/broadcom/cnic.c index 84536292b031..f7f10cfb3476 100644 --- a/drivers/net/ethernet/broadcom/cnic.c +++ b/drivers/net/ethernet/broadcom/cnic.c @@ -3009,10 +3009,10 @@ static int cnic_service_bnx2(void *data, void *status_blk) return cnic_service_bnx2_queues(dev); } -static void cnic_service_bnx2_msix(unsigned long data) +static void cnic_service_bnx2_msix(struct tasklet_struct *t) { - struct cnic_dev *dev = (struct cnic_dev *) data; - struct cnic_local *cp = dev->cnic_priv; + struct cnic_local *cp = from_tasklet(cp, t, cnic_irq_task); + struct cnic_dev *dev = cp->dev; cp->last_status_idx = cnic_service_bnx2_queues(dev); @@ -3134,10 +3134,10 @@ static u32 cnic_service_bnx2x_kcq(struct cnic_dev *dev, struct kcq_info *info) return last_status; } -static void cnic_service_bnx2x_bh(unsigned long data) +static void cnic_service_bnx2x_bh(struct tasklet_struct *t) { - struct cnic_dev *dev = (struct cnic_dev *) data; - struct cnic_local *cp = dev->cnic_priv; + struct cnic_local *cp = from_tasklet(cp, t, cnic_irq_task); + struct cnic_dev *dev = cp->dev; struct bnx2x *bp = netdev_priv(dev->netdev); u32 status_idx, new_status_idx; @@ -4458,8 +4458,7 @@ static int cnic_init_bnx2_irq(struct cnic_dev *dev) CNIC_WR(dev, base + BNX2_HC_CMD_TICKS_OFF, (64 << 16) | 220); cp->last_status_idx = cp->status_blk.bnx2->status_idx; - tasklet_init(&cp->cnic_irq_task, cnic_service_bnx2_msix, - (unsigned long) dev); + tasklet_setup(&cp->cnic_irq_task, cnic_service_bnx2_msix); err = cnic_request_irq(dev); if (err) return err; @@ -4868,8 +4867,7 @@ static int cnic_init_bnx2x_irq(struct cnic_dev *dev) struct cnic_eth_dev *ethdev = cp->ethdev; int err = 0; - tasklet_init(&cp->cnic_irq_task, cnic_service_bnx2x_bh, - (unsigned long) dev); + tasklet_setup(&cp->cnic_irq_task, cnic_service_bnx2x_bh); if (ethdev->drv_state & CNIC_DRV_STATE_USING_MSIX) err = cnic_request_irq(dev); diff --git a/drivers/net/ethernet/brocade/bna/bfa_cee.c b/drivers/net/ethernet/brocade/bna/bfa_cee.c index 09fb9315d1ae..06f221c44802 100644 --- a/drivers/net/ethernet/brocade/bna/bfa_cee.c +++ b/drivers/net/ethernet/brocade/bna/bfa_cee.c @@ -102,14 +102,10 @@ bfa_cee_get_stats_isr(struct bfa_cee *cee, enum bfa_status status) } /** - * bfa_cee_get_attr_isr() + * bfa_cee_reset_stats_isr - CEE ISR for reset-stats responses from f/w * - * @brief CEE ISR for reset-stats responses from f/w - * - * @param[in] cee - Pointer to the CEE module - * status - Return status from the f/w - * - * @return void + * @cee: Input Pointer to the CEE module + * @status: Return status from the f/w */ static void bfa_cee_reset_stats_isr(struct bfa_cee *cee, enum bfa_status status) @@ -148,9 +144,12 @@ bfa_nw_cee_mem_claim(struct bfa_cee *cee, u8 *dma_kva, u64 dma_pa) } /** - * bfa_cee_get_attr - Send the request to the f/w to fetch CEE attributes. + * bfa_nw_cee_get_attr - Send the request to the f/w to fetch CEE attributes. * * @cee: Pointer to the CEE module data structure. + * @attr: attribute requested + * @cbfn: function pointer + * @cbarg: function pointer arguments * * Return: status */ @@ -181,7 +180,9 @@ bfa_nw_cee_get_attr(struct bfa_cee *cee, struct bfa_cee_attr *attr, } /** - * bfa_cee_isrs - Handles Mail-box interrupts for CEE module. + * bfa_cee_isr - Handles Mail-box interrupts for CEE module. + * @cbarg: argument passed containing pointer to the CEE module data structure. + * @m: message pointer */ static void @@ -210,6 +211,7 @@ bfa_cee_isr(void *cbarg, struct bfi_mbmsg *m) /** * bfa_cee_notify - CEE module heart-beat failure handler. * + * @arg: argument passed containing pointer to the CEE module data structure. * @event: IOC event type */ diff --git a/drivers/net/ethernet/brocade/bna/bfa_ioc.c b/drivers/net/ethernet/brocade/bna/bfa_ioc.c index b9dd06b12945..cd933817a0b8 100644 --- a/drivers/net/ethernet/brocade/bna/bfa_ioc.c +++ b/drivers/net/ethernet/brocade/bna/bfa_ioc.c @@ -269,7 +269,7 @@ bfa_ioc_sm_enabling(struct bfa_ioc *ioc, enum ioc_event event) break; case IOC_E_PFFAILED: - /* !!! fall through !!! */ + fallthrough; case IOC_E_HWERROR: ioc->cbfn->enable_cbfn(ioc->bfa, BFA_STATUS_IOC_FAILURE); bfa_fsm_set_state(ioc, bfa_ioc_sm_fail); @@ -365,7 +365,8 @@ bfa_ioc_sm_op(struct bfa_ioc *ioc, enum ioc_event event) case IOC_E_PFFAILED: case IOC_E_HWERROR: bfa_ioc_hb_stop(ioc); - /* !!! fall through !!! */ + fallthrough; + case IOC_E_HBFAIL: if (ioc->iocpf.auto_recover) bfa_fsm_set_state(ioc, bfa_ioc_sm_fail_retry); @@ -1763,7 +1764,7 @@ bfa_ioc_flash_fwver_cmp(struct bfa_ioc *ioc, return BFI_IOC_IMG_VER_INCOMP; } -/** +/* * Returns TRUE if driver is willing to work with current smem f/w version. */ bool @@ -2469,6 +2470,7 @@ bfa_ioc_isr(struct bfa_ioc *ioc, struct bfi_mbmsg *m) * * @ioc: memory for IOC * @bfa: driver instance structure + * @cbfn: callback function */ void bfa_nw_ioc_attach(struct bfa_ioc *ioc, void *bfa, struct bfa_ioc_cbfn *cbfn) @@ -2500,7 +2502,9 @@ bfa_nw_ioc_detach(struct bfa_ioc *ioc) /** * bfa_nw_ioc_pci_init - Setup IOC PCI properties. * + * @ioc: memory for IOC * @pcidev: PCI device information for this IOC + * @clscode: class code */ void bfa_nw_ioc_pci_init(struct bfa_ioc *ioc, struct bfa_pcidev *pcidev, @@ -2569,6 +2573,7 @@ bfa_nw_ioc_pci_init(struct bfa_ioc *ioc, struct bfa_pcidev *pcidev, /** * bfa_nw_ioc_mem_claim - Initialize IOC dma memory * + * @ioc: memory for IOC * @dm_kva: kernel virtual address of IOC dma memory * @dm_pa: physical address of IOC dma memory */ @@ -2636,6 +2641,8 @@ bfa_nw_ioc_mbox_regisr(struct bfa_ioc *ioc, enum bfi_mclass mc, * * @ioc: IOC instance * @cmd: Mailbox command + * @cbfn: callback function + * @cbarg: arguments to callback * * Waits if mailbox is busy. Responsibility of caller to serialize */ diff --git a/drivers/net/ethernet/brocade/bna/bnad.c b/drivers/net/ethernet/brocade/bna/bnad.c index cc80bbbefe87..7e4e831d720f 100644 --- a/drivers/net/ethernet/brocade/bna/bnad.c +++ b/drivers/net/ethernet/brocade/bna/bnad.c @@ -3277,7 +3277,7 @@ bnad_change_mtu(struct net_device *netdev, int new_mtu) { int err, mtu; struct bnad *bnad = netdev_priv(netdev); - u32 rx_count = 0, frame, new_frame; + u32 frame, new_frame; mutex_lock(&bnad->conf_mutex); @@ -3293,12 +3293,9 @@ bnad_change_mtu(struct net_device *netdev, int new_mtu) /* only when transition is over 4K */ if ((frame <= 4096 && new_frame > 4096) || (frame > 4096 && new_frame <= 4096)) - rx_count = bnad_reinit_rx(bnad); + bnad_reinit_rx(bnad); } - /* rx_count > 0 - new rx created - * - Linux set err = 0 and return - */ err = bnad_mtu_set(bnad, new_frame); if (err) err = -EBUSY; diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index 4f1b41569260..5de47f6fde5a 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -7,6 +7,7 @@ #ifndef _MACB_H #define _MACB_H +#include <linux/clk.h> #include <linux/phylink.h> #include <linux/ptp_clock_kernel.h> #include <linux/net_tstamp.h> @@ -365,6 +366,8 @@ #define MACB_ISR_RLE_SIZE 1 #define MACB_TXERR_OFFSET 6 /* EN TX frame corrupt from error interrupt */ #define MACB_TXERR_SIZE 1 +#define MACB_RM9200_TBRE_OFFSET 6 /* EN may send new frame interrupt (RM9200) */ +#define MACB_RM9200_TBRE_SIZE 1 #define MACB_TCOMP_OFFSET 7 /* Enable transmit complete interrupt */ #define MACB_TCOMP_SIZE 1 #define MACB_ISR_LINK_OFFSET 9 /* Enable link change interrupt */ @@ -1204,10 +1207,10 @@ struct macb { phy_interface_t phy_interface; - /* AT91RM9200 transmit */ - struct sk_buff *skb; /* holds skb until xmit interrupt completes */ - dma_addr_t skb_physaddr; /* phys addr from pci_map_single */ - int skb_length; /* saved skb length for pci_unmap_single */ + /* AT91RM9200 transmit queue (1 on wire + 1 queued) */ + struct macb_tx_skb rm9200_txq[2]; + unsigned int rm9200_tx_tail; + unsigned int rm9200_tx_len; unsigned int max_tx_length; u64 ethtool_stats[GEM_STATS_LEN + QUEUE_STATS_LEN * MACB_MAX_QUEUES]; @@ -1298,4 +1301,14 @@ static inline bool gem_has_ptp(struct macb *bp) return !!(bp->caps & MACB_CAPS_GEM_HAS_PTP); } +/** + * struct macb_platform_data - platform data for MACB Ethernet used for PCI registration + * @pclk: platform clock + * @hclk: AHB clock + */ +struct macb_platform_data { + struct clk *pclk; + struct clk *hclk; +}; + #endif /* _MACB_H */ diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 9179f7b0b900..883e47c5b1a7 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -23,7 +23,6 @@ #include <linux/netdevice.h> #include <linux/etherdevice.h> #include <linux/dma-mapping.h> -#include <linux/platform_data/macb.h> #include <linux/platform_device.h> #include <linux/phylink.h> #include <linux/of.h> @@ -458,9 +457,9 @@ static void macb_init_buffers(struct macb *bp) /** * macb_set_tx_clk() - Set a clock to a new frequency - * @clk Pointer to the clock to change - * @rate New frequency in Hz - * @dev Pointer to the struct net_device + * @clk: Pointer to the clock to change + * @speed: New frequency in Hz + * @dev: Pointer to the struct net_device */ static void macb_set_tx_clk(struct clk *clk, int speed, struct net_device *dev) { @@ -1465,9 +1464,9 @@ static int macb_poll(struct napi_struct *napi, int budget) return work_done; } -static void macb_hresp_error_task(unsigned long data) +static void macb_hresp_error_task(struct tasklet_struct *t) { - struct macb *bp = (struct macb *)data; + struct macb *bp = from_tasklet(bp, t, hresp_err_tasklet); struct net_device *dev = bp->dev; struct macb_queue *queue; unsigned int q; @@ -3909,6 +3908,7 @@ static int at91ether_start(struct macb *lp) MACB_BIT(ISR_TUND) | MACB_BIT(ISR_RLE) | MACB_BIT(TCOMP) | + MACB_BIT(RM9200_TBRE) | MACB_BIT(ISR_ROVR) | MACB_BIT(HRESP)); @@ -3925,6 +3925,7 @@ static void at91ether_stop(struct macb *lp) MACB_BIT(ISR_TUND) | MACB_BIT(ISR_RLE) | MACB_BIT(TCOMP) | + MACB_BIT(RM9200_TBRE) | MACB_BIT(ISR_ROVR) | MACB_BIT(HRESP)); @@ -3994,24 +3995,34 @@ static netdev_tx_t at91ether_start_xmit(struct sk_buff *skb, struct net_device *dev) { struct macb *lp = netdev_priv(dev); + unsigned long flags; - if (macb_readl(lp, TSR) & MACB_BIT(RM9200_BNQ)) { - netif_stop_queue(dev); + if (lp->rm9200_tx_len < 2) { + int desc = lp->rm9200_tx_tail; /* Store packet information (to free when Tx completed) */ - lp->skb = skb; - lp->skb_length = skb->len; - lp->skb_physaddr = dma_map_single(&lp->pdev->dev, skb->data, - skb->len, DMA_TO_DEVICE); - if (dma_mapping_error(&lp->pdev->dev, lp->skb_physaddr)) { + lp->rm9200_txq[desc].skb = skb; + lp->rm9200_txq[desc].size = skb->len; + lp->rm9200_txq[desc].mapping = dma_map_single(&lp->pdev->dev, skb->data, + skb->len, DMA_TO_DEVICE); + if (dma_mapping_error(&lp->pdev->dev, lp->rm9200_txq[desc].mapping)) { dev_kfree_skb_any(skb); dev->stats.tx_dropped++; netdev_err(dev, "%s: DMA mapping error\n", __func__); return NETDEV_TX_OK; } + spin_lock_irqsave(&lp->lock, flags); + + lp->rm9200_tx_tail = (desc + 1) & 1; + lp->rm9200_tx_len++; + if (lp->rm9200_tx_len > 1) + netif_stop_queue(dev); + + spin_unlock_irqrestore(&lp->lock, flags); + /* Set address of the data in the Transmit Address register */ - macb_writel(lp, TAR, lp->skb_physaddr); + macb_writel(lp, TAR, lp->rm9200_txq[desc].mapping); /* Set length of the packet in the Transmit Control register */ macb_writel(lp, TCR, skb->len); @@ -4074,6 +4085,9 @@ static irqreturn_t at91ether_interrupt(int irq, void *dev_id) struct net_device *dev = dev_id; struct macb *lp = netdev_priv(dev); u32 intstatus, ctl; + unsigned int desc; + unsigned int qlen; + u32 tsr; /* MAC Interrupt Status register indicates what interrupts are pending. * It is automatically cleared once read. @@ -4085,20 +4099,39 @@ static irqreturn_t at91ether_interrupt(int irq, void *dev_id) at91ether_rx(dev); /* Transmit complete */ - if (intstatus & MACB_BIT(TCOMP)) { + if (intstatus & (MACB_BIT(TCOMP) | MACB_BIT(RM9200_TBRE))) { /* The TCOM bit is set even if the transmission failed */ if (intstatus & (MACB_BIT(ISR_TUND) | MACB_BIT(ISR_RLE))) dev->stats.tx_errors++; - if (lp->skb) { - dev_consume_skb_irq(lp->skb); - lp->skb = NULL; - dma_unmap_single(&lp->pdev->dev, lp->skb_physaddr, - lp->skb_length, DMA_TO_DEVICE); + spin_lock(&lp->lock); + + tsr = macb_readl(lp, TSR); + + /* we have three possibilities here: + * - all pending packets transmitted (TGO, implies BNQ) + * - only first packet transmitted (!TGO && BNQ) + * - two frames pending (!TGO && !BNQ) + * Note that TGO ("transmit go") is called "IDLE" on RM9200. + */ + qlen = (tsr & MACB_BIT(TGO)) ? 0 : + (tsr & MACB_BIT(RM9200_BNQ)) ? 1 : 2; + + while (lp->rm9200_tx_len > qlen) { + desc = (lp->rm9200_tx_tail - lp->rm9200_tx_len) & 1; + dev_consume_skb_irq(lp->rm9200_txq[desc].skb); + lp->rm9200_txq[desc].skb = NULL; + dma_unmap_single(&lp->pdev->dev, lp->rm9200_txq[desc].mapping, + lp->rm9200_txq[desc].size, DMA_TO_DEVICE); dev->stats.tx_packets++; - dev->stats.tx_bytes += lp->skb_length; + dev->stats.tx_bytes += lp->rm9200_txq[desc].size; + lp->rm9200_tx_len--; } - netif_wake_queue(dev); + + if (lp->rm9200_tx_len < 2 && netif_queue_stopped(dev)) + netif_wake_queue(dev); + + spin_unlock(&lp->lock); } /* Work-around for EMAC Errata section 41.3.1 */ @@ -4559,8 +4592,7 @@ static int macb_probe(struct platform_device *pdev) goto err_out_unregister_mdio; } - tasklet_init(&bp->hresp_err_tasklet, macb_hresp_error_task, - (unsigned long)bp); + tasklet_setup(&bp->hresp_err_tasklet, macb_hresp_error_task); netdev_info(dev, "Cadence %s rev 0x%08x at 0x%08lx irq %d (%pM)\n", macb_is_gem(bp) ? "GEM" : "MACB", macb_readl(bp, MID), diff --git a/drivers/net/ethernet/cadence/macb_pci.c b/drivers/net/ethernet/cadence/macb_pci.c index cd7d0332cba3..353393dea639 100644 --- a/drivers/net/ethernet/cadence/macb_pci.c +++ b/drivers/net/ethernet/cadence/macb_pci.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0-only /** - * Cadence GEM PCI wrapper. + * DOC: Cadence GEM PCI wrapper. * * Copyright (C) 2016 Cadence Design Systems - https://www.cadence.com * @@ -13,7 +13,6 @@ #include <linux/etherdevice.h> #include <linux/module.h> #include <linux/pci.h> -#include <linux/platform_data/macb.h> #include <linux/platform_device.h> #include "macb.h" diff --git a/drivers/net/ethernet/calxeda/xgmac.c b/drivers/net/ethernet/calxeda/xgmac.c index 05a3d067c3fc..bbb453c6a5f7 100644 --- a/drivers/net/ethernet/calxeda/xgmac.c +++ b/drivers/net/ethernet/calxeda/xgmac.c @@ -1246,6 +1246,8 @@ static int xgmac_poll(struct napi_struct *napi, int budget) /** * xgmac_tx_timeout * @dev : Pointer to net device structure + * @txqueue: index of the hung transmit queue + * * Description: this function is called when a packet transmission fails to * complete within a reasonable tmrate. The driver will mark the error in the * netdev structure and arrange for the device to be reset to a sane state diff --git a/drivers/net/ethernet/cavium/common/cavium_ptp.c b/drivers/net/ethernet/cavium/common/cavium_ptp.c index 81ff9ac73f9a..9fd717b9cf69 100644 --- a/drivers/net/ethernet/cavium/common/cavium_ptp.c +++ b/drivers/net/ethernet/cavium/common/cavium_ptp.c @@ -86,7 +86,7 @@ EXPORT_SYMBOL(cavium_ptp_put); /** * cavium_ptp_adjfine() - Adjust ptp frequency - * @ptp: PTP clock info + * @ptp_info: PTP clock info * @scaled_ppm: how much to adjust by, in parts per million, but with a * 16 bit binary fractional field */ @@ -134,7 +134,7 @@ static int cavium_ptp_adjfine(struct ptp_clock_info *ptp_info, long scaled_ppm) /** * cavium_ptp_adjtime() - Adjust ptp time - * @ptp: PTP clock info + * @ptp_info: PTP clock info * @delta: how much to adjust by, in nanosecs */ static int cavium_ptp_adjtime(struct ptp_clock_info *ptp_info, s64 delta) @@ -155,7 +155,7 @@ static int cavium_ptp_adjtime(struct ptp_clock_info *ptp_info, s64 delta) /** * cavium_ptp_gettime() - Get hardware clock time with adjustment - * @ptp: PTP clock info + * @ptp_info: PTP clock info * @ts: timespec */ static int cavium_ptp_gettime(struct ptp_clock_info *ptp_info, @@ -177,7 +177,7 @@ static int cavium_ptp_gettime(struct ptp_clock_info *ptp_info, /** * cavium_ptp_settime() - Set hardware clock time. Reset adjustment - * @ptp: PTP clock info + * @ptp_info: PTP clock info * @ts: timespec */ static int cavium_ptp_settime(struct ptp_clock_info *ptp_info, @@ -199,7 +199,7 @@ static int cavium_ptp_settime(struct ptp_clock_info *ptp_info, /** * cavium_ptp_enable() - Request to enable or disable an ancillary feature. - * @ptp: PTP clock info + * @ptp_info: PTP clock info * @rq: request * @on: is it on */ diff --git a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c index 50b533ff58e6..2a6d1cadac9e 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c +++ b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c @@ -25,7 +25,9 @@ #include "octeon_main.h" #include "cn66xx_regs.h" #include "cn66xx_device.h" +#include "cn68xx_device.h" #include "cn68xx_regs.h" +#include "cn68xx_device.h" static void lio_cn68xx_set_dpi_regs(struct octeon_device *oct) { diff --git a/drivers/net/ethernet/cavium/liquidio/lio_core.c b/drivers/net/ethernet/cavium/liquidio/lio_core.c index e40c64b79f66..9ef172976b35 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_core.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_core.c @@ -32,8 +32,8 @@ #define OCTNIC_MAX_SG MAX_SKB_FRAGS /** - * \brief Delete gather lists - * @param lio per-network private data + * lio_delete_glists - Delete gather lists + * @lio: per-network private data */ void lio_delete_glists(struct lio *lio) { @@ -73,8 +73,10 @@ void lio_delete_glists(struct lio *lio) } /** - * \brief Setup gather lists - * @param lio per-network private data + * lio_setup_glists - Setup gather lists + * @oct: octeon_device + * @lio: per-network private data + * @num_iqs: count of iqs to allocate */ int lio_setup_glists(struct octeon_device *oct, struct lio *lio, int num_iqs) { @@ -521,12 +523,12 @@ static void lio_update_txq_status(struct octeon_device *oct, int iq_num) } /** - * \brief Setup output queue - * @param oct octeon device - * @param q_no which queue - * @param num_descs how many descriptors - * @param desc_size size of each descriptor - * @param app_ctx application context + * octeon_setup_droq - Setup output queue + * @oct: octeon device + * @q_no: which queue + * @num_descs: how many descriptors + * @desc_size: size of each descriptor + * @app_ctx: application context */ static int octeon_setup_droq(struct octeon_device *oct, int q_no, int num_descs, int desc_size, void *app_ctx) @@ -555,16 +557,17 @@ static int octeon_setup_droq(struct octeon_device *oct, int q_no, int num_descs, return ret_val; } -/** Routine to push packets arriving on Octeon interface upto network layer. - * @param oct_id - octeon device id. - * @param skbuff - skbuff struct to be passed to network layer. - * @param len - size of total data received. - * @param rh - Control header associated with the packet - * @param param - additional control data with the packet - * @param arg - farg registered in droq_ops +/** + * liquidio_push_packet - Routine to push packets arriving on Octeon interface upto network layer. + * @octeon_id:octeon device id. + * @skbuff: skbuff struct to be passed to network layer. + * @len: size of total data received. + * @rh: Control header associated with the packet + * @param: additional control data with the packet + * @arg: farg registered in droq_ops */ static void -liquidio_push_packet(u32 octeon_id __attribute__((unused)), +liquidio_push_packet(u32 __maybe_unused octeon_id, void *skbuff, u32 len, union octeon_rh *rh, @@ -698,8 +701,8 @@ liquidio_push_packet(u32 octeon_id __attribute__((unused)), } /** - * \brief wrapper for calling napi_schedule - * @param param parameters to pass to napi_schedule + * napi_schedule_wrapper - wrapper for calling napi_schedule + * @param: parameters to pass to napi_schedule * * Used when scheduling on different CPUs */ @@ -711,8 +714,8 @@ static void napi_schedule_wrapper(void *param) } /** - * \brief callback when receive interrupt occurs and we are in NAPI mode - * @param arg pointer to octeon output queue + * liquidio_napi_drv_callback - callback when receive interrupt occurs and we are in NAPI mode + * @arg: pointer to octeon output queue */ static void liquidio_napi_drv_callback(void *arg) { @@ -737,9 +740,9 @@ static void liquidio_napi_drv_callback(void *arg) } /** - * \brief Entry point for NAPI polling - * @param napi NAPI structure - * @param budget maximum number of items to process + * liquidio_napi_poll - Entry point for NAPI polling + * @napi: NAPI structure + * @budget: maximum number of items to process */ static int liquidio_napi_poll(struct napi_struct *napi, int budget) { @@ -792,9 +795,11 @@ static int liquidio_napi_poll(struct napi_struct *napi, int budget) } /** - * \brief Setup input and output queues - * @param octeon_dev octeon device - * @param ifidx Interface index + * liquidio_setup_io_queues - Setup input and output queues + * @octeon_dev: octeon device + * @ifidx: Interface index + * @num_iqs: input io queue count + * @num_oqs: output io queue count * * Note: Queues are with respect to the octeon device. Thus * an input queue is for egress packets, and output queues @@ -927,7 +932,7 @@ int liquidio_schedule_msix_droq_pkt_handler(struct octeon_droq *droq, u64 ret) } irqreturn_t -liquidio_msix_intr_handler(int irq __attribute__((unused)), void *dev) +liquidio_msix_intr_handler(int __maybe_unused irq, void *dev) { struct octeon_ioq_vector *ioq_vector = (struct octeon_ioq_vector *)dev; struct octeon_device *oct = ioq_vector->oct_dev; @@ -943,8 +948,8 @@ liquidio_msix_intr_handler(int irq __attribute__((unused)), void *dev) } /** - * \brief Droq packet processor sceduler - * @param oct octeon device + * liquidio_schedule_droq_pkt_handlers - Droq packet processor sceduler + * @oct: octeon device */ static void liquidio_schedule_droq_pkt_handlers(struct octeon_device *oct) { @@ -972,13 +977,12 @@ static void liquidio_schedule_droq_pkt_handlers(struct octeon_device *oct) } /** - * \brief Interrupt handler for octeon - * @param irq unused - * @param dev octeon device + * liquidio_legacy_intr_handler - Interrupt handler for octeon + * @irq: unused + * @dev: octeon device */ static -irqreturn_t liquidio_legacy_intr_handler(int irq __attribute__((unused)), - void *dev) +irqreturn_t liquidio_legacy_intr_handler(int __maybe_unused irq, void *dev) { struct octeon_device *oct = (struct octeon_device *)dev; irqreturn_t ret; @@ -999,8 +1003,9 @@ irqreturn_t liquidio_legacy_intr_handler(int irq __attribute__((unused)), } /** - * \brief Setup interrupt for octeon device - * @param oct octeon device + * octeon_setup_interrupt - Setup interrupt for octeon device + * @oct: octeon device + * @num_ioqs: number of queues * * Enable interrupt in Octeon device as given in the PCI interrupt mask. */ @@ -1083,7 +1088,7 @@ int octeon_setup_interrupt(struct octeon_device *oct, u32 num_ioqs) dev_dbg(&oct->pci_dev->dev, "OCTEON: Enough MSI-X interrupts are allocated...\n"); num_ioq_vectors = oct->num_msix_irqs; - /** For PF, there is one non-ioq interrupt handler */ + /* For PF, there is one non-ioq interrupt handler */ if (OCTEON_CN23XX_PF(oct)) { num_ioq_vectors -= 1; @@ -1126,13 +1131,13 @@ int octeon_setup_interrupt(struct octeon_device *oct, u32 num_ioqs) dev_err(&oct->pci_dev->dev, "Request_irq failed for MSIX interrupt Error: %d\n", irqret); - /** Freeing the non-ioq irq vector here . */ + /* Freeing the non-ioq irq vector here . */ free_irq(msix_entries[num_ioq_vectors].vector, oct); while (i) { i--; - /** clearing affinity mask. */ + /* clearing affinity mask. */ irq_set_affinity_hint( msix_entries[i].vector, NULL); @@ -1197,8 +1202,9 @@ int octeon_setup_interrupt(struct octeon_device *oct, u32 num_ioqs) } /** - * \brief Net device change_mtu - * @param netdev network device + * liquidio_change_mtu - Net device change_mtu + * @netdev: network device + * @new_mtu: the new max transmit unit size */ int liquidio_change_mtu(struct net_device *netdev, int new_mtu) { diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 8e0ed01e7f03..7d00d3a8ded4 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -69,9 +69,9 @@ MODULE_PARM_DESC(console_bitmask, "Bitmask indicating which consoles have debug output redirected to syslog."); /** - * \brief determines if a given console has debug enabled. - * @param console console to check - * @returns 1 = enabled. 0 otherwise + * octeon_console_debug_enabled - determines if a given console has debug enabled. + * @console: console to check + * Return: 1 = enabled. 0 otherwise */ static int octeon_console_debug_enabled(u32 console) { @@ -126,7 +126,7 @@ union tx_info { } s; }; -/** Octeon device properties to be used by the NIC module. +/* Octeon device properties to be used by the NIC module. * Each octeon device in the system will be represented * by this structure in the NIC module. */ @@ -161,13 +161,13 @@ static int liquidio_set_vf_link_state(struct net_device *netdev, int vfidx, static struct handshake handshake[MAX_OCTEON_DEVICES]; static struct completion first_stage; -static void octeon_droq_bh(unsigned long pdev) +static void octeon_droq_bh(struct tasklet_struct *t) { int q_no; int reschedule = 0; - struct octeon_device *oct = (struct octeon_device *)pdev; - struct octeon_device_priv *oct_priv = - (struct octeon_device_priv *)oct->priv; + struct octeon_device_priv *oct_priv = from_tasklet(oct_priv, t, + droq_tasklet); + struct octeon_device *oct = oct_priv->dev; for (q_no = 0; q_no < MAX_OCTEON_OUTPUT_QUEUES(oct); q_no++) { if (!(oct->io_qmask.oq & BIT_ULL(q_no))) @@ -222,8 +222,8 @@ static int lio_wait_for_oq_pkts(struct octeon_device *oct) } /** - * \brief Forces all IO queues off on a given device - * @param oct Pointer to Octeon device + * force_io_queues_off - Forces all IO queues off on a given device + * @oct: Pointer to Octeon device */ static void force_io_queues_off(struct octeon_device *oct) { @@ -238,8 +238,8 @@ static void force_io_queues_off(struct octeon_device *oct) } /** - * \brief Cause device to go quiet so it can be safely removed/reset/etc - * @param oct Pointer to Octeon device + * pcierror_quiesce_device - Cause device to go quiet so it can be safely removed/reset/etc + * @oct: Pointer to Octeon device */ static inline void pcierror_quiesce_device(struct octeon_device *oct) { @@ -283,8 +283,8 @@ static inline void pcierror_quiesce_device(struct octeon_device *oct) } /** - * \brief Cleanup PCI AER uncorrectable error status - * @param dev Pointer to PCI device + * cleanup_aer_uncorrect_error_status - Cleanup PCI AER uncorrectable error status + * @dev: Pointer to PCI device */ static void cleanup_aer_uncorrect_error_status(struct pci_dev *dev) { @@ -303,8 +303,8 @@ static void cleanup_aer_uncorrect_error_status(struct pci_dev *dev) } /** - * \brief Stop all PCI IO to a given device - * @param dev Pointer to Octeon device + * stop_pci_io - Stop all PCI IO to a given device + * @oct: Pointer to Octeon device */ static void stop_pci_io(struct octeon_device *oct) { @@ -332,9 +332,9 @@ static void stop_pci_io(struct octeon_device *oct) } /** - * \brief called when PCI error is detected - * @param pdev Pointer to PCI device - * @param state The current pci connection state + * liquidio_pcie_error_detected - called when PCI error is detected + * @pdev: Pointer to PCI device + * @state: The current pci connection state * * This function is called after a PCI bus error affecting * this device has been detected. @@ -362,11 +362,10 @@ static pci_ers_result_t liquidio_pcie_error_detected(struct pci_dev *pdev, } /** - * \brief mmio handler - * @param pdev Pointer to PCI device + * liquidio_pcie_mmio_enabled - mmio handler + * @pdev: Pointer to PCI device */ -static pci_ers_result_t liquidio_pcie_mmio_enabled( - struct pci_dev *pdev __attribute__((unused))) +static pci_ers_result_t liquidio_pcie_mmio_enabled(struct pci_dev __maybe_unused *pdev) { /* We should never hit this since we never ask for a reset for a Fatal * Error. We always return DISCONNECT in io_error above. @@ -376,14 +375,13 @@ static pci_ers_result_t liquidio_pcie_mmio_enabled( } /** - * \brief called after the pci bus has been reset. - * @param pdev Pointer to PCI device + * liquidio_pcie_slot_reset - called after the pci bus has been reset. + * @pdev: Pointer to PCI device * * Restart the card from scratch, as if from a cold-boot. Implementation * resembles the first-half of the octeon_resume routine. */ -static pci_ers_result_t liquidio_pcie_slot_reset( - struct pci_dev *pdev __attribute__((unused))) +static pci_ers_result_t liquidio_pcie_slot_reset(struct pci_dev __maybe_unused *pdev) { /* We should never hit this since we never ask for a reset for a Fatal * Error. We always return DISCONNECT in io_error above. @@ -393,14 +391,14 @@ static pci_ers_result_t liquidio_pcie_slot_reset( } /** - * \brief called when traffic can start flowing again. - * @param pdev Pointer to PCI device + * liquidio_pcie_resume - called when traffic can start flowing again. + * @pdev: Pointer to PCI device * * This callback is called when the error recovery driver tells us that * its OK to resume normal operation. Implementation resembles the * second-half of the octeon_resume routine. */ -static void liquidio_pcie_resume(struct pci_dev *pdev __attribute__((unused))) +static void liquidio_pcie_resume(struct pci_dev __maybe_unused *pdev) { /* Nothing to be done here. */ } @@ -447,7 +445,7 @@ static struct pci_driver liquidio_pci_driver = { }; /** - * \brief register PCI driver + * liquidio_init_pci - register PCI driver */ static int liquidio_init_pci(void) { @@ -455,7 +453,7 @@ static int liquidio_init_pci(void) } /** - * \brief unregister PCI driver + * liquidio_deinit_pci - unregister PCI driver */ static void liquidio_deinit_pci(void) { @@ -463,9 +461,9 @@ static void liquidio_deinit_pci(void) } /** - * \brief Check Tx queue status, and take appropriate action - * @param lio per-network private data - * @returns 0 if full, number of queues woken up otherwise + * check_txq_status - Check Tx queue status, and take appropriate action + * @lio: per-network private data + * Return: 0 if full, number of queues woken up otherwise */ static inline int check_txq_status(struct lio *lio) { @@ -491,8 +489,8 @@ static inline int check_txq_status(struct lio *lio) } /** - * \brief Print link information - * @param netdev network device + * print_link_info - Print link information + * @netdev: network device */ static void print_link_info(struct net_device *netdev) { @@ -513,8 +511,8 @@ static void print_link_info(struct net_device *netdev) } /** - * \brief Routine to notify MTU change - * @param work work_struct data structure + * octnet_link_status_change - Routine to notify MTU change + * @work: work_struct data structure */ static void octnet_link_status_change(struct work_struct *work) { @@ -531,8 +529,8 @@ static void octnet_link_status_change(struct work_struct *work) } /** - * \brief Sets up the mtu status change work - * @param netdev network device + * setup_link_status_change_wq - Sets up the mtu status change work + * @netdev: network device */ static inline int setup_link_status_change_wq(struct net_device *netdev) { @@ -563,9 +561,9 @@ static inline void cleanup_link_status_change_wq(struct net_device *netdev) } /** - * \brief Update link status - * @param netdev network device - * @param ls link status structure + * update_link_status - Update link status + * @netdev: network device + * @ls: link status structure * * Called on receipt of a link status response from the core application to * update each interface's link status. @@ -663,10 +661,9 @@ static void lio_sync_octeon_time(struct work_struct *work) } /** - * setup_sync_octeon_time_wq - Sets up the work to periodically update - * local time to octeon firmware + * setup_sync_octeon_time_wq - prepare work to periodically update local time to octeon firmware * - * @netdev - network device which should send time update to firmware + * @netdev: network device which should send time update to firmware **/ static inline int setup_sync_octeon_time_wq(struct net_device *netdev) { @@ -690,10 +687,12 @@ static inline int setup_sync_octeon_time_wq(struct net_device *netdev) } /** - * cleanup_sync_octeon_time_wq - stop scheduling and destroy the work created - * to periodically update local time to octeon firmware + * cleanup_sync_octeon_time_wq - destroy wq * - * @netdev - network device which should send time update to firmware + * @netdev: network device which should send time update to firmware + * + * Stop scheduling and destroy the work created to periodically update local + * time to octeon firmware. **/ static inline void cleanup_sync_octeon_time_wq(struct net_device *netdev) { @@ -828,13 +827,12 @@ static int liquidio_watchdog(void *param) } /** - * \brief PCI probe handler - * @param pdev PCI device structure - * @param ent unused + * liquidio_probe - PCI probe handler + * @pdev: PCI device structure + * @ent: unused */ static int -liquidio_probe(struct pci_dev *pdev, - const struct pci_device_id *ent __attribute__((unused))) +liquidio_probe(struct pci_dev *pdev, const struct pci_device_id __maybe_unused *ent) { struct octeon_device *oct_dev = NULL; struct handshake *hs; @@ -924,8 +922,8 @@ static bool fw_type_is_auto(void) } /** - * \brief PCI FLR for each Octeon device. - * @param oct octeon device + * octeon_pci_flr - PCI FLR for each Octeon device. + * @oct: octeon device */ static void octeon_pci_flr(struct octeon_device *oct) { @@ -951,9 +949,8 @@ static void octeon_pci_flr(struct octeon_device *oct) } /** - *\brief Destroy resources associated with octeon device - * @param pdev PCI device structure - * @param ent unused + * octeon_destroy_resources - Destroy resources associated with octeon device + * @oct: octeon device */ static void octeon_destroy_resources(struct octeon_device *oct) { @@ -1152,9 +1149,9 @@ static void octeon_destroy_resources(struct octeon_device *oct) } /** - * \brief Send Rx control command - * @param lio per-network private data - * @param start_stop whether to start or stop + * send_rx_ctrl_cmd - Send Rx control command + * @lio: per-network private data + * @start_stop: whether to start or stop */ static void send_rx_ctrl_cmd(struct lio *lio, int start_stop) { @@ -1210,9 +1207,9 @@ static void send_rx_ctrl_cmd(struct lio *lio, int start_stop) } /** - * \brief Destroy NIC device interface - * @param oct octeon device - * @param ifidx which interface to destroy + * liquidio_destroy_nic_device - Destroy NIC device interface + * @oct: octeon device + * @ifidx: which interface to destroy * * Cleanup associated with each interface for an Octeon device when NIC * module is being unloaded or if initialization fails during load. @@ -1272,8 +1269,8 @@ static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx) } /** - * \brief Stop complete NIC functionality - * @param oct octeon device + * liquidio_stop_nic_module - Stop complete NIC functionality + * @oct: octeon device */ static int liquidio_stop_nic_module(struct octeon_device *oct) { @@ -1313,8 +1310,8 @@ static int liquidio_stop_nic_module(struct octeon_device *oct) } /** - * \brief Cleans up resources at unload time - * @param pdev PCI device structure + * liquidio_remove - Cleans up resources at unload time + * @pdev: PCI device structure */ static void liquidio_remove(struct pci_dev *pdev) { @@ -1346,8 +1343,8 @@ static void liquidio_remove(struct pci_dev *pdev) } /** - * \brief Identify the Octeon device and to map the BAR address space - * @param oct octeon device + * octeon_chip_specific_setup - Identify the Octeon device and to map the BAR address space + * @oct: octeon device */ static int octeon_chip_specific_setup(struct octeon_device *oct) { @@ -1390,8 +1387,8 @@ static int octeon_chip_specific_setup(struct octeon_device *oct) } /** - * \brief PCI initialization for each Octeon device. - * @param oct octeon device + * octeon_pci_os_setup - PCI initialization for each Octeon device. + * @oct: octeon device */ static int octeon_pci_os_setup(struct octeon_device *oct) { @@ -1414,8 +1411,8 @@ static int octeon_pci_os_setup(struct octeon_device *oct) } /** - * \brief Unmap and free network buffer - * @param buf buffer + * free_netbuf - Unmap and free network buffer + * @buf: buffer */ static void free_netbuf(void *buf) { @@ -1434,8 +1431,8 @@ static void free_netbuf(void *buf) } /** - * \brief Unmap and free gather buffer - * @param buf buffer + * free_netsgbuf - Unmap and free gather buffer + * @buf: buffer */ static void free_netsgbuf(void *buf) { @@ -1474,8 +1471,8 @@ static void free_netsgbuf(void *buf) } /** - * \brief Unmap and free gather buffer with response - * @param buf buffer + * free_netsgbuf_with_resp - Unmap and free gather buffer with response + * @buf: buffer */ static void free_netsgbuf_with_resp(void *buf) { @@ -1518,9 +1515,9 @@ static void free_netsgbuf_with_resp(void *buf) } /** - * \brief Adjust ptp frequency - * @param ptp PTP clock info - * @param ppb how much to adjust by, in parts-per-billion + * liquidio_ptp_adjfreq - Adjust ptp frequency + * @ptp: PTP clock info + * @ppb: how much to adjust by, in parts-per-billion */ static int liquidio_ptp_adjfreq(struct ptp_clock_info *ptp, s32 ppb) { @@ -1555,9 +1552,9 @@ static int liquidio_ptp_adjfreq(struct ptp_clock_info *ptp, s32 ppb) } /** - * \brief Adjust ptp time - * @param ptp PTP clock info - * @param delta how much to adjust by, in nanosecs + * liquidio_ptp_adjtime - Adjust ptp time + * @ptp: PTP clock info + * @delta: how much to adjust by, in nanosecs */ static int liquidio_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta) { @@ -1572,9 +1569,9 @@ static int liquidio_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta) } /** - * \brief Get hardware clock time, including any adjustment - * @param ptp PTP clock info - * @param ts timespec + * liquidio_ptp_gettime - Get hardware clock time, including any adjustment + * @ptp: PTP clock info + * @ts: timespec */ static int liquidio_ptp_gettime(struct ptp_clock_info *ptp, struct timespec64 *ts) @@ -1595,9 +1592,9 @@ static int liquidio_ptp_gettime(struct ptp_clock_info *ptp, } /** - * \brief Set hardware clock time. Reset adjustment - * @param ptp PTP clock info - * @param ts timespec + * liquidio_ptp_settime - Set hardware clock time. Reset adjustment + * @ptp: PTP clock info + * @ts: timespec */ static int liquidio_ptp_settime(struct ptp_clock_info *ptp, const struct timespec64 *ts) @@ -1618,22 +1615,22 @@ static int liquidio_ptp_settime(struct ptp_clock_info *ptp, } /** - * \brief Check if PTP is enabled - * @param ptp PTP clock info - * @param rq request - * @param on is it on + * liquidio_ptp_enable - Check if PTP is enabled + * @ptp: PTP clock info + * @rq: request + * @on: is it on */ static int -liquidio_ptp_enable(struct ptp_clock_info *ptp __attribute__((unused)), - struct ptp_clock_request *rq __attribute__((unused)), - int on __attribute__((unused))) +liquidio_ptp_enable(struct ptp_clock_info __maybe_unused *ptp, + struct ptp_clock_request __maybe_unused *rq, + int __maybe_unused on) { return -EOPNOTSUPP; } /** - * \brief Open PTP clock source - * @param netdev network device + * oct_ptp_open - Open PTP clock source + * @netdev: network device */ static void oct_ptp_open(struct net_device *netdev) { @@ -1665,8 +1662,8 @@ static void oct_ptp_open(struct net_device *netdev) } /** - * \brief Init PTP clock - * @param oct octeon device + * liquidio_ptp_init - Init PTP clock + * @oct: octeon device */ static void liquidio_ptp_init(struct octeon_device *oct) { @@ -1682,8 +1679,8 @@ static void liquidio_ptp_init(struct octeon_device *oct) } /** - * \brief Load firmware to device - * @param oct octeon device + * load_firmware - Load firmware to device + * @oct: octeon device * * Maps device to firmware filename, requests firmware, and downloads it */ @@ -1721,8 +1718,8 @@ static int load_firmware(struct octeon_device *oct) } /** - * \brief Poll routine for checking transmit queue status - * @param work work_struct data structure + * octnet_poll_check_txq_status - Poll routine for checking transmit queue status + * @work: work_struct data structure */ static void octnet_poll_check_txq_status(struct work_struct *work) { @@ -1738,8 +1735,8 @@ static void octnet_poll_check_txq_status(struct work_struct *work) } /** - * \brief Sets up the txq poll check - * @param netdev network device + * setup_tx_poll_fn - Sets up the txq poll check + * @netdev: network device */ static inline int setup_tx_poll_fn(struct net_device *netdev) { @@ -1771,8 +1768,8 @@ static inline void cleanup_tx_poll_fn(struct net_device *netdev) } /** - * \brief Net device open for LiquidIO - * @param netdev network device + * liquidio_open - Net device open for LiquidIO + * @netdev: network device */ static int liquidio_open(struct net_device *netdev) { @@ -1831,8 +1828,8 @@ static int liquidio_open(struct net_device *netdev) } /** - * \brief Net device stop for LiquidIO - * @param netdev network device + * liquidio_stop - Net device stop for LiquidIO + * @netdev: network device */ static int liquidio_stop(struct net_device *netdev) { @@ -1896,8 +1893,8 @@ static int liquidio_stop(struct net_device *netdev) } /** - * \brief Converts a mask based on net device flags - * @param netdev network device + * get_new_flags - Converts a mask based on net device flags + * @netdev: network device * * This routine generates a octnet_ifflags mask from the net device flags * received from the OS. @@ -1929,8 +1926,8 @@ static inline enum octnet_ifflags get_new_flags(struct net_device *netdev) } /** - * \brief Net device set_multicast_list - * @param netdev network device + * liquidio_set_mcast_list - Net device set_multicast_list + * @netdev: network device */ static void liquidio_set_mcast_list(struct net_device *netdev) { @@ -1977,8 +1974,9 @@ static void liquidio_set_mcast_list(struct net_device *netdev) } /** - * \brief Net device set_mac_address - * @param netdev network device + * liquidio_set_mac - Net device set_mac_address + * @netdev: network device + * @p: pointer to sockaddr */ static int liquidio_set_mac(struct net_device *netdev, void *p) { @@ -2096,10 +2094,9 @@ liquidio_get_stats64(struct net_device *netdev, } /** - * \brief Handler for SIOCSHWTSTAMP ioctl - * @param netdev network device - * @param ifr interface request - * @param cmd command + * hwtstamp_ioctl - Handler for SIOCSHWTSTAMP ioctl + * @netdev: network device + * @ifr: interface request */ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr) { @@ -2154,10 +2151,10 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr) } /** - * \brief ioctl handler - * @param netdev network device - * @param ifr interface request - * @param cmd command + * liquidio_ioctl - ioctl handler + * @netdev: network device + * @ifr: interface request + * @cmd: command */ static int liquidio_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) { @@ -2174,9 +2171,10 @@ static int liquidio_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) } /** - * \brief handle a Tx timestamp response - * @param status response status - * @param buf pointer to skb + * handle_timestamp - handle a Tx timestamp response + * @oct: octeon device + * @status: response status + * @buf: pointer to skb */ static void handle_timestamp(struct octeon_device *oct, u32 status, @@ -2217,10 +2215,12 @@ static void handle_timestamp(struct octeon_device *oct, tx_buffer_free(skb); } -/* \brief Send a data packet that will be timestamped - * @param oct octeon device - * @param ndata pointer to network data - * @param finfo pointer to private network data +/** + * send_nic_timestamp_pkt - Send a data packet that will be timestamped + * @oct: octeon device + * @ndata: pointer to network data + * @finfo: pointer to private network data + * @xmit_more: more is coming */ static inline int send_nic_timestamp_pkt(struct octeon_device *oct, struct octnic_data_pkt *ndata, @@ -2276,10 +2276,12 @@ static inline int send_nic_timestamp_pkt(struct octeon_device *oct, return retval; } -/** \brief Transmit networks packets to the Octeon interface - * @param skbuff skbuff struct to be passed to network layer. - * @param netdev pointer to network device - * @returns whether the packet was transmitted to the device okay or not +/** + * liquidio_xmit - Transmit networks packets to the Octeon interface + * @skb: skbuff struct to be passed to network layer. + * @netdev: pointer to network device + * + * Return: whether the packet was transmitted to the device okay or not * (NETDEV_TX_OK or NETDEV_TX_BUSY) */ static netdev_tx_t liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) @@ -2524,8 +2526,10 @@ lio_xmit_failed: return NETDEV_TX_OK; } -/** \brief Network device Tx timeout - * @param netdev pointer to network device +/** + * liquidio_tx_timeout - Network device Tx timeout + * @netdev: pointer to network device + * @txqueue: index of the hung transmit queue */ static void liquidio_tx_timeout(struct net_device *netdev, unsigned int txqueue) { @@ -2597,12 +2601,12 @@ static int liquidio_vlan_rx_kill_vid(struct net_device *netdev, return ret; } -/** Sending command to enable/disable RX checksum offload - * @param netdev pointer to network device - * @param command OCTNET_CMD_TNL_RX_CSUM_CTL - * @param rx_cmd_bit OCTNET_CMD_RXCSUM_ENABLE/ - * OCTNET_CMD_RXCSUM_DISABLE - * @returns SUCCESS or FAILURE +/** + * liquidio_set_rxcsum_command - Sending command to enable/disable RX checksum offload + * @netdev: pointer to network device + * @command: OCTNET_CMD_TNL_RX_CSUM_CTL + * @rx_cmd: OCTNET_CMD_RXCSUM_ENABLE/OCTNET_CMD_RXCSUM_DISABLE + * Returns: SUCCESS or FAILURE */ static int liquidio_set_rxcsum_command(struct net_device *netdev, int command, u8 rx_cmd) @@ -2632,13 +2636,14 @@ static int liquidio_set_rxcsum_command(struct net_device *netdev, int command, return ret; } -/** Sending command to add/delete VxLAN UDP port to firmware - * @param netdev pointer to network device - * @param command OCTNET_CMD_VXLAN_PORT_CONFIG - * @param vxlan_port VxLAN port to be added or deleted - * @param vxlan_cmd_bit OCTNET_CMD_VXLAN_PORT_ADD, +/** + * liquidio_vxlan_port_command - Sending command to add/delete VxLAN UDP port to firmware + * @netdev: pointer to network device + * @command: OCTNET_CMD_VXLAN_PORT_CONFIG + * @vxlan_port: VxLAN port to be added or deleted + * @vxlan_cmd_bit: OCTNET_CMD_VXLAN_PORT_ADD, * OCTNET_CMD_VXLAN_PORT_DEL - * @returns SUCCESS or FAILURE + * Return: SUCCESS or FAILURE */ static int liquidio_vxlan_port_command(struct net_device *netdev, int command, u16 vxlan_port, u8 vxlan_cmd_bit) @@ -2698,10 +2703,11 @@ static const struct udp_tunnel_nic_info liquidio_udp_tunnels = { }, }; -/** \brief Net device fix features - * @param netdev pointer to network device - * @param request features requested - * @returns updated features list +/** + * liquidio_fix_features - Net device fix features + * @netdev: pointer to network device + * @request: features requested + * Return: updated features list */ static netdev_features_t liquidio_fix_features(struct net_device *netdev, netdev_features_t request) @@ -2737,9 +2743,10 @@ static netdev_features_t liquidio_fix_features(struct net_device *netdev, return request; } -/** \brief Net device set features - * @param netdev pointer to network device - * @param features features to enable/disable +/** + * liquidio_set_features - Net device set features + * @netdev: pointer to network device + * @features: features to enable/disable */ static int liquidio_set_features(struct net_device *netdev, netdev_features_t features) @@ -3224,7 +3231,8 @@ static const struct net_device_ops lionetdevops = { .ndo_get_port_parent_id = liquidio_get_port_parent_id, }; -/** \brief Entry point for the liquidio module +/** + * liquidio_init - Entry point for the liquidio module */ static int __init liquidio_init(void) { @@ -3307,8 +3315,8 @@ nic_info_err: } /** - * \brief Setup network interfaces - * @param octeon_dev octeon device + * setup_nic_devices - Setup network interfaces + * @octeon_dev: octeon device * * Called during init time for each device. It assumes the NIC * is already up and running. The link information for each @@ -3872,8 +3880,8 @@ static int liquidio_enable_sriov(struct pci_dev *dev, int num_vfs) #endif /** - * \brief initialize the NIC - * @param oct octeon device + * liquidio_init_nic_module - initialize the NIC + * @oct: octeon device * * This initialization routine is called once the Octeon device application is * up and running @@ -3928,9 +3936,10 @@ octnet_init_failure: } /** - * \brief starter callback that invokes the remaining initialization work after - * the NIC is up and running. - * @param octptr work struct work_struct + * nic_starter - finish init + * @work: work struct work_struct + * + * starter callback that invokes the remaining initialization work after the NIC is up and running. */ static void nic_starter(struct work_struct *work) { @@ -4023,8 +4032,8 @@ octeon_recv_vf_drv_notice(struct octeon_recv_info *recv_info, void *buf) } /** - * \brief Device initialization for each Octeon device that is probed - * @param octeon_dev octeon device + * octeon_device_init - Device initialization for each Octeon device that is probed + * @octeon_dev: octeon device */ static int octeon_device_init(struct octeon_device *octeon_dev) { @@ -4193,8 +4202,7 @@ static int octeon_device_init(struct octeon_device *octeon_dev) /* Initialize the tasklet that handles output queue packet processing.*/ dev_dbg(&octeon_dev->pci_dev->dev, "Initializing droq tasklet\n"); - tasklet_init(&oct_priv->droq_tasklet, octeon_droq_bh, - (unsigned long)octeon_dev); + tasklet_setup(&oct_priv->droq_tasklet, octeon_droq_bh); /* Setup the interrupt handler and record the INT SUM register address */ @@ -4298,16 +4306,17 @@ static int octeon_device_init(struct octeon_device *octeon_dev) complete(&handshake[octeon_dev->octeon_id].init); atomic_set(&octeon_dev->status, OCT_DEV_HOST_OK); + oct_priv->dev = octeon_dev; return 0; } /** - * \brief Debug console print function - * @param octeon_dev octeon device - * @param console_num console number - * @param prefix first portion of line to display - * @param suffix second portion of line to display + * octeon_dbg_console_print - Debug console print function + * @oct: octeon device + * @console_num: console number + * @prefix: first portion of line to display + * @suffix: second portion of line to display * * The OCTEON debug console outputs entire lines (excluding '\n'). * Normally, the line will be passed in the 'prefix' parameter. @@ -4330,7 +4339,7 @@ static int octeon_dbg_console_print(struct octeon_device *oct, u32 console_num, } /** - * \brief Exits the module + * liquidio_exit - Exits the module */ static void __exit liquidio_exit(void) { diff --git a/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c b/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c index 8c5879e31240..103440f97bc8 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_vf_main.c @@ -99,8 +99,8 @@ static int lio_wait_for_oq_pkts(struct octeon_device *oct) } /** - * \brief Cause device to go quiet so it can be safely removed/reset/etc - * @param oct Pointer to Octeon device + * pcierror_quiesce_device - Cause device to go quiet so it can be safely removed/reset/etc + * @oct: Pointer to Octeon device */ static void pcierror_quiesce_device(struct octeon_device *oct) { @@ -143,8 +143,8 @@ static void pcierror_quiesce_device(struct octeon_device *oct) } /** - * \brief Cleanup PCI AER uncorrectable error status - * @param dev Pointer to PCI device + * cleanup_aer_uncorrect_error_status - Cleanup PCI AER uncorrectable error status + * @dev: Pointer to PCI device */ static void cleanup_aer_uncorrect_error_status(struct pci_dev *dev) { @@ -163,8 +163,8 @@ static void cleanup_aer_uncorrect_error_status(struct pci_dev *dev) } /** - * \brief Stop all PCI IO to a given device - * @param dev Pointer to Octeon device + * stop_pci_io - Stop all PCI IO to a given device + * @oct: Pointer to Octeon device */ static void stop_pci_io(struct octeon_device *oct) { @@ -205,9 +205,9 @@ static void stop_pci_io(struct octeon_device *oct) } /** - * \brief called when PCI error is detected - * @param pdev Pointer to PCI device - * @param state The current pci connection state + * liquidio_pcie_error_detected - called when PCI error is detected + * @pdev: Pointer to PCI device + * @state: The current pci connection state * * This function is called after a PCI bus error affecting * this device has been detected. @@ -256,8 +256,8 @@ static struct pci_driver liquidio_vf_pci_driver = { }; /** - * \brief Print link information - * @param netdev network device + * print_link_info - Print link information + * @netdev: network device */ static void print_link_info(struct net_device *netdev) { @@ -278,8 +278,8 @@ static void print_link_info(struct net_device *netdev) } /** - * \brief Routine to notify MTU change - * @param work work_struct data structure + * octnet_link_status_change - Routine to notify MTU change + * @work: work_struct data structure */ static void octnet_link_status_change(struct work_struct *work) { @@ -296,8 +296,8 @@ static void octnet_link_status_change(struct work_struct *work) } /** - * \brief Sets up the mtu status change work - * @param netdev network device + * setup_link_status_change_wq - Sets up the mtu status change work + * @netdev: network device */ static int setup_link_status_change_wq(struct net_device *netdev) { @@ -328,9 +328,9 @@ static void cleanup_link_status_change_wq(struct net_device *netdev) } /** - * \brief Update link status - * @param netdev network device - * @param ls link status structure + * update_link_status - Update link status + * @netdev: network device + * @ls: link status structure * * Called on receipt of a link status response from the core application to * update each interface's link status. @@ -374,13 +374,13 @@ static void update_link_status(struct net_device *netdev, } /** - * \brief PCI probe handler - * @param pdev PCI device structure - * @param ent unused + * liquidio_vf_probe - PCI probe handler + * @pdev: PCI device structure + * @ent: unused */ static int liquidio_vf_probe(struct pci_dev *pdev, - const struct pci_device_id *ent __attribute__((unused))) + const struct pci_device_id __maybe_unused *ent) { struct octeon_device *oct_dev = NULL; @@ -416,8 +416,8 @@ liquidio_vf_probe(struct pci_dev *pdev, } /** - * \brief PCI FLR for each Octeon device. - * @param oct octeon device + * octeon_pci_flr - PCI FLR for each Octeon device. + * @oct: octeon device */ static void octeon_pci_flr(struct octeon_device *oct) { @@ -437,9 +437,8 @@ static void octeon_pci_flr(struct octeon_device *oct) } /** - *\brief Destroy resources associated with octeon device - * @param pdev PCI device structure - * @param ent unused + * octeon_destroy_resources - Destroy resources associated with octeon device + * @oct: octeon device */ static void octeon_destroy_resources(struct octeon_device *oct) { @@ -592,9 +591,9 @@ static void octeon_destroy_resources(struct octeon_device *oct) } /** - * \brief Send Rx control command - * @param lio per-network private data - * @param start_stop whether to start or stop + * send_rx_ctrl_cmd - Send Rx control command + * @lio: per-network private data + * @start_stop: whether to start or stop */ static void send_rx_ctrl_cmd(struct lio *lio, int start_stop) { @@ -644,9 +643,9 @@ static void send_rx_ctrl_cmd(struct lio *lio, int start_stop) } /** - * \brief Destroy NIC device interface - * @param oct octeon device - * @param ifidx which interface to destroy + * liquidio_destroy_nic_device - Destroy NIC device interface + * @oct: octeon device + * @ifidx: which interface to destroy * * Cleanup associated with each interface for an Octeon device when NIC * module is being unloaded or if initialization fails during load. @@ -704,8 +703,8 @@ static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx) } /** - * \brief Stop complete NIC functionality - * @param oct octeon device + * liquidio_stop_nic_module - Stop complete NIC functionality + * @oct: octeon device */ static int liquidio_stop_nic_module(struct octeon_device *oct) { @@ -737,8 +736,8 @@ static int liquidio_stop_nic_module(struct octeon_device *oct) } /** - * \brief Cleans up resources at unload time - * @param pdev PCI device structure + * liquidio_vf_remove - Cleans up resources at unload time + * @pdev: PCI device structure */ static void liquidio_vf_remove(struct pci_dev *pdev) { @@ -763,8 +762,8 @@ static void liquidio_vf_remove(struct pci_dev *pdev) } /** - * \brief PCI initialization for each Octeon device. - * @param oct octeon device + * octeon_pci_os_setup - PCI initialization for each Octeon device. + * @oct: octeon device */ static int octeon_pci_os_setup(struct octeon_device *oct) { @@ -792,8 +791,8 @@ static int octeon_pci_os_setup(struct octeon_device *oct) } /** - * \brief Unmap and free network buffer - * @param buf buffer + * free_netbuf - Unmap and free network buffer + * @buf: buffer */ static void free_netbuf(void *buf) { @@ -812,8 +811,8 @@ static void free_netbuf(void *buf) } /** - * \brief Unmap and free gather buffer - * @param buf buffer + * free_netsgbuf - Unmap and free gather buffer + * @buf: buffer */ static void free_netsgbuf(void *buf) { @@ -853,8 +852,8 @@ static void free_netsgbuf(void *buf) } /** - * \brief Unmap and free gather buffer with response - * @param buf buffer + * free_netsgbuf_with_resp - Unmap and free gather buffer with response + * @buf: buffer */ static void free_netsgbuf_with_resp(void *buf) { @@ -897,8 +896,8 @@ static void free_netsgbuf_with_resp(void *buf) } /** - * \brief Net device open for LiquidIO - * @param netdev network device + * liquidio_open - Net device open for LiquidIO + * @netdev: network device */ static int liquidio_open(struct net_device *netdev) { @@ -941,8 +940,8 @@ static int liquidio_open(struct net_device *netdev) } /** - * \brief Net device stop for LiquidIO - * @param netdev network device + * liquidio_stop - jNet device stop for LiquidIO + * @netdev: network device */ static int liquidio_stop(struct net_device *netdev) { @@ -991,8 +990,8 @@ static int liquidio_stop(struct net_device *netdev) } /** - * \brief Converts a mask based on net device flags - * @param netdev network device + * get_new_flags - Converts a mask based on net device flags + * @netdev: network device * * This routine generates a octnet_ifflags mask from the net device flags * received from the OS. @@ -1060,8 +1059,8 @@ static void liquidio_set_uc_list(struct net_device *netdev) } /** - * \brief Net device set_multicast_list - * @param netdev network device + * liquidio_set_mcast_list - Net device set_multicast_list + * @netdev: network device */ static void liquidio_set_mcast_list(struct net_device *netdev) { @@ -1110,8 +1109,9 @@ static void liquidio_set_mcast_list(struct net_device *netdev) } /** - * \brief Net device set_mac_address - * @param netdev network device + * liquidio_set_mac - Net device set_mac_address + * @netdev: network device + * @p: opaque pointer to sockaddr */ static int liquidio_set_mac(struct net_device *netdev, void *p) { @@ -1229,10 +1229,9 @@ liquidio_get_stats64(struct net_device *netdev, } /** - * \brief Handler for SIOCSHWTSTAMP ioctl - * @param netdev network device - * @param ifr interface request - * @param cmd command + * hwtstamp_ioctl - Handler for SIOCSHWTSTAMP ioctl + * @netdev: network device + * @ifr: interface request */ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr) { @@ -1287,10 +1286,10 @@ static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr) } /** - * \brief ioctl handler - * @param netdev network device - * @param ifr interface request - * @param cmd command + * liquidio_ioctl - ioctl handler + * @netdev: network device + * @ifr: interface request + * @cmd: command */ static int liquidio_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) { @@ -1339,10 +1338,10 @@ static void handle_timestamp(struct octeon_device *oct, u32 status, void *buf) tx_buffer_free(skb); } -/* \brief Send a data packet that will be timestamped - * @param oct octeon device - * @param ndata pointer to network data - * @param finfo pointer to private network data +/* send_nic_timestamp_pkt - Send a data packet that will be timestamped + * @oct: octeon device + * @ndata: pointer to network data + * @finfo: pointer to private network data */ static int send_nic_timestamp_pkt(struct octeon_device *oct, struct octnic_data_pkt *ndata, @@ -1393,9 +1392,10 @@ static int send_nic_timestamp_pkt(struct octeon_device *oct, return retval; } -/** \brief Transmit networks packets to the Octeon interface - * @param skbuff skbuff struct to be passed to network layer. - * @param netdev pointer to network device +/** + * liquidio_xmit - Transmit networks packets to the Octeon interface + * @skb: skbuff struct to be passed to network layer. + * @netdev: pointer to network device * @returns whether the packet was transmitted to the device okay or not * (NETDEV_TX_OK or NETDEV_TX_BUSY) */ @@ -1623,8 +1623,10 @@ lio_xmit_failed: return NETDEV_TX_OK; } -/** \brief Network device Tx timeout - * @param netdev pointer to network device +/** + * liquidio_tx_timeout - Network device Tx timeout + * @netdev: pointer to network device + * @txqueue: index of the hung transmit queue */ static void liquidio_tx_timeout(struct net_device *netdev, unsigned int txqueue) { @@ -1917,8 +1919,8 @@ nic_info_err: } /** - * \brief Setup network interfaces - * @param octeon_dev octeon device + * setup_nic_devices - Setup network interfaces + * @octeon_dev: octeon device * * Called during init time for each device. It assumes the NIC * is already up and running. The link information for each @@ -2229,8 +2231,8 @@ setup_nic_dev_done: } /** - * \brief initialize the NIC - * @param oct octeon device + * liquidio_init_nic_module - initialize the NIC + * @oct: octeon device * * This initialization routine is called once the Octeon device application is * up and running @@ -2270,8 +2272,8 @@ octnet_init_failure: } /** - * \brief Device initialization for each Octeon device that is probed - * @param octeon_dev octeon device + * octeon_device_init - Device initialization for each Octeon device that is probed + * @oct: octeon device */ static int octeon_device_init(struct octeon_device *oct) { diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_console.c b/drivers/net/ethernet/cavium/liquidio/octeon_console.c index 0d2831d10f65..28feabec8fbb 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_console.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_console.c @@ -15,7 +15,7 @@ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or * NONINFRINGEMENT. See the GNU General Public License for more details. ***********************************************************************/ -/** +/* * @file octeon_console.c */ #include <linux/moduleparam.h> @@ -131,7 +131,7 @@ struct octeon_pci_console_desc { /* Implicit storage for console_addr_array */ }; -/** +/* * This function is the implementation of the get macros defined * for individual structure members. The argument are generated * by the macros inorder to read only the needed memory. @@ -160,7 +160,7 @@ static inline u64 __cvmx_bootmem_desc_get(struct octeon_device *oct, } } -/** +/* * This function retrieves the string name of a named block. It is * more complicated than a simple memcpy() since the named block * descriptor may not be directly accessible. @@ -182,7 +182,7 @@ static void CVMX_BOOTMEM_NAMED_GET_NAME(struct octeon_device *oct, /* See header file for descriptions of functions */ -/** +/* * Check the version information on the bootmem descriptor * * @param exact_match @@ -323,7 +323,7 @@ static u64 cvmx_bootmem_phy_named_block_find(struct octeon_device *oct, return result; } -/** +/* * Find a named block on the remote Octeon * * @param name Name of block to find @@ -707,7 +707,7 @@ int octeon_add_console(struct octeon_device *oct, u32 console_num, return ret; } -/** +/* * Removes all consoles * * @param oct octeon device diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c index ac32facaa427..387a57cbfb73 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c @@ -1307,7 +1307,7 @@ struct octeon_config *octeon_get_conf(struct octeon_device *oct) /* scratch register address is same in all the OCT-II and CN70XX models */ #define CNXX_SLI_SCRATCH1 0x3C0 -/** Get the octeon device pointer. +/* Get the octeon device pointer. * @param octeon_id - The id for which the octeon device pointer is required. * @return Success: Octeon device pointer. * @return Failure: NULL. @@ -1324,7 +1324,7 @@ u64 lio_pci_readq(struct octeon_device *oct, u64 addr) { u64 val64; unsigned long flags; - u32 val32, addrhi; + u32 addrhi; spin_lock_irqsave(&oct->pci_win_lock, flags); @@ -1339,10 +1339,10 @@ u64 lio_pci_readq(struct octeon_device *oct, u64 addr) writel(addrhi, oct->reg_list.pci_win_rd_addr_hi); /* Read back to preserve ordering of writes */ - val32 = readl(oct->reg_list.pci_win_rd_addr_hi); + readl(oct->reg_list.pci_win_rd_addr_hi); writel(addr & 0xffffffff, oct->reg_list.pci_win_rd_addr_lo); - val32 = readl(oct->reg_list.pci_win_rd_addr_lo); + readl(oct->reg_list.pci_win_rd_addr_lo); val64 = readq(oct->reg_list.pci_win_rd_data); @@ -1355,7 +1355,6 @@ void lio_pci_writeq(struct octeon_device *oct, u64 val, u64 addr) { - u32 val32; unsigned long flags; spin_lock_irqsave(&oct->pci_win_lock, flags); @@ -1365,7 +1364,7 @@ void lio_pci_writeq(struct octeon_device *oct, /* The write happens when the LSB is written. So write MSB first. */ writel(val >> 32, oct->reg_list.pci_win_wr_data_hi); /* Read the MSB to ensure ordering of writes. */ - val32 = readl(oct->reg_list.pci_win_wr_data_hi); + readl(oct->reg_list.pci_win_wr_data_hi); writel(val & 0xffffffff, oct->reg_list.pci_win_wr_data_lo); @@ -1411,7 +1410,7 @@ int octeon_wait_for_ddr_init(struct octeon_device *oct, u32 *timeout) return ret; } -/** Get the octeon id assigned to the octeon device passed as argument. +/* Get the octeon id assigned to the octeon device passed as argument. * This function is exported to other modules. * @param dev - octeon device pointer passed as a void *. * @return octeon device id diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c index 017169023cca..d4080bddcb6b 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c @@ -280,13 +280,10 @@ int octeon_init_droq(struct octeon_device *oct, dev_dbg(&oct->pci_dev->dev, "droq[%d]: num_desc: %d\n", q_no, droq->max_count); - droq->recv_buf_list = (struct octeon_recv_buffer *) - vzalloc_node(array_size(droq->max_count, OCT_DROQ_RECVBUF_SIZE), - numa_node); + droq->recv_buf_list = vzalloc_node(array_size(droq->max_count, OCT_DROQ_RECVBUF_SIZE), + numa_node); if (!droq->recv_buf_list) - droq->recv_buf_list = (struct octeon_recv_buffer *) - vzalloc(array_size(droq->max_count, - OCT_DROQ_RECVBUF_SIZE)); + droq->recv_buf_list = vzalloc(array_size(droq->max_count, OCT_DROQ_RECVBUF_SIZE)); if (!droq->recv_buf_list) { dev_err(&oct->pci_dev->dev, "Output queue recv buf list alloc failed\n"); goto init_droq_fail; @@ -777,7 +774,7 @@ octeon_droq_process_packets(struct octeon_device *oct, return 0; } -/** +/* * Utility function to poll for packets. check_hw_for_packets must be * called before calling this routine. */ diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c b/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c index 614d07be7181..ad685f5d0a13 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c @@ -28,7 +28,7 @@ /** * octeon_mbox_read: - * @oct: Pointer mailbox + * @mbox: Pointer mailbox * * Reads the 8-bytes of data from the mbox register * Writes back the acknowldgement inidcating completion of read @@ -285,7 +285,8 @@ static int octeon_mbox_process_cmd(struct octeon_mbox *mbox, } /** - *octeon_mbox_process_message: + * octeon_mbox_process_message + * @mbox: mailbox * * Process the received mbox message. */ diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_main.h b/drivers/net/ethernet/cavium/liquidio/octeon_main.h index 073d0647b439..5b4cb725f60f 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_main.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_main.h @@ -39,6 +39,7 @@ struct octeon_device_priv { /** Tasklet structures for this device. */ struct tasklet_struct droq_tasklet; unsigned long napi_mask; + struct octeon_device *dev; }; /** This structure is used by NIC driver to store information required diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c index 4c85ae643b7b..7ccab36143c1 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c @@ -22,6 +22,7 @@ #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" +#include "octeon_mem_ops.h" #define MEMOPS_IDX BAR1_INDEX_DYNAMIC_MAP diff --git a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c index 6cb2162a75d4..5e50bb19bf26 100644 --- a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c +++ b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c @@ -315,9 +315,9 @@ static void octeon_mgmt_clean_tx_buffers(struct octeon_mgmt *p) netif_wake_queue(p->netdev); } -static void octeon_mgmt_clean_tx_tasklet(unsigned long arg) +static void octeon_mgmt_clean_tx_tasklet(struct tasklet_struct *t) { - struct octeon_mgmt *p = (struct octeon_mgmt *)arg; + struct octeon_mgmt *p = from_tasklet(p, t, tx_clean_tasklet); octeon_mgmt_clean_tx_buffers(p); octeon_mgmt_enable_tx_irq(p); } @@ -1491,8 +1491,8 @@ static int octeon_mgmt_probe(struct platform_device *pdev) skb_queue_head_init(&p->tx_list); skb_queue_head_init(&p->rx_list); - tasklet_init(&p->tx_clean_tasklet, - octeon_mgmt_clean_tx_tasklet, (unsigned long)p); + tasklet_setup(&p->tx_clean_tasklet, + octeon_mgmt_clean_tx_tasklet); netdev->priv_flags |= IFF_UNICAST_FLT; diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c index 063e560d9c1b..f3b7b443f964 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c @@ -985,9 +985,9 @@ static int nicvf_poll(struct napi_struct *napi, int budget) * * As of now only CQ errors are handled */ -static void nicvf_handle_qs_err(unsigned long data) +static void nicvf_handle_qs_err(struct tasklet_struct *t) { - struct nicvf *nic = (struct nicvf *)data; + struct nicvf *nic = from_tasklet(nic, t, qs_err_task); struct queue_set *qs = nic->qs; int qidx; u64 status; @@ -1493,12 +1493,10 @@ int nicvf_open(struct net_device *netdev) } /* Init tasklet for handling Qset err interrupt */ - tasklet_init(&nic->qs_err_task, nicvf_handle_qs_err, - (unsigned long)nic); + tasklet_setup(&nic->qs_err_task, nicvf_handle_qs_err); /* Init RBDR tasklet which will refill RBDR */ - tasklet_init(&nic->rbdr_task, nicvf_rbdr_task, - (unsigned long)nic); + tasklet_setup(&nic->rbdr_task, nicvf_rbdr_task); INIT_DELAYED_WORK(&nic->rbdr_work, nicvf_rbdr_work); /* Configure CPI alorithm */ @@ -2067,8 +2065,8 @@ static void nicvf_set_rx_mode(struct net_device *netdev) mode |= BGX_XCAST_MCAST_FILTER; /* here we need to copy mc addrs */ if (netdev_mc_count(netdev)) { - mc_list = kmalloc(offsetof(typeof(*mc_list), - mc[netdev_mc_count(netdev)]), + mc_list = kmalloc(struct_size(mc_list, mc, + netdev_mc_count(netdev)), GFP_ATOMIC); if (unlikely(!mc_list)) return; diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c index a45223f0cca5..7a141ce32e86 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c @@ -460,9 +460,9 @@ void nicvf_rbdr_work(struct work_struct *work) } /* In Softirq context, alloc rcv buffers in atomic mode */ -void nicvf_rbdr_task(unsigned long data) +void nicvf_rbdr_task(struct tasklet_struct *t) { - struct nicvf *nic = (struct nicvf *)data; + struct nicvf *nic = from_tasklet(nic, t, rbdr_task); nicvf_refill_rbdr(nic, GFP_ATOMIC); if (nic->rb_alloc_fail) { diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.h b/drivers/net/ethernet/cavium/thunder/nicvf_queues.h index 2460451fc48f..8453defc296c 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.h +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.h @@ -348,7 +348,7 @@ void nicvf_xdp_sq_doorbell(struct nicvf *nic, struct snd_queue *sq, int sq_num); struct sk_buff *nicvf_get_rcv_skb(struct nicvf *nic, struct cqe_rx_t *cqe_rx, bool xdp); -void nicvf_rbdr_task(unsigned long data); +void nicvf_rbdr_task(struct tasklet_struct *t); void nicvf_rbdr_work(struct work_struct *work); void nicvf_enable_intr(struct nicvf *nic, int int_type, int q_idx); diff --git a/drivers/net/ethernet/chelsio/Kconfig b/drivers/net/ethernet/chelsio/Kconfig index f6f3ef9a93cf..87cc0ef68b31 100644 --- a/drivers/net/ethernet/chelsio/Kconfig +++ b/drivers/net/ethernet/chelsio/Kconfig @@ -134,4 +134,6 @@ config CHELSIO_LIB help Common library for Chelsio drivers. +source "drivers/net/ethernet/chelsio/inline_crypto/Kconfig" + endif # NET_VENDOR_CHELSIO diff --git a/drivers/net/ethernet/chelsio/Makefile b/drivers/net/ethernet/chelsio/Makefile index c0f978d2e8a7..1a6fd8b2bb7d 100644 --- a/drivers/net/ethernet/chelsio/Makefile +++ b/drivers/net/ethernet/chelsio/Makefile @@ -8,3 +8,4 @@ obj-$(CONFIG_CHELSIO_T3) += cxgb3/ obj-$(CONFIG_CHELSIO_T4) += cxgb4/ obj-$(CONFIG_CHELSIO_T4VF) += cxgb4vf/ obj-$(CONFIG_CHELSIO_LIB) += libcxgb/ +obj-$(CONFIG_CHELSIO_INLINE_CRYPTO) += inline_crypto/ diff --git a/drivers/net/ethernet/chelsio/cxgb/cxgb2.c b/drivers/net/ethernet/chelsio/cxgb/cxgb2.c index 99736796e1a0..0e4a0f413960 100644 --- a/drivers/net/ethernet/chelsio/cxgb/cxgb2.c +++ b/drivers/net/ethernet/chelsio/cxgb/cxgb2.c @@ -997,17 +997,17 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) goto out_disable_pdev; } - if (!pci_set_dma_mask(pdev, DMA_BIT_MASK(64))) { + if (!dma_set_mask(&pdev->dev, DMA_BIT_MASK(64))) { pci_using_dac = 1; - if (pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64))) { - pr_err("%s: unable to obtain 64-bit DMA for " - "consistent allocations\n", pci_name(pdev)); + if (dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(64))) { + pr_err("%s: unable to obtain 64-bit DMA for coherent allocations\n", + pci_name(pdev)); err = -ENODEV; goto out_disable_pdev; } - } else if ((err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) != 0) { + } else if ((err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32))) != 0) { pr_err("%s: no usable DMA configuration\n", pci_name(pdev)); goto out_disable_pdev; } diff --git a/drivers/net/ethernet/chelsio/cxgb/sge.c b/drivers/net/ethernet/chelsio/cxgb/sge.c index 47b5c8e2104b..2d9c2b5a690a 100644 --- a/drivers/net/ethernet/chelsio/cxgb/sge.c +++ b/drivers/net/ethernet/chelsio/cxgb/sge.c @@ -239,8 +239,10 @@ struct sched { unsigned int num; /* num skbs in per port queues */ struct sched_port p[MAX_NPORTS]; struct tasklet_struct sched_tsk;/* tasklet used to run scheduler */ + struct sge *sge; }; -static void restart_sched(unsigned long); + +static void restart_sched(struct tasklet_struct *t); /* @@ -378,7 +380,8 @@ static int tx_sched_init(struct sge *sge) return -ENOMEM; pr_debug("tx_sched_init\n"); - tasklet_init(&s->sched_tsk, restart_sched, (unsigned long) sge); + tasklet_setup(&s->sched_tsk, restart_sched); + s->sge = sge; sge->tx_sched = s; for (i = 0; i < MAX_NPORTS; i++) { @@ -509,9 +512,8 @@ static void free_freelQ_buffers(struct pci_dev *pdev, struct freelQ *q) while (q->credits--) { struct freelQ_ce *ce = &q->centries[cidx]; - pci_unmap_single(pdev, dma_unmap_addr(ce, dma_addr), - dma_unmap_len(ce, dma_len), - PCI_DMA_FROMDEVICE); + dma_unmap_single(&pdev->dev, dma_unmap_addr(ce, dma_addr), + dma_unmap_len(ce, dma_len), DMA_FROM_DEVICE); dev_kfree_skb(ce->skb); ce->skb = NULL; if (++cidx == q->size) @@ -529,8 +531,8 @@ static void free_rx_resources(struct sge *sge) if (sge->respQ.entries) { size = sizeof(struct respQ_e) * sge->respQ.size; - pci_free_consistent(pdev, size, sge->respQ.entries, - sge->respQ.dma_addr); + dma_free_coherent(&pdev->dev, size, sge->respQ.entries, + sge->respQ.dma_addr); } for (i = 0; i < SGE_FREELQ_N; i++) { @@ -542,8 +544,8 @@ static void free_rx_resources(struct sge *sge) } if (q->entries) { size = sizeof(struct freelQ_e) * q->size; - pci_free_consistent(pdev, size, q->entries, - q->dma_addr); + dma_free_coherent(&pdev->dev, size, q->entries, + q->dma_addr); } } } @@ -564,7 +566,8 @@ static int alloc_rx_resources(struct sge *sge, struct sge_params *p) q->size = p->freelQ_size[i]; q->dma_offset = sge->rx_pkt_pad ? 0 : NET_IP_ALIGN; size = sizeof(struct freelQ_e) * q->size; - q->entries = pci_alloc_consistent(pdev, size, &q->dma_addr); + q->entries = dma_alloc_coherent(&pdev->dev, size, + &q->dma_addr, GFP_KERNEL); if (!q->entries) goto err_no_mem; @@ -601,7 +604,8 @@ static int alloc_rx_resources(struct sge *sge, struct sge_params *p) sge->respQ.credits = 0; size = sizeof(struct respQ_e) * sge->respQ.size; sge->respQ.entries = - pci_alloc_consistent(pdev, size, &sge->respQ.dma_addr); + dma_alloc_coherent(&pdev->dev, size, &sge->respQ.dma_addr, + GFP_KERNEL); if (!sge->respQ.entries) goto err_no_mem; return 0; @@ -624,9 +628,10 @@ static void free_cmdQ_buffers(struct sge *sge, struct cmdQ *q, unsigned int n) ce = &q->centries[cidx]; while (n--) { if (likely(dma_unmap_len(ce, dma_len))) { - pci_unmap_single(pdev, dma_unmap_addr(ce, dma_addr), + dma_unmap_single(&pdev->dev, + dma_unmap_addr(ce, dma_addr), dma_unmap_len(ce, dma_len), - PCI_DMA_TODEVICE); + DMA_TO_DEVICE); if (q->sop) q->sop = 0; } @@ -663,8 +668,8 @@ static void free_tx_resources(struct sge *sge) } if (q->entries) { size = sizeof(struct cmdQ_e) * q->size; - pci_free_consistent(pdev, size, q->entries, - q->dma_addr); + dma_free_coherent(&pdev->dev, size, q->entries, + q->dma_addr); } } } @@ -689,7 +694,8 @@ static int alloc_tx_resources(struct sge *sge, struct sge_params *p) q->stop_thres = 0; spin_lock_init(&q->lock); size = sizeof(struct cmdQ_e) * q->size; - q->entries = pci_alloc_consistent(pdev, size, &q->dma_addr); + q->entries = dma_alloc_coherent(&pdev->dev, size, + &q->dma_addr, GFP_KERNEL); if (!q->entries) goto err_no_mem; @@ -837,8 +843,8 @@ static void refill_free_list(struct sge *sge, struct freelQ *q) break; skb_reserve(skb, q->dma_offset); - mapping = pci_map_single(pdev, skb->data, dma_len, - PCI_DMA_FROMDEVICE); + mapping = dma_map_single(&pdev->dev, skb->data, dma_len, + DMA_FROM_DEVICE); skb_reserve(skb, sge->rx_pkt_pad); ce->skb = skb; @@ -1049,15 +1055,15 @@ static inline struct sk_buff *get_packet(struct adapter *adapter, goto use_orig_buf; skb_put(skb, len); - pci_dma_sync_single_for_cpu(pdev, - dma_unmap_addr(ce, dma_addr), - dma_unmap_len(ce, dma_len), - PCI_DMA_FROMDEVICE); + dma_sync_single_for_cpu(&pdev->dev, + dma_unmap_addr(ce, dma_addr), + dma_unmap_len(ce, dma_len), + DMA_FROM_DEVICE); skb_copy_from_linear_data(ce->skb, skb->data, len); - pci_dma_sync_single_for_device(pdev, - dma_unmap_addr(ce, dma_addr), - dma_unmap_len(ce, dma_len), - PCI_DMA_FROMDEVICE); + dma_sync_single_for_device(&pdev->dev, + dma_unmap_addr(ce, dma_addr), + dma_unmap_len(ce, dma_len), + DMA_FROM_DEVICE); recycle_fl_buf(fl, fl->cidx); return skb; } @@ -1068,8 +1074,8 @@ use_orig_buf: return NULL; } - pci_unmap_single(pdev, dma_unmap_addr(ce, dma_addr), - dma_unmap_len(ce, dma_len), PCI_DMA_FROMDEVICE); + dma_unmap_single(&pdev->dev, dma_unmap_addr(ce, dma_addr), + dma_unmap_len(ce, dma_len), DMA_FROM_DEVICE); skb = ce->skb; prefetch(skb->data); @@ -1091,8 +1097,9 @@ static void unexpected_offload(struct adapter *adapter, struct freelQ *fl) struct freelQ_ce *ce = &fl->centries[fl->cidx]; struct sk_buff *skb = ce->skb; - pci_dma_sync_single_for_cpu(adapter->pdev, dma_unmap_addr(ce, dma_addr), - dma_unmap_len(ce, dma_len), PCI_DMA_FROMDEVICE); + dma_sync_single_for_cpu(&adapter->pdev->dev, + dma_unmap_addr(ce, dma_addr), + dma_unmap_len(ce, dma_len), DMA_FROM_DEVICE); pr_err("%s: unexpected offload packet, cmd %u\n", adapter->name, *skb->data); recycle_fl_buf(fl, fl->cidx); @@ -1209,8 +1216,8 @@ static inline void write_tx_descs(struct adapter *adapter, struct sk_buff *skb, e = e1 = &q->entries[pidx]; ce = &q->centries[pidx]; - mapping = pci_map_single(adapter->pdev, skb->data, - skb_headlen(skb), PCI_DMA_TODEVICE); + mapping = dma_map_single(&adapter->pdev->dev, skb->data, + skb_headlen(skb), DMA_TO_DEVICE); desc_mapping = mapping; desc_len = skb_headlen(skb); @@ -1301,9 +1308,10 @@ static inline void reclaim_completed_tx(struct sge *sge, struct cmdQ *q) * Called from tasklet. Checks the scheduler for any * pending skbs that can be sent. */ -static void restart_sched(unsigned long arg) +static void restart_sched(struct tasklet_struct *t) { - struct sge *sge = (struct sge *) arg; + struct sched *s = from_tasklet(s, t, sched_tsk); + struct sge *sge = s->sge; struct adapter *adapter = sge->adapter; struct cmdQ *q = &sge->cmdQ[0]; struct sk_buff *skb; diff --git a/drivers/net/ethernet/chelsio/cxgb3/adapter.h b/drivers/net/ethernet/chelsio/cxgb3/adapter.h index 087ff0ffb597..f80fbd81b609 100644 --- a/drivers/net/ethernet/chelsio/cxgb3/adapter.h +++ b/drivers/net/ethernet/chelsio/cxgb3/adapter.h @@ -313,6 +313,7 @@ void t3_os_link_fault(struct adapter *adapter, int port_id, int state); void t3_os_link_fault_handler(struct adapter *adapter, int port_id); void t3_sge_start(struct adapter *adap); +void t3_sge_stop_dma(struct adapter *adap); void t3_sge_stop(struct adapter *adap); void t3_start_sge_timers(struct adapter *adap); void t3_stop_sge_timers(struct adapter *adap); diff --git a/drivers/net/ethernet/chelsio/cxgb3/ael1002.c b/drivers/net/ethernet/chelsio/cxgb3/ael1002.c index dadf11e3dddb..9d591f0ddfc5 100644 --- a/drivers/net/ethernet/chelsio/cxgb3/ael1002.c +++ b/drivers/net/ethernet/chelsio/cxgb3/ael1002.c @@ -815,17 +815,12 @@ static const struct cphy_ops ael2020_ops = { int t3_ael2020_phy_prep(struct cphy *phy, struct adapter *adapter, int phy_addr, const struct mdio_ops *mdio_ops) { - int err; - cphy_init(phy, adapter, phy_addr, &ael2020_ops, mdio_ops, SUPPORTED_10000baseT_Full | SUPPORTED_AUI | SUPPORTED_FIBRE | SUPPORTED_IRQ, "10GBASE-R"); msleep(125); - err = set_phy_regs(phy, ael2020_reset_regs); - if (err) - return err; - return 0; + return set_phy_regs(phy, ael2020_reset_regs); } /* diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c index 387c357e1b8e..84ad7261e243 100644 --- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c +++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c @@ -148,7 +148,7 @@ struct workqueue_struct *cxgb3_wq; /** * link_report - show link status and link speed/duplex - * @p: the port whose settings are to be reported + * @dev: the port whose settings are to be reported * * Shows the link status, speed, and duplex of a port. */ @@ -304,8 +304,8 @@ void t3_os_link_changed(struct adapter *adapter, int port_id, int link_stat, /** * t3_os_phymod_changed - handle PHY module changes - * @phy: the PHY reporting the module change - * @mod_type: new module type + * @adap: the adapter associated with the link change + * @port_id: the port index whose limk status has changed * * This is the OS-dependent handler for PHY module changes. It is * invoked when a PHY module is removed or inserted for any OS-specific @@ -1200,7 +1200,7 @@ static void cxgb_vlan_mode(struct net_device *dev, netdev_features_t features) /** * cxgb_up - enable the adapter - * @adapter: adapter being enabled + * @adap: adapter being enabled * * Called when the first port is enabled, this function performs the * actions necessary to make an adapter operational, such as completing @@ -2996,7 +2996,7 @@ void t3_fatal_err(struct adapter *adapter) unsigned int fw_status[4]; if (adapter->flags & FULL_INIT_DONE) { - t3_sge_stop(adapter); + t3_sge_stop_dma(adapter); t3_write_reg(adapter, A_XGM_TX_CTRL, 0); t3_write_reg(adapter, A_XGM_RX_CTRL, 0); t3_write_reg(adapter, XGM_REG(A_XGM_TX_CTRL, 1), 0); diff --git a/drivers/net/ethernet/chelsio/cxgb3/sge.c b/drivers/net/ethernet/chelsio/cxgb3/sge.c index 6dabbf1502c7..e18e9ce27f94 100644 --- a/drivers/net/ethernet/chelsio/cxgb3/sge.c +++ b/drivers/net/ethernet/chelsio/cxgb3/sge.c @@ -372,7 +372,7 @@ static void clear_rx_desc(struct pci_dev *pdev, const struct sge_fl *q, /** * free_rx_bufs - free the Rx buffers on an SGE free list * @pdev: the PCI device associated with the adapter - * @rxq: the SGE free list to clean up + * @q: the SGE free list to clean up * * Release the buffers on an SGE free-buffer Rx queue. HW fetching from * this queue should be stopped before calling this function. @@ -493,7 +493,7 @@ static inline void ring_fl_db(struct adapter *adap, struct sge_fl *q) /** * refill_fl - refill an SGE free-buffer list - * @adapter: the adapter + * @adap: the adapter * @q: the free-list to refill * @n: the number of new buffers to allocate * @gfp: the gfp flags for allocating new buffers @@ -568,7 +568,7 @@ static inline void __refill_fl(struct adapter *adap, struct sge_fl *fl) /** * recycle_rx_buf - recycle a receive buffer - * @adapter: the adapter + * @adap: the adapter * @q: the SGE free list * @idx: index of buffer to recycle * @@ -825,6 +825,7 @@ use_orig_buf: * get_packet_pg - return the next ingress packet buffer from a free list * @adap: the adapter that received the packet * @fl: the SGE free list holding the packet + * @q: the queue * @len: the packet length including any SGE padding * @drop_thres: # of remaining buffers before we start dropping packets * @@ -1173,6 +1174,7 @@ static void write_wr_hdr_sgl(unsigned int ndesc, struct sk_buff *skb, * @q: the Tx queue * @ndesc: number of descriptors the packet will occupy * @compl: the value of the COMPL bit to use + * @addr: address * * Generate a TX_PKT work request to send the supplied packet. */ @@ -1516,14 +1518,14 @@ static int ctrl_xmit(struct adapter *adap, struct sge_txq *q, /** * restart_ctrlq - restart a suspended control queue - * @qs: the queue set cotaining the control queue + * @t: pointer to the tasklet associated with this handler * * Resumes transmission on a suspended Tx control queue. */ -static void restart_ctrlq(unsigned long data) +static void restart_ctrlq(struct tasklet_struct *t) { struct sk_buff *skb; - struct sge_qset *qs = (struct sge_qset *)data; + struct sge_qset *qs = from_tasklet(qs, t, txq[TXQ_CTRL].qresume_tsk); struct sge_txq *q = &qs->txq[TXQ_CTRL]; spin_lock(&q->lock); @@ -1622,6 +1624,7 @@ static void setup_deferred_unmapping(struct sk_buff *skb, struct pci_dev *pdev, * @pidx: index of the first Tx descriptor to write * @gen: the generation value to use * @ndesc: number of descriptors the packet will occupy + * @addr: the address * * Write an offload work request to send the supplied packet. The packet * data already carry the work request with most fields populated. @@ -1733,14 +1736,14 @@ again: reclaim_completed_tx(adap, q, TX_RECLAIM_CHUNK); /** * restart_offloadq - restart a suspended offload queue - * @qs: the queue set cotaining the offload queue + * @t: pointer to the tasklet associated with this handler * * Resumes transmission on a suspended Tx offload queue. */ -static void restart_offloadq(unsigned long data) +static void restart_offloadq(struct tasklet_struct *t) { struct sk_buff *skb; - struct sge_qset *qs = (struct sge_qset *)data; + struct sge_qset *qs = from_tasklet(qs, t, txq[TXQ_OFLD].qresume_tsk); struct sge_txq *q = &qs->txq[TXQ_OFLD]; const struct port_info *pi = netdev_priv(qs->netdev); struct adapter *adap = pi->adapter; @@ -1883,7 +1886,7 @@ static inline void deliver_partial_bundle(struct t3cdev *tdev, /** * ofld_poll - NAPI handler for offload packets in interrupt mode - * @dev: the network device doing the polling + * @napi: the network device doing the polling * @budget: polling budget * * The NAPI handler for offload packets when a response queue is serviced @@ -2007,7 +2010,7 @@ static void restart_tx(struct sge_qset *qs) /** * cxgb3_arp_process - process an ARP request probing a private IP address - * @adapter: the adapter + * @pi: the port info * @skb: the skbuff containing the ARP request * * Check if the ARP request is probing the private IP address @@ -2069,7 +2072,8 @@ static void cxgb3_process_iscsi_prov_pack(struct port_info *pi, * @adap: the adapter * @rq: the response queue that received the packet * @skb: the packet - * @pad: amount of padding at the start of the buffer + * @pad: padding + * @lro: large receive offload * * Process an ingress ethernet pakcet and deliver it to the stack. * The padding is 2 if the packet was delivered in an Rx buffer and 0 @@ -2239,7 +2243,7 @@ static inline void handle_rsp_cntrl_info(struct sge_qset *qs, u32 flags) /** * check_ring_db - check if we need to ring any doorbells - * @adapter: the adapter + * @adap: the adapter * @qs: the queue set whose Tx queues are to be examined * @sleeping: indicates which Tx queue sent GTS * @@ -2372,10 +2376,7 @@ no_mem: if (fl->use_pages) { void *addr = fl->sdesc[fl->cidx].pg_chunk.va; - prefetch(addr); -#if L1_CACHE_BYTES < 128 - prefetch(addr + L1_CACHE_BYTES); -#endif + net_prefetch(addr); __refill_fl(adap, fl); if (lro > 0) { lro_add_page(adap, qs, fl, @@ -2902,7 +2903,7 @@ void t3_sge_err_intr_handler(struct adapter *adapter) /** * sge_timer_tx - perform periodic maintenance of an SGE qset - * @data: the SGE queue set to maintain + * @t: a timer list containing the SGE queue set to maintain * * Runs periodically from a timer to perform maintenance of an SGE queue * set. It performs two tasks: @@ -2946,7 +2947,7 @@ static void sge_timer_tx(struct timer_list *t) /** * sge_timer_rx - perform periodic maintenance of an SGE qset - * @data: the SGE queue set to maintain + * @t: the timer list containing the SGE queue set to maintain * * a) Replenishes Rx queues that have run out due to memory shortage. * Normally new Rx buffers are added when existing ones are consumed but @@ -3024,7 +3025,7 @@ void t3_update_qset_coalesce(struct sge_qset *qs, const struct qset_params *p) * @irq_vec_idx: the IRQ vector index for response queue interrupts * @p: configuration parameters for this queue set * @ntxq: number of Tx queues for the queue set - * @netdev: net device associated with this queue set + * @dev: net device associated with this queue set * @netdevq: net device TX queue associated with this queue set * * Allocate resources and initialize an SGE queue set. A queue set @@ -3084,10 +3085,8 @@ int t3_sge_alloc_qset(struct adapter *adapter, unsigned int id, int nports, skb_queue_head_init(&q->txq[i].sendq); } - tasklet_init(&q->txq[TXQ_OFLD].qresume_tsk, restart_offloadq, - (unsigned long)q); - tasklet_init(&q->txq[TXQ_CTRL].qresume_tsk, restart_ctrlq, - (unsigned long)q); + tasklet_setup(&q->txq[TXQ_OFLD].qresume_tsk, restart_offloadq); + tasklet_setup(&q->txq[TXQ_CTRL].qresume_tsk, restart_ctrlq); q->fl[0].gen = q->fl[1].gen = 1; q->fl[0].size = p->fl_size; @@ -3271,30 +3270,40 @@ void t3_sge_start(struct adapter *adap) } /** - * t3_sge_stop - disable SGE operation + * t3_sge_stop_dma - Disable SGE DMA engine operation * @adap: the adapter * - * Disables the DMA engine. This can be called in emeregencies (e.g., - * from error interrupts) or from normal process context. In the latter - * case it also disables any pending queue restart tasklets. Note that - * if it is called in interrupt context it cannot disable the restart - * tasklets as it cannot wait, however the tasklets will have no effect - * since the doorbells are disabled and the driver will call this again - * later from process context, at which time the tasklets will be stopped - * if they are still running. + * Can be invoked from interrupt context e.g. error handler. + * + * Note that this function cannot disable the restart of tasklets as + * it cannot wait if called from interrupt context, however the + * tasklets will have no effect since the doorbells are disabled. The + * driver will call tg3_sge_stop() later from process context, at + * which time the tasklets will be stopped if they are still running. */ -void t3_sge_stop(struct adapter *adap) +void t3_sge_stop_dma(struct adapter *adap) { t3_set_reg_field(adap, A_SG_CONTROL, F_GLOBALENABLE, 0); - if (!in_interrupt()) { - int i; +} - for (i = 0; i < SGE_QSETS; ++i) { - struct sge_qset *qs = &adap->sge.qs[i]; +/** + * t3_sge_stop - disable SGE operation completly + * @adap: the adapter + * + * Called from process context. Disables the DMA engine and any + * pending queue restart tasklets. + */ +void t3_sge_stop(struct adapter *adap) +{ + int i; - tasklet_kill(&qs->txq[TXQ_OFLD].qresume_tsk); - tasklet_kill(&qs->txq[TXQ_CTRL].qresume_tsk); - } + t3_sge_stop_dma(adap); + + for (i = 0; i < SGE_QSETS; ++i) { + struct sge_qset *qs = &adap->sge.qs[i]; + + tasklet_kill(&qs->txq[TXQ_OFLD].qresume_tsk); + tasklet_kill(&qs->txq[TXQ_CTRL].qresume_tsk); } } diff --git a/drivers/net/ethernet/chelsio/cxgb3/t3_hw.c b/drivers/net/ethernet/chelsio/cxgb3/t3_hw.c index 0a9f2c596624..7ff31d1026fb 100644 --- a/drivers/net/ethernet/chelsio/cxgb3/t3_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb3/t3_hw.c @@ -2195,7 +2195,7 @@ static int t3_sge_write_context(struct adapter *adapter, unsigned int id, /** * clear_sge_ctxt - completely clear an SGE context - * @adapter: the adapter + * @adap: the adapter * @id: the context id * @type: the context type * @@ -2484,6 +2484,7 @@ int t3_sge_disable_cqcntxt(struct adapter *adapter, unsigned int id) * @adapter: the adapter * @id: the context id * @op: the operation to perform + * @credits: credit value to write * * Perform the selected operation on an SGE completion queue context. * The caller is responsible for ensuring only one context operation @@ -2885,7 +2886,7 @@ static void init_cong_ctrl(unsigned short *a, unsigned short *b) * t3_load_mtus - write the MTU and congestion control HW tables * @adap: the adapter * @mtus: the unrestricted values for the MTU table - * @alphs: the values for the congestion control alpha parameter + * @alpha: the values for the congestion control alpha parameter * @beta: the values for the congestion control beta parameter * @mtu_cap: the maximum permitted effective MTU * @@ -2966,7 +2967,7 @@ static void ulp_config(struct adapter *adap, const struct tp_params *p) /** * t3_set_proto_sram - set the contents of the protocol sram - * @adapter: the adapter + * @adap: the adapter * @data: the protocol image * * Write the contents of the protocol SRAM. @@ -3483,7 +3484,7 @@ static void get_pci_mode(struct adapter *adapter, struct pci_params *p) /** * init_link_config - initialize a link's SW state * @lc: structure holding the link state - * @ai: information about the current card + * @caps: information about the current card * * Initializes the SW state maintained for each link, including the link's * capabilities and default speed/duplex/flow-control/autonegotiation diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h index 9cb8b229c1b3..3352dad6ca99 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h @@ -146,6 +146,11 @@ enum { CXGB4_ETHTOOL_FLASH_BOOTCFG = 4 }; +enum cxgb4_netdev_tls_ops { + CXGB4_TLSDEV_OPS = 1, + CXGB4_XFRMDEV_OPS +}; + struct cxgb4_bootcfg_data { __le16 signature; __u8 reserved[2]; @@ -1196,6 +1201,12 @@ struct adapter { struct cxgb4_tc_u32_table *tc_u32; struct chcr_ktls chcr_ktls; struct chcr_stats_debug chcr_stats; +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) + struct ch_ktls_stats_debug ch_ktls_stats; +#endif +#if IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE) + struct ch_ipsec_stats_debug ch_ipsec_stats; +#endif /* TC flower offload */ bool tc_flower_initialized; @@ -2100,7 +2111,7 @@ void free_tx_desc(struct adapter *adap, struct sge_txq *q, void cxgb4_eosw_txq_free_desc(struct adapter *adap, struct sge_eosw_txq *txq, u32 ndesc); int cxgb4_ethofld_send_flowc(struct net_device *dev, u32 eotid, u32 tc); -void cxgb4_ethofld_restart(unsigned long data); +void cxgb4_ethofld_restart(struct tasklet_struct *t); int cxgb4_ethofld_rx_handler(struct sge_rspq *q, const __be64 *rsp, const struct pkt_gl *si); void free_txq(struct adapter *adap, struct sge_txq *q); @@ -2169,7 +2180,7 @@ void cxgb4_enable_rx(struct adapter *adap, struct sge_rspq *q); void cxgb4_quiesce_rx(struct sge_rspq *q); int cxgb4_port_mirror_alloc(struct net_device *dev); void cxgb4_port_mirror_free(struct net_device *dev); -#ifdef CONFIG_CHELSIO_TLS_DEVICE +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) int cxgb4_set_ktls_feature(struct adapter *adap, bool enable); #endif #endif /* __CXGB4_H__ */ diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c index 05f33b7e3677..0273f40b85f7 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c @@ -3527,6 +3527,10 @@ DEFINE_SHOW_ATTRIBUTE(meminfo); static int chcr_stats_show(struct seq_file *seq, void *v) { +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) + struct ch_ktls_port_stats_debug *ktls_port; + int i = 0; +#endif struct adapter *adap = seq->private; seq_puts(seq, "Chelsio Crypto Accelerator Stats \n"); @@ -3542,52 +3546,45 @@ static int chcr_stats_show(struct seq_file *seq, void *v) atomic_read(&adap->chcr_stats.error)); seq_printf(seq, "Fallback: %10u \n", atomic_read(&adap->chcr_stats.fallback)); - seq_printf(seq, "IPSec PDU: %10u\n", - atomic_read(&adap->chcr_stats.ipsec_cnt)); seq_printf(seq, "TLS PDU Tx: %10u\n", atomic_read(&adap->chcr_stats.tls_pdu_tx)); seq_printf(seq, "TLS PDU Rx: %10u\n", atomic_read(&adap->chcr_stats.tls_pdu_rx)); seq_printf(seq, "TLS Keys (DDR) Count: %10u\n", atomic_read(&adap->chcr_stats.tls_key)); -#ifdef CONFIG_CHELSIO_TLS_DEVICE +#if IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE) + seq_puts(seq, "\nChelsio Inline IPsec Crypto Accelerator Stats\n"); + seq_printf(seq, "IPSec PDU: %10u\n", + atomic_read(&adap->ch_ipsec_stats.ipsec_cnt)); +#endif +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) seq_puts(seq, "\nChelsio KTLS Crypto Accelerator Stats\n"); seq_printf(seq, "Tx TLS offload refcount: %20u\n", refcount_read(&adap->chcr_ktls.ktls_refcount)); - seq_printf(seq, "Tx HW offload contexts added: %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_ctx)); - seq_printf(seq, "Tx connection created: %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_connection_open)); - seq_printf(seq, "Tx connection failed: %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_connection_fail)); - seq_printf(seq, "Tx connection closed: %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_connection_close)); - seq_printf(seq, "Packets passed for encryption : %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_encrypted_packets)); - seq_printf(seq, "Bytes passed for encryption : %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_encrypted_bytes)); seq_printf(seq, "Tx records send: %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_send_records)); + atomic64_read(&adap->ch_ktls_stats.ktls_tx_send_records)); seq_printf(seq, "Tx partial start of records: %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_start_pkts)); + atomic64_read(&adap->ch_ktls_stats.ktls_tx_start_pkts)); seq_printf(seq, "Tx partial middle of records: %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_middle_pkts)); + atomic64_read(&adap->ch_ktls_stats.ktls_tx_middle_pkts)); seq_printf(seq, "Tx partial end of record: %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_end_pkts)); + atomic64_read(&adap->ch_ktls_stats.ktls_tx_end_pkts)); seq_printf(seq, "Tx complete records: %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_complete_pkts)); + atomic64_read(&adap->ch_ktls_stats.ktls_tx_complete_pkts)); seq_printf(seq, "TX trim pkts : %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_trimmed_pkts)); - seq_printf(seq, "Tx out of order packets: %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_ooo)); - seq_printf(seq, "Tx drop pkts before HW offload: %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_skip_no_sync_data)); - seq_printf(seq, "Tx drop not synced packets: %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_drop_no_sync_data)); - seq_printf(seq, "Tx drop bypass req: %20llu\n", - atomic64_read(&adap->chcr_stats.ktls_tx_drop_bypass_req)); + atomic64_read(&adap->ch_ktls_stats.ktls_tx_trimmed_pkts)); + while (i < MAX_NPORTS) { + ktls_port = &adap->ch_ktls_stats.ktls_port[i]; + seq_printf(seq, "Port %d\n", i); + seq_printf(seq, "Tx connection created: %20llu\n", + atomic64_read(&ktls_port->ktls_tx_connection_open)); + seq_printf(seq, "Tx connection failed: %20llu\n", + atomic64_read(&ktls_port->ktls_tx_connection_fail)); + seq_printf(seq, "Tx connection closed: %20llu\n", + atomic64_read(&ktls_port->ktls_tx_connection_close)); + i++; + } #endif - return 0; } DEFINE_SHOW_ATTRIBUTE(chcr_stats); diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c index 9f3173f86eed..61ea3ec5c3fc 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c @@ -117,15 +117,7 @@ static const char stats_strings[][ETH_GSTRING_LEN] = { "vlan_insertions ", "gro_packets ", "gro_merged ", -}; - -static char adapter_stats_strings[][ETH_GSTRING_LEN] = { - "db_drop ", - "db_full ", - "db_empty ", - "write_coal_success ", - "write_coal_fail ", -#ifdef CONFIG_CHELSIO_TLS_DEVICE +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) "tx_tls_encrypted_packets", "tx_tls_encrypted_bytes ", "tx_tls_ctx ", @@ -136,6 +128,14 @@ static char adapter_stats_strings[][ETH_GSTRING_LEN] = { #endif }; +static char adapter_stats_strings[][ETH_GSTRING_LEN] = { + "db_drop ", + "db_full ", + "db_empty ", + "write_coal_success ", + "write_coal_fail ", +}; + static char loopback_stats_strings[][ETH_GSTRING_LEN] = { "-------Loopback----------- ", "octets_ok ", @@ -257,15 +257,7 @@ struct queue_port_stats { u64 vlan_ins; u64 gro_pkts; u64 gro_merged; -}; - -struct adapter_stats { - u64 db_drop; - u64 db_full; - u64 db_empty; - u64 wc_success; - u64 wc_fail; -#ifdef CONFIG_CHELSIO_TLS_DEVICE +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) u64 tx_tls_encrypted_packets; u64 tx_tls_encrypted_bytes; u64 tx_tls_ctx; @@ -276,12 +268,23 @@ struct adapter_stats { #endif }; +struct adapter_stats { + u64 db_drop; + u64 db_full; + u64 db_empty; + u64 wc_success; + u64 wc_fail; +}; + static void collect_sge_port_stats(const struct adapter *adap, const struct port_info *p, struct queue_port_stats *s) { const struct sge_eth_txq *tx = &adap->sge.ethtxq[p->first_qset]; const struct sge_eth_rxq *rx = &adap->sge.ethrxq[p->first_qset]; +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) + const struct ch_ktls_port_stats_debug *ktls_stats; +#endif struct sge_eohw_txq *eohw_tx; unsigned int i; @@ -306,6 +309,21 @@ static void collect_sge_port_stats(const struct adapter *adap, s->vlan_ins += eohw_tx->vlan_ins; } } +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) + ktls_stats = &adap->ch_ktls_stats.ktls_port[p->port_id]; + s->tx_tls_encrypted_packets = + atomic64_read(&ktls_stats->ktls_tx_encrypted_packets); + s->tx_tls_encrypted_bytes = + atomic64_read(&ktls_stats->ktls_tx_encrypted_bytes); + s->tx_tls_ctx = atomic64_read(&ktls_stats->ktls_tx_ctx); + s->tx_tls_ooo = atomic64_read(&ktls_stats->ktls_tx_ooo); + s->tx_tls_skip_no_sync_data = + atomic64_read(&ktls_stats->ktls_tx_skip_no_sync_data); + s->tx_tls_drop_no_sync_data = + atomic64_read(&ktls_stats->ktls_tx_drop_no_sync_data); + s->tx_tls_drop_bypass_req = + atomic64_read(&ktls_stats->ktls_tx_drop_bypass_req); +#endif } static void collect_adapter_stats(struct adapter *adap, struct adapter_stats *s) diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c index 481498585ead..6ec5f2f26f05 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c @@ -604,17 +604,14 @@ int cxgb4_get_free_ftid(struct net_device *dev, u8 family, bool hash_en, /* If the new rule wants to get inserted into * HPFILTER region, but its prio is greater * than the rule with the highest prio in HASH - * region, then reject the rule. - */ - if (t->tc_hash_tids_max_prio && - tc_prio > t->tc_hash_tids_max_prio) - break; - - /* If there's not enough slots available - * in HPFILTER region, then move on to - * normal FILTER region immediately. + * region, or if there's not enough slots + * available in HPFILTER region, then skip + * trying to insert this rule into HPFILTER + * region and directly go to the next region. */ - if (ftid + n > t->nhpftids) { + if ((t->tc_hash_tids_max_prio && + tc_prio > t->tc_hash_tids_max_prio) || + (ftid + n) > t->nhpftids) { ftid = t->nhpftids; continue; } diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index de078a5bf23e..a952fe198eb9 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -66,7 +66,7 @@ #include <linux/crash_dump.h> #include <net/udp_tunnel.h> #include <net/xfrm.h> -#if defined(CONFIG_CHELSIO_TLS_DEVICE) +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) #include <net/tls.h> #endif @@ -6396,7 +6396,50 @@ static int cxgb4_iov_configure(struct pci_dev *pdev, int num_vfs) } #endif /* CONFIG_PCI_IOV */ -#if defined(CONFIG_CHELSIO_TLS_DEVICE) +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) || IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE) + +static int chcr_offload_state(struct adapter *adap, + enum cxgb4_netdev_tls_ops op_val) +{ + switch (op_val) { +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) + case CXGB4_TLSDEV_OPS: + if (!adap->uld[CXGB4_ULD_KTLS].handle) { + dev_dbg(adap->pdev_dev, "ch_ktls driver is not loaded\n"); + return -EOPNOTSUPP; + } + if (!adap->uld[CXGB4_ULD_KTLS].tlsdev_ops) { + dev_dbg(adap->pdev_dev, + "ch_ktls driver has no registered tlsdev_ops\n"); + return -EOPNOTSUPP; + } + break; +#endif /* CONFIG_CHELSIO_TLS_DEVICE */ +#if IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE) + case CXGB4_XFRMDEV_OPS: + if (!adap->uld[CXGB4_ULD_IPSEC].handle) { + dev_dbg(adap->pdev_dev, "chipsec driver is not loaded\n"); + return -EOPNOTSUPP; + } + if (!adap->uld[CXGB4_ULD_IPSEC].xfrmdev_ops) { + dev_dbg(adap->pdev_dev, + "chipsec driver has no registered xfrmdev_ops\n"); + return -EOPNOTSUPP; + } + break; +#endif /* CONFIG_CHELSIO_IPSEC_INLINE */ + default: + dev_dbg(adap->pdev_dev, + "driver has no support for offload %d\n", op_val); + return -EOPNOTSUPP; + } + + return 0; +} + +#endif /* CONFIG_CHELSIO_TLS_DEVICE || CONFIG_CHELSIO_IPSEC_INLINE */ + +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) static int cxgb4_ktls_dev_add(struct net_device *netdev, struct sock *sk, enum tls_offload_ctx_dir direction, @@ -6404,30 +6447,21 @@ static int cxgb4_ktls_dev_add(struct net_device *netdev, struct sock *sk, u32 tcp_sn) { struct adapter *adap = netdev2adap(netdev); - int ret = 0; + int ret; mutex_lock(&uld_mutex); - if (!adap->uld[CXGB4_ULD_CRYPTO].handle) { - dev_err(adap->pdev_dev, "chcr driver is not loaded\n"); - ret = -EOPNOTSUPP; - goto out_unlock; - } - - if (!adap->uld[CXGB4_ULD_CRYPTO].tlsdev_ops) { - dev_err(adap->pdev_dev, - "chcr driver has no registered tlsdev_ops()\n"); - ret = -EOPNOTSUPP; + ret = chcr_offload_state(adap, CXGB4_TLSDEV_OPS); + if (ret) goto out_unlock; - } ret = cxgb4_set_ktls_feature(adap, FW_PARAMS_PARAM_DEV_KTLS_HW_ENABLE); if (ret) goto out_unlock; - ret = adap->uld[CXGB4_ULD_CRYPTO].tlsdev_ops->tls_dev_add(netdev, sk, - direction, - crypto_info, - tcp_sn); + ret = adap->uld[CXGB4_ULD_KTLS].tlsdev_ops->tls_dev_add(netdev, sk, + direction, + crypto_info, + tcp_sn); /* if there is a failure, clear the refcount */ if (ret) cxgb4_set_ktls_feature(adap, @@ -6444,19 +6478,11 @@ static void cxgb4_ktls_dev_del(struct net_device *netdev, struct adapter *adap = netdev2adap(netdev); mutex_lock(&uld_mutex); - if (!adap->uld[CXGB4_ULD_CRYPTO].handle) { - dev_err(adap->pdev_dev, "chcr driver is not loaded\n"); + if (chcr_offload_state(adap, CXGB4_TLSDEV_OPS)) goto out_unlock; - } - if (!adap->uld[CXGB4_ULD_CRYPTO].tlsdev_ops) { - dev_err(adap->pdev_dev, - "chcr driver has no registered tlsdev_ops\n"); - goto out_unlock; - } - - adap->uld[CXGB4_ULD_CRYPTO].tlsdev_ops->tls_dev_del(netdev, tls_ctx, - direction); + adap->uld[CXGB4_ULD_KTLS].tlsdev_ops->tls_dev_del(netdev, tls_ctx, + direction); cxgb4_set_ktls_feature(adap, FW_PARAMS_PARAM_DEV_KTLS_HW_DISABLE); out_unlock: @@ -6469,6 +6495,114 @@ static const struct tlsdev_ops cxgb4_ktls_ops = { }; #endif /* CONFIG_CHELSIO_TLS_DEVICE */ +#if IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE) + +static int cxgb4_xfrm_add_state(struct xfrm_state *x) +{ + struct adapter *adap = netdev2adap(x->xso.dev); + int ret; + + if (!mutex_trylock(&uld_mutex)) { + dev_dbg(adap->pdev_dev, + "crypto uld critical resource is under use\n"); + return -EBUSY; + } + ret = chcr_offload_state(adap, CXGB4_XFRMDEV_OPS); + if (ret) + goto out_unlock; + + ret = adap->uld[CXGB4_ULD_IPSEC].xfrmdev_ops->xdo_dev_state_add(x); + +out_unlock: + mutex_unlock(&uld_mutex); + + return ret; +} + +static void cxgb4_xfrm_del_state(struct xfrm_state *x) +{ + struct adapter *adap = netdev2adap(x->xso.dev); + + if (!mutex_trylock(&uld_mutex)) { + dev_dbg(adap->pdev_dev, + "crypto uld critical resource is under use\n"); + return; + } + if (chcr_offload_state(adap, CXGB4_XFRMDEV_OPS)) + goto out_unlock; + + adap->uld[CXGB4_ULD_IPSEC].xfrmdev_ops->xdo_dev_state_delete(x); + +out_unlock: + mutex_unlock(&uld_mutex); +} + +static void cxgb4_xfrm_free_state(struct xfrm_state *x) +{ + struct adapter *adap = netdev2adap(x->xso.dev); + + if (!mutex_trylock(&uld_mutex)) { + dev_dbg(adap->pdev_dev, + "crypto uld critical resource is under use\n"); + return; + } + if (chcr_offload_state(adap, CXGB4_XFRMDEV_OPS)) + goto out_unlock; + + adap->uld[CXGB4_ULD_IPSEC].xfrmdev_ops->xdo_dev_state_free(x); + +out_unlock: + mutex_unlock(&uld_mutex); +} + +static bool cxgb4_ipsec_offload_ok(struct sk_buff *skb, struct xfrm_state *x) +{ + struct adapter *adap = netdev2adap(x->xso.dev); + bool ret = false; + + if (!mutex_trylock(&uld_mutex)) { + dev_dbg(adap->pdev_dev, + "crypto uld critical resource is under use\n"); + return ret; + } + if (chcr_offload_state(adap, CXGB4_XFRMDEV_OPS)) + goto out_unlock; + + ret = adap->uld[CXGB4_ULD_IPSEC].xfrmdev_ops->xdo_dev_offload_ok(skb, x); + +out_unlock: + mutex_unlock(&uld_mutex); + return ret; +} + +static void cxgb4_advance_esn_state(struct xfrm_state *x) +{ + struct adapter *adap = netdev2adap(x->xso.dev); + + if (!mutex_trylock(&uld_mutex)) { + dev_dbg(adap->pdev_dev, + "crypto uld critical resource is under use\n"); + return; + } + if (chcr_offload_state(adap, CXGB4_XFRMDEV_OPS)) + goto out_unlock; + + adap->uld[CXGB4_ULD_IPSEC].xfrmdev_ops->xdo_dev_state_advance_esn(x); + +out_unlock: + mutex_unlock(&uld_mutex); +} + +static const struct xfrmdev_ops cxgb4_xfrmdev_ops = { + .xdo_dev_state_add = cxgb4_xfrm_add_state, + .xdo_dev_state_delete = cxgb4_xfrm_del_state, + .xdo_dev_state_free = cxgb4_xfrm_free_state, + .xdo_dev_offload_ok = cxgb4_ipsec_offload_ok, + .xdo_dev_state_advance_esn = cxgb4_advance_esn_state, +}; + +#endif /* CONFIG_CHELSIO_IPSEC_INLINE */ + static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) { struct net_device *netdev; @@ -6721,14 +6855,22 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) netdev->hw_features |= NETIF_F_HIGHDMA; netdev->features |= netdev->hw_features; netdev->vlan_features = netdev->features & VLAN_FEAT; -#if defined(CONFIG_CHELSIO_TLS_DEVICE) +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) if (pi->adapter->params.crypto & FW_CAPS_CONFIG_TLS_HW) { netdev->hw_features |= NETIF_F_HW_TLS_TX; netdev->tlsdev_ops = &cxgb4_ktls_ops; /* initialize the refcount */ refcount_set(&pi->adapter->chcr_ktls.ktls_refcount, 0); } -#endif +#endif /* CONFIG_CHELSIO_TLS_DEVICE */ +#if IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE) + if (pi->adapter->params.crypto & FW_CAPS_CONFIG_IPSEC_INLINE) { + netdev->hw_enc_features |= NETIF_F_HW_ESP; + netdev->features |= NETIF_F_HW_ESP; + netdev->xfrmdev_ops = &cxgb4_xfrmdev_ops; + } +#endif /* CONFIG_CHELSIO_IPSEC_INLINE */ + netdev->priv_flags |= IFF_UNICAST_FLT; /* MTU range: 81 - 9600 */ diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c index f642c1b475c4..1b88bd1c2dbe 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c @@ -60,6 +60,89 @@ static struct ch_tc_pedit_fields pedits[] = { PEDIT_FIELDS(IP6_, DST_127_96, 4, nat_lip, 12), }; +static const struct cxgb4_natmode_config cxgb4_natmode_config_array[] = { + /* Default supported NAT modes */ + { + .chip = CHELSIO_T5, + .flags = CXGB4_ACTION_NATMODE_NONE, + .natmode = NAT_MODE_NONE, + }, + { + .chip = CHELSIO_T5, + .flags = CXGB4_ACTION_NATMODE_DIP, + .natmode = NAT_MODE_DIP, + }, + { + .chip = CHELSIO_T5, + .flags = CXGB4_ACTION_NATMODE_DIP | CXGB4_ACTION_NATMODE_DPORT, + .natmode = NAT_MODE_DIP_DP, + }, + { + .chip = CHELSIO_T5, + .flags = CXGB4_ACTION_NATMODE_DIP | CXGB4_ACTION_NATMODE_DPORT | + CXGB4_ACTION_NATMODE_SIP, + .natmode = NAT_MODE_DIP_DP_SIP, + }, + { + .chip = CHELSIO_T5, + .flags = CXGB4_ACTION_NATMODE_DIP | CXGB4_ACTION_NATMODE_DPORT | + CXGB4_ACTION_NATMODE_SPORT, + .natmode = NAT_MODE_DIP_DP_SP, + }, + { + .chip = CHELSIO_T5, + .flags = CXGB4_ACTION_NATMODE_SIP | CXGB4_ACTION_NATMODE_SPORT, + .natmode = NAT_MODE_SIP_SP, + }, + { + .chip = CHELSIO_T5, + .flags = CXGB4_ACTION_NATMODE_DIP | CXGB4_ACTION_NATMODE_SIP | + CXGB4_ACTION_NATMODE_SPORT, + .natmode = NAT_MODE_DIP_SIP_SP, + }, + { + .chip = CHELSIO_T5, + .flags = CXGB4_ACTION_NATMODE_DIP | CXGB4_ACTION_NATMODE_SIP | + CXGB4_ACTION_NATMODE_DPORT | + CXGB4_ACTION_NATMODE_SPORT, + .natmode = NAT_MODE_ALL, + }, + /* T6+ can ignore L4 ports when they're disabled. */ + { + .chip = CHELSIO_T6, + .flags = CXGB4_ACTION_NATMODE_SIP, + .natmode = NAT_MODE_SIP_SP, + }, + { + .chip = CHELSIO_T6, + .flags = CXGB4_ACTION_NATMODE_DIP | CXGB4_ACTION_NATMODE_SPORT, + .natmode = NAT_MODE_DIP_DP_SP, + }, + { + .chip = CHELSIO_T6, + .flags = CXGB4_ACTION_NATMODE_DIP | CXGB4_ACTION_NATMODE_SIP, + .natmode = NAT_MODE_ALL, + }, +}; + +static void cxgb4_action_natmode_tweak(struct ch_filter_specification *fs, + u8 natmode_flags) +{ + u8 i = 0; + + /* Translate the enabled NAT 4-tuple fields to one of the + * hardware supported NAT mode configurations. This ensures + * that we pick a valid combination, where the disabled fields + * do not get overwritten to 0. + */ + for (i = 0; i < ARRAY_SIZE(cxgb4_natmode_config_array); i++) { + if (cxgb4_natmode_config_array[i].flags == natmode_flags) { + fs->nat_mode = cxgb4_natmode_config_array[i].natmode; + return; + } + } +} + static struct ch_tc_flower_entry *allocate_flower_entry(void) { struct ch_tc_flower_entry *new = kzalloc(sizeof(*new), GFP_KERNEL); @@ -289,7 +372,8 @@ static void offload_pedit(struct ch_filter_specification *fs, u32 val, u32 mask, } static void process_pedit_field(struct ch_filter_specification *fs, u32 val, - u32 mask, u32 offset, u8 htype) + u32 mask, u32 offset, u8 htype, + u8 *natmode_flags) { switch (htype) { case FLOW_ACT_MANGLE_HDR_TYPE_ETH: @@ -314,60 +398,94 @@ static void process_pedit_field(struct ch_filter_specification *fs, u32 val, switch (offset) { case PEDIT_IP4_SRC: offload_pedit(fs, val, mask, IP4_SRC); + *natmode_flags |= CXGB4_ACTION_NATMODE_SIP; break; case PEDIT_IP4_DST: offload_pedit(fs, val, mask, IP4_DST); + *natmode_flags |= CXGB4_ACTION_NATMODE_DIP; } - fs->nat_mode = NAT_MODE_ALL; break; case FLOW_ACT_MANGLE_HDR_TYPE_IP6: switch (offset) { case PEDIT_IP6_SRC_31_0: offload_pedit(fs, val, mask, IP6_SRC_31_0); + *natmode_flags |= CXGB4_ACTION_NATMODE_SIP; break; case PEDIT_IP6_SRC_63_32: offload_pedit(fs, val, mask, IP6_SRC_63_32); + *natmode_flags |= CXGB4_ACTION_NATMODE_SIP; break; case PEDIT_IP6_SRC_95_64: offload_pedit(fs, val, mask, IP6_SRC_95_64); + *natmode_flags |= CXGB4_ACTION_NATMODE_SIP; break; case PEDIT_IP6_SRC_127_96: offload_pedit(fs, val, mask, IP6_SRC_127_96); + *natmode_flags |= CXGB4_ACTION_NATMODE_SIP; break; case PEDIT_IP6_DST_31_0: offload_pedit(fs, val, mask, IP6_DST_31_0); + *natmode_flags |= CXGB4_ACTION_NATMODE_DIP; break; case PEDIT_IP6_DST_63_32: offload_pedit(fs, val, mask, IP6_DST_63_32); + *natmode_flags |= CXGB4_ACTION_NATMODE_DIP; break; case PEDIT_IP6_DST_95_64: offload_pedit(fs, val, mask, IP6_DST_95_64); + *natmode_flags |= CXGB4_ACTION_NATMODE_DIP; break; case PEDIT_IP6_DST_127_96: offload_pedit(fs, val, mask, IP6_DST_127_96); + *natmode_flags |= CXGB4_ACTION_NATMODE_DIP; } - fs->nat_mode = NAT_MODE_ALL; break; case FLOW_ACT_MANGLE_HDR_TYPE_TCP: switch (offset) { case PEDIT_TCP_SPORT_DPORT: - if (~mask & PEDIT_TCP_UDP_SPORT_MASK) + if (~mask & PEDIT_TCP_UDP_SPORT_MASK) { fs->nat_fport = val; - else + *natmode_flags |= CXGB4_ACTION_NATMODE_SPORT; + } else { fs->nat_lport = val >> 16; + *natmode_flags |= CXGB4_ACTION_NATMODE_DPORT; + } } - fs->nat_mode = NAT_MODE_ALL; break; case FLOW_ACT_MANGLE_HDR_TYPE_UDP: switch (offset) { case PEDIT_UDP_SPORT_DPORT: - if (~mask & PEDIT_TCP_UDP_SPORT_MASK) + if (~mask & PEDIT_TCP_UDP_SPORT_MASK) { fs->nat_fport = val; - else + *natmode_flags |= CXGB4_ACTION_NATMODE_SPORT; + } else { fs->nat_lport = val >> 16; + *natmode_flags |= CXGB4_ACTION_NATMODE_DPORT; + } } - fs->nat_mode = NAT_MODE_ALL; + break; + } +} + +static int cxgb4_action_natmode_validate(struct adapter *adap, u8 natmode_flags, + struct netlink_ext_ack *extack) +{ + u8 i = 0; + + /* Extract the NAT mode to enable based on what 4-tuple fields + * are enabled to be overwritten. This ensures that the + * disabled fields don't get overwritten to 0. + */ + for (i = 0; i < ARRAY_SIZE(cxgb4_natmode_config_array); i++) { + const struct cxgb4_natmode_config *c; + + c = &cxgb4_natmode_config_array[i]; + if (CHELSIO_CHIP_VERSION(adap->params.chip) >= c->chip && + natmode_flags == c->flags) + return 0; } + NL_SET_ERR_MSG_MOD(extack, "Unsupported NAT mode 4-tuple combination"); + return -EOPNOTSUPP; } void cxgb4_process_flow_actions(struct net_device *in, @@ -375,6 +493,7 @@ void cxgb4_process_flow_actions(struct net_device *in, struct ch_filter_specification *fs) { struct flow_action_entry *act; + u8 natmode_flags = 0; int i; flow_action_for_each(i, act, actions) { @@ -426,7 +545,8 @@ void cxgb4_process_flow_actions(struct net_device *in, val = act->mangle.val; offset = act->mangle.offset; - process_pedit_field(fs, val, mask, offset, htype); + process_pedit_field(fs, val, mask, offset, htype, + &natmode_flags); } break; case FLOW_ACTION_QUEUE: @@ -438,6 +558,9 @@ void cxgb4_process_flow_actions(struct net_device *in, break; } } + if (natmode_flags) + cxgb4_action_natmode_tweak(fs, natmode_flags); + } static bool valid_l4_mask(u32 mask) @@ -454,7 +577,8 @@ static bool valid_l4_mask(u32 mask) } static bool valid_pedit_action(struct net_device *dev, - const struct flow_action_entry *act) + const struct flow_action_entry *act, + u8 *natmode_flags) { u32 mask, offset; u8 htype; @@ -479,7 +603,10 @@ static bool valid_pedit_action(struct net_device *dev, case FLOW_ACT_MANGLE_HDR_TYPE_IP4: switch (offset) { case PEDIT_IP4_SRC: + *natmode_flags |= CXGB4_ACTION_NATMODE_SIP; + break; case PEDIT_IP4_DST: + *natmode_flags |= CXGB4_ACTION_NATMODE_DIP; break; default: netdev_err(dev, "%s: Unsupported pedit field\n", @@ -493,10 +620,13 @@ static bool valid_pedit_action(struct net_device *dev, case PEDIT_IP6_SRC_63_32: case PEDIT_IP6_SRC_95_64: case PEDIT_IP6_SRC_127_96: + *natmode_flags |= CXGB4_ACTION_NATMODE_SIP; + break; case PEDIT_IP6_DST_31_0: case PEDIT_IP6_DST_63_32: case PEDIT_IP6_DST_95_64: case PEDIT_IP6_DST_127_96: + *natmode_flags |= CXGB4_ACTION_NATMODE_DIP; break; default: netdev_err(dev, "%s: Unsupported pedit field\n", @@ -512,6 +642,10 @@ static bool valid_pedit_action(struct net_device *dev, __func__); return false; } + if (~mask & PEDIT_TCP_UDP_SPORT_MASK) + *natmode_flags |= CXGB4_ACTION_NATMODE_SPORT; + else + *natmode_flags |= CXGB4_ACTION_NATMODE_DPORT; break; default: netdev_err(dev, "%s: Unsupported pedit field\n", @@ -527,6 +661,10 @@ static bool valid_pedit_action(struct net_device *dev, __func__); return false; } + if (~mask & PEDIT_TCP_UDP_SPORT_MASK) + *natmode_flags |= CXGB4_ACTION_NATMODE_SPORT; + else + *natmode_flags |= CXGB4_ACTION_NATMODE_DPORT; break; default: netdev_err(dev, "%s: Unsupported pedit field\n", @@ -546,10 +684,12 @@ int cxgb4_validate_flow_actions(struct net_device *dev, struct netlink_ext_ack *extack, u8 matchall_filter) { + struct adapter *adap = netdev2adap(dev); struct flow_action_entry *act; bool act_redir = false; bool act_pedit = false; bool act_vlan = false; + u8 natmode_flags = 0; int i; if (!flow_action_basic_hw_stats_check(actions, extack)) @@ -563,7 +703,6 @@ int cxgb4_validate_flow_actions(struct net_device *dev, break; case FLOW_ACTION_MIRRED: case FLOW_ACTION_REDIRECT: { - struct adapter *adap = netdev2adap(dev); struct net_device *n_dev, *target_dev; bool found = false; unsigned int i; @@ -620,7 +759,8 @@ int cxgb4_validate_flow_actions(struct net_device *dev, } break; case FLOW_ACTION_MANGLE: { - bool pedit_valid = valid_pedit_action(dev, act); + bool pedit_valid = valid_pedit_action(dev, act, + &natmode_flags); if (!pedit_valid) return -EOPNOTSUPP; @@ -642,6 +782,15 @@ int cxgb4_validate_flow_actions(struct net_device *dev, return -EINVAL; } + if (act_pedit) { + int ret; + + ret = cxgb4_action_natmode_validate(adap, natmode_flags, + extack); + if (ret) + return ret; + } + return 0; } diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.h index 6296e1d5a12b..3a2fa00c8cde 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.h @@ -108,6 +108,21 @@ struct ch_tc_pedit_fields { #define PEDIT_TCP_SPORT_DPORT 0x0 #define PEDIT_UDP_SPORT_DPORT 0x0 +enum cxgb4_action_natmode_flags { + CXGB4_ACTION_NATMODE_NONE = 0, + CXGB4_ACTION_NATMODE_DIP = (1 << 0), + CXGB4_ACTION_NATMODE_SIP = (1 << 1), + CXGB4_ACTION_NATMODE_DPORT = (1 << 2), + CXGB4_ACTION_NATMODE_SPORT = (1 << 3), +}; + +/* TC PEDIT action to NATMODE translation entry */ +struct cxgb4_natmode_config { + enum chip_type chip; + u8 flags; + u8 natmode; +}; + void cxgb4_process_flow_actions(struct net_device *in, struct flow_action *actions, struct ch_filter_specification *fs); diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c index ae7123a9de8e..6c259de96f96 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c @@ -114,8 +114,7 @@ static int cxgb4_init_eosw_txq(struct net_device *dev, eosw_txq->cred = adap->params.ofldq_wr_cred; eosw_txq->hwqid = hwqid; eosw_txq->netdev = dev; - tasklet_init(&eosw_txq->qresume_tsk, cxgb4_ethofld_restart, - (unsigned long)eosw_txq); + tasklet_setup(&eosw_txq->qresume_tsk, cxgb4_ethofld_restart); return 0; } diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c index 08439e215efe..743af9e654aa 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c @@ -663,7 +663,7 @@ static int uld_attach(struct adapter *adap, unsigned int uld) return 0; } -#ifdef CONFIG_CHELSIO_TLS_DEVICE +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) static bool cxgb4_uld_in_use(struct adapter *adap) { const struct tid_info *t = &adap->tids; @@ -690,8 +690,8 @@ int cxgb4_set_ktls_feature(struct adapter *adap, bool enable) * ULD is/are already active, return failure. */ if (cxgb4_uld_in_use(adap)) { - dev_warn(adap->pdev_dev, - "ULD connections (tid/stid) active. Can't enable kTLS\n"); + dev_dbg(adap->pdev_dev, + "ULD connections (tid/stid) active. Can't enable kTLS\n"); return -EINVAL; } ret = t4_set_params(adap, adap->mbox, adap->pf, @@ -699,7 +699,7 @@ int cxgb4_set_ktls_feature(struct adapter *adap, bool enable) if (ret) return ret; refcount_set(&adap->chcr_ktls.ktls_refcount, 1); - pr_info("kTLS has been enabled. Restrictions placed on ULD support\n"); + pr_debug("kTLS has been enabled. Restrictions placed on ULD support\n"); } else { /* ktls settings already up, just increment refcount. */ refcount_inc(&adap->chcr_ktls.ktls_refcount); @@ -716,7 +716,7 @@ int cxgb4_set_ktls_feature(struct adapter *adap, bool enable) 0, 1, ¶ms, ¶ms); if (ret) return ret; - pr_info("kTLS is disabled. Restrictions on ULD support removed\n"); + pr_debug("kTLS is disabled. Restrictions on ULD support removed\n"); } } diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h index a963fd0b4540..b169776ab484 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h @@ -40,9 +40,11 @@ #include <linux/skbuff.h> #include <linux/inetdevice.h> #include <linux/atomic.h> +#include <net/tls.h> #include "cxgb4.h" #define MAX_ULD_QSETS 16 +#define MAX_ULD_NPORTS 4 /* CPL message priority levels */ enum { @@ -302,7 +304,9 @@ enum cxgb4_uld { CXGB4_ULD_ISCSI, CXGB4_ULD_ISCSIT, CXGB4_ULD_CRYPTO, + CXGB4_ULD_IPSEC, CXGB4_ULD_TLS, + CXGB4_ULD_KTLS, CXGB4_ULD_MAX }; @@ -361,28 +365,11 @@ struct cxgb4_virt_res { /* virtualized HW resources */ struct cxgb4_range ppod_edram; }; -struct chcr_stats_debug { - atomic_t cipher_rqst; - atomic_t digest_rqst; - atomic_t aead_rqst; - atomic_t complete; - atomic_t error; - atomic_t fallback; - atomic_t ipsec_cnt; - atomic_t tls_pdu_tx; - atomic_t tls_pdu_rx; - atomic_t tls_key; -#ifdef CONFIG_CHELSIO_TLS_DEVICE +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) +struct ch_ktls_port_stats_debug { atomic64_t ktls_tx_connection_open; atomic64_t ktls_tx_connection_fail; atomic64_t ktls_tx_connection_close; - atomic64_t ktls_tx_send_records; - atomic64_t ktls_tx_end_pkts; - atomic64_t ktls_tx_start_pkts; - atomic64_t ktls_tx_middle_pkts; - atomic64_t ktls_tx_retransmit_pkts; - atomic64_t ktls_tx_complete_pkts; - atomic64_t ktls_tx_trimmed_pkts; atomic64_t ktls_tx_encrypted_packets; atomic64_t ktls_tx_encrypted_bytes; atomic64_t ktls_tx_ctx; @@ -390,10 +377,38 @@ struct chcr_stats_debug { atomic64_t ktls_tx_skip_no_sync_data; atomic64_t ktls_tx_drop_no_sync_data; atomic64_t ktls_tx_drop_bypass_req; +}; +struct ch_ktls_stats_debug { + struct ch_ktls_port_stats_debug ktls_port[MAX_ULD_NPORTS]; + atomic64_t ktls_tx_send_records; + atomic64_t ktls_tx_end_pkts; + atomic64_t ktls_tx_start_pkts; + atomic64_t ktls_tx_middle_pkts; + atomic64_t ktls_tx_retransmit_pkts; + atomic64_t ktls_tx_complete_pkts; + atomic64_t ktls_tx_trimmed_pkts; +}; #endif + +struct chcr_stats_debug { + atomic_t cipher_rqst; + atomic_t digest_rqst; + atomic_t aead_rqst; + atomic_t complete; + atomic_t error; + atomic_t fallback; + atomic_t tls_pdu_tx; + atomic_t tls_pdu_rx; + atomic_t tls_key; }; +#if IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE) +struct ch_ipsec_stats_debug { + atomic_t ipsec_cnt; +}; +#endif + #define OCQ_WIN_OFFSET(pdev, vres) \ (pci_resource_len((pdev), 2) - roundup_pow_of_two((vres)->ocq.size)) @@ -470,9 +485,12 @@ struct cxgb4_uld_info { struct napi_struct *napi); void (*lro_flush)(struct t4_lro_mgr *); int (*tx_handler)(struct sk_buff *skb, struct net_device *dev); -#if IS_ENABLED(CONFIG_TLS_DEVICE) +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) const struct tlsdev_ops *tlsdev_ops; #endif +#if IS_ENABLED(CONFIG_XFRM_OFFLOAD) + const struct xfrmdev_ops *xfrmdev_ops; +#endif }; void cxgb4_uld_enable(struct adapter *adap); diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c index 869431a1eedd..a9e9c7ae565d 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/sge.c +++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c @@ -1416,14 +1416,14 @@ static netdev_tx_t cxgb4_eth_xmit(struct sk_buff *skb, struct net_device *dev) pi = netdev_priv(dev); adap = pi->adapter; ssi = skb_shinfo(skb); -#ifdef CONFIG_CHELSIO_IPSEC_INLINE +#if IS_ENABLED(CONFIG_CHELSIO_IPSEC_INLINE) if (xfrm_offload(skb) && !ssi->gso_size) - return adap->uld[CXGB4_ULD_CRYPTO].tx_handler(skb, dev); + return adap->uld[CXGB4_ULD_IPSEC].tx_handler(skb, dev); #endif /* CHELSIO_IPSEC_INLINE */ -#ifdef CONFIG_CHELSIO_TLS_DEVICE +#if IS_ENABLED(CONFIG_CHELSIO_TLS_DEVICE) if (skb->decrypted) - return adap->uld[CXGB4_ULD_CRYPTO].tx_handler(skb, dev); + return adap->uld[CXGB4_ULD_KTLS].tx_handler(skb, dev); #endif /* CHELSIO_TLS_DEVICE */ qidx = skb_get_queue_mapping(skb); @@ -2660,15 +2660,15 @@ static int ctrl_xmit(struct sge_ctrl_txq *q, struct sk_buff *skb) /** * restart_ctrlq - restart a suspended control queue - * @data: the control queue to restart + * @t: pointer to the tasklet associated with this handler * * Resumes transmission on a suspended Tx control queue. */ -static void restart_ctrlq(unsigned long data) +static void restart_ctrlq(struct tasklet_struct *t) { struct sk_buff *skb; unsigned int written = 0; - struct sge_ctrl_txq *q = (struct sge_ctrl_txq *)data; + struct sge_ctrl_txq *q = from_tasklet(q, t, qresume_tsk); spin_lock(&q->sendq.lock); reclaim_completed_tx_imm(&q->q); @@ -2961,13 +2961,13 @@ static int ofld_xmit(struct sge_uld_txq *q, struct sk_buff *skb) /** * restart_ofldq - restart a suspended offload queue - * @data: the offload queue to restart + * @t: pointer to the tasklet associated with this handler * * Resumes transmission on a suspended Tx offload queue. */ -static void restart_ofldq(unsigned long data) +static void restart_ofldq(struct tasklet_struct *t) { - struct sge_uld_txq *q = (struct sge_uld_txq *)data; + struct sge_uld_txq *q = from_tasklet(q, t, qresume_tsk); spin_lock(&q->sendq.lock); q->full = 0; /* the queue actually is completely empty now */ @@ -3887,9 +3887,10 @@ static int napi_rx_handler(struct napi_struct *napi, int budget) return work_done; } -void cxgb4_ethofld_restart(unsigned long data) +void cxgb4_ethofld_restart(struct tasklet_struct *t) { - struct sge_eosw_txq *eosw_txq = (struct sge_eosw_txq *)data; + struct sge_eosw_txq *eosw_txq = from_tasklet(eosw_txq, t, + qresume_tsk); int pktcount; spin_lock(&eosw_txq->lock); @@ -4580,7 +4581,7 @@ int t4_sge_alloc_ctrl_txq(struct adapter *adap, struct sge_ctrl_txq *txq, init_txq(adap, &txq->q, FW_EQ_CTRL_CMD_EQID_G(ntohl(c.cmpliqid_eqid))); txq->adap = adap; skb_queue_head_init(&txq->sendq); - tasklet_init(&txq->qresume_tsk, restart_ctrlq, (unsigned long)txq); + tasklet_setup(&txq->qresume_tsk, restart_ctrlq); txq->full = 0; return 0; } @@ -4670,7 +4671,7 @@ int t4_sge_alloc_uld_txq(struct adapter *adap, struct sge_uld_txq *txq, txq->q.q_type = CXGB4_TXQ_ULD; txq->adap = adap; skb_queue_head_init(&txq->sendq); - tasklet_init(&txq->qresume_tsk, restart_ofldq, (unsigned long)txq); + tasklet_setup(&txq->qresume_tsk, restart_ofldq); txq->full = 0; txq->mapping_err = 0; return 0; @@ -4872,9 +4873,6 @@ void t4_sge_stop(struct adapter *adap) int i; struct sge *s = &adap->sge; - if (in_interrupt()) /* actions below require waiting */ - return; - if (s->rx_timer.function) del_timer_sync(&s->rx_timer); if (s->tx_timer.function) diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c index fa3367966f4b..98d01a7497ec 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c @@ -4745,9 +4745,11 @@ static void le_intr_handler(struct adapter *adap) static struct intr_info t6_le_intr_info[] = { { T6_LIPMISS_F, "LE LIP miss", -1, 0 }, { T6_LIP0_F, "LE 0 LIP error", -1, 0 }, + { CMDTIDERR_F, "LE cmd tid error", -1, 1 }, { TCAMINTPERR_F, "LE parity error", -1, 1 }, { T6_UNKNOWNCMD_F, "LE unknown command", -1, 1 }, { SSRAMINTPERR_F, "LE request queue parity error", -1, 1 }, + { HASHTBLMEMCRCERR_F, "LE hash table mem crc error", -1, 0 }, { 0 } }; diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h index 065c01c654ff..b11a172b5174 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h @@ -3017,6 +3017,14 @@ #define REV_V(x) ((x) << REV_S) #define REV_G(x) (((x) >> REV_S) & REV_M) +#define HASHTBLMEMCRCERR_S 27 +#define HASHTBLMEMCRCERR_V(x) ((x) << HASHTBLMEMCRCERR_S) +#define HASHTBLMEMCRCERR_F HASHTBLMEMCRCERR_V(1U) + +#define CMDTIDERR_S 22 +#define CMDTIDERR_V(x) ((x) << CMDTIDERR_S) +#define CMDTIDERR_F CMDTIDERR_V(1U) + #define T6_UNKNOWNCMD_S 3 #define T6_UNKNOWNCMD_V(x) ((x) << T6_UNKNOWNCMD_S) #define T6_UNKNOWNCMD_F T6_UNKNOWNCMD_V(1U) diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c index e2fe78e2e242..2820a0bb971b 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c @@ -2017,33 +2017,14 @@ static void mboxlog_stop(struct seq_file *seq, void *v) { } -static const struct seq_operations mboxlog_seq_ops = { +static const struct seq_operations mboxlog_sops = { .start = mboxlog_start, .next = mboxlog_next, .stop = mboxlog_stop, .show = mboxlog_show }; -static int mboxlog_open(struct inode *inode, struct file *file) -{ - int res = seq_open(file, &mboxlog_seq_ops); - - if (!res) { - struct seq_file *seq = file->private_data; - - seq->private = inode->i_private; - } - return res; -} - -static const struct file_operations mboxlog_fops = { - .owner = THIS_MODULE, - .open = mboxlog_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - +DEFINE_SEQ_ATTRIBUTE(mboxlog); /* * Show SGE Queue Set information. We display QPL Queues Sets per line. */ @@ -2171,31 +2152,14 @@ static void *sge_queue_next(struct seq_file *seq, void *v, loff_t *pos) return *pos < entries ? (void *)((uintptr_t)*pos + 1) : NULL; } -static const struct seq_operations sge_qinfo_seq_ops = { +static const struct seq_operations sge_qinfo_sops = { .start = sge_queue_start, .next = sge_queue_next, .stop = sge_queue_stop, .show = sge_qinfo_show }; -static int sge_qinfo_open(struct inode *inode, struct file *file) -{ - int res = seq_open(file, &sge_qinfo_seq_ops); - - if (!res) { - struct seq_file *seq = file->private_data; - seq->private = inode->i_private; - } - return res; -} - -static const struct file_operations sge_qinfo_debugfs_fops = { - .owner = THIS_MODULE, - .open = sge_qinfo_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; +DEFINE_SEQ_ATTRIBUTE(sge_qinfo); /* * Show SGE Queue Set statistics. We display QPL Queues Sets per line. @@ -2317,31 +2281,14 @@ static void *sge_qstats_next(struct seq_file *seq, void *v, loff_t *pos) return *pos < entries ? (void *)((uintptr_t)*pos + 1) : NULL; } -static const struct seq_operations sge_qstats_seq_ops = { +static const struct seq_operations sge_qstats_sops = { .start = sge_qstats_start, .next = sge_qstats_next, .stop = sge_qstats_stop, .show = sge_qstats_show }; -static int sge_qstats_open(struct inode *inode, struct file *file) -{ - int res = seq_open(file, &sge_qstats_seq_ops); - - if (res == 0) { - struct seq_file *seq = file->private_data; - seq->private = inode->i_private; - } - return res; -} - -static const struct file_operations sge_qstats_proc_fops = { - .owner = THIS_MODULE, - .open = sge_qstats_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; +DEFINE_SEQ_ATTRIBUTE(sge_qstats); /* * Show PCI-E SR-IOV Virtual Function Resource Limits. @@ -2415,31 +2362,14 @@ static void interfaces_stop(struct seq_file *seq, void *v) { } -static const struct seq_operations interfaces_seq_ops = { +static const struct seq_operations interfaces_sops = { .start = interfaces_start, .next = interfaces_next, .stop = interfaces_stop, .show = interfaces_show }; -static int interfaces_open(struct inode *inode, struct file *file) -{ - int res = seq_open(file, &interfaces_seq_ops); - - if (res == 0) { - struct seq_file *seq = file->private_data; - seq->private = inode->i_private; - } - return res; -} - -static const struct file_operations interfaces_proc_fops = { - .owner = THIS_MODULE, - .open = interfaces_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; +DEFINE_SEQ_ATTRIBUTE(interfaces); /* * /sys/kernel/debugfs/cxgb4vf/ files list. @@ -2452,10 +2382,10 @@ struct cxgb4vf_debugfs_entry { static struct cxgb4vf_debugfs_entry debugfs_files[] = { { "mboxlog", 0444, &mboxlog_fops }, - { "sge_qinfo", 0444, &sge_qinfo_debugfs_fops }, - { "sge_qstats", 0444, &sge_qstats_proc_fops }, + { "sge_qinfo", 0444, &sge_qinfo_fops }, + { "sge_qstats", 0444, &sge_qstats_fops }, { "resources", 0444, &resources_fops }, - { "interfaces", 0444, &interfaces_proc_fops }, + { "interfaces", 0444, &interfaces_fops }, }; /* diff --git a/drivers/net/ethernet/chelsio/inline_crypto/Kconfig b/drivers/net/ethernet/chelsio/inline_crypto/Kconfig new file mode 100644 index 000000000000..7dfa57348d54 --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/Kconfig @@ -0,0 +1,52 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +# Chelsio inline crypto configuration +# + +config CHELSIO_INLINE_CRYPTO + bool "Chelsio Inline Crypto support" + depends on CHELSIO_T4 + default y + help + Enable support for inline crypto. + Allows enable/disable from list of inline crypto drivers. + +if CHELSIO_INLINE_CRYPTO + +config CRYPTO_DEV_CHELSIO_TLS + tristate "Chelsio Crypto Inline TLS Driver" + depends on CHELSIO_T4 + depends on TLS_TOE + help + Support Chelsio Inline TLS with Chelsio crypto accelerator. + Enable inline TLS support for Tx and Rx. + + To compile this driver as a module, choose M here: the module + will be called chtls. + +config CHELSIO_IPSEC_INLINE + tristate "Chelsio IPSec XFRM Tx crypto offload" + depends on CHELSIO_T4 + depends on XFRM_OFFLOAD + depends on INET_ESP_OFFLOAD || INET6_ESP_OFFLOAD + help + Support Chelsio Inline IPsec with Chelsio crypto accelerator. + Enable inline IPsec support for Tx. + + To compile this driver as a module, choose M here: the module + will be called ch_ipsec. + +config CHELSIO_TLS_DEVICE + tristate "Chelsio Inline KTLS Offload" + depends on CHELSIO_T4 + depends on TLS + depends on TLS_DEVICE + help + This flag enables support for kernel tls offload over Chelsio T6 + crypto accelerator. CONFIG_CHELSIO_TLS_DEVICE flag can be enabled + only if CONFIG_TLS and CONFIG_TLS_DEVICE flags are enabled. + + To compile this driver as a module, choose M here: the module + will be called ch_ktls. + +endif # CHELSIO_INLINE_CRYPTO diff --git a/drivers/net/ethernet/chelsio/inline_crypto/Makefile b/drivers/net/ethernet/chelsio/inline_crypto/Makefile new file mode 100644 index 000000000000..27e6d7e2f1eb --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/Makefile @@ -0,0 +1,4 @@ +# SPDX-License-Identifier: GPL-2.0-only +obj-$(CONFIG_CRYPTO_DEV_CHELSIO_TLS) += chtls/ +obj-$(CONFIG_CHELSIO_IPSEC_INLINE) += ch_ipsec/ +obj-$(CONFIG_CHELSIO_TLS_DEVICE) += ch_ktls/ diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/Makefile b/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/Makefile new file mode 100644 index 000000000000..efdcaaebc455 --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/Makefile @@ -0,0 +1,8 @@ +# SPDX-License-Identifier: GPL-2.0-only +ccflags-y := -I $(srctree)/drivers/net/ethernet/chelsio/cxgb4 \ + -I $(srctree)/drivers/crypto/chelsio + +obj-$(CONFIG_CHELSIO_IPSEC_INLINE) += ch_ipsec.o +ch_ipsec-objs := chcr_ipsec.o + + diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/chcr_ipsec.c b/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/chcr_ipsec.c new file mode 100644 index 000000000000..072299b14b8d --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/chcr_ipsec.c @@ -0,0 +1,825 @@ +/* + * This file is part of the Chelsio T6 Crypto driver for Linux. + * + * Copyright (c) 2003-2017 Chelsio Communications, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + * + * Written and Maintained by: + * Atul Gupta (atul.gupta@chelsio.com) + */ + +#define pr_fmt(fmt) "ch_ipsec: " fmt + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/crypto.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/highmem.h> +#include <linux/if_vlan.h> +#include <linux/ip.h> +#include <linux/netdevice.h> +#include <net/esp.h> +#include <net/xfrm.h> +#include <crypto/aes.h> +#include <crypto/algapi.h> +#include <crypto/hash.h> +#include <crypto/sha.h> +#include <crypto/authenc.h> +#include <crypto/internal/aead.h> +#include <crypto/null.h> +#include <crypto/internal/skcipher.h> +#include <crypto/aead.h> +#include <crypto/scatterwalk.h> +#include <crypto/internal/hash.h> + +#include "chcr_ipsec.h" + +/* + * Max Tx descriptor space we allow for an Ethernet packet to be inlined + * into a WR. + */ +#define MAX_IMM_TX_PKT_LEN 256 +#define GCM_ESP_IV_SIZE 8 + +static LIST_HEAD(uld_ctx_list); +static DEFINE_MUTEX(dev_mutex); + +static bool ch_ipsec_offload_ok(struct sk_buff *skb, struct xfrm_state *x); +static int ch_ipsec_uld_state_change(void *handle, enum cxgb4_state new_state); +static int ch_ipsec_xmit(struct sk_buff *skb, struct net_device *dev); +static void *ch_ipsec_uld_add(const struct cxgb4_lld_info *infop); +static void ch_ipsec_advance_esn_state(struct xfrm_state *x); +static void ch_ipsec_xfrm_free_state(struct xfrm_state *x); +static void ch_ipsec_xfrm_del_state(struct xfrm_state *x); +static int ch_ipsec_xfrm_add_state(struct xfrm_state *x); + +static const struct xfrmdev_ops ch_ipsec_xfrmdev_ops = { + .xdo_dev_state_add = ch_ipsec_xfrm_add_state, + .xdo_dev_state_delete = ch_ipsec_xfrm_del_state, + .xdo_dev_state_free = ch_ipsec_xfrm_free_state, + .xdo_dev_offload_ok = ch_ipsec_offload_ok, + .xdo_dev_state_advance_esn = ch_ipsec_advance_esn_state, +}; + +static struct cxgb4_uld_info ch_ipsec_uld_info = { + .name = CHIPSEC_DRV_MODULE_NAME, + .nrxq = MAX_ULD_QSETS, + /* Max ntxq will be derived from fw config file*/ + .rxq_size = 1024, + .add = ch_ipsec_uld_add, + .state_change = ch_ipsec_uld_state_change, + .tx_handler = ch_ipsec_xmit, + .xfrmdev_ops = &ch_ipsec_xfrmdev_ops, +}; + +static void *ch_ipsec_uld_add(const struct cxgb4_lld_info *infop) +{ + struct ipsec_uld_ctx *u_ctx; + + pr_info_once("%s - version %s\n", CHIPSEC_DRV_DESC, + CHIPSEC_DRV_VERSION); + u_ctx = kzalloc(sizeof(*u_ctx), GFP_KERNEL); + if (!u_ctx) { + u_ctx = ERR_PTR(-ENOMEM); + goto out; + } + u_ctx->lldi = *infop; +out: + return u_ctx; +} + +static int ch_ipsec_uld_state_change(void *handle, enum cxgb4_state new_state) +{ + struct ipsec_uld_ctx *u_ctx = handle; + + pr_debug("new_state %u\n", new_state); + switch (new_state) { + case CXGB4_STATE_UP: + pr_info("%s: Up\n", pci_name(u_ctx->lldi.pdev)); + mutex_lock(&dev_mutex); + list_add_tail(&u_ctx->entry, &uld_ctx_list); + mutex_unlock(&dev_mutex); + break; + case CXGB4_STATE_START_RECOVERY: + case CXGB4_STATE_DOWN: + case CXGB4_STATE_DETACH: + pr_info("%s: Down\n", pci_name(u_ctx->lldi.pdev)); + list_del(&u_ctx->entry); + break; + default: + break; + } + + return 0; +} + +static int ch_ipsec_setauthsize(struct xfrm_state *x, + struct ipsec_sa_entry *sa_entry) +{ + int hmac_ctrl; + int authsize = x->aead->alg_icv_len / 8; + + sa_entry->authsize = authsize; + + switch (authsize) { + case ICV_8: + hmac_ctrl = CHCR_SCMD_HMAC_CTRL_DIV2; + break; + case ICV_12: + hmac_ctrl = CHCR_SCMD_HMAC_CTRL_IPSEC_96BIT; + break; + case ICV_16: + hmac_ctrl = CHCR_SCMD_HMAC_CTRL_NO_TRUNC; + break; + default: + return -EINVAL; + } + return hmac_ctrl; +} + +static int ch_ipsec_setkey(struct xfrm_state *x, + struct ipsec_sa_entry *sa_entry) +{ + int keylen = (x->aead->alg_key_len + 7) / 8; + unsigned char *key = x->aead->alg_key; + int ck_size, key_ctx_size = 0; + unsigned char ghash_h[AEAD_H_SIZE]; + struct crypto_aes_ctx aes; + int ret = 0; + + if (keylen > 3) { + keylen -= 4; /* nonce/salt is present in the last 4 bytes */ + memcpy(sa_entry->salt, key + keylen, 4); + } + + if (keylen == AES_KEYSIZE_128) { + ck_size = CHCR_KEYCTX_CIPHER_KEY_SIZE_128; + } else if (keylen == AES_KEYSIZE_192) { + ck_size = CHCR_KEYCTX_CIPHER_KEY_SIZE_192; + } else if (keylen == AES_KEYSIZE_256) { + ck_size = CHCR_KEYCTX_CIPHER_KEY_SIZE_256; + } else { + pr_err("GCM: Invalid key length %d\n", keylen); + ret = -EINVAL; + goto out; + } + + memcpy(sa_entry->key, key, keylen); + sa_entry->enckey_len = keylen; + key_ctx_size = sizeof(struct _key_ctx) + + ((DIV_ROUND_UP(keylen, 16)) << 4) + + AEAD_H_SIZE; + + sa_entry->key_ctx_hdr = FILL_KEY_CTX_HDR(ck_size, + CHCR_KEYCTX_MAC_KEY_SIZE_128, + 0, 0, + key_ctx_size >> 4); + + /* Calculate the H = CIPH(K, 0 repeated 16 times). + * It will go in key context + */ + ret = aes_expandkey(&aes, key, keylen); + if (ret) { + sa_entry->enckey_len = 0; + goto out; + } + memset(ghash_h, 0, AEAD_H_SIZE); + aes_encrypt(&aes, ghash_h, ghash_h); + memzero_explicit(&aes, sizeof(aes)); + + memcpy(sa_entry->key + (DIV_ROUND_UP(sa_entry->enckey_len, 16) * + 16), ghash_h, AEAD_H_SIZE); + sa_entry->kctx_len = ((DIV_ROUND_UP(sa_entry->enckey_len, 16)) << 4) + + AEAD_H_SIZE; +out: + return ret; +} + +/* + * ch_ipsec_xfrm_add_state + * returns 0 on success, negative error if failed to send message to FPGA + * positive error if FPGA returned a bad response + */ +static int ch_ipsec_xfrm_add_state(struct xfrm_state *x) +{ + struct ipsec_sa_entry *sa_entry; + int res = 0; + + if (x->props.aalgo != SADB_AALG_NONE) { + pr_debug("Cannot offload authenticated xfrm states\n"); + return -EINVAL; + } + if (x->props.calgo != SADB_X_CALG_NONE) { + pr_debug("Cannot offload compressed xfrm states\n"); + return -EINVAL; + } + if (x->props.family != AF_INET && + x->props.family != AF_INET6) { + pr_debug("Only IPv4/6 xfrm state offloaded\n"); + return -EINVAL; + } + if (x->props.mode != XFRM_MODE_TRANSPORT && + x->props.mode != XFRM_MODE_TUNNEL) { + pr_debug("Only transport and tunnel xfrm offload\n"); + return -EINVAL; + } + if (x->id.proto != IPPROTO_ESP) { + pr_debug("Only ESP xfrm state offloaded\n"); + return -EINVAL; + } + if (x->encap) { + pr_debug("Encapsulated xfrm state not offloaded\n"); + return -EINVAL; + } + if (!x->aead) { + pr_debug("Cannot offload xfrm states without aead\n"); + return -EINVAL; + } + if (x->aead->alg_icv_len != 128 && + x->aead->alg_icv_len != 96) { + pr_debug("Cannot offload xfrm states with AEAD ICV length other than 96b & 128b\n"); + return -EINVAL; + } + if ((x->aead->alg_key_len != 128 + 32) && + (x->aead->alg_key_len != 256 + 32)) { + pr_debug("cannot offload xfrm states with AEAD key length other than 128/256 bit\n"); + return -EINVAL; + } + if (x->tfcpad) { + pr_debug("Cannot offload xfrm states with tfc padding\n"); + return -EINVAL; + } + if (!x->geniv) { + pr_debug("Cannot offload xfrm states without geniv\n"); + return -EINVAL; + } + if (strcmp(x->geniv, "seqiv")) { + pr_debug("Cannot offload xfrm states with geniv other than seqiv\n"); + return -EINVAL; + } + + sa_entry = kzalloc(sizeof(*sa_entry), GFP_KERNEL); + if (!sa_entry) { + res = -ENOMEM; + goto out; + } + + sa_entry->hmac_ctrl = ch_ipsec_setauthsize(x, sa_entry); + if (x->props.flags & XFRM_STATE_ESN) + sa_entry->esn = 1; + ch_ipsec_setkey(x, sa_entry); + x->xso.offload_handle = (unsigned long)sa_entry; + try_module_get(THIS_MODULE); +out: + return res; +} + +static void ch_ipsec_xfrm_del_state(struct xfrm_state *x) +{ + /* do nothing */ + if (!x->xso.offload_handle) + return; +} + +static void ch_ipsec_xfrm_free_state(struct xfrm_state *x) +{ + struct ipsec_sa_entry *sa_entry; + + if (!x->xso.offload_handle) + return; + + sa_entry = (struct ipsec_sa_entry *)x->xso.offload_handle; + kfree(sa_entry); + module_put(THIS_MODULE); +} + +static bool ch_ipsec_offload_ok(struct sk_buff *skb, struct xfrm_state *x) +{ + if (x->props.family == AF_INET) { + /* Offload with IP options is not supported yet */ + if (ip_hdr(skb)->ihl > 5) + return false; + } else { + /* Offload with IPv6 extension headers is not support yet */ + if (ipv6_ext_hdr(ipv6_hdr(skb)->nexthdr)) + return false; + } + return true; +} + +static void ch_ipsec_advance_esn_state(struct xfrm_state *x) +{ + /* do nothing */ + if (!x->xso.offload_handle) + return; +} + +static int is_eth_imm(const struct sk_buff *skb, + struct ipsec_sa_entry *sa_entry) +{ + unsigned int kctx_len; + int hdrlen; + + kctx_len = sa_entry->kctx_len; + hdrlen = sizeof(struct fw_ulptx_wr) + + sizeof(struct chcr_ipsec_req) + kctx_len; + + hdrlen += sizeof(struct cpl_tx_pkt); + if (sa_entry->esn) + hdrlen += (DIV_ROUND_UP(sizeof(struct chcr_ipsec_aadiv), 16) + << 4); + if (skb->len <= MAX_IMM_TX_PKT_LEN - hdrlen) + return hdrlen; + return 0; +} + +static unsigned int calc_tx_sec_flits(const struct sk_buff *skb, + struct ipsec_sa_entry *sa_entry, + bool *immediate) +{ + unsigned int kctx_len; + unsigned int flits; + int aadivlen; + int hdrlen; + + kctx_len = sa_entry->kctx_len; + hdrlen = is_eth_imm(skb, sa_entry); + aadivlen = sa_entry->esn ? DIV_ROUND_UP(sizeof(struct chcr_ipsec_aadiv), + 16) : 0; + aadivlen <<= 4; + + /* If the skb is small enough, we can pump it out as a work request + * with only immediate data. In that case we just have to have the + * TX Packet header plus the skb data in the Work Request. + */ + + if (hdrlen) { + *immediate = true; + return DIV_ROUND_UP(skb->len + hdrlen, sizeof(__be64)); + } + + flits = sgl_len(skb_shinfo(skb)->nr_frags + 1); + + /* Otherwise, we're going to have to construct a Scatter gather list + * of the skb body and fragments. We also include the flits necessary + * for the TX Packet Work Request and CPL. We always have a firmware + * Write Header (incorporated as part of the cpl_tx_pkt_lso and + * cpl_tx_pkt structures), followed by either a TX Packet Write CPL + * message or, if we're doing a Large Send Offload, an LSO CPL message + * with an embedded TX Packet Write CPL message. + */ + flits += (sizeof(struct fw_ulptx_wr) + + sizeof(struct chcr_ipsec_req) + + kctx_len + + sizeof(struct cpl_tx_pkt_core) + + aadivlen) / sizeof(__be64); + return flits; +} + +static void *copy_esn_pktxt(struct sk_buff *skb, + struct net_device *dev, + void *pos, + struct ipsec_sa_entry *sa_entry) +{ + struct chcr_ipsec_aadiv *aadiv; + struct ulptx_idata *sc_imm; + struct ip_esp_hdr *esphdr; + struct xfrm_offload *xo; + struct sge_eth_txq *q; + struct adapter *adap; + struct port_info *pi; + __be64 seqno; + u32 qidx; + u32 seqlo; + u8 *iv; + int eoq; + int len; + + pi = netdev_priv(dev); + adap = pi->adapter; + qidx = skb->queue_mapping; + q = &adap->sge.ethtxq[qidx + pi->first_qset]; + + /* end of queue, reset pos to start of queue */ + eoq = (void *)q->q.stat - pos; + if (!eoq) + pos = q->q.desc; + + len = DIV_ROUND_UP(sizeof(struct chcr_ipsec_aadiv), 16) << 4; + memset(pos, 0, len); + aadiv = (struct chcr_ipsec_aadiv *)pos; + esphdr = (struct ip_esp_hdr *)skb_transport_header(skb); + iv = skb_transport_header(skb) + sizeof(struct ip_esp_hdr); + xo = xfrm_offload(skb); + + aadiv->spi = (esphdr->spi); + seqlo = ntohl(esphdr->seq_no); + seqno = cpu_to_be64(seqlo + ((u64)xo->seq.hi << 32)); + memcpy(aadiv->seq_no, &seqno, 8); + iv = skb_transport_header(skb) + sizeof(struct ip_esp_hdr); + memcpy(aadiv->iv, iv, 8); + + if (is_eth_imm(skb, sa_entry) && !skb_is_nonlinear(skb)) { + sc_imm = (struct ulptx_idata *)(pos + + (DIV_ROUND_UP(sizeof(struct chcr_ipsec_aadiv), + sizeof(__be64)) << 3)); + sc_imm->cmd_more = FILL_CMD_MORE(0); + sc_imm->len = cpu_to_be32(skb->len); + } + pos += len; + return pos; +} + +static void *copy_cpltx_pktxt(struct sk_buff *skb, + struct net_device *dev, + void *pos, + struct ipsec_sa_entry *sa_entry) +{ + struct cpl_tx_pkt_core *cpl; + struct sge_eth_txq *q; + struct adapter *adap; + struct port_info *pi; + u32 ctrl0, qidx; + u64 cntrl = 0; + int left; + + pi = netdev_priv(dev); + adap = pi->adapter; + qidx = skb->queue_mapping; + q = &adap->sge.ethtxq[qidx + pi->first_qset]; + + left = (void *)q->q.stat - pos; + if (!left) + pos = q->q.desc; + + cpl = (struct cpl_tx_pkt_core *)pos; + + cntrl = TXPKT_L4CSUM_DIS_F | TXPKT_IPCSUM_DIS_F; + ctrl0 = TXPKT_OPCODE_V(CPL_TX_PKT_XT) | TXPKT_INTF_V(pi->tx_chan) | + TXPKT_PF_V(adap->pf); + if (skb_vlan_tag_present(skb)) { + q->vlan_ins++; + cntrl |= TXPKT_VLAN_VLD_F | TXPKT_VLAN_V(skb_vlan_tag_get(skb)); + } + + cpl->ctrl0 = htonl(ctrl0); + cpl->pack = htons(0); + cpl->len = htons(skb->len); + cpl->ctrl1 = cpu_to_be64(cntrl); + + pos += sizeof(struct cpl_tx_pkt_core); + /* Copy ESN info for HW */ + if (sa_entry->esn) + pos = copy_esn_pktxt(skb, dev, pos, sa_entry); + return pos; +} + +static void *copy_key_cpltx_pktxt(struct sk_buff *skb, + struct net_device *dev, + void *pos, + struct ipsec_sa_entry *sa_entry) +{ + struct _key_ctx *key_ctx; + int left, eoq, key_len; + struct sge_eth_txq *q; + struct adapter *adap; + struct port_info *pi; + unsigned int qidx; + + pi = netdev_priv(dev); + adap = pi->adapter; + qidx = skb->queue_mapping; + q = &adap->sge.ethtxq[qidx + pi->first_qset]; + key_len = sa_entry->kctx_len; + + /* end of queue, reset pos to start of queue */ + eoq = (void *)q->q.stat - pos; + left = eoq; + if (!eoq) { + pos = q->q.desc; + left = 64 * q->q.size; + } + + /* Copy the Key context header */ + key_ctx = (struct _key_ctx *)pos; + key_ctx->ctx_hdr = sa_entry->key_ctx_hdr; + memcpy(key_ctx->salt, sa_entry->salt, MAX_SALT); + pos += sizeof(struct _key_ctx); + left -= sizeof(struct _key_ctx); + + if (likely(key_len <= left)) { + memcpy(key_ctx->key, sa_entry->key, key_len); + pos += key_len; + } else { + memcpy(pos, sa_entry->key, left); + memcpy(q->q.desc, sa_entry->key + left, + key_len - left); + pos = (u8 *)q->q.desc + (key_len - left); + } + /* Copy CPL TX PKT XT */ + pos = copy_cpltx_pktxt(skb, dev, pos, sa_entry); + + return pos; +} + +static void *ch_ipsec_crypto_wreq(struct sk_buff *skb, + struct net_device *dev, + void *pos, + int credits, + struct ipsec_sa_entry *sa_entry) +{ + struct port_info *pi = netdev_priv(dev); + struct adapter *adap = pi->adapter; + unsigned int ivsize = GCM_ESP_IV_SIZE; + struct chcr_ipsec_wr *wr; + bool immediate = false; + u16 immdatalen = 0; + unsigned int flits; + u32 ivinoffset; + u32 aadstart; + u32 aadstop; + u32 ciphstart; + u16 sc_more = 0; + u32 ivdrop = 0; + u32 esnlen = 0; + u32 wr_mid; + u16 ndesc; + int qidx = skb_get_queue_mapping(skb); + struct sge_eth_txq *q = &adap->sge.ethtxq[qidx + pi->first_qset]; + unsigned int kctx_len = sa_entry->kctx_len; + int qid = q->q.cntxt_id; + + atomic_inc(&adap->ch_ipsec_stats.ipsec_cnt); + + flits = calc_tx_sec_flits(skb, sa_entry, &immediate); + ndesc = DIV_ROUND_UP(flits, 2); + if (sa_entry->esn) + ivdrop = 1; + + if (immediate) + immdatalen = skb->len; + + if (sa_entry->esn) { + esnlen = sizeof(struct chcr_ipsec_aadiv); + if (!skb_is_nonlinear(skb)) + sc_more = 1; + } + + /* WR Header */ + wr = (struct chcr_ipsec_wr *)pos; + wr->wreq.op_to_compl = htonl(FW_WR_OP_V(FW_ULPTX_WR)); + wr_mid = FW_CRYPTO_LOOKASIDE_WR_LEN16_V(ndesc); + + if (unlikely(credits < ETHTXQ_STOP_THRES)) { + netif_tx_stop_queue(q->txq); + q->q.stops++; + if (!q->dbqt) + wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F; + } + wr_mid |= FW_ULPTX_WR_DATA_F; + wr->wreq.flowid_len16 = htonl(wr_mid); + + /* ULPTX */ + wr->req.ulptx.cmd_dest = FILL_ULPTX_CMD_DEST(pi->port_id, qid); + wr->req.ulptx.len = htonl(ndesc - 1); + + /* Sub-command */ + wr->req.sc_imm.cmd_more = FILL_CMD_MORE(!immdatalen || sc_more); + wr->req.sc_imm.len = cpu_to_be32(sizeof(struct cpl_tx_sec_pdu) + + sizeof(wr->req.key_ctx) + + kctx_len + + sizeof(struct cpl_tx_pkt_core) + + esnlen + + (esnlen ? 0 : immdatalen)); + + /* CPL_SEC_PDU */ + ivinoffset = sa_entry->esn ? (ESN_IV_INSERT_OFFSET + 1) : + (skb_transport_offset(skb) + + sizeof(struct ip_esp_hdr) + 1); + wr->req.sec_cpl.op_ivinsrtofst = htonl( + CPL_TX_SEC_PDU_OPCODE_V(CPL_TX_SEC_PDU) | + CPL_TX_SEC_PDU_CPLLEN_V(2) | + CPL_TX_SEC_PDU_PLACEHOLDER_V(1) | + CPL_TX_SEC_PDU_IVINSRTOFST_V( + ivinoffset)); + + wr->req.sec_cpl.pldlen = htonl(skb->len + esnlen); + aadstart = sa_entry->esn ? 1 : (skb_transport_offset(skb) + 1); + aadstop = sa_entry->esn ? ESN_IV_INSERT_OFFSET : + (skb_transport_offset(skb) + + sizeof(struct ip_esp_hdr)); + ciphstart = skb_transport_offset(skb) + sizeof(struct ip_esp_hdr) + + GCM_ESP_IV_SIZE + 1; + ciphstart += sa_entry->esn ? esnlen : 0; + + wr->req.sec_cpl.aadstart_cipherstop_hi = FILL_SEC_CPL_CIPHERSTOP_HI( + aadstart, + aadstop, + ciphstart, 0); + + wr->req.sec_cpl.cipherstop_lo_authinsert = + FILL_SEC_CPL_AUTHINSERT(0, ciphstart, + sa_entry->authsize, + sa_entry->authsize); + wr->req.sec_cpl.seqno_numivs = + FILL_SEC_CPL_SCMD0_SEQNO(CHCR_ENCRYPT_OP, 1, + CHCR_SCMD_CIPHER_MODE_AES_GCM, + CHCR_SCMD_AUTH_MODE_GHASH, + sa_entry->hmac_ctrl, + ivsize >> 1); + wr->req.sec_cpl.ivgen_hdrlen = FILL_SEC_CPL_IVGEN_HDRLEN(0, 0, 1, + 0, ivdrop, 0); + + pos += sizeof(struct fw_ulptx_wr) + + sizeof(struct ulp_txpkt) + + sizeof(struct ulptx_idata) + + sizeof(struct cpl_tx_sec_pdu); + + pos = copy_key_cpltx_pktxt(skb, dev, pos, sa_entry); + + return pos; +} + +/** + * flits_to_desc - returns the num of Tx descriptors for the given flits + * @n: the number of flits + * + * Returns the number of Tx descriptors needed for the supplied number + * of flits. + */ +static unsigned int flits_to_desc(unsigned int n) +{ + WARN_ON(n > SGE_MAX_WR_LEN / 8); + return DIV_ROUND_UP(n, 8); +} + +static unsigned int txq_avail(const struct sge_txq *q) +{ + return q->size - 1 - q->in_use; +} + +static void eth_txq_stop(struct sge_eth_txq *q) +{ + netif_tx_stop_queue(q->txq); + q->q.stops++; +} + +static void txq_advance(struct sge_txq *q, unsigned int n) +{ + q->in_use += n; + q->pidx += n; + if (q->pidx >= q->size) + q->pidx -= q->size; +} + +/* + * ch_ipsec_xmit called from ULD Tx handler + */ +int ch_ipsec_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct xfrm_state *x = xfrm_input_state(skb); + unsigned int last_desc, ndesc, flits = 0; + struct ipsec_sa_entry *sa_entry; + u64 *pos, *end, *before, *sgl; + struct tx_sw_desc *sgl_sdesc; + int qidx, left, credits; + bool immediate = false; + struct sge_eth_txq *q; + struct adapter *adap; + struct port_info *pi; + struct sec_path *sp; + + if (!x->xso.offload_handle) + return NETDEV_TX_BUSY; + + sa_entry = (struct ipsec_sa_entry *)x->xso.offload_handle; + + sp = skb_sec_path(skb); + if (sp->len != 1) { +out_free: dev_kfree_skb_any(skb); + return NETDEV_TX_OK; + } + + pi = netdev_priv(dev); + adap = pi->adapter; + qidx = skb->queue_mapping; + q = &adap->sge.ethtxq[qidx + pi->first_qset]; + + cxgb4_reclaim_completed_tx(adap, &q->q, true); + + flits = calc_tx_sec_flits(skb, sa_entry, &immediate); + ndesc = flits_to_desc(flits); + credits = txq_avail(&q->q) - ndesc; + + if (unlikely(credits < 0)) { + eth_txq_stop(q); + dev_err(adap->pdev_dev, + "%s: Tx ring %u full while queue awake! cred:%d %d %d flits:%d\n", + dev->name, qidx, credits, ndesc, txq_avail(&q->q), + flits); + return NETDEV_TX_BUSY; + } + + last_desc = q->q.pidx + ndesc - 1; + if (last_desc >= q->q.size) + last_desc -= q->q.size; + sgl_sdesc = &q->q.sdesc[last_desc]; + + if (!immediate && + unlikely(cxgb4_map_skb(adap->pdev_dev, skb, sgl_sdesc->addr) < 0)) { + memset(sgl_sdesc->addr, 0, sizeof(sgl_sdesc->addr)); + q->mapping_err++; + goto out_free; + } + + pos = (u64 *)&q->q.desc[q->q.pidx]; + before = (u64 *)pos; + end = (u64 *)pos + flits; + /* Setup IPSec CPL */ + pos = (void *)ch_ipsec_crypto_wreq(skb, dev, (void *)pos, + credits, sa_entry); + if (before > (u64 *)pos) { + left = (u8 *)end - (u8 *)q->q.stat; + end = (void *)q->q.desc + left; + } + if (pos == (u64 *)q->q.stat) { + left = (u8 *)end - (u8 *)q->q.stat; + end = (void *)q->q.desc + left; + pos = (void *)q->q.desc; + } + + sgl = (void *)pos; + if (immediate) { + cxgb4_inline_tx_skb(skb, &q->q, sgl); + dev_consume_skb_any(skb); + } else { + cxgb4_write_sgl(skb, &q->q, (void *)sgl, end, + 0, sgl_sdesc->addr); + skb_orphan(skb); + sgl_sdesc->skb = skb; + } + txq_advance(&q->q, ndesc); + + cxgb4_ring_tx_db(adap, &q->q, ndesc); + return NETDEV_TX_OK; +} + +static int __init ch_ipsec_init(void) +{ + cxgb4_register_uld(CXGB4_ULD_IPSEC, &ch_ipsec_uld_info); + + return 0; +} + +static void __exit ch_ipsec_exit(void) +{ + struct ipsec_uld_ctx *u_ctx, *tmp; + struct adapter *adap; + + mutex_lock(&dev_mutex); + list_for_each_entry_safe(u_ctx, tmp, &uld_ctx_list, entry) { + adap = pci_get_drvdata(u_ctx->lldi.pdev); + atomic_set(&adap->ch_ipsec_stats.ipsec_cnt, 0); + list_del(&u_ctx->entry); + kfree(u_ctx); + } + mutex_unlock(&dev_mutex); + cxgb4_unregister_uld(CXGB4_ULD_IPSEC); +} + +module_init(ch_ipsec_init); +module_exit(ch_ipsec_exit); + +MODULE_DESCRIPTION("Crypto IPSEC for Chelsio Terminator cards."); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Chelsio Communications"); +MODULE_VERSION(CHIPSEC_DRV_VERSION); + diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/chcr_ipsec.h b/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/chcr_ipsec.h new file mode 100644 index 000000000000..1d110d2edd64 --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ipsec/chcr_ipsec.h @@ -0,0 +1,58 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2018 Chelsio Communications, Inc. */ + +#ifndef __CHCR_IPSEC_H__ +#define __CHCR_IPSEC_H__ + +#include <crypto/algapi.h> +#include "t4_hw.h" +#include "cxgb4.h" +#include "t4_msg.h" +#include "cxgb4_uld.h" + +#include "chcr_core.h" +#include "chcr_algo.h" +#include "chcr_crypto.h" + +#define CHIPSEC_DRV_MODULE_NAME "ch_ipsec" +#define CHIPSEC_DRV_VERSION "1.0.0.0-ko" +#define CHIPSEC_DRV_DESC "Chelsio T6 Crypto Ipsec offload Driver" + +struct ipsec_uld_ctx { + struct list_head entry; + struct cxgb4_lld_info lldi; +}; + +struct chcr_ipsec_req { + struct ulp_txpkt ulptx; + struct ulptx_idata sc_imm; + struct cpl_tx_sec_pdu sec_cpl; + struct _key_ctx key_ctx; +}; + +struct chcr_ipsec_wr { + struct fw_ulptx_wr wreq; + struct chcr_ipsec_req req; +}; + +#define ESN_IV_INSERT_OFFSET 12 +struct chcr_ipsec_aadiv { + __be32 spi; + u8 seq_no[8]; + u8 iv[8]; +}; + +struct ipsec_sa_entry { + int hmac_ctrl; + u16 esn; + u16 resv; + unsigned int enckey_len; + unsigned int kctx_len; + unsigned int authsize; + __be32 key_ctx_hdr; + char salt[MAX_SALT]; + char key[2 * AES_MAX_KEY_SIZE]; +}; + +#endif /* __CHCR_IPSEC_H__ */ + diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/Makefile b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/Makefile new file mode 100644 index 000000000000..5e7d161c3199 --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/Makefile @@ -0,0 +1,5 @@ +# SPDX-License-Identifier: GPL-2.0-only +ccflags-y := -I $(srctree)/drivers/net/ethernet/chelsio/cxgb4 + +obj-$(CONFIG_CHELSIO_TLS_DEVICE) += ch_ktls.o +ch_ktls-objs := chcr_ktls.o diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_common.h b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_common.h new file mode 100644 index 000000000000..38319f4c3121 --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_common.h @@ -0,0 +1,111 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (C) 2020 Chelsio Communications. All rights reserved. */ + +#ifndef __CHCR_COMMON_H__ +#define __CHCR_COMMON_H__ + +#include "cxgb4.h" + +#define CHCR_MAX_SALT 4 +#define CHCR_KEYCTX_MAC_KEY_SIZE_128 0 +#define CHCR_KEYCTX_CIPHER_KEY_SIZE_128 0 +#define CHCR_SCMD_CIPHER_MODE_AES_GCM 2 +#define CHCR_SCMD_CIPHER_MODE_AES_CTR 3 +#define CHCR_CPL_TX_SEC_PDU_LEN_64BIT 2 +#define CHCR_SCMD_SEQ_NO_CTRL_64BIT 3 +#define CHCR_SCMD_PROTO_VERSION_TLS 0 +#define CHCR_SCMD_PROTO_VERSION_GENERIC 4 +#define CHCR_SCMD_AUTH_MODE_GHASH 4 +#define AES_BLOCK_LEN 16 + +struct ktls_key_ctx { + __be32 ctx_hdr; + u8 salt[CHCR_MAX_SALT]; + __be64 iv_to_auth; + unsigned char key[TLS_CIPHER_AES_GCM_128_KEY_SIZE + + TLS_CIPHER_AES_GCM_256_TAG_SIZE]; +}; + +/* Crypto key context */ +#define KEY_CONTEXT_CTX_LEN_S 24 +#define KEY_CONTEXT_CTX_LEN_V(x) ((x) << KEY_CONTEXT_CTX_LEN_S) + +#define KEY_CONTEXT_SALT_PRESENT_S 10 +#define KEY_CONTEXT_SALT_PRESENT_V(x) ((x) << KEY_CONTEXT_SALT_PRESENT_S) +#define KEY_CONTEXT_SALT_PRESENT_F KEY_CONTEXT_SALT_PRESENT_V(1U) + +#define KEY_CONTEXT_VALID_S 0 +#define KEY_CONTEXT_VALID_V(x) ((x) << KEY_CONTEXT_VALID_S) +#define KEY_CONTEXT_VALID_F KEY_CONTEXT_VALID_V(1U) + +#define KEY_CONTEXT_CK_SIZE_S 6 +#define KEY_CONTEXT_CK_SIZE_V(x) ((x) << KEY_CONTEXT_CK_SIZE_S) + +#define KEY_CONTEXT_MK_SIZE_S 2 +#define KEY_CONTEXT_MK_SIZE_V(x) ((x) << KEY_CONTEXT_MK_SIZE_S) + +#define KEY_CONTEXT_OPAD_PRESENT_S 11 +#define KEY_CONTEXT_OPAD_PRESENT_V(x) ((x) << KEY_CONTEXT_OPAD_PRESENT_S) +#define KEY_CONTEXT_OPAD_PRESENT_F KEY_CONTEXT_OPAD_PRESENT_V(1U) + +#define FILL_KEY_CTX_HDR(ck_size, mk_size, ctx_len) \ + htonl(KEY_CONTEXT_MK_SIZE_V(mk_size) | \ + KEY_CONTEXT_CK_SIZE_V(ck_size) | \ + KEY_CONTEXT_VALID_F | \ + KEY_CONTEXT_SALT_PRESENT_F | \ + KEY_CONTEXT_CTX_LEN_V((ctx_len))) + +static inline void *chcr_copy_to_txd(const void *src, const struct sge_txq *q, + void *pos, int length) +{ + int left = (void *)q->stat - pos; + u64 *p; + + if (likely(length <= left)) { + memcpy(pos, src, length); + pos += length; + } else { + memcpy(pos, src, left); + memcpy(q->desc, src + left, length - left); + pos = (void *)q->desc + (length - left); + } + /* 0-pad to multiple of 16 */ + p = PTR_ALIGN(pos, 8); + if ((uintptr_t)p & 8) { + *p = 0; + return p + 1; + } + return p; +} + +static inline unsigned int chcr_txq_avail(const struct sge_txq *q) +{ + return q->size - 1 - q->in_use; +} + +static inline void chcr_txq_advance(struct sge_txq *q, unsigned int n) +{ + q->in_use += n; + q->pidx += n; + if (q->pidx >= q->size) + q->pidx -= q->size; +} + +static inline void chcr_eth_txq_stop(struct sge_eth_txq *q) +{ + netif_tx_stop_queue(q->txq); + q->q.stops++; +} + +static inline unsigned int chcr_sgl_len(unsigned int n) +{ + n--; + return (3 * n) / 2 + (n & 1) + 2; +} + +static inline unsigned int chcr_flits_to_desc(unsigned int n) +{ + WARN_ON(n > SGE_MAX_WR_LEN / 8); + return DIV_ROUND_UP(n, 8); +} +#endif /* __CHCR_COMMON_H__ */ diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c new file mode 100644 index 000000000000..5195f692f14d --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c @@ -0,0 +1,2154 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (C) 2020 Chelsio Communications. All rights reserved. */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/skbuff.h> +#include <linux/module.h> +#include <linux/highmem.h> +#include <linux/ip.h> +#include <net/ipv6.h> +#include <linux/netdevice.h> +#include "chcr_ktls.h" + +static LIST_HEAD(uld_ctx_list); +static DEFINE_MUTEX(dev_mutex); + +static int chcr_init_tcb_fields(struct chcr_ktls_info *tx_info); +/* + * chcr_ktls_save_keys: calculate and save crypto keys. + * @tx_info - driver specific tls info. + * @crypto_info - tls crypto information. + * @direction - TX/RX direction. + * return - SUCCESS/FAILURE. + */ +static int chcr_ktls_save_keys(struct chcr_ktls_info *tx_info, + struct tls_crypto_info *crypto_info, + enum tls_offload_ctx_dir direction) +{ + int ck_size, key_ctx_size, mac_key_size, keylen, ghash_size, ret; + unsigned char ghash_h[TLS_CIPHER_AES_GCM_256_TAG_SIZE]; + struct tls12_crypto_info_aes_gcm_128 *info_128_gcm; + struct ktls_key_ctx *kctx = &tx_info->key_ctx; + struct crypto_cipher *cipher; + unsigned char *key, *salt; + + switch (crypto_info->cipher_type) { + case TLS_CIPHER_AES_GCM_128: + info_128_gcm = + (struct tls12_crypto_info_aes_gcm_128 *)crypto_info; + keylen = TLS_CIPHER_AES_GCM_128_KEY_SIZE; + ck_size = CHCR_KEYCTX_CIPHER_KEY_SIZE_128; + tx_info->salt_size = TLS_CIPHER_AES_GCM_128_SALT_SIZE; + mac_key_size = CHCR_KEYCTX_MAC_KEY_SIZE_128; + tx_info->iv_size = TLS_CIPHER_AES_GCM_128_IV_SIZE; + tx_info->iv = be64_to_cpu(*(__be64 *)info_128_gcm->iv); + + ghash_size = TLS_CIPHER_AES_GCM_128_TAG_SIZE; + key = info_128_gcm->key; + salt = info_128_gcm->salt; + tx_info->record_no = *(u64 *)info_128_gcm->rec_seq; + + /* The SCMD fields used when encrypting a full TLS + * record. Its a one time calculation till the + * connection exists. + */ + tx_info->scmd0_seqno_numivs = + SCMD_SEQ_NO_CTRL_V(CHCR_SCMD_SEQ_NO_CTRL_64BIT) | + SCMD_CIPH_AUTH_SEQ_CTRL_F | + SCMD_PROTO_VERSION_V(CHCR_SCMD_PROTO_VERSION_TLS) | + SCMD_CIPH_MODE_V(CHCR_SCMD_CIPHER_MODE_AES_GCM) | + SCMD_AUTH_MODE_V(CHCR_SCMD_AUTH_MODE_GHASH) | + SCMD_IV_SIZE_V(TLS_CIPHER_AES_GCM_128_IV_SIZE >> 1) | + SCMD_NUM_IVS_V(1); + + /* keys will be sent inline. */ + tx_info->scmd0_ivgen_hdrlen = SCMD_KEY_CTX_INLINE_F; + + /* The SCMD fields used when encrypting a partial TLS + * record (no trailer and possibly a truncated payload). + */ + tx_info->scmd0_short_seqno_numivs = + SCMD_CIPH_AUTH_SEQ_CTRL_F | + SCMD_PROTO_VERSION_V(CHCR_SCMD_PROTO_VERSION_GENERIC) | + SCMD_CIPH_MODE_V(CHCR_SCMD_CIPHER_MODE_AES_CTR) | + SCMD_IV_SIZE_V(AES_BLOCK_LEN >> 1); + + tx_info->scmd0_short_ivgen_hdrlen = + tx_info->scmd0_ivgen_hdrlen | SCMD_AADIVDROP_F; + + break; + + default: + pr_err("GCM: cipher type 0x%x not supported\n", + crypto_info->cipher_type); + ret = -EINVAL; + goto out; + } + + key_ctx_size = CHCR_KTLS_KEY_CTX_LEN + + roundup(keylen, 16) + ghash_size; + /* Calculate the H = CIPH(K, 0 repeated 16 times). + * It will go in key context + */ + cipher = crypto_alloc_cipher("aes", 0, 0); + if (IS_ERR(cipher)) { + ret = -ENOMEM; + goto out; + } + + ret = crypto_cipher_setkey(cipher, key, keylen); + if (ret) + goto out1; + + memset(ghash_h, 0, ghash_size); + crypto_cipher_encrypt_one(cipher, ghash_h, ghash_h); + + /* fill the Key context */ + if (direction == TLS_OFFLOAD_CTX_DIR_TX) { + kctx->ctx_hdr = FILL_KEY_CTX_HDR(ck_size, + mac_key_size, + key_ctx_size >> 4); + } else { + ret = -EINVAL; + goto out1; + } + + memcpy(kctx->salt, salt, tx_info->salt_size); + memcpy(kctx->key, key, keylen); + memcpy(kctx->key + keylen, ghash_h, ghash_size); + tx_info->key_ctx_len = key_ctx_size; + +out1: + crypto_free_cipher(cipher); +out: + return ret; +} + +/* + * chcr_ktls_act_open_req: creates TCB entry for ipv4 connection. + * @sk - tcp socket. + * @tx_info - driver specific tls info. + * @atid - connection active tid. + * return - send success/failure. + */ +static int chcr_ktls_act_open_req(struct sock *sk, + struct chcr_ktls_info *tx_info, + int atid) +{ + struct inet_sock *inet = inet_sk(sk); + struct cpl_t6_act_open_req *cpl6; + struct cpl_act_open_req *cpl; + struct sk_buff *skb; + unsigned int len; + int qid_atid; + u64 options; + + len = sizeof(*cpl6); + skb = alloc_skb(len, GFP_KERNEL); + if (unlikely(!skb)) + return -ENOMEM; + /* mark it a control pkt */ + set_wr_txq(skb, CPL_PRIORITY_CONTROL, tx_info->port_id); + + cpl6 = __skb_put_zero(skb, len); + cpl = (struct cpl_act_open_req *)cpl6; + INIT_TP_WR(cpl6, 0); + qid_atid = TID_QID_V(tx_info->rx_qid) | + TID_TID_V(atid); + OPCODE_TID(cpl) = htonl(MK_OPCODE_TID(CPL_ACT_OPEN_REQ, qid_atid)); + cpl->local_port = inet->inet_sport; + cpl->peer_port = inet->inet_dport; + cpl->local_ip = inet->inet_rcv_saddr; + cpl->peer_ip = inet->inet_daddr; + + /* fill first 64 bit option field. */ + options = TCAM_BYPASS_F | ULP_MODE_V(ULP_MODE_NONE) | NON_OFFLOAD_F | + SMAC_SEL_V(tx_info->smt_idx) | TX_CHAN_V(tx_info->tx_chan); + cpl->opt0 = cpu_to_be64(options); + + /* next 64 bit option field. */ + options = + TX_QUEUE_V(tx_info->adap->params.tp.tx_modq[tx_info->tx_chan]); + cpl->opt2 = htonl(options); + + return cxgb4_l2t_send(tx_info->netdev, skb, tx_info->l2te); +} + +#if IS_ENABLED(CONFIG_IPV6) +/* + * chcr_ktls_act_open_req6: creates TCB entry for ipv6 connection. + * @sk - tcp socket. + * @tx_info - driver specific tls info. + * @atid - connection active tid. + * return - send success/failure. + */ +static int chcr_ktls_act_open_req6(struct sock *sk, + struct chcr_ktls_info *tx_info, + int atid) +{ + struct inet_sock *inet = inet_sk(sk); + struct cpl_t6_act_open_req6 *cpl6; + struct cpl_act_open_req6 *cpl; + struct sk_buff *skb; + unsigned int len; + int qid_atid; + u64 options; + + len = sizeof(*cpl6); + skb = alloc_skb(len, GFP_KERNEL); + if (unlikely(!skb)) + return -ENOMEM; + /* mark it a control pkt */ + set_wr_txq(skb, CPL_PRIORITY_CONTROL, tx_info->port_id); + + cpl6 = __skb_put_zero(skb, len); + cpl = (struct cpl_act_open_req6 *)cpl6; + INIT_TP_WR(cpl6, 0); + qid_atid = TID_QID_V(tx_info->rx_qid) | TID_TID_V(atid); + OPCODE_TID(cpl) = htonl(MK_OPCODE_TID(CPL_ACT_OPEN_REQ6, qid_atid)); + cpl->local_port = inet->inet_sport; + cpl->peer_port = inet->inet_dport; + cpl->local_ip_hi = *(__be64 *)&sk->sk_v6_rcv_saddr.in6_u.u6_addr8[0]; + cpl->local_ip_lo = *(__be64 *)&sk->sk_v6_rcv_saddr.in6_u.u6_addr8[8]; + cpl->peer_ip_hi = *(__be64 *)&sk->sk_v6_daddr.in6_u.u6_addr8[0]; + cpl->peer_ip_lo = *(__be64 *)&sk->sk_v6_daddr.in6_u.u6_addr8[8]; + + /* first 64 bit option field. */ + options = TCAM_BYPASS_F | ULP_MODE_V(ULP_MODE_NONE) | NON_OFFLOAD_F | + SMAC_SEL_V(tx_info->smt_idx) | TX_CHAN_V(tx_info->tx_chan); + cpl->opt0 = cpu_to_be64(options); + /* next 64 bit option field. */ + options = + TX_QUEUE_V(tx_info->adap->params.tp.tx_modq[tx_info->tx_chan]); + cpl->opt2 = htonl(options); + + return cxgb4_l2t_send(tx_info->netdev, skb, tx_info->l2te); +} +#endif /* #if IS_ENABLED(CONFIG_IPV6) */ + +/* + * chcr_setup_connection: create a TCB entry so that TP will form tcp packets. + * @sk - tcp socket. + * @tx_info - driver specific tls info. + * return: NET_TX_OK/NET_XMIT_DROP + */ +static int chcr_setup_connection(struct sock *sk, + struct chcr_ktls_info *tx_info) +{ + struct tid_info *t = &tx_info->adap->tids; + int atid, ret = 0; + + atid = cxgb4_alloc_atid(t, tx_info); + if (atid == -1) + return -EINVAL; + + tx_info->atid = atid; + + if (tx_info->ip_family == AF_INET) { + ret = chcr_ktls_act_open_req(sk, tx_info, atid); +#if IS_ENABLED(CONFIG_IPV6) + } else { + ret = cxgb4_clip_get(tx_info->netdev, (const u32 *) + &sk->sk_v6_rcv_saddr, + 1); + if (ret) + return ret; + ret = chcr_ktls_act_open_req6(sk, tx_info, atid); +#endif + } + + /* if return type is NET_XMIT_CN, msg will be sent but delayed, mark ret + * success, if any other return type clear atid and return that failure. + */ + if (ret) { + if (ret == NET_XMIT_CN) { + ret = 0; + } else { +#if IS_ENABLED(CONFIG_IPV6) + /* clear clip entry */ + if (tx_info->ip_family == AF_INET6) + cxgb4_clip_release(tx_info->netdev, + (const u32 *) + &sk->sk_v6_rcv_saddr, + 1); +#endif + cxgb4_free_atid(t, atid); + } + } + + return ret; +} + +/* + * chcr_set_tcb_field: update tcb fields. + * @tx_info - driver specific tls info. + * @word - TCB word. + * @mask - TCB word related mask. + * @val - TCB word related value. + * @no_reply - set 1 if not looking for TP response. + */ +static int chcr_set_tcb_field(struct chcr_ktls_info *tx_info, u16 word, + u64 mask, u64 val, int no_reply) +{ + struct cpl_set_tcb_field *req; + struct sk_buff *skb; + + skb = alloc_skb(sizeof(struct cpl_set_tcb_field), GFP_ATOMIC); + if (!skb) + return -ENOMEM; + + req = (struct cpl_set_tcb_field *)__skb_put_zero(skb, sizeof(*req)); + INIT_TP_WR_CPL(req, CPL_SET_TCB_FIELD, tx_info->tid); + req->reply_ctrl = htons(QUEUENO_V(tx_info->rx_qid) | + NO_REPLY_V(no_reply)); + req->word_cookie = htons(TCB_WORD_V(word)); + req->mask = cpu_to_be64(mask); + req->val = cpu_to_be64(val); + + set_wr_txq(skb, CPL_PRIORITY_CONTROL, tx_info->port_id); + return cxgb4_ofld_send(tx_info->netdev, skb); +} + +/* + * chcr_ktls_mark_tcb_close: mark tcb state to CLOSE + * @tx_info - driver specific tls info. + * return: NET_TX_OK/NET_XMIT_DROP. + */ +static int chcr_ktls_mark_tcb_close(struct chcr_ktls_info *tx_info) +{ + return chcr_set_tcb_field(tx_info, TCB_T_STATE_W, + TCB_T_STATE_V(TCB_T_STATE_M), + CHCR_TCB_STATE_CLOSED, 1); +} + +/* + * chcr_ktls_dev_del: call back for tls_dev_del. + * Remove the tid and l2t entry and close the connection. + * it per connection basis. + * @netdev - net device. + * @tls_cts - tls context. + * @direction - TX/RX crypto direction + */ +static void chcr_ktls_dev_del(struct net_device *netdev, + struct tls_context *tls_ctx, + enum tls_offload_ctx_dir direction) +{ + struct chcr_ktls_ofld_ctx_tx *tx_ctx = + chcr_get_ktls_tx_context(tls_ctx); + struct chcr_ktls_info *tx_info = tx_ctx->chcr_info; + struct ch_ktls_port_stats_debug *port_stats; + + if (!tx_info) + return; + + /* clear l2t entry */ + if (tx_info->l2te) + cxgb4_l2t_release(tx_info->l2te); + +#if IS_ENABLED(CONFIG_IPV6) + /* clear clip entry */ + if (tx_info->ip_family == AF_INET6) + cxgb4_clip_release(netdev, (const u32 *) + &tx_info->sk->sk_v6_rcv_saddr, + 1); +#endif + + /* clear tid */ + if (tx_info->tid != -1) { + /* clear tcb state and then release tid */ + chcr_ktls_mark_tcb_close(tx_info); + cxgb4_remove_tid(&tx_info->adap->tids, tx_info->tx_chan, + tx_info->tid, tx_info->ip_family); + } + + port_stats = &tx_info->adap->ch_ktls_stats.ktls_port[tx_info->port_id]; + atomic64_inc(&port_stats->ktls_tx_connection_close); + kvfree(tx_info); + tx_ctx->chcr_info = NULL; + /* release module refcount */ + module_put(THIS_MODULE); +} + +/* + * chcr_ktls_dev_add: call back for tls_dev_add. + * Create a tcb entry for TP. Also add l2t entry for the connection. And + * generate keys & save those keys locally. + * @netdev - net device. + * @tls_cts - tls context. + * @direction - TX/RX crypto direction + * return: SUCCESS/FAILURE. + */ +static int chcr_ktls_dev_add(struct net_device *netdev, struct sock *sk, + enum tls_offload_ctx_dir direction, + struct tls_crypto_info *crypto_info, + u32 start_offload_tcp_sn) +{ + struct tls_context *tls_ctx = tls_get_ctx(sk); + struct ch_ktls_port_stats_debug *port_stats; + struct chcr_ktls_ofld_ctx_tx *tx_ctx; + struct chcr_ktls_info *tx_info; + struct dst_entry *dst; + struct adapter *adap; + struct port_info *pi; + struct neighbour *n; + u8 daaddr[16]; + int ret = -1; + + tx_ctx = chcr_get_ktls_tx_context(tls_ctx); + + pi = netdev_priv(netdev); + adap = pi->adapter; + port_stats = &adap->ch_ktls_stats.ktls_port[pi->port_id]; + atomic64_inc(&port_stats->ktls_tx_connection_open); + + if (direction == TLS_OFFLOAD_CTX_DIR_RX) { + pr_err("not expecting for RX direction\n"); + goto out; + } + + if (tx_ctx->chcr_info) + goto out; + + tx_info = kvzalloc(sizeof(*tx_info), GFP_KERNEL); + if (!tx_info) + goto out; + + tx_info->sk = sk; + spin_lock_init(&tx_info->lock); + /* initialize tid and atid to -1, 0 is a also a valid id. */ + tx_info->tid = -1; + tx_info->atid = -1; + + tx_info->adap = adap; + tx_info->netdev = netdev; + tx_info->first_qset = pi->first_qset; + tx_info->tx_chan = pi->tx_chan; + tx_info->smt_idx = pi->smt_idx; + tx_info->port_id = pi->port_id; + tx_info->prev_ack = 0; + tx_info->prev_win = 0; + + tx_info->rx_qid = chcr_get_first_rx_qid(adap); + if (unlikely(tx_info->rx_qid < 0)) + goto free_tx_info; + + tx_info->prev_seq = start_offload_tcp_sn; + tx_info->tcp_start_seq_number = start_offload_tcp_sn; + + /* save crypto keys */ + ret = chcr_ktls_save_keys(tx_info, crypto_info, direction); + if (ret < 0) + goto free_tx_info; + + /* get peer ip */ + if (sk->sk_family == AF_INET) { + memcpy(daaddr, &sk->sk_daddr, 4); + tx_info->ip_family = AF_INET; +#if IS_ENABLED(CONFIG_IPV6) + } else { + if (!sk->sk_ipv6only && + ipv6_addr_type(&sk->sk_v6_daddr) == IPV6_ADDR_MAPPED) { + memcpy(daaddr, &sk->sk_daddr, 4); + tx_info->ip_family = AF_INET; + } else { + memcpy(daaddr, sk->sk_v6_daddr.in6_u.u6_addr8, 16); + tx_info->ip_family = AF_INET6; + } +#endif + } + + /* get the l2t index */ + dst = sk_dst_get(sk); + if (!dst) { + pr_err("DST entry not found\n"); + goto free_tx_info; + } + n = dst_neigh_lookup(dst, daaddr); + if (!n || !n->dev) { + pr_err("neighbour not found\n"); + dst_release(dst); + goto free_tx_info; + } + tx_info->l2te = cxgb4_l2t_get(adap->l2t, n, n->dev, 0); + + neigh_release(n); + dst_release(dst); + + if (!tx_info->l2te) { + pr_err("l2t entry not found\n"); + goto free_tx_info; + } + + /* Driver shouldn't be removed until any single connection exists */ + if (!try_module_get(THIS_MODULE)) + goto free_l2t; + + init_completion(&tx_info->completion); + /* create a filter and call cxgb4_l2t_send to send the packet out, which + * will take care of updating l2t entry in hw if not already done. + */ + tx_info->open_state = CH_KTLS_OPEN_PENDING; + + if (chcr_setup_connection(sk, tx_info)) + goto put_module; + + /* Wait for reply */ + wait_for_completion_timeout(&tx_info->completion, 30 * HZ); + spin_lock_bh(&tx_info->lock); + if (tx_info->open_state) { + /* need to wait for hw response, can't free tx_info yet. */ + if (tx_info->open_state == CH_KTLS_OPEN_PENDING) + tx_info->pending_close = true; + /* free the lock after the cleanup */ + goto put_module; + } + spin_unlock_bh(&tx_info->lock); + + /* initialize tcb */ + reinit_completion(&tx_info->completion); + /* mark it pending for hw response */ + tx_info->open_state = CH_KTLS_OPEN_PENDING; + + if (chcr_init_tcb_fields(tx_info)) + goto free_tid; + + /* Wait for reply */ + wait_for_completion_timeout(&tx_info->completion, 30 * HZ); + spin_lock_bh(&tx_info->lock); + if (tx_info->open_state) { + /* need to wait for hw response, can't free tx_info yet. */ + tx_info->pending_close = true; + /* free the lock after cleanup */ + goto free_tid; + } + spin_unlock_bh(&tx_info->lock); + + if (!cxgb4_check_l2t_valid(tx_info->l2te)) + goto free_tid; + + atomic64_inc(&port_stats->ktls_tx_ctx); + tx_ctx->chcr_info = tx_info; + + return 0; + +free_tid: + chcr_ktls_mark_tcb_close(tx_info); +#if IS_ENABLED(CONFIG_IPV6) + /* clear clip entry */ + if (tx_info->ip_family == AF_INET6) + cxgb4_clip_release(netdev, (const u32 *) + &sk->sk_v6_rcv_saddr, + 1); +#endif + cxgb4_remove_tid(&tx_info->adap->tids, tx_info->tx_chan, + tx_info->tid, tx_info->ip_family); + +put_module: + /* release module refcount */ + module_put(THIS_MODULE); +free_l2t: + cxgb4_l2t_release(tx_info->l2te); +free_tx_info: + if (tx_info->pending_close) + spin_unlock_bh(&tx_info->lock); + else + kvfree(tx_info); +out: + atomic64_inc(&port_stats->ktls_tx_connection_fail); + return -1; +} + +/* + * chcr_init_tcb_fields: Initialize tcb fields to handle TCP seq number + * handling. + * @tx_info - driver specific tls info. + * return: NET_TX_OK/NET_XMIT_DROP + */ +static int chcr_init_tcb_fields(struct chcr_ktls_info *tx_info) +{ + int ret = 0; + + /* set tcb in offload and bypass */ + ret = + chcr_set_tcb_field(tx_info, TCB_T_FLAGS_W, + TCB_T_FLAGS_V(TF_CORE_BYPASS_F | TF_NON_OFFLOAD_F), + TCB_T_FLAGS_V(TF_CORE_BYPASS_F), 1); + if (ret) + return ret; + /* reset snd_una and snd_next fields in tcb */ + ret = chcr_set_tcb_field(tx_info, TCB_SND_UNA_RAW_W, + TCB_SND_NXT_RAW_V(TCB_SND_NXT_RAW_M) | + TCB_SND_UNA_RAW_V(TCB_SND_UNA_RAW_M), + 0, 1); + if (ret) + return ret; + + /* reset send max */ + ret = chcr_set_tcb_field(tx_info, TCB_SND_MAX_RAW_W, + TCB_SND_MAX_RAW_V(TCB_SND_MAX_RAW_M), + 0, 1); + if (ret) + return ret; + + /* update l2t index and request for tp reply to confirm tcb is + * initialised to handle tx traffic. + */ + ret = chcr_set_tcb_field(tx_info, TCB_L2T_IX_W, + TCB_L2T_IX_V(TCB_L2T_IX_M), + TCB_L2T_IX_V(tx_info->l2te->idx), 0); + return ret; +} + +/* + * chcr_ktls_cpl_act_open_rpl: connection reply received from TP. + */ +static int chcr_ktls_cpl_act_open_rpl(struct adapter *adap, + unsigned char *input) +{ + const struct cpl_act_open_rpl *p = (void *)input; + struct chcr_ktls_info *tx_info = NULL; + unsigned int atid, tid, status; + struct tid_info *t; + + tid = GET_TID(p); + status = AOPEN_STATUS_G(ntohl(p->atid_status)); + atid = TID_TID_G(AOPEN_ATID_G(ntohl(p->atid_status))); + + t = &adap->tids; + tx_info = lookup_atid(t, atid); + + if (!tx_info || tx_info->atid != atid) { + pr_err("%s: incorrect tx_info or atid\n", __func__); + return -1; + } + + cxgb4_free_atid(t, atid); + tx_info->atid = -1; + + spin_lock(&tx_info->lock); + /* HW response is very close, finish pending cleanup */ + if (tx_info->pending_close) { + spin_unlock(&tx_info->lock); + if (!status) { + /* it's a late success, tcb status is establised, + * mark it close. + */ + chcr_ktls_mark_tcb_close(tx_info); + cxgb4_remove_tid(&tx_info->adap->tids, tx_info->tx_chan, + tid, tx_info->ip_family); + } + kvfree(tx_info); + return 0; + } + + if (!status) { + tx_info->tid = tid; + cxgb4_insert_tid(t, tx_info, tx_info->tid, tx_info->ip_family); + tx_info->open_state = CH_KTLS_OPEN_SUCCESS; + } else { + tx_info->open_state = CH_KTLS_OPEN_FAILURE; + } + spin_unlock(&tx_info->lock); + + complete(&tx_info->completion); + return 0; +} + +/* + * chcr_ktls_cpl_set_tcb_rpl: TCB reply received from TP. + */ +static int chcr_ktls_cpl_set_tcb_rpl(struct adapter *adap, unsigned char *input) +{ + const struct cpl_set_tcb_rpl *p = (void *)input; + struct chcr_ktls_info *tx_info = NULL; + struct tid_info *t; + u32 tid; + + tid = GET_TID(p); + + t = &adap->tids; + tx_info = lookup_tid(t, tid); + + if (!tx_info || tx_info->tid != tid) { + pr_err("%s: incorrect tx_info or tid\n", __func__); + return -1; + } + + spin_lock(&tx_info->lock); + if (tx_info->pending_close) { + spin_unlock(&tx_info->lock); + kvfree(tx_info); + return 0; + } + tx_info->open_state = false; + spin_unlock(&tx_info->lock); + + complete(&tx_info->completion); + return 0; +} + +static void *__chcr_write_cpl_set_tcb_ulp(struct chcr_ktls_info *tx_info, + u32 tid, void *pos, u16 word, u64 mask, + u64 val, u32 reply) +{ + struct cpl_set_tcb_field_core *cpl; + struct ulptx_idata *idata; + struct ulp_txpkt *txpkt; + + /* ULP_TXPKT */ + txpkt = pos; + txpkt->cmd_dest = htonl(ULPTX_CMD_V(ULP_TX_PKT) | ULP_TXPKT_DEST_V(0)); + txpkt->len = htonl(DIV_ROUND_UP(CHCR_SET_TCB_FIELD_LEN, 16)); + + /* ULPTX_IDATA sub-command */ + idata = (struct ulptx_idata *)(txpkt + 1); + idata->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_IMM)); + idata->len = htonl(sizeof(*cpl)); + pos = idata + 1; + + cpl = pos; + /* CPL_SET_TCB_FIELD */ + OPCODE_TID(cpl) = htonl(MK_OPCODE_TID(CPL_SET_TCB_FIELD, tid)); + cpl->reply_ctrl = htons(QUEUENO_V(tx_info->rx_qid) | + NO_REPLY_V(!reply)); + cpl->word_cookie = htons(TCB_WORD_V(word)); + cpl->mask = cpu_to_be64(mask); + cpl->val = cpu_to_be64(val); + + /* ULPTX_NOOP */ + idata = (struct ulptx_idata *)(cpl + 1); + idata->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_NOOP)); + idata->len = htonl(0); + pos = idata + 1; + + return pos; +} + + +/* + * chcr_write_cpl_set_tcb_ulp: update tcb values. + * TCB is responsible to create tcp headers, so all the related values + * should be correctly updated. + * @tx_info - driver specific tls info. + * @q - tx queue on which packet is going out. + * @tid - TCB identifier. + * @pos - current index where should we start writing. + * @word - TCB word. + * @mask - TCB word related mask. + * @val - TCB word related value. + * @reply - set 1 if looking for TP response. + * return - next position to write. + */ +static void *chcr_write_cpl_set_tcb_ulp(struct chcr_ktls_info *tx_info, + struct sge_eth_txq *q, u32 tid, + void *pos, u16 word, u64 mask, + u64 val, u32 reply) +{ + int left = (void *)q->q.stat - pos; + + if (unlikely(left < CHCR_SET_TCB_FIELD_LEN)) { + if (!left) { + pos = q->q.desc; + } else { + u8 buf[48] = {0}; + + __chcr_write_cpl_set_tcb_ulp(tx_info, tid, buf, word, + mask, val, reply); + + return chcr_copy_to_txd(buf, &q->q, pos, + CHCR_SET_TCB_FIELD_LEN); + } + } + + pos = __chcr_write_cpl_set_tcb_ulp(tx_info, tid, pos, word, + mask, val, reply); + + /* check again if we are at the end of the queue */ + if (left == CHCR_SET_TCB_FIELD_LEN) + pos = q->q.desc; + + return pos; +} + +/* + * chcr_ktls_xmit_tcb_cpls: update tcb entry so that TP will create the header + * with updated values like tcp seq, ack, window etc. + * @tx_info - driver specific tls info. + * @q - TX queue. + * @tcp_seq + * @tcp_ack + * @tcp_win + * return: NETDEV_TX_BUSY/NET_TX_OK. + */ +static int chcr_ktls_xmit_tcb_cpls(struct chcr_ktls_info *tx_info, + struct sge_eth_txq *q, u64 tcp_seq, + u64 tcp_ack, u64 tcp_win) +{ + bool first_wr = ((tx_info->prev_ack == 0) && (tx_info->prev_win == 0)); + struct ch_ktls_port_stats_debug *port_stats; + u32 len, cpl = 0, ndesc, wr_len; + struct fw_ulptx_wr *wr; + int credits; + void *pos; + + wr_len = sizeof(*wr); + /* there can be max 4 cpls, check if we have enough credits */ + len = wr_len + 4 * roundup(CHCR_SET_TCB_FIELD_LEN, 16); + ndesc = DIV_ROUND_UP(len, 64); + + credits = chcr_txq_avail(&q->q) - ndesc; + if (unlikely(credits < 0)) { + chcr_eth_txq_stop(q); + return NETDEV_TX_BUSY; + } + + pos = &q->q.desc[q->q.pidx]; + /* make space for WR, we'll fill it later when we know all the cpls + * being sent out and have complete length. + */ + wr = pos; + pos += wr_len; + /* update tx_max if its a re-transmit or the first wr */ + if (first_wr || tcp_seq != tx_info->prev_seq) { + pos = chcr_write_cpl_set_tcb_ulp(tx_info, q, tx_info->tid, pos, + TCB_TX_MAX_W, + TCB_TX_MAX_V(TCB_TX_MAX_M), + TCB_TX_MAX_V(tcp_seq), 0); + cpl++; + } + /* reset snd una if it's a re-transmit pkt */ + if (tcp_seq != tx_info->prev_seq) { + /* reset snd_una */ + port_stats = + &tx_info->adap->ch_ktls_stats.ktls_port[tx_info->port_id]; + pos = chcr_write_cpl_set_tcb_ulp(tx_info, q, tx_info->tid, pos, + TCB_SND_UNA_RAW_W, + TCB_SND_UNA_RAW_V + (TCB_SND_UNA_RAW_M), + TCB_SND_UNA_RAW_V(0), 0); + atomic64_inc(&port_stats->ktls_tx_ooo); + cpl++; + } + /* update ack */ + if (first_wr || tx_info->prev_ack != tcp_ack) { + pos = chcr_write_cpl_set_tcb_ulp(tx_info, q, tx_info->tid, pos, + TCB_RCV_NXT_W, + TCB_RCV_NXT_V(TCB_RCV_NXT_M), + TCB_RCV_NXT_V(tcp_ack), 0); + tx_info->prev_ack = tcp_ack; + cpl++; + } + /* update receive window */ + if (first_wr || tx_info->prev_win != tcp_win) { + pos = chcr_write_cpl_set_tcb_ulp(tx_info, q, tx_info->tid, pos, + TCB_RCV_WND_W, + TCB_RCV_WND_V(TCB_RCV_WND_M), + TCB_RCV_WND_V(tcp_win), 0); + tx_info->prev_win = tcp_win; + cpl++; + } + + if (cpl) { + /* get the actual length */ + len = wr_len + cpl * roundup(CHCR_SET_TCB_FIELD_LEN, 16); + /* ULPTX wr */ + wr->op_to_compl = htonl(FW_WR_OP_V(FW_ULPTX_WR)); + wr->cookie = 0; + /* fill len in wr field */ + wr->flowid_len16 = htonl(FW_WR_LEN16_V(DIV_ROUND_UP(len, 16))); + + ndesc = DIV_ROUND_UP(len, 64); + chcr_txq_advance(&q->q, ndesc); + cxgb4_ring_tx_db(tx_info->adap, &q->q, ndesc); + } + return 0; +} + +/* + * chcr_ktls_skb_copy + * @nskb - new skb where the frags to be added. + * @skb - old skb from which frags will be copied. + */ +static void chcr_ktls_skb_copy(struct sk_buff *skb, struct sk_buff *nskb) +{ + int i; + + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { + skb_shinfo(nskb)->frags[i] = skb_shinfo(skb)->frags[i]; + __skb_frag_ref(&skb_shinfo(nskb)->frags[i]); + } + + skb_shinfo(nskb)->nr_frags = skb_shinfo(skb)->nr_frags; + nskb->len += skb->data_len; + nskb->data_len = skb->data_len; + nskb->truesize += skb->data_len; +} + +/* + * chcr_ktls_get_tx_flits + * returns number of flits to be sent out, it includes key context length, WR + * size and skb fragments. + */ +static unsigned int +chcr_ktls_get_tx_flits(const struct sk_buff *skb, unsigned int key_ctx_len) +{ + return chcr_sgl_len(skb_shinfo(skb)->nr_frags) + + DIV_ROUND_UP(key_ctx_len + CHCR_KTLS_WR_SIZE, 8); +} + +/* + * chcr_ktls_check_tcp_options: To check if there is any TCP option availbale + * other than timestamp. + * @skb - skb contains partial record.. + * return: 1 / 0 + */ +static int +chcr_ktls_check_tcp_options(struct tcphdr *tcp) +{ + int cnt, opt, optlen; + u_char *cp; + + cp = (u_char *)(tcp + 1); + cnt = (tcp->doff << 2) - sizeof(struct tcphdr); + for (; cnt > 0; cnt -= optlen, cp += optlen) { + opt = cp[0]; + if (opt == TCPOPT_EOL) + break; + if (opt == TCPOPT_NOP) { + optlen = 1; + } else { + if (cnt < 2) + break; + optlen = cp[1]; + if (optlen < 2 || optlen > cnt) + break; + } + switch (opt) { + case TCPOPT_NOP: + break; + default: + return 1; + } + } + return 0; +} + +/* + * chcr_ktls_write_tcp_options : TP can't send out all the options, we need to + * send out separately. + * @tx_info - driver specific tls info. + * @skb - skb contains partial record.. + * @q - TX queue. + * @tx_chan - channel number. + * return: NETDEV_TX_OK/NETDEV_TX_BUSY. + */ +static int +chcr_ktls_write_tcp_options(struct chcr_ktls_info *tx_info, struct sk_buff *skb, + struct sge_eth_txq *q, uint32_t tx_chan) +{ + struct fw_eth_tx_pkt_wr *wr; + struct cpl_tx_pkt_core *cpl; + u32 ctrl, iplen, maclen; +#if IS_ENABLED(CONFIG_IPV6) + struct ipv6hdr *ip6; +#endif + unsigned int ndesc; + struct tcphdr *tcp; + int len16, pktlen; + struct iphdr *ip; + int credits; + u8 buf[150]; + void *pos; + + iplen = skb_network_header_len(skb); + maclen = skb_mac_header_len(skb); + + /* packet length = eth hdr len + ip hdr len + tcp hdr len + * (including options). + */ + pktlen = skb->len - skb->data_len; + + ctrl = sizeof(*cpl) + pktlen; + len16 = DIV_ROUND_UP(sizeof(*wr) + ctrl, 16); + /* check how many descriptors needed */ + ndesc = DIV_ROUND_UP(len16, 4); + + credits = chcr_txq_avail(&q->q) - ndesc; + if (unlikely(credits < 0)) { + chcr_eth_txq_stop(q); + return NETDEV_TX_BUSY; + } + + pos = &q->q.desc[q->q.pidx]; + wr = pos; + + /* Firmware work request header */ + wr->op_immdlen = htonl(FW_WR_OP_V(FW_ETH_TX_PKT_WR) | + FW_WR_IMMDLEN_V(ctrl)); + + wr->equiq_to_len16 = htonl(FW_WR_LEN16_V(len16)); + wr->r3 = 0; + + cpl = (void *)(wr + 1); + + /* CPL header */ + cpl->ctrl0 = htonl(TXPKT_OPCODE_V(CPL_TX_PKT) | TXPKT_INTF_V(tx_chan) | + TXPKT_PF_V(tx_info->adap->pf)); + cpl->pack = 0; + cpl->len = htons(pktlen); + /* checksum offload */ + cpl->ctrl1 = 0; + + pos = cpl + 1; + + memcpy(buf, skb->data, pktlen); + if (tx_info->ip_family == AF_INET) { + /* we need to correct ip header len */ + ip = (struct iphdr *)(buf + maclen); + ip->tot_len = htons(pktlen - maclen); +#if IS_ENABLED(CONFIG_IPV6) + } else { + ip6 = (struct ipv6hdr *)(buf + maclen); + ip6->payload_len = htons(pktlen - maclen - iplen); +#endif + } + /* now take care of the tcp header, if fin is not set then clear push + * bit as well, and if fin is set, it will be sent at the last so we + * need to update the tcp sequence number as per the last packet. + */ + tcp = (struct tcphdr *)(buf + maclen + iplen); + + if (!tcp->fin) + tcp->psh = 0; + else + tcp->seq = htonl(tx_info->prev_seq); + + chcr_copy_to_txd(buf, &q->q, pos, pktlen); + + chcr_txq_advance(&q->q, ndesc); + cxgb4_ring_tx_db(tx_info->adap, &q->q, ndesc); + return 0; +} + +/* chcr_ktls_skb_shift - Shifts request length paged data from skb to another. + * @tgt- buffer into which tail data gets added + * @skb- buffer from which the paged data comes from + * @shiftlen- shift up to this many bytes + */ +static int chcr_ktls_skb_shift(struct sk_buff *tgt, struct sk_buff *skb, + int shiftlen) +{ + skb_frag_t *fragfrom, *fragto; + int from, to, todo; + + WARN_ON(shiftlen > skb->data_len); + + todo = shiftlen; + from = 0; + to = 0; + fragfrom = &skb_shinfo(skb)->frags[from]; + + while ((todo > 0) && (from < skb_shinfo(skb)->nr_frags)) { + fragfrom = &skb_shinfo(skb)->frags[from]; + fragto = &skb_shinfo(tgt)->frags[to]; + + if (todo >= skb_frag_size(fragfrom)) { + *fragto = *fragfrom; + todo -= skb_frag_size(fragfrom); + from++; + to++; + + } else { + __skb_frag_ref(fragfrom); + skb_frag_page_copy(fragto, fragfrom); + skb_frag_off_copy(fragto, fragfrom); + skb_frag_size_set(fragto, todo); + + skb_frag_off_add(fragfrom, todo); + skb_frag_size_sub(fragfrom, todo); + todo = 0; + + to++; + break; + } + } + + /* Ready to "commit" this state change to tgt */ + skb_shinfo(tgt)->nr_frags = to; + + /* Reposition in the original skb */ + to = 0; + while (from < skb_shinfo(skb)->nr_frags) + skb_shinfo(skb)->frags[to++] = skb_shinfo(skb)->frags[from++]; + + skb_shinfo(skb)->nr_frags = to; + + WARN_ON(todo > 0 && !skb_shinfo(skb)->nr_frags); + + skb->len -= shiftlen; + skb->data_len -= shiftlen; + skb->truesize -= shiftlen; + tgt->len += shiftlen; + tgt->data_len += shiftlen; + tgt->truesize += shiftlen; + + return shiftlen; +} + +/* + * chcr_ktls_xmit_wr_complete: This sends out the complete record. If an skb + * received has partial end part of the record, send out the complete record, so + * that crypto block will be able to generate TAG/HASH. + * @skb - segment which has complete or partial end part. + * @tx_info - driver specific tls info. + * @q - TX queue. + * @tcp_seq + * @tcp_push - tcp push bit. + * @mss - segment size. + * return: NETDEV_TX_BUSY/NET_TX_OK. + */ +static int chcr_ktls_xmit_wr_complete(struct sk_buff *skb, + struct chcr_ktls_info *tx_info, + struct sge_eth_txq *q, u32 tcp_seq, + bool tcp_push, u32 mss) +{ + u32 len16, wr_mid = 0, flits = 0, ndesc, cipher_start; + struct adapter *adap = tx_info->adap; + int credits, left, last_desc; + struct tx_sw_desc *sgl_sdesc; + struct cpl_tx_data *tx_data; + struct cpl_tx_sec_pdu *cpl; + struct ulptx_idata *idata; + struct ulp_txpkt *ulptx; + struct fw_ulptx_wr *wr; + void *pos; + u64 *end; + + /* get the number of flits required */ + flits = chcr_ktls_get_tx_flits(skb, tx_info->key_ctx_len); + /* number of descriptors */ + ndesc = chcr_flits_to_desc(flits); + /* check if enough credits available */ + credits = chcr_txq_avail(&q->q) - ndesc; + if (unlikely(credits < 0)) { + chcr_eth_txq_stop(q); + return NETDEV_TX_BUSY; + } + + if (unlikely(credits < ETHTXQ_STOP_THRES)) { + /* Credits are below the threshold vaues, stop the queue after + * injecting the Work Request for this packet. + */ + chcr_eth_txq_stop(q); + wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F; + } + + last_desc = q->q.pidx + ndesc - 1; + if (last_desc >= q->q.size) + last_desc -= q->q.size; + sgl_sdesc = &q->q.sdesc[last_desc]; + + if (unlikely(cxgb4_map_skb(adap->pdev_dev, skb, sgl_sdesc->addr) < 0)) { + memset(sgl_sdesc->addr, 0, sizeof(sgl_sdesc->addr)); + q->mapping_err++; + return NETDEV_TX_BUSY; + } + + pos = &q->q.desc[q->q.pidx]; + end = (u64 *)pos + flits; + /* FW_ULPTX_WR */ + wr = pos; + /* WR will need len16 */ + len16 = DIV_ROUND_UP(flits, 2); + wr->op_to_compl = htonl(FW_WR_OP_V(FW_ULPTX_WR)); + wr->flowid_len16 = htonl(wr_mid | FW_WR_LEN16_V(len16)); + wr->cookie = 0; + pos += sizeof(*wr); + /* ULP_TXPKT */ + ulptx = pos; + ulptx->cmd_dest = htonl(ULPTX_CMD_V(ULP_TX_PKT) | + ULP_TXPKT_CHANNELID_V(tx_info->port_id) | + ULP_TXPKT_FID_V(q->q.cntxt_id) | + ULP_TXPKT_RO_F); + ulptx->len = htonl(len16 - 1); + /* ULPTX_IDATA sub-command */ + idata = (struct ulptx_idata *)(ulptx + 1); + idata->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_IMM) | ULP_TX_SC_MORE_F); + /* idata length will include cpl_tx_sec_pdu + key context size + + * cpl_tx_data header. + */ + idata->len = htonl(sizeof(*cpl) + tx_info->key_ctx_len + + sizeof(*tx_data)); + /* SEC CPL */ + cpl = (struct cpl_tx_sec_pdu *)(idata + 1); + cpl->op_ivinsrtofst = + htonl(CPL_TX_SEC_PDU_OPCODE_V(CPL_TX_SEC_PDU) | + CPL_TX_SEC_PDU_CPLLEN_V(CHCR_CPL_TX_SEC_PDU_LEN_64BIT) | + CPL_TX_SEC_PDU_PLACEHOLDER_V(1) | + CPL_TX_SEC_PDU_IVINSRTOFST_V(TLS_HEADER_SIZE + 1)); + cpl->pldlen = htonl(skb->data_len); + + /* encryption should start after tls header size + iv size */ + cipher_start = TLS_HEADER_SIZE + tx_info->iv_size + 1; + + cpl->aadstart_cipherstop_hi = + htonl(CPL_TX_SEC_PDU_AADSTART_V(1) | + CPL_TX_SEC_PDU_AADSTOP_V(TLS_HEADER_SIZE) | + CPL_TX_SEC_PDU_CIPHERSTART_V(cipher_start)); + + /* authentication will also start after tls header + iv size */ + cpl->cipherstop_lo_authinsert = + htonl(CPL_TX_SEC_PDU_AUTHSTART_V(cipher_start) | + CPL_TX_SEC_PDU_AUTHSTOP_V(TLS_CIPHER_AES_GCM_128_TAG_SIZE) | + CPL_TX_SEC_PDU_AUTHINSERT_V(TLS_CIPHER_AES_GCM_128_TAG_SIZE)); + + /* These two flits are actually a CPL_TLS_TX_SCMD_FMT. */ + cpl->seqno_numivs = htonl(tx_info->scmd0_seqno_numivs); + cpl->ivgen_hdrlen = htonl(tx_info->scmd0_ivgen_hdrlen); + cpl->scmd1 = cpu_to_be64(tx_info->record_no); + + pos = cpl + 1; + /* check if space left to fill the keys */ + left = (void *)q->q.stat - pos; + if (!left) { + left = (void *)end - (void *)q->q.stat; + pos = q->q.desc; + end = pos + left; + } + + pos = chcr_copy_to_txd(&tx_info->key_ctx, &q->q, pos, + tx_info->key_ctx_len); + left = (void *)q->q.stat - pos; + + if (!left) { + left = (void *)end - (void *)q->q.stat; + pos = q->q.desc; + end = pos + left; + } + /* CPL_TX_DATA */ + tx_data = (void *)pos; + OPCODE_TID(tx_data) = htonl(MK_OPCODE_TID(CPL_TX_DATA, tx_info->tid)); + tx_data->len = htonl(TX_DATA_MSS_V(mss) | TX_LENGTH_V(skb->data_len)); + + tx_data->rsvd = htonl(tcp_seq); + + tx_data->flags = htonl(TX_BYPASS_F); + if (tcp_push) + tx_data->flags |= htonl(TX_PUSH_F | TX_SHOVE_F); + + /* check left again, it might go beyond queue limit */ + pos = tx_data + 1; + left = (void *)q->q.stat - pos; + + /* check the position again */ + if (!left) { + left = (void *)end - (void *)q->q.stat; + pos = q->q.desc; + end = pos + left; + } + + /* send the complete packet except the header */ + cxgb4_write_sgl(skb, &q->q, pos, end, skb->len - skb->data_len, + sgl_sdesc->addr); + sgl_sdesc->skb = skb; + + chcr_txq_advance(&q->q, ndesc); + cxgb4_ring_tx_db(adap, &q->q, ndesc); + atomic64_inc(&adap->ch_ktls_stats.ktls_tx_send_records); + + return 0; +} + +/* + * chcr_ktls_xmit_wr_short: This is to send out partial records. If its + * a middle part of a record, fetch the prior data to make it 16 byte aligned + * and then only send it out. + * + * @skb - skb contains partial record.. + * @tx_info - driver specific tls info. + * @q - TX queue. + * @tcp_seq + * @tcp_push - tcp push bit. + * @mss - segment size. + * @tls_rec_offset - offset from start of the tls record. + * @perior_data - data before the current segment, required to make this record + * 16 byte aligned. + * @prior_data_len - prior_data length (less than 16) + * return: NETDEV_TX_BUSY/NET_TX_OK. + */ +static int chcr_ktls_xmit_wr_short(struct sk_buff *skb, + struct chcr_ktls_info *tx_info, + struct sge_eth_txq *q, + u32 tcp_seq, bool tcp_push, u32 mss, + u32 tls_rec_offset, u8 *prior_data, + u32 prior_data_len) +{ + struct adapter *adap = tx_info->adap; + u32 len16, wr_mid = 0, cipher_start; + unsigned int flits = 0, ndesc; + int credits, left, last_desc; + struct tx_sw_desc *sgl_sdesc; + struct cpl_tx_data *tx_data; + struct cpl_tx_sec_pdu *cpl; + struct ulptx_idata *idata; + struct ulp_txpkt *ulptx; + struct fw_ulptx_wr *wr; + __be64 iv_record; + void *pos; + u64 *end; + + /* get the number of flits required, it's a partial record so 2 flits + * (AES_BLOCK_SIZE) will be added. + */ + flits = chcr_ktls_get_tx_flits(skb, tx_info->key_ctx_len) + 2; + /* get the correct 8 byte IV of this record */ + iv_record = cpu_to_be64(tx_info->iv + tx_info->record_no); + /* If it's a middle record and not 16 byte aligned to run AES CTR, need + * to make it 16 byte aligned. So atleadt 2 extra flits of immediate + * data will be added. + */ + if (prior_data_len) + flits += 2; + /* number of descriptors */ + ndesc = chcr_flits_to_desc(flits); + /* check if enough credits available */ + credits = chcr_txq_avail(&q->q) - ndesc; + if (unlikely(credits < 0)) { + chcr_eth_txq_stop(q); + return NETDEV_TX_BUSY; + } + + if (unlikely(credits < ETHTXQ_STOP_THRES)) { + chcr_eth_txq_stop(q); + wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F; + } + + last_desc = q->q.pidx + ndesc - 1; + if (last_desc >= q->q.size) + last_desc -= q->q.size; + sgl_sdesc = &q->q.sdesc[last_desc]; + + if (unlikely(cxgb4_map_skb(adap->pdev_dev, skb, sgl_sdesc->addr) < 0)) { + memset(sgl_sdesc->addr, 0, sizeof(sgl_sdesc->addr)); + q->mapping_err++; + return NETDEV_TX_BUSY; + } + + pos = &q->q.desc[q->q.pidx]; + end = (u64 *)pos + flits; + /* FW_ULPTX_WR */ + wr = pos; + /* WR will need len16 */ + len16 = DIV_ROUND_UP(flits, 2); + wr->op_to_compl = htonl(FW_WR_OP_V(FW_ULPTX_WR)); + wr->flowid_len16 = htonl(wr_mid | FW_WR_LEN16_V(len16)); + wr->cookie = 0; + pos += sizeof(*wr); + /* ULP_TXPKT */ + ulptx = pos; + ulptx->cmd_dest = htonl(ULPTX_CMD_V(ULP_TX_PKT) | + ULP_TXPKT_CHANNELID_V(tx_info->port_id) | + ULP_TXPKT_FID_V(q->q.cntxt_id) | + ULP_TXPKT_RO_F); + ulptx->len = htonl(len16 - 1); + /* ULPTX_IDATA sub-command */ + idata = (struct ulptx_idata *)(ulptx + 1); + idata->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_IMM) | ULP_TX_SC_MORE_F); + /* idata length will include cpl_tx_sec_pdu + key context size + + * cpl_tx_data header. + */ + idata->len = htonl(sizeof(*cpl) + tx_info->key_ctx_len + + sizeof(*tx_data) + AES_BLOCK_LEN + prior_data_len); + /* SEC CPL */ + cpl = (struct cpl_tx_sec_pdu *)(idata + 1); + /* cipher start will have tls header + iv size extra if its a header + * part of tls record. else only 16 byte IV will be added. + */ + cipher_start = + AES_BLOCK_LEN + 1 + + (!tls_rec_offset ? TLS_HEADER_SIZE + tx_info->iv_size : 0); + + cpl->op_ivinsrtofst = + htonl(CPL_TX_SEC_PDU_OPCODE_V(CPL_TX_SEC_PDU) | + CPL_TX_SEC_PDU_CPLLEN_V(CHCR_CPL_TX_SEC_PDU_LEN_64BIT) | + CPL_TX_SEC_PDU_IVINSRTOFST_V(1)); + cpl->pldlen = htonl(skb->data_len + AES_BLOCK_LEN + prior_data_len); + cpl->aadstart_cipherstop_hi = + htonl(CPL_TX_SEC_PDU_CIPHERSTART_V(cipher_start)); + cpl->cipherstop_lo_authinsert = 0; + /* These two flits are actually a CPL_TLS_TX_SCMD_FMT. */ + cpl->seqno_numivs = htonl(tx_info->scmd0_short_seqno_numivs); + cpl->ivgen_hdrlen = htonl(tx_info->scmd0_short_ivgen_hdrlen); + cpl->scmd1 = 0; + + pos = cpl + 1; + /* check if space left to fill the keys */ + left = (void *)q->q.stat - pos; + if (!left) { + left = (void *)end - (void *)q->q.stat; + pos = q->q.desc; + end = pos + left; + } + + pos = chcr_copy_to_txd(&tx_info->key_ctx, &q->q, pos, + tx_info->key_ctx_len); + left = (void *)q->q.stat - pos; + + if (!left) { + left = (void *)end - (void *)q->q.stat; + pos = q->q.desc; + end = pos + left; + } + /* CPL_TX_DATA */ + tx_data = (void *)pos; + OPCODE_TID(tx_data) = htonl(MK_OPCODE_TID(CPL_TX_DATA, tx_info->tid)); + tx_data->len = htonl(TX_DATA_MSS_V(mss) | + TX_LENGTH_V(skb->data_len + prior_data_len)); + tx_data->rsvd = htonl(tcp_seq); + tx_data->flags = htonl(TX_BYPASS_F); + if (tcp_push) + tx_data->flags |= htonl(TX_PUSH_F | TX_SHOVE_F); + + /* check left again, it might go beyond queue limit */ + pos = tx_data + 1; + left = (void *)q->q.stat - pos; + + /* check the position again */ + if (!left) { + left = (void *)end - (void *)q->q.stat; + pos = q->q.desc; + end = pos + left; + } + /* copy the 16 byte IV for AES-CTR, which includes 4 bytes of salt, 8 + * bytes of actual IV and 4 bytes of 16 byte-sequence. + */ + memcpy(pos, tx_info->key_ctx.salt, tx_info->salt_size); + memcpy(pos + tx_info->salt_size, &iv_record, tx_info->iv_size); + *(__be32 *)(pos + tx_info->salt_size + tx_info->iv_size) = + htonl(2 + (tls_rec_offset ? ((tls_rec_offset - + (TLS_HEADER_SIZE + tx_info->iv_size)) / AES_BLOCK_LEN) : 0)); + + pos += 16; + /* Prior_data_len will always be less than 16 bytes, fill the + * prio_data_len after AES_CTRL_BLOCK and clear the remaining length + * to 0. + */ + if (prior_data_len) + pos = chcr_copy_to_txd(prior_data, &q->q, pos, 16); + /* send the complete packet except the header */ + cxgb4_write_sgl(skb, &q->q, pos, end, skb->len - skb->data_len, + sgl_sdesc->addr); + sgl_sdesc->skb = skb; + + chcr_txq_advance(&q->q, ndesc); + cxgb4_ring_tx_db(adap, &q->q, ndesc); + + return 0; +} + +/* + * chcr_ktls_tx_plaintxt: This handler will take care of the records which has + * only plain text (only tls header and iv) + * @tx_info - driver specific tls info. + * @skb - skb contains partial record.. + * @tcp_seq + * @mss - segment size. + * @tcp_push - tcp push bit. + * @q - TX queue. + * @port_id : port number + * @perior_data - data before the current segment, required to make this record + * 16 byte aligned. + * @prior_data_len - prior_data length (less than 16) + * return: NETDEV_TX_BUSY/NET_TX_OK. + */ +static int chcr_ktls_tx_plaintxt(struct chcr_ktls_info *tx_info, + struct sk_buff *skb, u32 tcp_seq, u32 mss, + bool tcp_push, struct sge_eth_txq *q, + u32 port_id, u8 *prior_data, + u32 prior_data_len) +{ + int credits, left, len16, last_desc; + unsigned int flits = 0, ndesc; + struct tx_sw_desc *sgl_sdesc; + struct cpl_tx_data *tx_data; + struct ulptx_idata *idata; + struct ulp_txpkt *ulptx; + struct fw_ulptx_wr *wr; + u32 wr_mid = 0; + void *pos; + u64 *end; + + flits = DIV_ROUND_UP(CHCR_PLAIN_TX_DATA_LEN, 8); + flits += chcr_sgl_len(skb_shinfo(skb)->nr_frags); + if (prior_data_len) + flits += 2; + /* WR will need len16 */ + len16 = DIV_ROUND_UP(flits, 2); + /* check how many descriptors needed */ + ndesc = DIV_ROUND_UP(flits, 8); + + credits = chcr_txq_avail(&q->q) - ndesc; + if (unlikely(credits < 0)) { + chcr_eth_txq_stop(q); + return NETDEV_TX_BUSY; + } + + if (unlikely(credits < ETHTXQ_STOP_THRES)) { + chcr_eth_txq_stop(q); + wr_mid |= FW_WR_EQUEQ_F | FW_WR_EQUIQ_F; + } + + last_desc = q->q.pidx + ndesc - 1; + if (last_desc >= q->q.size) + last_desc -= q->q.size; + sgl_sdesc = &q->q.sdesc[last_desc]; + + if (unlikely(cxgb4_map_skb(tx_info->adap->pdev_dev, skb, + sgl_sdesc->addr) < 0)) { + memset(sgl_sdesc->addr, 0, sizeof(sgl_sdesc->addr)); + q->mapping_err++; + return NETDEV_TX_BUSY; + } + + pos = &q->q.desc[q->q.pidx]; + end = (u64 *)pos + flits; + /* FW_ULPTX_WR */ + wr = pos; + wr->op_to_compl = htonl(FW_WR_OP_V(FW_ULPTX_WR)); + wr->flowid_len16 = htonl(wr_mid | FW_WR_LEN16_V(len16)); + wr->cookie = 0; + pos += sizeof(*wr); + /* ULP_TXPKT */ + ulptx = (struct ulp_txpkt *)(wr + 1); + ulptx->cmd_dest = htonl(ULPTX_CMD_V(ULP_TX_PKT) | + ULP_TXPKT_DATAMODIFY_V(0) | + ULP_TXPKT_CHANNELID_V(tx_info->port_id) | + ULP_TXPKT_DEST_V(0) | + ULP_TXPKT_FID_V(q->q.cntxt_id) | ULP_TXPKT_RO_V(1)); + ulptx->len = htonl(len16 - 1); + /* ULPTX_IDATA sub-command */ + idata = (struct ulptx_idata *)(ulptx + 1); + idata->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_IMM) | ULP_TX_SC_MORE_F); + idata->len = htonl(sizeof(*tx_data) + prior_data_len); + /* CPL_TX_DATA */ + tx_data = (struct cpl_tx_data *)(idata + 1); + OPCODE_TID(tx_data) = htonl(MK_OPCODE_TID(CPL_TX_DATA, tx_info->tid)); + tx_data->len = htonl(TX_DATA_MSS_V(mss) | + TX_LENGTH_V(skb->data_len + prior_data_len)); + /* set tcp seq number */ + tx_data->rsvd = htonl(tcp_seq); + tx_data->flags = htonl(TX_BYPASS_F); + if (tcp_push) + tx_data->flags |= htonl(TX_PUSH_F | TX_SHOVE_F); + + pos = tx_data + 1; + /* apart from prior_data_len, we should set remaining part of 16 bytes + * to be zero. + */ + if (prior_data_len) + pos = chcr_copy_to_txd(prior_data, &q->q, pos, 16); + + /* check left again, it might go beyond queue limit */ + left = (void *)q->q.stat - pos; + + /* check the position again */ + if (!left) { + left = (void *)end - (void *)q->q.stat; + pos = q->q.desc; + end = pos + left; + } + /* send the complete packet including the header */ + cxgb4_write_sgl(skb, &q->q, pos, end, skb->len - skb->data_len, + sgl_sdesc->addr); + sgl_sdesc->skb = skb; + + chcr_txq_advance(&q->q, ndesc); + cxgb4_ring_tx_db(tx_info->adap, &q->q, ndesc); + return 0; +} + +/* + * chcr_ktls_copy_record_in_skb + * @nskb - new skb where the frags to be added. + * @record - specific record which has complete 16k record in frags. + */ +static void chcr_ktls_copy_record_in_skb(struct sk_buff *nskb, + struct tls_record_info *record) +{ + int i = 0; + + for (i = 0; i < record->num_frags; i++) { + skb_shinfo(nskb)->frags[i] = record->frags[i]; + /* increase the frag ref count */ + __skb_frag_ref(&skb_shinfo(nskb)->frags[i]); + } + + skb_shinfo(nskb)->nr_frags = record->num_frags; + nskb->data_len = record->len; + nskb->len += record->len; + nskb->truesize += record->len; +} + +/* + * chcr_ktls_update_snd_una: Reset the SEND_UNA. It will be done to avoid + * sending the same segment again. It will discard the segment which is before + * the current tx max. + * @tx_info - driver specific tls info. + * @q - TX queue. + * return: NET_TX_OK/NET_XMIT_DROP. + */ +static int chcr_ktls_update_snd_una(struct chcr_ktls_info *tx_info, + struct sge_eth_txq *q) +{ + struct fw_ulptx_wr *wr; + unsigned int ndesc; + int credits; + void *pos; + u32 len; + + len = sizeof(*wr) + roundup(CHCR_SET_TCB_FIELD_LEN, 16); + ndesc = DIV_ROUND_UP(len, 64); + + credits = chcr_txq_avail(&q->q) - ndesc; + if (unlikely(credits < 0)) { + chcr_eth_txq_stop(q); + return NETDEV_TX_BUSY; + } + + pos = &q->q.desc[q->q.pidx]; + + wr = pos; + /* ULPTX wr */ + wr->op_to_compl = htonl(FW_WR_OP_V(FW_ULPTX_WR)); + wr->cookie = 0; + /* fill len in wr field */ + wr->flowid_len16 = htonl(FW_WR_LEN16_V(DIV_ROUND_UP(len, 16))); + + pos += sizeof(*wr); + + pos = chcr_write_cpl_set_tcb_ulp(tx_info, q, tx_info->tid, pos, + TCB_SND_UNA_RAW_W, + TCB_SND_UNA_RAW_V(TCB_SND_UNA_RAW_M), + TCB_SND_UNA_RAW_V(0), 0); + + chcr_txq_advance(&q->q, ndesc); + cxgb4_ring_tx_db(tx_info->adap, &q->q, ndesc); + + return 0; +} + +/* + * chcr_end_part_handler: This handler will handle the record which + * is complete or if record's end part is received. T6 adapter has a issue that + * it can't send out TAG with partial record so if its an end part then we have + * to send TAG as well and for which we need to fetch the complete record and + * send it to crypto module. + * @tx_info - driver specific tls info. + * @skb - skb contains partial record. + * @record - complete record of 16K size. + * @tcp_seq + * @mss - segment size in which TP needs to chop a packet. + * @tcp_push_no_fin - tcp push if fin is not set. + * @q - TX queue. + * @tls_end_offset - offset from end of the record. + * @last wr : check if this is the last part of the skb going out. + * return: NETDEV_TX_OK/NETDEV_TX_BUSY. + */ +static int chcr_end_part_handler(struct chcr_ktls_info *tx_info, + struct sk_buff *skb, + struct tls_record_info *record, + u32 tcp_seq, int mss, bool tcp_push_no_fin, + struct sge_eth_txq *q, + u32 tls_end_offset, bool last_wr) +{ + struct sk_buff *nskb = NULL; + /* check if it is a complete record */ + if (tls_end_offset == record->len) { + nskb = skb; + atomic64_inc(&tx_info->adap->ch_ktls_stats.ktls_tx_complete_pkts); + } else { + dev_kfree_skb_any(skb); + + nskb = alloc_skb(0, GFP_KERNEL); + if (!nskb) + return NETDEV_TX_BUSY; + /* copy complete record in skb */ + chcr_ktls_copy_record_in_skb(nskb, record); + /* packet is being sent from the beginning, update the tcp_seq + * accordingly. + */ + tcp_seq = tls_record_start_seq(record); + /* reset snd una, so the middle record won't send the already + * sent part. + */ + if (chcr_ktls_update_snd_una(tx_info, q)) + goto out; + atomic64_inc(&tx_info->adap->ch_ktls_stats.ktls_tx_end_pkts); + } + + if (chcr_ktls_xmit_wr_complete(nskb, tx_info, q, tcp_seq, + (last_wr && tcp_push_no_fin), + mss)) { + goto out; + } + return 0; +out: + dev_kfree_skb_any(nskb); + return NETDEV_TX_BUSY; +} + +/* + * chcr_short_record_handler: This handler will take care of the records which + * doesn't have end part (1st part or the middle part(/s) of a record). In such + * cases, AES CTR will be used in place of AES GCM to send out partial packet. + * This partial record might be the first part of the record, or the middle + * part. In case of middle record we should fetch the prior data to make it 16 + * byte aligned. If it has a partial tls header or iv then get to the start of + * tls header. And if it has partial TAG, then remove the complete TAG and send + * only the payload. + * There is one more possibility that it gets a partial header, send that + * portion as a plaintext. + * @tx_info - driver specific tls info. + * @skb - skb contains partial record.. + * @record - complete record of 16K size. + * @tcp_seq + * @mss - segment size in which TP needs to chop a packet. + * @tcp_push_no_fin - tcp push if fin is not set. + * @q - TX queue. + * @tls_end_offset - offset from end of the record. + * return: NETDEV_TX_OK/NETDEV_TX_BUSY. + */ +static int chcr_short_record_handler(struct chcr_ktls_info *tx_info, + struct sk_buff *skb, + struct tls_record_info *record, + u32 tcp_seq, int mss, bool tcp_push_no_fin, + struct sge_eth_txq *q, u32 tls_end_offset) +{ + u32 tls_rec_offset = tcp_seq - tls_record_start_seq(record); + u8 prior_data[16] = {0}; + u32 prior_data_len = 0; + u32 data_len; + + /* check if the skb is ending in middle of tag/HASH, its a big + * trouble, send the packet before the HASH. + */ + int remaining_record = tls_end_offset - skb->data_len; + + if (remaining_record > 0 && + remaining_record < TLS_CIPHER_AES_GCM_128_TAG_SIZE) { + int trimmed_len = skb->data_len - + (TLS_CIPHER_AES_GCM_128_TAG_SIZE - remaining_record); + struct sk_buff *tmp_skb = NULL; + /* don't process the pkt if it is only a partial tag */ + if (skb->data_len < TLS_CIPHER_AES_GCM_128_TAG_SIZE) + goto out; + + WARN_ON(trimmed_len > skb->data_len); + + /* shift to those many bytes */ + tmp_skb = alloc_skb(0, GFP_KERNEL); + if (unlikely(!tmp_skb)) + goto out; + + chcr_ktls_skb_shift(tmp_skb, skb, trimmed_len); + /* free the last trimmed portion */ + dev_kfree_skb_any(skb); + skb = tmp_skb; + atomic64_inc(&tx_info->adap->ch_ktls_stats.ktls_tx_trimmed_pkts); + } + data_len = skb->data_len; + /* check if the middle record's start point is 16 byte aligned. CTR + * needs 16 byte aligned start point to start encryption. + */ + if (tls_rec_offset) { + /* there is an offset from start, means its a middle record */ + int remaining = 0; + + if (tls_rec_offset < (TLS_HEADER_SIZE + tx_info->iv_size)) { + prior_data_len = tls_rec_offset; + tls_rec_offset = 0; + remaining = 0; + } else { + prior_data_len = + (tls_rec_offset - + (TLS_HEADER_SIZE + tx_info->iv_size)) + % AES_BLOCK_LEN; + remaining = tls_rec_offset - prior_data_len; + } + + /* if prior_data_len is not zero, means we need to fetch prior + * data to make this record 16 byte aligned, or we need to reach + * to start offset. + */ + if (prior_data_len) { + int i = 0; + u8 *data = NULL; + skb_frag_t *f; + u8 *vaddr; + int frag_size = 0, frag_delta = 0; + + while (remaining > 0) { + frag_size = skb_frag_size(&record->frags[i]); + if (remaining < frag_size) + break; + + remaining -= frag_size; + i++; + } + f = &record->frags[i]; + vaddr = kmap_atomic(skb_frag_page(f)); + + data = vaddr + skb_frag_off(f) + remaining; + frag_delta = skb_frag_size(f) - remaining; + + if (frag_delta >= prior_data_len) { + memcpy(prior_data, data, prior_data_len); + kunmap_atomic(vaddr); + } else { + memcpy(prior_data, data, frag_delta); + kunmap_atomic(vaddr); + /* get the next page */ + f = &record->frags[i + 1]; + vaddr = kmap_atomic(skb_frag_page(f)); + data = vaddr + skb_frag_off(f); + memcpy(prior_data + frag_delta, + data, (prior_data_len - frag_delta)); + kunmap_atomic(vaddr); + } + /* reset tcp_seq as per the prior_data_required len */ + tcp_seq -= prior_data_len; + /* include prio_data_len for further calculation. + */ + data_len += prior_data_len; + } + /* reset snd una, so the middle record won't send the already + * sent part. + */ + if (chcr_ktls_update_snd_una(tx_info, q)) + goto out; + atomic64_inc(&tx_info->adap->ch_ktls_stats.ktls_tx_middle_pkts); + } else { + /* Else means, its a partial first part of the record. Check if + * its only the header, don't need to send for encryption then. + */ + if (data_len <= TLS_HEADER_SIZE + tx_info->iv_size) { + if (chcr_ktls_tx_plaintxt(tx_info, skb, tcp_seq, mss, + tcp_push_no_fin, q, + tx_info->port_id, + prior_data, + prior_data_len)) { + goto out; + } + return 0; + } + atomic64_inc(&tx_info->adap->ch_ktls_stats.ktls_tx_start_pkts); + } + + if (chcr_ktls_xmit_wr_short(skb, tx_info, q, tcp_seq, tcp_push_no_fin, + mss, tls_rec_offset, prior_data, + prior_data_len)) { + goto out; + } + + return 0; +out: + dev_kfree_skb_any(skb); + return NETDEV_TX_BUSY; +} + +/* nic tls TX handler */ +static int chcr_ktls_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct ch_ktls_port_stats_debug *port_stats; + struct chcr_ktls_ofld_ctx_tx *tx_ctx; + struct ch_ktls_stats_debug *stats; + struct tcphdr *th = tcp_hdr(skb); + int data_len, qidx, ret = 0, mss; + struct tls_record_info *record; + struct chcr_ktls_info *tx_info; + u32 tls_end_offset, tcp_seq; + struct tls_context *tls_ctx; + struct sk_buff *local_skb; + struct sge_eth_txq *q; + struct adapter *adap; + unsigned long flags; + + tcp_seq = ntohl(th->seq); + + mss = skb_is_gso(skb) ? skb_shinfo(skb)->gso_size : skb->data_len; + + /* check if we haven't set it for ktls offload */ + if (!skb->sk || !tls_is_sk_tx_device_offloaded(skb->sk)) + goto out; + + tls_ctx = tls_get_ctx(skb->sk); + if (unlikely(tls_ctx->netdev != dev)) + goto out; + + tx_ctx = chcr_get_ktls_tx_context(tls_ctx); + tx_info = tx_ctx->chcr_info; + + if (unlikely(!tx_info)) + goto out; + + /* don't touch the original skb, make a new skb to extract each records + * and send them separately. + */ + local_skb = alloc_skb(0, GFP_KERNEL); + + if (unlikely(!local_skb)) + return NETDEV_TX_BUSY; + + adap = tx_info->adap; + stats = &adap->ch_ktls_stats; + port_stats = &stats->ktls_port[tx_info->port_id]; + + qidx = skb->queue_mapping; + q = &adap->sge.ethtxq[qidx + tx_info->first_qset]; + cxgb4_reclaim_completed_tx(adap, &q->q, true); + /* if tcp options are set but finish is not send the options first */ + if (!th->fin && chcr_ktls_check_tcp_options(th)) { + ret = chcr_ktls_write_tcp_options(tx_info, skb, q, + tx_info->tx_chan); + if (ret) + return NETDEV_TX_BUSY; + } + /* update tcb */ + ret = chcr_ktls_xmit_tcb_cpls(tx_info, q, ntohl(th->seq), + ntohl(th->ack_seq), + ntohs(th->window)); + if (ret) { + dev_kfree_skb_any(local_skb); + return NETDEV_TX_BUSY; + } + + /* copy skb contents into local skb */ + chcr_ktls_skb_copy(skb, local_skb); + + /* go through the skb and send only one record at a time. */ + data_len = skb->data_len; + /* TCP segments can be in received either complete or partial. + * chcr_end_part_handler will handle cases if complete record or end + * part of the record is received. Incase of partial end part of record, + * we will send the complete record again. + */ + + do { + int i; + + cxgb4_reclaim_completed_tx(adap, &q->q, true); + /* lock taken */ + spin_lock_irqsave(&tx_ctx->base.lock, flags); + /* fetch the tls record */ + record = tls_get_record(&tx_ctx->base, tcp_seq, + &tx_info->record_no); + /* By the time packet reached to us, ACK is received, and record + * won't be found in that case, handle it gracefully. + */ + if (unlikely(!record)) { + spin_unlock_irqrestore(&tx_ctx->base.lock, flags); + atomic64_inc(&port_stats->ktls_tx_drop_no_sync_data); + goto out; + } + + if (unlikely(tls_record_is_start_marker(record))) { + spin_unlock_irqrestore(&tx_ctx->base.lock, flags); + atomic64_inc(&port_stats->ktls_tx_skip_no_sync_data); + goto out; + } + + /* increase page reference count of the record, so that there + * won't be any chance of page free in middle if in case stack + * receives ACK and try to delete the record. + */ + for (i = 0; i < record->num_frags; i++) + __skb_frag_ref(&record->frags[i]); + /* lock cleared */ + spin_unlock_irqrestore(&tx_ctx->base.lock, flags); + + tls_end_offset = record->end_seq - tcp_seq; + + pr_debug("seq 0x%x, end_seq 0x%x prev_seq 0x%x, datalen 0x%x\n", + tcp_seq, record->end_seq, tx_info->prev_seq, data_len); + /* if a tls record is finishing in this SKB */ + if (tls_end_offset <= data_len) { + struct sk_buff *nskb = NULL; + + if (tls_end_offset < data_len) { + nskb = alloc_skb(0, GFP_KERNEL); + if (unlikely(!nskb)) { + ret = -ENOMEM; + goto clear_ref; + } + + chcr_ktls_skb_shift(nskb, local_skb, + tls_end_offset); + } else { + /* its the only record in this skb, directly + * point it. + */ + nskb = local_skb; + } + ret = chcr_end_part_handler(tx_info, nskb, record, + tcp_seq, mss, + (!th->fin && th->psh), q, + tls_end_offset, + (nskb == local_skb)); + + if (ret && nskb != local_skb) + dev_kfree_skb_any(local_skb); + + data_len -= tls_end_offset; + /* tcp_seq increment is required to handle next record. + */ + tcp_seq += tls_end_offset; + } else { + ret = chcr_short_record_handler(tx_info, local_skb, + record, tcp_seq, mss, + (!th->fin && th->psh), + q, tls_end_offset); + data_len = 0; + } +clear_ref: + /* clear the frag ref count which increased locally before */ + for (i = 0; i < record->num_frags; i++) { + /* clear the frag ref count */ + __skb_frag_unref(&record->frags[i]); + } + /* if any failure, come out from the loop. */ + if (ret) + goto out; + /* length should never be less than 0 */ + WARN_ON(data_len < 0); + + } while (data_len > 0); + + tx_info->prev_seq = ntohl(th->seq) + skb->data_len; + atomic64_inc(&port_stats->ktls_tx_encrypted_packets); + atomic64_add(skb->data_len, &port_stats->ktls_tx_encrypted_bytes); + + /* tcp finish is set, send a separate tcp msg including all the options + * as well. + */ + if (th->fin) + chcr_ktls_write_tcp_options(tx_info, skb, q, tx_info->tx_chan); + +out: + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; +} + +static void *chcr_ktls_uld_add(const struct cxgb4_lld_info *lldi) +{ + struct chcr_ktls_uld_ctx *u_ctx; + + pr_info_once("%s - version %s\n", CHCR_KTLS_DRV_DESC, + CHCR_KTLS_DRV_VERSION); + u_ctx = kzalloc(sizeof(*u_ctx), GFP_KERNEL); + if (!u_ctx) { + u_ctx = ERR_PTR(-ENOMEM); + goto out; + } + u_ctx->lldi = *lldi; +out: + return u_ctx; +} + +static const struct tlsdev_ops chcr_ktls_ops = { + .tls_dev_add = chcr_ktls_dev_add, + .tls_dev_del = chcr_ktls_dev_del, +}; + +static chcr_handler_func work_handlers[NUM_CPL_CMDS] = { + [CPL_ACT_OPEN_RPL] = chcr_ktls_cpl_act_open_rpl, + [CPL_SET_TCB_RPL] = chcr_ktls_cpl_set_tcb_rpl, +}; + +static int chcr_ktls_uld_rx_handler(void *handle, const __be64 *rsp, + const struct pkt_gl *pgl) +{ + const struct cpl_act_open_rpl *rpl = (struct cpl_act_open_rpl *)rsp; + struct chcr_ktls_uld_ctx *u_ctx = handle; + u8 opcode = rpl->ot.opcode; + struct adapter *adap; + + adap = pci_get_drvdata(u_ctx->lldi.pdev); + + if (!work_handlers[opcode]) { + pr_err("Unsupported opcode %d received\n", opcode); + return 0; + } + + work_handlers[opcode](adap, (unsigned char *)&rsp[1]); + return 0; +} + +static int chcr_ktls_uld_state_change(void *handle, enum cxgb4_state new_state) +{ + struct chcr_ktls_uld_ctx *u_ctx = handle; + + switch (new_state) { + case CXGB4_STATE_UP: + pr_info("%s: Up\n", pci_name(u_ctx->lldi.pdev)); + mutex_lock(&dev_mutex); + list_add_tail(&u_ctx->entry, &uld_ctx_list); + mutex_unlock(&dev_mutex); + break; + case CXGB4_STATE_START_RECOVERY: + case CXGB4_STATE_DOWN: + case CXGB4_STATE_DETACH: + pr_info("%s: Down\n", pci_name(u_ctx->lldi.pdev)); + mutex_lock(&dev_mutex); + list_del(&u_ctx->entry); + mutex_unlock(&dev_mutex); + break; + default: + break; + } + + return 0; +} + +static struct cxgb4_uld_info chcr_ktls_uld_info = { + .name = CHCR_KTLS_DRV_MODULE_NAME, + .nrxq = 1, + .rxq_size = 1024, + .add = chcr_ktls_uld_add, + .tx_handler = chcr_ktls_xmit, + .rx_handler = chcr_ktls_uld_rx_handler, + .state_change = chcr_ktls_uld_state_change, + .tlsdev_ops = &chcr_ktls_ops, +}; + +static int __init chcr_ktls_init(void) +{ + cxgb4_register_uld(CXGB4_ULD_KTLS, &chcr_ktls_uld_info); + return 0; +} + +static void __exit chcr_ktls_exit(void) +{ + struct chcr_ktls_uld_ctx *u_ctx, *tmp; + struct adapter *adap; + + pr_info("driver unloaded\n"); + + mutex_lock(&dev_mutex); + list_for_each_entry_safe(u_ctx, tmp, &uld_ctx_list, entry) { + adap = pci_get_drvdata(u_ctx->lldi.pdev); + memset(&adap->ch_ktls_stats, 0, sizeof(adap->ch_ktls_stats)); + list_del(&u_ctx->entry); + kfree(u_ctx); + } + mutex_unlock(&dev_mutex); + cxgb4_unregister_uld(CXGB4_ULD_KTLS); +} + +module_init(chcr_ktls_init); +module_exit(chcr_ktls_exit); + +MODULE_DESCRIPTION("Chelsio NIC TLS ULD driver"); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Chelsio Communications"); +MODULE_VERSION(CHCR_KTLS_DRV_VERSION); diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.h b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.h new file mode 100644 index 000000000000..c1651b1431a0 --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.h @@ -0,0 +1,100 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (C) 2020 Chelsio Communications. All rights reserved. */ + +#ifndef __CHCR_KTLS_H__ +#define __CHCR_KTLS_H__ + +#include "cxgb4.h" +#include "t4_msg.h" +#include "t4_tcb.h" +#include "l2t.h" +#include "chcr_common.h" +#include "cxgb4_uld.h" +#include "clip_tbl.h" + +#define CHCR_KTLS_DRV_MODULE_NAME "ch_ktls" +#define CHCR_KTLS_DRV_VERSION "1.0.0.0-ko" +#define CHCR_KTLS_DRV_DESC "Chelsio NIC TLS ULD Driver" + +#define CHCR_TCB_STATE_CLOSED 0 +#define CHCR_KTLS_KEY_CTX_LEN 16 +#define CHCR_SET_TCB_FIELD_LEN sizeof(struct cpl_set_tcb_field) +#define CHCR_PLAIN_TX_DATA_LEN (sizeof(struct fw_ulptx_wr) +\ + sizeof(struct ulp_txpkt) +\ + sizeof(struct ulptx_idata) +\ + sizeof(struct cpl_tx_data)) + +#define CHCR_KTLS_WR_SIZE (CHCR_PLAIN_TX_DATA_LEN +\ + sizeof(struct cpl_tx_sec_pdu)) + +enum ch_ktls_open_state { + CH_KTLS_OPEN_SUCCESS = 0, + CH_KTLS_OPEN_PENDING = 1, + CH_KTLS_OPEN_FAILURE = 2, +}; + +struct chcr_ktls_info { + struct sock *sk; + spinlock_t lock; /* lock for pending_close */ + struct ktls_key_ctx key_ctx; + struct adapter *adap; + struct l2t_entry *l2te; + struct net_device *netdev; + struct completion completion; + u64 iv; + u64 record_no; + int tid; + int atid; + int rx_qid; + u32 iv_size; + u32 prev_seq; + u32 prev_ack; + u32 salt_size; + u32 key_ctx_len; + u32 scmd0_seqno_numivs; + u32 scmd0_ivgen_hdrlen; + u32 tcp_start_seq_number; + u32 scmd0_short_seqno_numivs; + u32 scmd0_short_ivgen_hdrlen; + u16 prev_win; + u8 tx_chan; + u8 smt_idx; + u8 port_id; + u8 ip_family; + u8 first_qset; + enum ch_ktls_open_state open_state; + bool pending_close; +}; + +struct chcr_ktls_ofld_ctx_tx { + struct tls_offload_context_tx base; + struct chcr_ktls_info *chcr_info; +}; + +struct chcr_ktls_uld_ctx { + struct list_head entry; + struct cxgb4_lld_info lldi; +}; + +static inline struct chcr_ktls_ofld_ctx_tx * +chcr_get_ktls_tx_context(struct tls_context *tls_ctx) +{ + BUILD_BUG_ON(sizeof(struct chcr_ktls_ofld_ctx_tx) > + TLS_OFFLOAD_CONTEXT_SIZE_TX); + return container_of(tls_offload_ctx_tx(tls_ctx), + struct chcr_ktls_ofld_ctx_tx, + base); +} + +static inline int chcr_get_first_rx_qid(struct adapter *adap) +{ + /* u_ctx is saved in adap, fetch it */ + struct chcr_ktls_uld_ctx *u_ctx = adap->uld[CXGB4_ULD_KTLS].handle; + + if (!u_ctx) + return -1; + return u_ctx->lldi.rxq_ids[0]; +} + +typedef int (*chcr_handler_func)(struct adapter *adap, unsigned char *input); +#endif /* __CHCR_KTLS_H__ */ diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/Makefile b/drivers/net/ethernet/chelsio/inline_crypto/chtls/Makefile new file mode 100644 index 000000000000..bc11495acdb3 --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/Makefile @@ -0,0 +1,6 @@ +# SPDX-License-Identifier: GPL-2.0-only +ccflags-y := -I $(srctree)/drivers/net/ethernet/chelsio/cxgb4 \ + -I $(srctree)/drivers/crypto/chelsio + +obj-$(CONFIG_CRYPTO_DEV_CHELSIO_TLS) += chtls.o +chtls-objs := chtls_main.o chtls_cm.o chtls_io.o chtls_hw.o diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls.h b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls.h new file mode 100644 index 000000000000..2d3dfdd2a716 --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls.h @@ -0,0 +1,580 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2018 Chelsio Communications, Inc. + */ + +#ifndef __CHTLS_H__ +#define __CHTLS_H__ + +#include <crypto/aes.h> +#include <crypto/algapi.h> +#include <crypto/hash.h> +#include <crypto/sha.h> +#include <crypto/authenc.h> +#include <crypto/ctr.h> +#include <crypto/gf128mul.h> +#include <crypto/internal/aead.h> +#include <crypto/null.h> +#include <crypto/internal/skcipher.h> +#include <crypto/aead.h> +#include <crypto/scatterwalk.h> +#include <crypto/internal/hash.h> +#include <linux/tls.h> +#include <net/tls.h> +#include <net/tls_toe.h> + +#include "t4fw_api.h" +#include "t4_msg.h" +#include "cxgb4.h" +#include "cxgb4_uld.h" +#include "l2t.h" +#include "chcr_algo.h" +#include "chcr_core.h" +#include "chcr_crypto.h" + +#define CHTLS_DRV_VERSION "1.0.0.0-ko" + +#define TLS_KEYCTX_RXFLIT_CNT_S 24 +#define TLS_KEYCTX_RXFLIT_CNT_V(x) ((x) << TLS_KEYCTX_RXFLIT_CNT_S) + +#define TLS_KEYCTX_RXPROT_VER_S 20 +#define TLS_KEYCTX_RXPROT_VER_M 0xf +#define TLS_KEYCTX_RXPROT_VER_V(x) ((x) << TLS_KEYCTX_RXPROT_VER_S) + +#define TLS_KEYCTX_RXCIPH_MODE_S 16 +#define TLS_KEYCTX_RXCIPH_MODE_M 0xf +#define TLS_KEYCTX_RXCIPH_MODE_V(x) ((x) << TLS_KEYCTX_RXCIPH_MODE_S) + +#define TLS_KEYCTX_RXAUTH_MODE_S 12 +#define TLS_KEYCTX_RXAUTH_MODE_M 0xf +#define TLS_KEYCTX_RXAUTH_MODE_V(x) ((x) << TLS_KEYCTX_RXAUTH_MODE_S) + +#define TLS_KEYCTX_RXCIAU_CTRL_S 11 +#define TLS_KEYCTX_RXCIAU_CTRL_V(x) ((x) << TLS_KEYCTX_RXCIAU_CTRL_S) + +#define TLS_KEYCTX_RX_SEQCTR_S 9 +#define TLS_KEYCTX_RX_SEQCTR_M 0x3 +#define TLS_KEYCTX_RX_SEQCTR_V(x) ((x) << TLS_KEYCTX_RX_SEQCTR_S) + +#define TLS_KEYCTX_RX_VALID_S 8 +#define TLS_KEYCTX_RX_VALID_V(x) ((x) << TLS_KEYCTX_RX_VALID_S) + +#define TLS_KEYCTX_RXCK_SIZE_S 3 +#define TLS_KEYCTX_RXCK_SIZE_M 0x7 +#define TLS_KEYCTX_RXCK_SIZE_V(x) ((x) << TLS_KEYCTX_RXCK_SIZE_S) + +#define TLS_KEYCTX_RXMK_SIZE_S 0 +#define TLS_KEYCTX_RXMK_SIZE_M 0x7 +#define TLS_KEYCTX_RXMK_SIZE_V(x) ((x) << TLS_KEYCTX_RXMK_SIZE_S) + +#define KEYCTX_TX_WR_IV_S 55 +#define KEYCTX_TX_WR_IV_M 0x1ffULL +#define KEYCTX_TX_WR_IV_V(x) ((x) << KEYCTX_TX_WR_IV_S) +#define KEYCTX_TX_WR_IV_G(x) \ + (((x) >> KEYCTX_TX_WR_IV_S) & KEYCTX_TX_WR_IV_M) + +#define KEYCTX_TX_WR_AAD_S 47 +#define KEYCTX_TX_WR_AAD_M 0xffULL +#define KEYCTX_TX_WR_AAD_V(x) ((x) << KEYCTX_TX_WR_AAD_S) +#define KEYCTX_TX_WR_AAD_G(x) (((x) >> KEYCTX_TX_WR_AAD_S) & \ + KEYCTX_TX_WR_AAD_M) + +#define KEYCTX_TX_WR_AADST_S 39 +#define KEYCTX_TX_WR_AADST_M 0xffULL +#define KEYCTX_TX_WR_AADST_V(x) ((x) << KEYCTX_TX_WR_AADST_S) +#define KEYCTX_TX_WR_AADST_G(x) \ + (((x) >> KEYCTX_TX_WR_AADST_S) & KEYCTX_TX_WR_AADST_M) + +#define KEYCTX_TX_WR_CIPHER_S 30 +#define KEYCTX_TX_WR_CIPHER_M 0x1ffULL +#define KEYCTX_TX_WR_CIPHER_V(x) ((x) << KEYCTX_TX_WR_CIPHER_S) +#define KEYCTX_TX_WR_CIPHER_G(x) \ + (((x) >> KEYCTX_TX_WR_CIPHER_S) & KEYCTX_TX_WR_CIPHER_M) + +#define KEYCTX_TX_WR_CIPHERST_S 23 +#define KEYCTX_TX_WR_CIPHERST_M 0x7f +#define KEYCTX_TX_WR_CIPHERST_V(x) ((x) << KEYCTX_TX_WR_CIPHERST_S) +#define KEYCTX_TX_WR_CIPHERST_G(x) \ + (((x) >> KEYCTX_TX_WR_CIPHERST_S) & KEYCTX_TX_WR_CIPHERST_M) + +#define KEYCTX_TX_WR_AUTH_S 14 +#define KEYCTX_TX_WR_AUTH_M 0x1ff +#define KEYCTX_TX_WR_AUTH_V(x) ((x) << KEYCTX_TX_WR_AUTH_S) +#define KEYCTX_TX_WR_AUTH_G(x) \ + (((x) >> KEYCTX_TX_WR_AUTH_S) & KEYCTX_TX_WR_AUTH_M) + +#define KEYCTX_TX_WR_AUTHST_S 7 +#define KEYCTX_TX_WR_AUTHST_M 0x7f +#define KEYCTX_TX_WR_AUTHST_V(x) ((x) << KEYCTX_TX_WR_AUTHST_S) +#define KEYCTX_TX_WR_AUTHST_G(x) \ + (((x) >> KEYCTX_TX_WR_AUTHST_S) & KEYCTX_TX_WR_AUTHST_M) + +#define KEYCTX_TX_WR_AUTHIN_S 0 +#define KEYCTX_TX_WR_AUTHIN_M 0x7f +#define KEYCTX_TX_WR_AUTHIN_V(x) ((x) << KEYCTX_TX_WR_AUTHIN_S) +#define KEYCTX_TX_WR_AUTHIN_G(x) \ + (((x) >> KEYCTX_TX_WR_AUTHIN_S) & KEYCTX_TX_WR_AUTHIN_M) + +struct sge_opaque_hdr { + void *dev; + dma_addr_t addr[MAX_SKB_FRAGS + 1]; +}; + +#define MAX_IVS_PAGE 256 +#define TLS_KEY_CONTEXT_SZ 64 +#define CIPHER_BLOCK_SIZE 16 +#define GCM_TAG_SIZE 16 +#define KEY_ON_MEM_SZ 16 +#define AEAD_EXPLICIT_DATA_SIZE 8 +#define TLS_HEADER_LENGTH 5 +#define SCMD_CIPH_MODE_AES_GCM 2 +/* Any MFS size should work and come from openssl */ +#define TLS_MFS 16384 + +#define RSS_HDR sizeof(struct rss_header) +#define TLS_WR_CPL_LEN \ + (sizeof(struct fw_tlstx_data_wr) + sizeof(struct cpl_tx_tls_sfo)) + +enum { + CHTLS_KEY_CONTEXT_DSGL, + CHTLS_KEY_CONTEXT_IMM, + CHTLS_KEY_CONTEXT_DDR, +}; + +enum { + CHTLS_LISTEN_START, + CHTLS_LISTEN_STOP, +}; + +/* Flags for return value of CPL message handlers */ +enum { + CPL_RET_BUF_DONE = 1, /* buffer processing done */ + CPL_RET_BAD_MSG = 2, /* bad CPL message */ + CPL_RET_UNKNOWN_TID = 4 /* unexpected unknown TID */ +}; + +#define LISTEN_INFO_HASH_SIZE 32 +#define RSPQ_HASH_BITS 5 +struct listen_info { + struct listen_info *next; /* Link to next entry */ + struct sock *sk; /* The listening socket */ + unsigned int stid; /* The server TID */ +}; + +enum { + T4_LISTEN_START_PENDING, + T4_LISTEN_STARTED +}; + +enum csk_flags { + CSK_CALLBACKS_CHKD, /* socket callbacks have been sanitized */ + CSK_ABORT_REQ_RCVD, /* received one ABORT_REQ_RSS message */ + CSK_TX_MORE_DATA, /* sending ULP data; don't set SHOVE bit */ + CSK_TX_WAIT_IDLE, /* suspend Tx until in-flight data is ACKed */ + CSK_ABORT_SHUTDOWN, /* shouldn't send more abort requests */ + CSK_ABORT_RPL_PENDING, /* expecting an abort reply */ + CSK_CLOSE_CON_REQUESTED,/* we've sent a close_conn_req */ + CSK_TX_DATA_SENT, /* sent a TX_DATA WR on this connection */ + CSK_TX_FAILOVER, /* Tx traffic failing over */ + CSK_UPDATE_RCV_WND, /* Need to update rcv window */ + CSK_RST_ABORTED, /* outgoing RST was aborted */ + CSK_TLS_HANDSHK, /* TLS Handshake */ + CSK_CONN_INLINE, /* Connection on HW */ +}; + +enum chtls_cdev_state { + CHTLS_CDEV_STATE_UP = 1 +}; + +struct listen_ctx { + struct sock *lsk; + struct chtls_dev *cdev; + struct sk_buff_head synq; + u32 state; +}; + +struct key_map { + unsigned long *addr; + unsigned int start; + unsigned int available; + unsigned int size; + spinlock_t lock; /* lock for key id request from map */ +} __packed; + +struct tls_scmd { + u32 seqno_numivs; + u32 ivgen_hdrlen; +}; + +struct chtls_dev { + struct tls_toe_device tlsdev; + struct list_head list; + struct cxgb4_lld_info *lldi; + struct pci_dev *pdev; + struct listen_info *listen_hash_tab[LISTEN_INFO_HASH_SIZE]; + spinlock_t listen_lock; /* lock for listen list */ + struct net_device **ports; + struct tid_info *tids; + unsigned int pfvf; + const unsigned short *mtus; + + struct idr hwtid_idr; + struct idr stid_idr; + + spinlock_t idr_lock ____cacheline_aligned_in_smp; + + struct net_device *egr_dev[NCHAN * 2]; + struct sk_buff *rspq_skb_cache[1 << RSPQ_HASH_BITS]; + struct sk_buff *askb; + + struct sk_buff_head deferq; + struct work_struct deferq_task; + + struct list_head list_node; + struct list_head rcu_node; + struct list_head na_node; + unsigned int send_page_order; + int max_host_sndbuf; + struct key_map kmap; + unsigned int cdev_state; +}; + +struct chtls_listen { + struct chtls_dev *cdev; + struct sock *sk; +}; + +struct chtls_hws { + struct sk_buff_head sk_recv_queue; + u8 txqid; + u8 ofld; + u16 type; + u16 rstate; + u16 keyrpl; + u16 pldlen; + u16 rcvpld; + u16 compute; + u16 expansion; + u16 keylen; + u16 pdus; + u16 adjustlen; + u16 ivsize; + u16 txleft; + u32 mfs; + s32 txkey; + s32 rxkey; + u32 fcplenmax; + u32 copied_seq; + u64 tx_seq_no; + struct tls_scmd scmd; + union { + struct tls12_crypto_info_aes_gcm_128 aes_gcm_128; + struct tls12_crypto_info_aes_gcm_256 aes_gcm_256; + } crypto_info; +}; + +struct chtls_sock { + struct sock *sk; + struct chtls_dev *cdev; + struct l2t_entry *l2t_entry; /* pointer to the L2T entry */ + struct net_device *egress_dev; /* TX_CHAN for act open retry */ + + struct sk_buff_head txq; + struct sk_buff *wr_skb_head; + struct sk_buff *wr_skb_tail; + struct sk_buff *ctrl_skb_cache; + struct sk_buff *txdata_skb_cache; /* abort path messages */ + struct kref kref; + unsigned long flags; + u32 opt2; + u32 wr_credits; + u32 wr_unacked; + u32 wr_max_credits; + u32 wr_nondata; + u32 hwtid; /* TCP Control Block ID */ + u32 txq_idx; + u32 rss_qid; + u32 tid; + u32 idr; + u32 mss; + u32 ulp_mode; + u32 tx_chan; + u32 rx_chan; + u32 sndbuf; + u32 txplen_max; + u32 mtu_idx; /* MTU table index */ + u32 smac_idx; + u8 port_id; + u8 tos; + u16 resv2; + u32 delack_mode; + u32 delack_seq; + u32 snd_win; + u32 rcv_win; + + void *passive_reap_next; /* placeholder for passive */ + struct chtls_hws tlshws; + struct synq { + struct sk_buff *next; + struct sk_buff *prev; + } synq; + struct listen_ctx *listen_ctx; +}; + +struct tls_hdr { + u8 type; + u16 version; + u16 length; +} __packed; + +struct tlsrx_cmp_hdr { + u8 type; + u16 version; + u16 length; + + u64 tls_seq; + u16 reserved1; + u8 res_to_mac_error; +} __packed; + +/* res_to_mac_error fields */ +#define TLSRX_HDR_PKT_INT_ERROR_S 4 +#define TLSRX_HDR_PKT_INT_ERROR_M 0x1 +#define TLSRX_HDR_PKT_INT_ERROR_V(x) \ + ((x) << TLSRX_HDR_PKT_INT_ERROR_S) +#define TLSRX_HDR_PKT_INT_ERROR_G(x) \ + (((x) >> TLSRX_HDR_PKT_INT_ERROR_S) & TLSRX_HDR_PKT_INT_ERROR_M) +#define TLSRX_HDR_PKT_INT_ERROR_F TLSRX_HDR_PKT_INT_ERROR_V(1U) + +#define TLSRX_HDR_PKT_SPP_ERROR_S 3 +#define TLSRX_HDR_PKT_SPP_ERROR_M 0x1 +#define TLSRX_HDR_PKT_SPP_ERROR_V(x) ((x) << TLSRX_HDR_PKT_SPP_ERROR) +#define TLSRX_HDR_PKT_SPP_ERROR_G(x) \ + (((x) >> TLSRX_HDR_PKT_SPP_ERROR_S) & TLSRX_HDR_PKT_SPP_ERROR_M) +#define TLSRX_HDR_PKT_SPP_ERROR_F TLSRX_HDR_PKT_SPP_ERROR_V(1U) + +#define TLSRX_HDR_PKT_CCDX_ERROR_S 2 +#define TLSRX_HDR_PKT_CCDX_ERROR_M 0x1 +#define TLSRX_HDR_PKT_CCDX_ERROR_V(x) ((x) << TLSRX_HDR_PKT_CCDX_ERROR_S) +#define TLSRX_HDR_PKT_CCDX_ERROR_G(x) \ + (((x) >> TLSRX_HDR_PKT_CCDX_ERROR_S) & TLSRX_HDR_PKT_CCDX_ERROR_M) +#define TLSRX_HDR_PKT_CCDX_ERROR_F TLSRX_HDR_PKT_CCDX_ERROR_V(1U) + +#define TLSRX_HDR_PKT_PAD_ERROR_S 1 +#define TLSRX_HDR_PKT_PAD_ERROR_M 0x1 +#define TLSRX_HDR_PKT_PAD_ERROR_V(x) ((x) << TLSRX_HDR_PKT_PAD_ERROR_S) +#define TLSRX_HDR_PKT_PAD_ERROR_G(x) \ + (((x) >> TLSRX_HDR_PKT_PAD_ERROR_S) & TLSRX_HDR_PKT_PAD_ERROR_M) +#define TLSRX_HDR_PKT_PAD_ERROR_F TLSRX_HDR_PKT_PAD_ERROR_V(1U) + +#define TLSRX_HDR_PKT_MAC_ERROR_S 0 +#define TLSRX_HDR_PKT_MAC_ERROR_M 0x1 +#define TLSRX_HDR_PKT_MAC_ERROR_V(x) ((x) << TLSRX_HDR_PKT_MAC_ERROR) +#define TLSRX_HDR_PKT_MAC_ERROR_G(x) \ + (((x) >> S_TLSRX_HDR_PKT_MAC_ERROR_S) & TLSRX_HDR_PKT_MAC_ERROR_M) +#define TLSRX_HDR_PKT_MAC_ERROR_F TLSRX_HDR_PKT_MAC_ERROR_V(1U) + +#define TLSRX_HDR_PKT_ERROR_M 0x1F +#define CONTENT_TYPE_ERROR 0x7F + +struct ulp_mem_rw { + __be32 cmd; + __be32 len16; /* command length */ + __be32 dlen; /* data length in 32-byte units */ + __be32 lock_addr; +}; + +struct tls_key_wr { + __be32 op_to_compl; + __be32 flowid_len16; + __be32 ftid; + u8 reneg_to_write_rx; + u8 protocol; + __be16 mfs; +}; + +struct tls_key_req { + struct tls_key_wr wr; + struct ulp_mem_rw req; + struct ulptx_idata sc_imm; +}; + +/* + * This lives in skb->cb and is used to chain WRs in a linked list. + */ +struct wr_skb_cb { + struct l2t_skb_cb l2t; /* reserve space for l2t CB */ + struct sk_buff *next_wr; /* next write request */ +}; + +/* Per-skb backlog handler. Run when a socket's backlog is processed. */ +struct blog_skb_cb { + void (*backlog_rcv)(struct sock *sk, struct sk_buff *skb); + struct chtls_dev *cdev; +}; + +/* + * Similar to tcp_skb_cb but with ULP elements added to support TLS, + * etc. + */ +struct ulp_skb_cb { + struct wr_skb_cb wr; /* reserve space for write request */ + u16 flags; /* TCP-like flags */ + u8 psh; + u8 ulp_mode; /* ULP mode/submode of sk_buff */ + u32 seq; /* TCP sequence number */ + union { /* ULP-specific fields */ + struct { + u8 type; + u8 ofld; + u8 iv; + } tls; + } ulp; +}; + +#define ULP_SKB_CB(skb) ((struct ulp_skb_cb *)&((skb)->cb[0])) +#define BLOG_SKB_CB(skb) ((struct blog_skb_cb *)(skb)->cb) + +/* + * Flags for ulp_skb_cb.flags. + */ +enum { + ULPCB_FLAG_NEED_HDR = 1 << 0, /* packet needs a TX_DATA_WR header */ + ULPCB_FLAG_NO_APPEND = 1 << 1, /* don't grow this skb */ + ULPCB_FLAG_BARRIER = 1 << 2, /* set TX_WAIT_IDLE after sending */ + ULPCB_FLAG_HOLD = 1 << 3, /* skb not ready for Tx yet */ + ULPCB_FLAG_COMPL = 1 << 4, /* request WR completion */ + ULPCB_FLAG_URG = 1 << 5, /* urgent data */ + ULPCB_FLAG_TLS_HDR = 1 << 6, /* payload with tls hdr */ + ULPCB_FLAG_NO_HDR = 1 << 7, /* not a ofld wr */ +}; + +/* The ULP mode/submode of an skbuff */ +#define skb_ulp_mode(skb) (ULP_SKB_CB(skb)->ulp_mode) +#define TCP_PAGE(sk) (sk->sk_frag.page) +#define TCP_OFF(sk) (sk->sk_frag.offset) + +static inline struct chtls_dev *to_chtls_dev(struct tls_toe_device *tlsdev) +{ + return container_of(tlsdev, struct chtls_dev, tlsdev); +} + +static inline void csk_set_flag(struct chtls_sock *csk, + enum csk_flags flag) +{ + __set_bit(flag, &csk->flags); +} + +static inline void csk_reset_flag(struct chtls_sock *csk, + enum csk_flags flag) +{ + __clear_bit(flag, &csk->flags); +} + +static inline bool csk_conn_inline(const struct chtls_sock *csk) +{ + return test_bit(CSK_CONN_INLINE, &csk->flags); +} + +static inline int csk_flag(const struct sock *sk, enum csk_flags flag) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + + if (!csk_conn_inline(csk)) + return 0; + return test_bit(flag, &csk->flags); +} + +static inline int csk_flag_nochk(const struct chtls_sock *csk, + enum csk_flags flag) +{ + return test_bit(flag, &csk->flags); +} + +static inline void *cplhdr(struct sk_buff *skb) +{ + return skb->data; +} + +static inline int is_neg_adv(unsigned int status) +{ + return status == CPL_ERR_RTX_NEG_ADVICE || + status == CPL_ERR_KEEPALV_NEG_ADVICE || + status == CPL_ERR_PERSIST_NEG_ADVICE; +} + +static inline void process_cpl_msg(void (*fn)(struct sock *, struct sk_buff *), + struct sock *sk, + struct sk_buff *skb) +{ + skb_reset_mac_header(skb); + skb_reset_network_header(skb); + skb_reset_transport_header(skb); + + bh_lock_sock(sk); + if (unlikely(sock_owned_by_user(sk))) { + BLOG_SKB_CB(skb)->backlog_rcv = fn; + __sk_add_backlog(sk, skb); + } else { + fn(sk, skb); + } + bh_unlock_sock(sk); +} + +static inline void chtls_sock_free(struct kref *ref) +{ + struct chtls_sock *csk = container_of(ref, struct chtls_sock, + kref); + kfree(csk); +} + +static inline void __chtls_sock_put(const char *fn, struct chtls_sock *csk) +{ + kref_put(&csk->kref, chtls_sock_free); +} + +static inline void __chtls_sock_get(const char *fn, + struct chtls_sock *csk) +{ + kref_get(&csk->kref); +} + +static inline void send_or_defer(struct sock *sk, struct tcp_sock *tp, + struct sk_buff *skb, int through_l2t) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + + if (through_l2t) { + /* send through L2T */ + cxgb4_l2t_send(csk->egress_dev, skb, csk->l2t_entry); + } else { + /* send directly */ + cxgb4_ofld_send(csk->egress_dev, skb); + } +} + +typedef int (*chtls_handler_func)(struct chtls_dev *, struct sk_buff *); +extern chtls_handler_func chtls_handlers[NUM_CPL_CMDS]; +void chtls_install_cpl_ops(struct sock *sk); +int chtls_init_kmap(struct chtls_dev *cdev, struct cxgb4_lld_info *lldi); +void chtls_listen_stop(struct chtls_dev *cdev, struct sock *sk); +int chtls_listen_start(struct chtls_dev *cdev, struct sock *sk); +void chtls_close(struct sock *sk, long timeout); +int chtls_disconnect(struct sock *sk, int flags); +void chtls_shutdown(struct sock *sk, int how); +void chtls_destroy_sock(struct sock *sk); +int chtls_sendmsg(struct sock *sk, struct msghdr *msg, size_t size); +int chtls_recvmsg(struct sock *sk, struct msghdr *msg, + size_t len, int nonblock, int flags, int *addr_len); +int chtls_sendpage(struct sock *sk, struct page *page, + int offset, size_t size, int flags); +int send_tx_flowc_wr(struct sock *sk, int compl, + u32 snd_nxt, u32 rcv_nxt); +void chtls_tcp_push(struct sock *sk, int flags); +int chtls_push_frames(struct chtls_sock *csk, int comp); +int chtls_set_tcb_tflag(struct sock *sk, unsigned int bit_pos, int val); +int chtls_setkey(struct chtls_sock *csk, u32 keylen, u32 mode, int cipher_type); +void skb_entail(struct sock *sk, struct sk_buff *skb, int flags); +unsigned int keyid_to_addr(int start_addr, int keyid); +void free_tls_keyid(struct sock *sk); +#endif diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c new file mode 100644 index 000000000000..05520dccd906 --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c @@ -0,0 +1,2327 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2018 Chelsio Communications, Inc. + * + * Written by: Atul Gupta (atul.gupta@chelsio.com) + */ + +#include <linux/module.h> +#include <linux/list.h> +#include <linux/workqueue.h> +#include <linux/skbuff.h> +#include <linux/timer.h> +#include <linux/notifier.h> +#include <linux/inetdevice.h> +#include <linux/ip.h> +#include <linux/tcp.h> +#include <linux/sched/signal.h> +#include <linux/kallsyms.h> +#include <linux/kprobes.h> +#include <linux/if_vlan.h> +#include <linux/ipv6.h> +#include <net/ipv6.h> +#include <net/transp_v6.h> +#include <net/ip6_route.h> +#include <net/inet_common.h> +#include <net/tcp.h> +#include <net/dst.h> +#include <net/tls.h> +#include <net/addrconf.h> +#include <net/secure_seq.h> + +#include "chtls.h" +#include "chtls_cm.h" +#include "clip_tbl.h" + +/* + * State transitions and actions for close. Note that if we are in SYN_SENT + * we remain in that state as we cannot control a connection while it's in + * SYN_SENT; such connections are allowed to establish and are then aborted. + */ +static unsigned char new_state[16] = { + /* current state: new state: action: */ + /* (Invalid) */ TCP_CLOSE, + /* TCP_ESTABLISHED */ TCP_FIN_WAIT1 | TCP_ACTION_FIN, + /* TCP_SYN_SENT */ TCP_SYN_SENT, + /* TCP_SYN_RECV */ TCP_FIN_WAIT1 | TCP_ACTION_FIN, + /* TCP_FIN_WAIT1 */ TCP_FIN_WAIT1, + /* TCP_FIN_WAIT2 */ TCP_FIN_WAIT2, + /* TCP_TIME_WAIT */ TCP_CLOSE, + /* TCP_CLOSE */ TCP_CLOSE, + /* TCP_CLOSE_WAIT */ TCP_LAST_ACK | TCP_ACTION_FIN, + /* TCP_LAST_ACK */ TCP_LAST_ACK, + /* TCP_LISTEN */ TCP_CLOSE, + /* TCP_CLOSING */ TCP_CLOSING, +}; + +static struct chtls_sock *chtls_sock_create(struct chtls_dev *cdev) +{ + struct chtls_sock *csk = kzalloc(sizeof(*csk), GFP_ATOMIC); + + if (!csk) + return NULL; + + csk->txdata_skb_cache = alloc_skb(TXDATA_SKB_LEN, GFP_ATOMIC); + if (!csk->txdata_skb_cache) { + kfree(csk); + return NULL; + } + + kref_init(&csk->kref); + csk->cdev = cdev; + skb_queue_head_init(&csk->txq); + csk->wr_skb_head = NULL; + csk->wr_skb_tail = NULL; + csk->mss = MAX_MSS; + csk->tlshws.ofld = 1; + csk->tlshws.txkey = -1; + csk->tlshws.rxkey = -1; + csk->tlshws.mfs = TLS_MFS; + skb_queue_head_init(&csk->tlshws.sk_recv_queue); + return csk; +} + +static void chtls_sock_release(struct kref *ref) +{ + struct chtls_sock *csk = + container_of(ref, struct chtls_sock, kref); + + kfree(csk); +} + +static struct net_device *chtls_find_netdev(struct chtls_dev *cdev, + struct sock *sk) +{ + struct net_device *ndev = cdev->ports[0]; +#if IS_ENABLED(CONFIG_IPV6) + struct net_device *temp; + int addr_type; +#endif + + switch (sk->sk_family) { + case PF_INET: + if (likely(!inet_sk(sk)->inet_rcv_saddr)) + return ndev; + ndev = __ip_dev_find(&init_net, inet_sk(sk)->inet_rcv_saddr, false); + break; +#if IS_ENABLED(CONFIG_IPV6) + case PF_INET6: + addr_type = ipv6_addr_type(&sk->sk_v6_rcv_saddr); + if (likely(addr_type == IPV6_ADDR_ANY)) + return ndev; + + for_each_netdev_rcu(&init_net, temp) { + if (ipv6_chk_addr(&init_net, (struct in6_addr *) + &sk->sk_v6_rcv_saddr, temp, 1)) { + ndev = temp; + break; + } + } + break; +#endif + default: + return NULL; + } + + if (!ndev) + return NULL; + + if (is_vlan_dev(ndev)) + return vlan_dev_real_dev(ndev); + return ndev; +} + +static void assign_rxopt(struct sock *sk, unsigned int opt) +{ + const struct chtls_dev *cdev; + struct chtls_sock *csk; + struct tcp_sock *tp; + + csk = rcu_dereference_sk_user_data(sk); + tp = tcp_sk(sk); + + cdev = csk->cdev; + tp->tcp_header_len = sizeof(struct tcphdr); + tp->rx_opt.mss_clamp = cdev->mtus[TCPOPT_MSS_G(opt)] - 40; + tp->mss_cache = tp->rx_opt.mss_clamp; + tp->rx_opt.tstamp_ok = TCPOPT_TSTAMP_G(opt); + tp->rx_opt.snd_wscale = TCPOPT_SACK_G(opt); + tp->rx_opt.wscale_ok = TCPOPT_WSCALE_OK_G(opt); + SND_WSCALE(tp) = TCPOPT_SND_WSCALE_G(opt); + if (!tp->rx_opt.wscale_ok) + tp->rx_opt.rcv_wscale = 0; + if (tp->rx_opt.tstamp_ok) { + tp->tcp_header_len += TCPOLEN_TSTAMP_ALIGNED; + tp->rx_opt.mss_clamp -= TCPOLEN_TSTAMP_ALIGNED; + } else if (csk->opt2 & TSTAMPS_EN_F) { + csk->opt2 &= ~TSTAMPS_EN_F; + csk->mtu_idx = TCPOPT_MSS_G(opt); + } +} + +static void chtls_purge_receive_queue(struct sock *sk) +{ + struct sk_buff *skb; + + while ((skb = __skb_dequeue(&sk->sk_receive_queue)) != NULL) { + skb_dst_set(skb, (void *)NULL); + kfree_skb(skb); + } +} + +static void chtls_purge_write_queue(struct sock *sk) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + struct sk_buff *skb; + + while ((skb = __skb_dequeue(&csk->txq))) { + sk->sk_wmem_queued -= skb->truesize; + __kfree_skb(skb); + } +} + +static void chtls_purge_recv_queue(struct sock *sk) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + struct chtls_hws *tlsk = &csk->tlshws; + struct sk_buff *skb; + + while ((skb = __skb_dequeue(&tlsk->sk_recv_queue)) != NULL) { + skb_dst_set(skb, NULL); + kfree_skb(skb); + } +} + +static void abort_arp_failure(void *handle, struct sk_buff *skb) +{ + struct cpl_abort_req *req = cplhdr(skb); + struct chtls_dev *cdev; + + cdev = (struct chtls_dev *)handle; + req->cmd = CPL_ABORT_NO_RST; + cxgb4_ofld_send(cdev->lldi->ports[0], skb); +} + +static struct sk_buff *alloc_ctrl_skb(struct sk_buff *skb, int len) +{ + if (likely(skb && !skb_shared(skb) && !skb_cloned(skb))) { + __skb_trim(skb, 0); + refcount_add(2, &skb->users); + } else { + skb = alloc_skb(len, GFP_KERNEL | __GFP_NOFAIL); + } + return skb; +} + +static void chtls_send_abort(struct sock *sk, int mode, struct sk_buff *skb) +{ + struct cpl_abort_req *req; + struct chtls_sock *csk; + struct tcp_sock *tp; + + csk = rcu_dereference_sk_user_data(sk); + tp = tcp_sk(sk); + + if (!skb) + skb = alloc_ctrl_skb(csk->txdata_skb_cache, sizeof(*req)); + + req = (struct cpl_abort_req *)skb_put(skb, sizeof(*req)); + INIT_TP_WR_CPL(req, CPL_ABORT_REQ, csk->tid); + skb_set_queue_mapping(skb, (csk->txq_idx << 1) | CPL_PRIORITY_DATA); + req->rsvd0 = htonl(tp->snd_nxt); + req->rsvd1 = !csk_flag_nochk(csk, CSK_TX_DATA_SENT); + req->cmd = mode; + t4_set_arp_err_handler(skb, csk->cdev, abort_arp_failure); + send_or_defer(sk, tp, skb, mode == CPL_ABORT_SEND_RST); +} + +static void chtls_send_reset(struct sock *sk, int mode, struct sk_buff *skb) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + + if (unlikely(csk_flag_nochk(csk, CSK_ABORT_SHUTDOWN) || + !csk->cdev)) { + if (sk->sk_state == TCP_SYN_RECV) + csk_set_flag(csk, CSK_RST_ABORTED); + goto out; + } + + if (!csk_flag_nochk(csk, CSK_TX_DATA_SENT)) { + struct tcp_sock *tp = tcp_sk(sk); + + if (send_tx_flowc_wr(sk, 0, tp->snd_nxt, tp->rcv_nxt) < 0) + WARN_ONCE(1, "send tx flowc error"); + csk_set_flag(csk, CSK_TX_DATA_SENT); + } + + csk_set_flag(csk, CSK_ABORT_RPL_PENDING); + chtls_purge_write_queue(sk); + + csk_set_flag(csk, CSK_ABORT_SHUTDOWN); + if (sk->sk_state != TCP_SYN_RECV) + chtls_send_abort(sk, mode, skb); + else + goto out; + + return; +out: + kfree_skb(skb); +} + +static void release_tcp_port(struct sock *sk) +{ + if (inet_csk(sk)->icsk_bind_hash) + inet_put_port(sk); +} + +static void tcp_uncork(struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); + + if (tp->nonagle & TCP_NAGLE_CORK) { + tp->nonagle &= ~TCP_NAGLE_CORK; + chtls_tcp_push(sk, 0); + } +} + +static void chtls_close_conn(struct sock *sk) +{ + struct cpl_close_con_req *req; + struct chtls_sock *csk; + struct sk_buff *skb; + unsigned int tid; + unsigned int len; + + len = roundup(sizeof(struct cpl_close_con_req), 16); + csk = rcu_dereference_sk_user_data(sk); + tid = csk->tid; + + skb = alloc_skb(len, GFP_KERNEL | __GFP_NOFAIL); + req = (struct cpl_close_con_req *)__skb_put(skb, len); + memset(req, 0, len); + req->wr.wr_hi = htonl(FW_WR_OP_V(FW_TP_WR) | + FW_WR_IMMDLEN_V(sizeof(*req) - + sizeof(req->wr))); + req->wr.wr_mid = htonl(FW_WR_LEN16_V(DIV_ROUND_UP(sizeof(*req), 16)) | + FW_WR_FLOWID_V(tid)); + + OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_CLOSE_CON_REQ, tid)); + + tcp_uncork(sk); + skb_entail(sk, skb, ULPCB_FLAG_NO_HDR | ULPCB_FLAG_NO_APPEND); + if (sk->sk_state != TCP_SYN_SENT) + chtls_push_frames(csk, 1); +} + +/* + * Perform a state transition during close and return the actions indicated + * for the transition. Do not make this function inline, the main reason + * it exists at all is to avoid multiple inlining of tcp_set_state. + */ +static int make_close_transition(struct sock *sk) +{ + int next = (int)new_state[sk->sk_state]; + + tcp_set_state(sk, next & TCP_STATE_MASK); + return next & TCP_ACTION_FIN; +} + +void chtls_close(struct sock *sk, long timeout) +{ + int data_lost, prev_state; + struct chtls_sock *csk; + + csk = rcu_dereference_sk_user_data(sk); + + lock_sock(sk); + sk->sk_shutdown |= SHUTDOWN_MASK; + + data_lost = skb_queue_len(&sk->sk_receive_queue); + data_lost |= skb_queue_len(&csk->tlshws.sk_recv_queue); + chtls_purge_recv_queue(sk); + chtls_purge_receive_queue(sk); + + if (sk->sk_state == TCP_CLOSE) { + goto wait; + } else if (data_lost || sk->sk_state == TCP_SYN_SENT) { + chtls_send_reset(sk, CPL_ABORT_SEND_RST, NULL); + release_tcp_port(sk); + goto unlock; + } else if (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime) { + sk->sk_prot->disconnect(sk, 0); + } else if (make_close_transition(sk)) { + chtls_close_conn(sk); + } +wait: + if (timeout) + sk_stream_wait_close(sk, timeout); + +unlock: + prev_state = sk->sk_state; + sock_hold(sk); + sock_orphan(sk); + + release_sock(sk); + + local_bh_disable(); + bh_lock_sock(sk); + + if (prev_state != TCP_CLOSE && sk->sk_state == TCP_CLOSE) + goto out; + + if (sk->sk_state == TCP_FIN_WAIT2 && tcp_sk(sk)->linger2 < 0 && + !csk_flag(sk, CSK_ABORT_SHUTDOWN)) { + struct sk_buff *skb; + + skb = alloc_skb(sizeof(struct cpl_abort_req), GFP_ATOMIC); + if (skb) + chtls_send_reset(sk, CPL_ABORT_SEND_RST, skb); + } + + if (sk->sk_state == TCP_CLOSE) + inet_csk_destroy_sock(sk); + +out: + bh_unlock_sock(sk); + local_bh_enable(); + sock_put(sk); +} + +/* + * Wait until a socket enters on of the given states. + */ +static int wait_for_states(struct sock *sk, unsigned int states) +{ + DECLARE_WAITQUEUE(wait, current); + struct socket_wq _sk_wq; + long current_timeo; + int err = 0; + + current_timeo = 200; + + /* + * We want this to work even when there's no associated struct socket. + * In that case we provide a temporary wait_queue_head_t. + */ + if (!sk->sk_wq) { + init_waitqueue_head(&_sk_wq.wait); + _sk_wq.fasync_list = NULL; + init_rcu_head_on_stack(&_sk_wq.rcu); + RCU_INIT_POINTER(sk->sk_wq, &_sk_wq); + } + + add_wait_queue(sk_sleep(sk), &wait); + while (!sk_in_state(sk, states)) { + if (!current_timeo) { + err = -EBUSY; + break; + } + if (signal_pending(current)) { + err = sock_intr_errno(current_timeo); + break; + } + set_current_state(TASK_UNINTERRUPTIBLE); + release_sock(sk); + if (!sk_in_state(sk, states)) + current_timeo = schedule_timeout(current_timeo); + __set_current_state(TASK_RUNNING); + lock_sock(sk); + } + remove_wait_queue(sk_sleep(sk), &wait); + + if (rcu_dereference(sk->sk_wq) == &_sk_wq) + sk->sk_wq = NULL; + return err; +} + +int chtls_disconnect(struct sock *sk, int flags) +{ + struct tcp_sock *tp; + int err; + + tp = tcp_sk(sk); + chtls_purge_recv_queue(sk); + chtls_purge_receive_queue(sk); + chtls_purge_write_queue(sk); + + if (sk->sk_state != TCP_CLOSE) { + sk->sk_err = ECONNRESET; + chtls_send_reset(sk, CPL_ABORT_SEND_RST, NULL); + err = wait_for_states(sk, TCPF_CLOSE); + if (err) + return err; + } + chtls_purge_recv_queue(sk); + chtls_purge_receive_queue(sk); + tp->max_window = 0xFFFF << (tp->rx_opt.snd_wscale); + return tcp_disconnect(sk, flags); +} + +#define SHUTDOWN_ELIGIBLE_STATE (TCPF_ESTABLISHED | \ + TCPF_SYN_RECV | TCPF_CLOSE_WAIT) +void chtls_shutdown(struct sock *sk, int how) +{ + if ((how & SEND_SHUTDOWN) && + sk_in_state(sk, SHUTDOWN_ELIGIBLE_STATE) && + make_close_transition(sk)) + chtls_close_conn(sk); +} + +void chtls_destroy_sock(struct sock *sk) +{ + struct chtls_sock *csk; + + csk = rcu_dereference_sk_user_data(sk); + chtls_purge_recv_queue(sk); + csk->ulp_mode = ULP_MODE_NONE; + chtls_purge_write_queue(sk); + free_tls_keyid(sk); + kref_put(&csk->kref, chtls_sock_release); + csk->cdev = NULL; + if (sk->sk_family == AF_INET) + sk->sk_prot = &tcp_prot; +#if IS_ENABLED(CONFIG_IPV6) + else + sk->sk_prot = &tcpv6_prot; +#endif + sk->sk_prot->destroy(sk); +} + +static void reset_listen_child(struct sock *child) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(child); + struct sk_buff *skb; + + skb = alloc_ctrl_skb(csk->txdata_skb_cache, + sizeof(struct cpl_abort_req)); + + chtls_send_reset(child, CPL_ABORT_SEND_RST, skb); + sock_orphan(child); + INC_ORPHAN_COUNT(child); + if (child->sk_state == TCP_CLOSE) + inet_csk_destroy_sock(child); +} + +static void chtls_disconnect_acceptq(struct sock *listen_sk) +{ + struct request_sock **pprev; + + pprev = ACCEPT_QUEUE(listen_sk); + while (*pprev) { + struct request_sock *req = *pprev; + + if (req->rsk_ops == &chtls_rsk_ops || + req->rsk_ops == &chtls_rsk_opsv6) { + struct sock *child = req->sk; + + *pprev = req->dl_next; + sk_acceptq_removed(listen_sk); + reqsk_put(req); + sock_hold(child); + local_bh_disable(); + bh_lock_sock(child); + release_tcp_port(child); + reset_listen_child(child); + bh_unlock_sock(child); + local_bh_enable(); + sock_put(child); + } else { + pprev = &req->dl_next; + } + } +} + +static int listen_hashfn(const struct sock *sk) +{ + return ((unsigned long)sk >> 10) & (LISTEN_INFO_HASH_SIZE - 1); +} + +static struct listen_info *listen_hash_add(struct chtls_dev *cdev, + struct sock *sk, + unsigned int stid) +{ + struct listen_info *p = kmalloc(sizeof(*p), GFP_KERNEL); + + if (p) { + int key = listen_hashfn(sk); + + p->sk = sk; + p->stid = stid; + spin_lock(&cdev->listen_lock); + p->next = cdev->listen_hash_tab[key]; + cdev->listen_hash_tab[key] = p; + spin_unlock(&cdev->listen_lock); + } + return p; +} + +static int listen_hash_find(struct chtls_dev *cdev, + struct sock *sk) +{ + struct listen_info *p; + int stid = -1; + int key; + + key = listen_hashfn(sk); + + spin_lock(&cdev->listen_lock); + for (p = cdev->listen_hash_tab[key]; p; p = p->next) + if (p->sk == sk) { + stid = p->stid; + break; + } + spin_unlock(&cdev->listen_lock); + return stid; +} + +static int listen_hash_del(struct chtls_dev *cdev, + struct sock *sk) +{ + struct listen_info *p, **prev; + int stid = -1; + int key; + + key = listen_hashfn(sk); + prev = &cdev->listen_hash_tab[key]; + + spin_lock(&cdev->listen_lock); + for (p = *prev; p; prev = &p->next, p = p->next) + if (p->sk == sk) { + stid = p->stid; + *prev = p->next; + kfree(p); + break; + } + spin_unlock(&cdev->listen_lock); + return stid; +} + +static void cleanup_syn_rcv_conn(struct sock *child, struct sock *parent) +{ + struct request_sock *req; + struct chtls_sock *csk; + + csk = rcu_dereference_sk_user_data(child); + req = csk->passive_reap_next; + + reqsk_queue_removed(&inet_csk(parent)->icsk_accept_queue, req); + __skb_unlink((struct sk_buff *)&csk->synq, &csk->listen_ctx->synq); + chtls_reqsk_free(req); + csk->passive_reap_next = NULL; +} + +static void chtls_reset_synq(struct listen_ctx *listen_ctx) +{ + struct sock *listen_sk = listen_ctx->lsk; + + while (!skb_queue_empty(&listen_ctx->synq)) { + struct chtls_sock *csk = + container_of((struct synq *)__skb_dequeue + (&listen_ctx->synq), struct chtls_sock, synq); + struct sock *child = csk->sk; + + cleanup_syn_rcv_conn(child, listen_sk); + sock_hold(child); + local_bh_disable(); + bh_lock_sock(child); + release_tcp_port(child); + reset_listen_child(child); + bh_unlock_sock(child); + local_bh_enable(); + sock_put(child); + } +} + +int chtls_listen_start(struct chtls_dev *cdev, struct sock *sk) +{ + struct net_device *ndev; +#if IS_ENABLED(CONFIG_IPV6) + bool clip_valid = false; +#endif + struct listen_ctx *ctx; + struct adapter *adap; + struct port_info *pi; + int ret = 0; + int stid; + + rcu_read_lock(); + ndev = chtls_find_netdev(cdev, sk); + rcu_read_unlock(); + if (!ndev) + return -EBADF; + + pi = netdev_priv(ndev); + adap = pi->adapter; + if (!(adap->flags & CXGB4_FULL_INIT_DONE)) + return -EBADF; + + if (listen_hash_find(cdev, sk) >= 0) /* already have it */ + return -EADDRINUSE; + + ctx = kmalloc(sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + + __module_get(THIS_MODULE); + ctx->lsk = sk; + ctx->cdev = cdev; + ctx->state = T4_LISTEN_START_PENDING; + skb_queue_head_init(&ctx->synq); + + stid = cxgb4_alloc_stid(cdev->tids, sk->sk_family, ctx); + if (stid < 0) + goto free_ctx; + + sock_hold(sk); + if (!listen_hash_add(cdev, sk, stid)) + goto free_stid; + + if (sk->sk_family == PF_INET) { + ret = cxgb4_create_server(ndev, stid, + inet_sk(sk)->inet_rcv_saddr, + inet_sk(sk)->inet_sport, 0, + cdev->lldi->rxq_ids[0]); +#if IS_ENABLED(CONFIG_IPV6) + } else { + int addr_type; + + addr_type = ipv6_addr_type(&sk->sk_v6_rcv_saddr); + if (addr_type != IPV6_ADDR_ANY) { + ret = cxgb4_clip_get(ndev, (const u32 *) + &sk->sk_v6_rcv_saddr, 1); + if (ret) + goto del_hash; + clip_valid = true; + } + ret = cxgb4_create_server6(ndev, stid, + &sk->sk_v6_rcv_saddr, + inet_sk(sk)->inet_sport, + cdev->lldi->rxq_ids[0]); +#endif + } + if (ret > 0) + ret = net_xmit_errno(ret); + if (ret) + goto del_hash; + return 0; +del_hash: +#if IS_ENABLED(CONFIG_IPV6) + if (clip_valid) + cxgb4_clip_release(ndev, (const u32 *)&sk->sk_v6_rcv_saddr, 1); +#endif + listen_hash_del(cdev, sk); +free_stid: + cxgb4_free_stid(cdev->tids, stid, sk->sk_family); + sock_put(sk); +free_ctx: + kfree(ctx); + module_put(THIS_MODULE); + return -EBADF; +} + +void chtls_listen_stop(struct chtls_dev *cdev, struct sock *sk) +{ + struct listen_ctx *listen_ctx; + int stid; + + stid = listen_hash_del(cdev, sk); + if (stid < 0) + return; + + listen_ctx = (struct listen_ctx *)lookup_stid(cdev->tids, stid); + chtls_reset_synq(listen_ctx); + + cxgb4_remove_server(cdev->lldi->ports[0], stid, + cdev->lldi->rxq_ids[0], sk->sk_family == PF_INET6); + +#if IS_ENABLED(CONFIG_IPV6) + if (sk->sk_family == PF_INET6) { + struct chtls_sock *csk; + int addr_type = 0; + + csk = rcu_dereference_sk_user_data(sk); + addr_type = ipv6_addr_type((const struct in6_addr *) + &sk->sk_v6_rcv_saddr); + if (addr_type != IPV6_ADDR_ANY) + cxgb4_clip_release(csk->egress_dev, (const u32 *) + &sk->sk_v6_rcv_saddr, 1); + } +#endif + chtls_disconnect_acceptq(sk); +} + +static int chtls_pass_open_rpl(struct chtls_dev *cdev, struct sk_buff *skb) +{ + struct cpl_pass_open_rpl *rpl = cplhdr(skb) + RSS_HDR; + unsigned int stid = GET_TID(rpl); + struct listen_ctx *listen_ctx; + + listen_ctx = (struct listen_ctx *)lookup_stid(cdev->tids, stid); + if (!listen_ctx) + return CPL_RET_BUF_DONE; + + if (listen_ctx->state == T4_LISTEN_START_PENDING) { + listen_ctx->state = T4_LISTEN_STARTED; + return CPL_RET_BUF_DONE; + } + + if (rpl->status != CPL_ERR_NONE) { + pr_info("Unexpected PASS_OPEN_RPL status %u for STID %u\n", + rpl->status, stid); + return CPL_RET_BUF_DONE; + } + cxgb4_free_stid(cdev->tids, stid, listen_ctx->lsk->sk_family); + sock_put(listen_ctx->lsk); + kfree(listen_ctx); + module_put(THIS_MODULE); + + return 0; +} + +static int chtls_close_listsrv_rpl(struct chtls_dev *cdev, struct sk_buff *skb) +{ + struct cpl_close_listsvr_rpl *rpl = cplhdr(skb) + RSS_HDR; + struct listen_ctx *listen_ctx; + unsigned int stid; + void *data; + + stid = GET_TID(rpl); + data = lookup_stid(cdev->tids, stid); + listen_ctx = (struct listen_ctx *)data; + + if (rpl->status != CPL_ERR_NONE) { + pr_info("Unexpected CLOSE_LISTSRV_RPL status %u for STID %u\n", + rpl->status, stid); + return CPL_RET_BUF_DONE; + } + + cxgb4_free_stid(cdev->tids, stid, listen_ctx->lsk->sk_family); + sock_put(listen_ctx->lsk); + kfree(listen_ctx); + module_put(THIS_MODULE); + + return 0; +} + +static void chtls_purge_wr_queue(struct sock *sk) +{ + struct sk_buff *skb; + + while ((skb = dequeue_wr(sk)) != NULL) + kfree_skb(skb); +} + +static void chtls_release_resources(struct sock *sk) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + struct chtls_dev *cdev = csk->cdev; + unsigned int tid = csk->tid; + struct tid_info *tids; + + if (!cdev) + return; + + tids = cdev->tids; + kfree_skb(csk->txdata_skb_cache); + csk->txdata_skb_cache = NULL; + + if (csk->wr_credits != csk->wr_max_credits) { + chtls_purge_wr_queue(sk); + chtls_reset_wr_list(csk); + } + + if (csk->l2t_entry) { + cxgb4_l2t_release(csk->l2t_entry); + csk->l2t_entry = NULL; + } + + if (sk->sk_state != TCP_SYN_SENT) { + cxgb4_remove_tid(tids, csk->port_id, tid, sk->sk_family); + sock_put(sk); + } +} + +static void chtls_conn_done(struct sock *sk) +{ + if (sock_flag(sk, SOCK_DEAD)) + chtls_purge_receive_queue(sk); + sk_wakeup_sleepers(sk, 0); + tcp_done(sk); +} + +static void do_abort_syn_rcv(struct sock *child, struct sock *parent) +{ + /* + * If the server is still open we clean up the child connection, + * otherwise the server already did the clean up as it was purging + * its SYN queue and the skb was just sitting in its backlog. + */ + if (likely(parent->sk_state == TCP_LISTEN)) { + cleanup_syn_rcv_conn(child, parent); + /* Without the below call to sock_orphan, + * we leak the socket resource with syn_flood test + * as inet_csk_destroy_sock will not be called + * in tcp_done since SOCK_DEAD flag is not set. + * Kernel handles this differently where new socket is + * created only after 3 way handshake is done. + */ + sock_orphan(child); + percpu_counter_inc((child)->sk_prot->orphan_count); + chtls_release_resources(child); + chtls_conn_done(child); + } else { + if (csk_flag(child, CSK_RST_ABORTED)) { + chtls_release_resources(child); + chtls_conn_done(child); + } + } +} + +static void pass_open_abort(struct sock *child, struct sock *parent, + struct sk_buff *skb) +{ + do_abort_syn_rcv(child, parent); + kfree_skb(skb); +} + +static void bl_pass_open_abort(struct sock *lsk, struct sk_buff *skb) +{ + pass_open_abort(skb->sk, lsk, skb); +} + +static void chtls_pass_open_arp_failure(struct sock *sk, + struct sk_buff *skb) +{ + const struct request_sock *oreq; + struct chtls_sock *csk; + struct chtls_dev *cdev; + struct sock *parent; + void *data; + + csk = rcu_dereference_sk_user_data(sk); + cdev = csk->cdev; + + /* + * If the connection is being aborted due to the parent listening + * socket going away there's nothing to do, the ABORT_REQ will close + * the connection. + */ + if (csk_flag(sk, CSK_ABORT_RPL_PENDING)) { + kfree_skb(skb); + return; + } + + oreq = csk->passive_reap_next; + data = lookup_stid(cdev->tids, oreq->ts_recent); + parent = ((struct listen_ctx *)data)->lsk; + + bh_lock_sock(parent); + if (!sock_owned_by_user(parent)) { + pass_open_abort(sk, parent, skb); + } else { + BLOG_SKB_CB(skb)->backlog_rcv = bl_pass_open_abort; + __sk_add_backlog(parent, skb); + } + bh_unlock_sock(parent); +} + +static void chtls_accept_rpl_arp_failure(void *handle, + struct sk_buff *skb) +{ + struct sock *sk = (struct sock *)handle; + + sock_hold(sk); + process_cpl_msg(chtls_pass_open_arp_failure, sk, skb); + sock_put(sk); +} + +static unsigned int chtls_select_mss(const struct chtls_sock *csk, + unsigned int pmtu, + struct cpl_pass_accept_req *req) +{ + struct chtls_dev *cdev; + struct dst_entry *dst; + unsigned int tcpoptsz; + unsigned int iphdrsz; + unsigned int mtu_idx; + struct tcp_sock *tp; + unsigned int mss; + struct sock *sk; + + mss = ntohs(req->tcpopt.mss); + sk = csk->sk; + dst = __sk_dst_get(sk); + cdev = csk->cdev; + tp = tcp_sk(sk); + tcpoptsz = 0; + +#if IS_ENABLED(CONFIG_IPV6) + if (sk->sk_family == AF_INET6) + iphdrsz = sizeof(struct ipv6hdr) + sizeof(struct tcphdr); + else +#endif + iphdrsz = sizeof(struct iphdr) + sizeof(struct tcphdr); + if (req->tcpopt.tstamp) + tcpoptsz += round_up(TCPOLEN_TIMESTAMP, 4); + + tp->advmss = dst_metric_advmss(dst); + if (USER_MSS(tp) && tp->advmss > USER_MSS(tp)) + tp->advmss = USER_MSS(tp); + if (tp->advmss > pmtu - iphdrsz) + tp->advmss = pmtu - iphdrsz; + if (mss && tp->advmss > mss) + tp->advmss = mss; + + tp->advmss = cxgb4_best_aligned_mtu(cdev->lldi->mtus, + iphdrsz + tcpoptsz, + tp->advmss - tcpoptsz, + 8, &mtu_idx); + tp->advmss -= iphdrsz; + + inet_csk(sk)->icsk_pmtu_cookie = pmtu; + return mtu_idx; +} + +static unsigned int select_rcv_wscale(int space, int wscale_ok, int win_clamp) +{ + int wscale = 0; + + if (space > MAX_RCV_WND) + space = MAX_RCV_WND; + if (win_clamp && win_clamp < space) + space = win_clamp; + + if (wscale_ok) { + while (wscale < 14 && (65535 << wscale) < space) + wscale++; + } + return wscale; +} + +static void chtls_pass_accept_rpl(struct sk_buff *skb, + struct cpl_pass_accept_req *req, + unsigned int tid) + +{ + struct cpl_t5_pass_accept_rpl *rpl5; + struct cxgb4_lld_info *lldi; + const struct tcphdr *tcph; + const struct tcp_sock *tp; + struct chtls_sock *csk; + unsigned int len; + struct sock *sk; + u32 opt2, hlen; + u64 opt0; + + sk = skb->sk; + tp = tcp_sk(sk); + csk = sk->sk_user_data; + csk->tid = tid; + lldi = csk->cdev->lldi; + len = roundup(sizeof(*rpl5), 16); + + rpl5 = __skb_put_zero(skb, len); + INIT_TP_WR(rpl5, tid); + + OPCODE_TID(rpl5) = cpu_to_be32(MK_OPCODE_TID(CPL_PASS_ACCEPT_RPL, + csk->tid)); + csk->mtu_idx = chtls_select_mss(csk, dst_mtu(__sk_dst_get(sk)), + req); + opt0 = TCAM_BYPASS_F | + WND_SCALE_V(RCV_WSCALE(tp)) | + MSS_IDX_V(csk->mtu_idx) | + L2T_IDX_V(csk->l2t_entry->idx) | + NAGLE_V(!(tp->nonagle & TCP_NAGLE_OFF)) | + TX_CHAN_V(csk->tx_chan) | + SMAC_SEL_V(csk->smac_idx) | + DSCP_V(csk->tos >> 2) | + ULP_MODE_V(ULP_MODE_TLS) | + RCV_BUFSIZ_V(min(tp->rcv_wnd >> 10, RCV_BUFSIZ_M)); + + opt2 = RX_CHANNEL_V(0) | + RSS_QUEUE_VALID_F | RSS_QUEUE_V(csk->rss_qid); + + if (!is_t5(lldi->adapter_type)) + opt2 |= RX_FC_DISABLE_F; + if (req->tcpopt.tstamp) + opt2 |= TSTAMPS_EN_F; + if (req->tcpopt.sack) + opt2 |= SACK_EN_F; + hlen = ntohl(req->hdr_len); + + tcph = (struct tcphdr *)((u8 *)(req + 1) + + T6_ETH_HDR_LEN_G(hlen) + T6_IP_HDR_LEN_G(hlen)); + if (tcph->ece && tcph->cwr) + opt2 |= CCTRL_ECN_V(1); + opt2 |= CONG_CNTRL_V(CONG_ALG_NEWRENO); + opt2 |= T5_ISS_F; + opt2 |= T5_OPT_2_VALID_F; + opt2 |= WND_SCALE_EN_V(WSCALE_OK(tp)); + rpl5->opt0 = cpu_to_be64(opt0); + rpl5->opt2 = cpu_to_be32(opt2); + rpl5->iss = cpu_to_be32((prandom_u32() & ~7UL) - 1); + set_wr_txq(skb, CPL_PRIORITY_SETUP, csk->port_id); + t4_set_arp_err_handler(skb, sk, chtls_accept_rpl_arp_failure); + cxgb4_l2t_send(csk->egress_dev, skb, csk->l2t_entry); +} + +static void inet_inherit_port(struct inet_hashinfo *hash_info, + struct sock *lsk, struct sock *newsk) +{ + local_bh_disable(); + __inet_inherit_port(lsk, newsk); + local_bh_enable(); +} + +static int chtls_backlog_rcv(struct sock *sk, struct sk_buff *skb) +{ + if (skb->protocol) { + kfree_skb(skb); + return 0; + } + BLOG_SKB_CB(skb)->backlog_rcv(sk, skb); + return 0; +} + +static void chtls_set_tcp_window(struct chtls_sock *csk) +{ + struct net_device *ndev = csk->egress_dev; + struct port_info *pi = netdev_priv(ndev); + unsigned int linkspeed; + u8 scale; + + linkspeed = pi->link_cfg.speed; + scale = linkspeed / SPEED_10000; +#define CHTLS_10G_RCVWIN (256 * 1024) + csk->rcv_win = CHTLS_10G_RCVWIN; + if (scale) + csk->rcv_win *= scale; +#define CHTLS_10G_SNDWIN (256 * 1024) + csk->snd_win = CHTLS_10G_SNDWIN; + if (scale) + csk->snd_win *= scale; +} + +static struct sock *chtls_recv_sock(struct sock *lsk, + struct request_sock *oreq, + void *network_hdr, + const struct cpl_pass_accept_req *req, + struct chtls_dev *cdev) +{ + struct neighbour *n = NULL; + struct inet_sock *newinet; + const struct iphdr *iph; + struct tls_context *ctx; + struct net_device *ndev; + struct chtls_sock *csk; + struct dst_entry *dst; + struct tcp_sock *tp; + struct sock *newsk; + u16 port_id; + int rxq_idx; + int step; + + iph = (const struct iphdr *)network_hdr; + newsk = tcp_create_openreq_child(lsk, oreq, cdev->askb); + if (!newsk) + goto free_oreq; + + if (lsk->sk_family == AF_INET) { + dst = inet_csk_route_child_sock(lsk, newsk, oreq); + if (!dst) + goto free_sk; + + n = dst_neigh_lookup(dst, &iph->saddr); +#if IS_ENABLED(CONFIG_IPV6) + } else { + const struct ipv6hdr *ip6h; + struct flowi6 fl6; + + ip6h = (const struct ipv6hdr *)network_hdr; + memset(&fl6, 0, sizeof(fl6)); + fl6.flowi6_proto = IPPROTO_TCP; + fl6.saddr = ip6h->daddr; + fl6.daddr = ip6h->saddr; + fl6.fl6_dport = inet_rsk(oreq)->ir_rmt_port; + fl6.fl6_sport = htons(inet_rsk(oreq)->ir_num); + security_req_classify_flow(oreq, flowi6_to_flowi(&fl6)); + dst = ip6_dst_lookup_flow(sock_net(lsk), lsk, &fl6, NULL); + if (IS_ERR(dst)) + goto free_sk; + n = dst_neigh_lookup(dst, &ip6h->saddr); +#endif + } + if (!n) + goto free_sk; + + ndev = n->dev; + if (!ndev) + goto free_dst; + port_id = cxgb4_port_idx(ndev); + + csk = chtls_sock_create(cdev); + if (!csk) + goto free_dst; + + csk->l2t_entry = cxgb4_l2t_get(cdev->lldi->l2t, n, ndev, 0); + if (!csk->l2t_entry) + goto free_csk; + + newsk->sk_user_data = csk; + newsk->sk_backlog_rcv = chtls_backlog_rcv; + + tp = tcp_sk(newsk); + newinet = inet_sk(newsk); + + if (iph->version == 0x4) { + newinet->inet_daddr = iph->saddr; + newinet->inet_rcv_saddr = iph->daddr; + newinet->inet_saddr = iph->daddr; +#if IS_ENABLED(CONFIG_IPV6) + } else { + struct tcp6_sock *newtcp6sk = (struct tcp6_sock *)newsk; + struct inet_request_sock *treq = inet_rsk(oreq); + struct ipv6_pinfo *newnp = inet6_sk(newsk); + struct ipv6_pinfo *np = inet6_sk(lsk); + + inet_sk(newsk)->pinet6 = &newtcp6sk->inet6; + memcpy(newnp, np, sizeof(struct ipv6_pinfo)); + newsk->sk_v6_daddr = treq->ir_v6_rmt_addr; + newsk->sk_v6_rcv_saddr = treq->ir_v6_loc_addr; + inet6_sk(newsk)->saddr = treq->ir_v6_loc_addr; + newnp->ipv6_fl_list = NULL; + newnp->pktoptions = NULL; + newsk->sk_bound_dev_if = treq->ir_iif; + newinet->inet_opt = NULL; + newinet->inet_daddr = LOOPBACK4_IPV6; + newinet->inet_saddr = LOOPBACK4_IPV6; +#endif + } + + oreq->ts_recent = PASS_OPEN_TID_G(ntohl(req->tos_stid)); + sk_setup_caps(newsk, dst); + ctx = tls_get_ctx(lsk); + newsk->sk_destruct = ctx->sk_destruct; + csk->sk = newsk; + csk->passive_reap_next = oreq; + csk->tx_chan = cxgb4_port_chan(ndev); + csk->port_id = port_id; + csk->egress_dev = ndev; + csk->tos = PASS_OPEN_TOS_G(ntohl(req->tos_stid)); + chtls_set_tcp_window(csk); + tp->rcv_wnd = csk->rcv_win; + csk->sndbuf = csk->snd_win; + csk->ulp_mode = ULP_MODE_TLS; + step = cdev->lldi->nrxq / cdev->lldi->nchan; + csk->rss_qid = cdev->lldi->rxq_ids[port_id * step]; + rxq_idx = port_id * step; + csk->txq_idx = (rxq_idx < cdev->lldi->ntxq) ? rxq_idx : + port_id * step; + csk->sndbuf = newsk->sk_sndbuf; + csk->smac_idx = ((struct port_info *)netdev_priv(ndev))->smt_idx; + RCV_WSCALE(tp) = select_rcv_wscale(tcp_full_space(newsk), + sock_net(newsk)-> + ipv4.sysctl_tcp_window_scaling, + tp->window_clamp); + neigh_release(n); + inet_inherit_port(&tcp_hashinfo, lsk, newsk); + csk_set_flag(csk, CSK_CONN_INLINE); + bh_unlock_sock(newsk); /* tcp_create_openreq_child ->sk_clone_lock */ + + return newsk; +free_csk: + chtls_sock_release(&csk->kref); +free_dst: + dst_release(dst); +free_sk: + inet_csk_prepare_forced_close(newsk); + tcp_done(newsk); +free_oreq: + chtls_reqsk_free(oreq); + return NULL; +} + +/* + * Populate a TID_RELEASE WR. The skb must be already propely sized. + */ +static void mk_tid_release(struct sk_buff *skb, + unsigned int chan, unsigned int tid) +{ + struct cpl_tid_release *req; + unsigned int len; + + len = roundup(sizeof(struct cpl_tid_release), 16); + req = (struct cpl_tid_release *)__skb_put(skb, len); + memset(req, 0, len); + set_wr_txq(skb, CPL_PRIORITY_SETUP, chan); + INIT_TP_WR_CPL(req, CPL_TID_RELEASE, tid); +} + +static int chtls_get_module(struct sock *sk) +{ + struct inet_connection_sock *icsk = inet_csk(sk); + + if (!try_module_get(icsk->icsk_ulp_ops->owner)) + return -1; + + return 0; +} + +static void chtls_pass_accept_request(struct sock *sk, + struct sk_buff *skb) +{ + struct cpl_t5_pass_accept_rpl *rpl; + struct cpl_pass_accept_req *req; + struct listen_ctx *listen_ctx; + struct vlan_ethhdr *vlan_eh; + struct request_sock *oreq; + struct sk_buff *reply_skb; + struct chtls_sock *csk; + struct chtls_dev *cdev; + struct ipv6hdr *ip6h; + struct tcphdr *tcph; + struct sock *newsk; + struct ethhdr *eh; + struct iphdr *iph; + void *network_hdr; + unsigned int stid; + unsigned int len; + unsigned int tid; + bool th_ecn, ect; + __u8 ip_dsfield; /* IPv4 tos or IPv6 dsfield */ + u16 eth_hdr_len; + bool ecn_ok; + + req = cplhdr(skb) + RSS_HDR; + tid = GET_TID(req); + cdev = BLOG_SKB_CB(skb)->cdev; + newsk = lookup_tid(cdev->tids, tid); + stid = PASS_OPEN_TID_G(ntohl(req->tos_stid)); + if (newsk) { + pr_info("tid (%d) already in use\n", tid); + return; + } + + len = roundup(sizeof(*rpl), 16); + reply_skb = alloc_skb(len, GFP_ATOMIC); + if (!reply_skb) { + cxgb4_remove_tid(cdev->tids, 0, tid, sk->sk_family); + kfree_skb(skb); + return; + } + + if (sk->sk_state != TCP_LISTEN) + goto reject; + + if (inet_csk_reqsk_queue_is_full(sk)) + goto reject; + + if (sk_acceptq_is_full(sk)) + goto reject; + + + eth_hdr_len = T6_ETH_HDR_LEN_G(ntohl(req->hdr_len)); + if (eth_hdr_len == ETH_HLEN) { + eh = (struct ethhdr *)(req + 1); + iph = (struct iphdr *)(eh + 1); + ip6h = (struct ipv6hdr *)(eh + 1); + network_hdr = (void *)(eh + 1); + } else { + vlan_eh = (struct vlan_ethhdr *)(req + 1); + iph = (struct iphdr *)(vlan_eh + 1); + ip6h = (struct ipv6hdr *)(vlan_eh + 1); + network_hdr = (void *)(vlan_eh + 1); + } + + if (iph->version == 0x4) { + tcph = (struct tcphdr *)(iph + 1); + skb_set_network_header(skb, (void *)iph - (void *)req); + oreq = inet_reqsk_alloc(&chtls_rsk_ops, sk, true); + } else { + tcph = (struct tcphdr *)(ip6h + 1); + skb_set_network_header(skb, (void *)ip6h - (void *)req); + oreq = inet_reqsk_alloc(&chtls_rsk_opsv6, sk, false); + } + + if (!oreq) + goto reject; + + oreq->rsk_rcv_wnd = 0; + oreq->rsk_window_clamp = 0; + oreq->syncookie = 0; + oreq->mss = 0; + oreq->ts_recent = 0; + + tcp_rsk(oreq)->tfo_listener = false; + tcp_rsk(oreq)->rcv_isn = ntohl(tcph->seq); + chtls_set_req_port(oreq, tcph->source, tcph->dest); + if (iph->version == 0x4) { + chtls_set_req_addr(oreq, iph->daddr, iph->saddr); + ip_dsfield = ipv4_get_dsfield(iph); +#if IS_ENABLED(CONFIG_IPV6) + } else { + inet_rsk(oreq)->ir_v6_rmt_addr = ipv6_hdr(skb)->saddr; + inet_rsk(oreq)->ir_v6_loc_addr = ipv6_hdr(skb)->daddr; + ip_dsfield = ipv6_get_dsfield(ipv6_hdr(skb)); +#endif + } + if (req->tcpopt.wsf <= 14 && + sock_net(sk)->ipv4.sysctl_tcp_window_scaling) { + inet_rsk(oreq)->wscale_ok = 1; + inet_rsk(oreq)->snd_wscale = req->tcpopt.wsf; + } + inet_rsk(oreq)->ir_iif = sk->sk_bound_dev_if; + th_ecn = tcph->ece && tcph->cwr; + if (th_ecn) { + ect = !INET_ECN_is_not_ect(ip_dsfield); + ecn_ok = sock_net(sk)->ipv4.sysctl_tcp_ecn; + if ((!ect && ecn_ok) || tcp_ca_needs_ecn(sk)) + inet_rsk(oreq)->ecn_ok = 1; + } + + newsk = chtls_recv_sock(sk, oreq, network_hdr, req, cdev); + if (!newsk) + goto free_oreq; + + if (chtls_get_module(newsk)) + goto reject; + inet_csk_reqsk_queue_added(sk); + reply_skb->sk = newsk; + chtls_install_cpl_ops(newsk); + cxgb4_insert_tid(cdev->tids, newsk, tid, newsk->sk_family); + csk = rcu_dereference_sk_user_data(newsk); + listen_ctx = (struct listen_ctx *)lookup_stid(cdev->tids, stid); + csk->listen_ctx = listen_ctx; + __skb_queue_tail(&listen_ctx->synq, (struct sk_buff *)&csk->synq); + chtls_pass_accept_rpl(reply_skb, req, tid); + kfree_skb(skb); + return; + +free_oreq: + chtls_reqsk_free(oreq); +reject: + mk_tid_release(reply_skb, 0, tid); + cxgb4_ofld_send(cdev->lldi->ports[0], reply_skb); + kfree_skb(skb); +} + +/* + * Handle a CPL_PASS_ACCEPT_REQ message. + */ +static int chtls_pass_accept_req(struct chtls_dev *cdev, struct sk_buff *skb) +{ + struct cpl_pass_accept_req *req = cplhdr(skb) + RSS_HDR; + struct listen_ctx *ctx; + unsigned int stid; + unsigned int tid; + struct sock *lsk; + void *data; + + stid = PASS_OPEN_TID_G(ntohl(req->tos_stid)); + tid = GET_TID(req); + + data = lookup_stid(cdev->tids, stid); + if (!data) + return 1; + + ctx = (struct listen_ctx *)data; + lsk = ctx->lsk; + + if (unlikely(tid_out_of_range(cdev->tids, tid))) { + pr_info("passive open TID %u too large\n", tid); + return 1; + } + + BLOG_SKB_CB(skb)->cdev = cdev; + process_cpl_msg(chtls_pass_accept_request, lsk, skb); + return 0; +} + +/* + * Completes some final bits of initialization for just established connections + * and changes their state to TCP_ESTABLISHED. + * + * snd_isn here is the ISN after the SYN, i.e., the true ISN + 1. + */ +static void make_established(struct sock *sk, u32 snd_isn, unsigned int opt) +{ + struct tcp_sock *tp = tcp_sk(sk); + + tp->pushed_seq = snd_isn; + tp->write_seq = snd_isn; + tp->snd_nxt = snd_isn; + tp->snd_una = snd_isn; + inet_sk(sk)->inet_id = prandom_u32(); + assign_rxopt(sk, opt); + + if (tp->rcv_wnd > (RCV_BUFSIZ_M << 10)) + tp->rcv_wup -= tp->rcv_wnd - (RCV_BUFSIZ_M << 10); + + smp_mb(); + tcp_set_state(sk, TCP_ESTABLISHED); +} + +static void chtls_abort_conn(struct sock *sk, struct sk_buff *skb) +{ + struct sk_buff *abort_skb; + + abort_skb = alloc_skb(sizeof(struct cpl_abort_req), GFP_ATOMIC); + if (abort_skb) + chtls_send_reset(sk, CPL_ABORT_SEND_RST, abort_skb); +} + +static struct sock *reap_list; +static DEFINE_SPINLOCK(reap_list_lock); + +/* + * Process the reap list. + */ +DECLARE_TASK_FUNC(process_reap_list, task_param) +{ + spin_lock_bh(&reap_list_lock); + while (reap_list) { + struct sock *sk = reap_list; + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + + reap_list = csk->passive_reap_next; + csk->passive_reap_next = NULL; + spin_unlock(&reap_list_lock); + sock_hold(sk); + + bh_lock_sock(sk); + chtls_abort_conn(sk, NULL); + sock_orphan(sk); + if (sk->sk_state == TCP_CLOSE) + inet_csk_destroy_sock(sk); + bh_unlock_sock(sk); + sock_put(sk); + spin_lock(&reap_list_lock); + } + spin_unlock_bh(&reap_list_lock); +} + +static DECLARE_WORK(reap_task, process_reap_list); + +static void add_to_reap_list(struct sock *sk) +{ + struct chtls_sock *csk = sk->sk_user_data; + + local_bh_disable(); + bh_lock_sock(sk); + release_tcp_port(sk); /* release the port immediately */ + + spin_lock(&reap_list_lock); + csk->passive_reap_next = reap_list; + reap_list = sk; + if (!csk->passive_reap_next) + schedule_work(&reap_task); + spin_unlock(&reap_list_lock); + bh_unlock_sock(sk); + local_bh_enable(); +} + +static void add_pass_open_to_parent(struct sock *child, struct sock *lsk, + struct chtls_dev *cdev) +{ + struct request_sock *oreq; + struct chtls_sock *csk; + + if (lsk->sk_state != TCP_LISTEN) + return; + + csk = child->sk_user_data; + oreq = csk->passive_reap_next; + csk->passive_reap_next = NULL; + + reqsk_queue_removed(&inet_csk(lsk)->icsk_accept_queue, oreq); + __skb_unlink((struct sk_buff *)&csk->synq, &csk->listen_ctx->synq); + + if (sk_acceptq_is_full(lsk)) { + chtls_reqsk_free(oreq); + add_to_reap_list(child); + } else { + refcount_set(&oreq->rsk_refcnt, 1); + inet_csk_reqsk_queue_add(lsk, oreq, child); + lsk->sk_data_ready(lsk); + } +} + +static void bl_add_pass_open_to_parent(struct sock *lsk, struct sk_buff *skb) +{ + struct sock *child = skb->sk; + + skb->sk = NULL; + add_pass_open_to_parent(child, lsk, BLOG_SKB_CB(skb)->cdev); + kfree_skb(skb); +} + +static int chtls_pass_establish(struct chtls_dev *cdev, struct sk_buff *skb) +{ + struct cpl_pass_establish *req = cplhdr(skb) + RSS_HDR; + struct chtls_sock *csk; + struct sock *lsk, *sk; + unsigned int hwtid; + + hwtid = GET_TID(req); + sk = lookup_tid(cdev->tids, hwtid); + if (!sk) + return (CPL_RET_UNKNOWN_TID | CPL_RET_BUF_DONE); + + bh_lock_sock(sk); + if (unlikely(sock_owned_by_user(sk))) { + kfree_skb(skb); + } else { + unsigned int stid; + void *data; + + csk = sk->sk_user_data; + csk->wr_max_credits = 64; + csk->wr_credits = 64; + csk->wr_unacked = 0; + make_established(sk, ntohl(req->snd_isn), ntohs(req->tcp_opt)); + stid = PASS_OPEN_TID_G(ntohl(req->tos_stid)); + sk->sk_state_change(sk); + if (unlikely(sk->sk_socket)) + sk_wake_async(sk, 0, POLL_OUT); + + data = lookup_stid(cdev->tids, stid); + lsk = ((struct listen_ctx *)data)->lsk; + + bh_lock_sock(lsk); + if (unlikely(skb_queue_empty(&csk->listen_ctx->synq))) { + /* removed from synq */ + bh_unlock_sock(lsk); + kfree_skb(skb); + goto unlock; + } + + if (likely(!sock_owned_by_user(lsk))) { + kfree_skb(skb); + add_pass_open_to_parent(sk, lsk, cdev); + } else { + skb->sk = sk; + BLOG_SKB_CB(skb)->cdev = cdev; + BLOG_SKB_CB(skb)->backlog_rcv = + bl_add_pass_open_to_parent; + __sk_add_backlog(lsk, skb); + } + bh_unlock_sock(lsk); + } +unlock: + bh_unlock_sock(sk); + return 0; +} + +/* + * Handle receipt of an urgent pointer. + */ +static void handle_urg_ptr(struct sock *sk, u32 urg_seq) +{ + struct tcp_sock *tp = tcp_sk(sk); + + urg_seq--; + if (tp->urg_data && !after(urg_seq, tp->urg_seq)) + return; /* duplicate pointer */ + + sk_send_sigurg(sk); + if (tp->urg_seq == tp->copied_seq && tp->urg_data && + !sock_flag(sk, SOCK_URGINLINE) && + tp->copied_seq != tp->rcv_nxt) { + struct sk_buff *skb = skb_peek(&sk->sk_receive_queue); + + tp->copied_seq++; + if (skb && tp->copied_seq - ULP_SKB_CB(skb)->seq >= skb->len) + chtls_free_skb(sk, skb); + } + + tp->urg_data = TCP_URG_NOTYET; + tp->urg_seq = urg_seq; +} + +static void check_sk_callbacks(struct chtls_sock *csk) +{ + struct sock *sk = csk->sk; + + if (unlikely(sk->sk_user_data && + !csk_flag_nochk(csk, CSK_CALLBACKS_CHKD))) + csk_set_flag(csk, CSK_CALLBACKS_CHKD); +} + +/* + * Handles Rx data that arrives in a state where the socket isn't accepting + * new data. + */ +static void handle_excess_rx(struct sock *sk, struct sk_buff *skb) +{ + if (!csk_flag(sk, CSK_ABORT_SHUTDOWN)) + chtls_abort_conn(sk, skb); + + kfree_skb(skb); +} + +static void chtls_recv_data(struct sock *sk, struct sk_buff *skb) +{ + struct cpl_rx_data *hdr = cplhdr(skb) + RSS_HDR; + struct chtls_sock *csk; + struct tcp_sock *tp; + + csk = rcu_dereference_sk_user_data(sk); + tp = tcp_sk(sk); + + if (unlikely(sk->sk_shutdown & RCV_SHUTDOWN)) { + handle_excess_rx(sk, skb); + return; + } + + ULP_SKB_CB(skb)->seq = ntohl(hdr->seq); + ULP_SKB_CB(skb)->psh = hdr->psh; + skb_ulp_mode(skb) = ULP_MODE_NONE; + + skb_reset_transport_header(skb); + __skb_pull(skb, sizeof(*hdr) + RSS_HDR); + if (!skb->data_len) + __skb_trim(skb, ntohs(hdr->len)); + + if (unlikely(hdr->urg)) + handle_urg_ptr(sk, tp->rcv_nxt + ntohs(hdr->urg)); + if (unlikely(tp->urg_data == TCP_URG_NOTYET && + tp->urg_seq - tp->rcv_nxt < skb->len)) + tp->urg_data = TCP_URG_VALID | + skb->data[tp->urg_seq - tp->rcv_nxt]; + + if (unlikely(hdr->dack_mode != csk->delack_mode)) { + csk->delack_mode = hdr->dack_mode; + csk->delack_seq = tp->rcv_nxt; + } + + tcp_hdr(skb)->fin = 0; + tp->rcv_nxt += skb->len; + + __skb_queue_tail(&sk->sk_receive_queue, skb); + + if (!sock_flag(sk, SOCK_DEAD)) { + check_sk_callbacks(csk); + sk->sk_data_ready(sk); + } +} + +static int chtls_rx_data(struct chtls_dev *cdev, struct sk_buff *skb) +{ + struct cpl_rx_data *req = cplhdr(skb) + RSS_HDR; + unsigned int hwtid = GET_TID(req); + struct sock *sk; + + sk = lookup_tid(cdev->tids, hwtid); + if (unlikely(!sk)) { + pr_err("can't find conn. for hwtid %u.\n", hwtid); + return -EINVAL; + } + skb_dst_set(skb, NULL); + process_cpl_msg(chtls_recv_data, sk, skb); + return 0; +} + +static void chtls_recv_pdu(struct sock *sk, struct sk_buff *skb) +{ + struct cpl_tls_data *hdr = cplhdr(skb); + struct chtls_sock *csk; + struct chtls_hws *tlsk; + struct tcp_sock *tp; + + csk = rcu_dereference_sk_user_data(sk); + tlsk = &csk->tlshws; + tp = tcp_sk(sk); + + if (unlikely(sk->sk_shutdown & RCV_SHUTDOWN)) { + handle_excess_rx(sk, skb); + return; + } + + ULP_SKB_CB(skb)->seq = ntohl(hdr->seq); + ULP_SKB_CB(skb)->flags = 0; + skb_ulp_mode(skb) = ULP_MODE_TLS; + + skb_reset_transport_header(skb); + __skb_pull(skb, sizeof(*hdr)); + if (!skb->data_len) + __skb_trim(skb, + CPL_TLS_DATA_LENGTH_G(ntohl(hdr->length_pkd))); + + if (unlikely(tp->urg_data == TCP_URG_NOTYET && tp->urg_seq - + tp->rcv_nxt < skb->len)) + tp->urg_data = TCP_URG_VALID | + skb->data[tp->urg_seq - tp->rcv_nxt]; + + tcp_hdr(skb)->fin = 0; + tlsk->pldlen = CPL_TLS_DATA_LENGTH_G(ntohl(hdr->length_pkd)); + __skb_queue_tail(&tlsk->sk_recv_queue, skb); +} + +static int chtls_rx_pdu(struct chtls_dev *cdev, struct sk_buff *skb) +{ + struct cpl_tls_data *req = cplhdr(skb); + unsigned int hwtid = GET_TID(req); + struct sock *sk; + + sk = lookup_tid(cdev->tids, hwtid); + if (unlikely(!sk)) { + pr_err("can't find conn. for hwtid %u.\n", hwtid); + return -EINVAL; + } + skb_dst_set(skb, NULL); + process_cpl_msg(chtls_recv_pdu, sk, skb); + return 0; +} + +static void chtls_set_hdrlen(struct sk_buff *skb, unsigned int nlen) +{ + struct tlsrx_cmp_hdr *tls_cmp_hdr = cplhdr(skb); + + skb->hdr_len = ntohs((__force __be16)tls_cmp_hdr->length); + tls_cmp_hdr->length = ntohs((__force __be16)nlen); +} + +static void chtls_rx_hdr(struct sock *sk, struct sk_buff *skb) +{ + struct tlsrx_cmp_hdr *tls_hdr_pkt; + struct cpl_rx_tls_cmp *cmp_cpl; + struct sk_buff *skb_rec; + struct chtls_sock *csk; + struct chtls_hws *tlsk; + struct tcp_sock *tp; + + cmp_cpl = cplhdr(skb); + csk = rcu_dereference_sk_user_data(sk); + tlsk = &csk->tlshws; + tp = tcp_sk(sk); + + ULP_SKB_CB(skb)->seq = ntohl(cmp_cpl->seq); + ULP_SKB_CB(skb)->flags = 0; + + skb_reset_transport_header(skb); + __skb_pull(skb, sizeof(*cmp_cpl)); + tls_hdr_pkt = (struct tlsrx_cmp_hdr *)skb->data; + if (tls_hdr_pkt->res_to_mac_error & TLSRX_HDR_PKT_ERROR_M) + tls_hdr_pkt->type = CONTENT_TYPE_ERROR; + if (!skb->data_len) + __skb_trim(skb, TLS_HEADER_LENGTH); + + tp->rcv_nxt += + CPL_RX_TLS_CMP_PDULENGTH_G(ntohl(cmp_cpl->pdulength_length)); + + ULP_SKB_CB(skb)->flags |= ULPCB_FLAG_TLS_HDR; + skb_rec = __skb_dequeue(&tlsk->sk_recv_queue); + if (!skb_rec) { + __skb_queue_tail(&sk->sk_receive_queue, skb); + } else { + chtls_set_hdrlen(skb, tlsk->pldlen); + tlsk->pldlen = 0; + __skb_queue_tail(&sk->sk_receive_queue, skb); + __skb_queue_tail(&sk->sk_receive_queue, skb_rec); + } + + if (!sock_flag(sk, SOCK_DEAD)) { + check_sk_callbacks(csk); + sk->sk_data_ready(sk); + } +} + +static int chtls_rx_cmp(struct chtls_dev *cdev, struct sk_buff *skb) +{ + struct cpl_rx_tls_cmp *req = cplhdr(skb); + unsigned int hwtid = GET_TID(req); + struct sock *sk; + + sk = lookup_tid(cdev->tids, hwtid); + if (unlikely(!sk)) { + pr_err("can't find conn. for hwtid %u.\n", hwtid); + return -EINVAL; + } + skb_dst_set(skb, NULL); + process_cpl_msg(chtls_rx_hdr, sk, skb); + + return 0; +} + +static void chtls_timewait(struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); + + tp->rcv_nxt++; + tp->rx_opt.ts_recent_stamp = ktime_get_seconds(); + tp->srtt_us = 0; + tcp_time_wait(sk, TCP_TIME_WAIT, 0); +} + +static void chtls_peer_close(struct sock *sk, struct sk_buff *skb) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + + if (csk_flag_nochk(csk, CSK_ABORT_RPL_PENDING)) + goto out; + + sk->sk_shutdown |= RCV_SHUTDOWN; + sock_set_flag(sk, SOCK_DONE); + + switch (sk->sk_state) { + case TCP_SYN_RECV: + case TCP_ESTABLISHED: + tcp_set_state(sk, TCP_CLOSE_WAIT); + break; + case TCP_FIN_WAIT1: + tcp_set_state(sk, TCP_CLOSING); + break; + case TCP_FIN_WAIT2: + chtls_release_resources(sk); + if (csk_flag_nochk(csk, CSK_ABORT_RPL_PENDING)) + chtls_conn_done(sk); + else + chtls_timewait(sk); + break; + default: + pr_info("cpl_peer_close in bad state %d\n", sk->sk_state); + } + + if (!sock_flag(sk, SOCK_DEAD)) { + sk->sk_state_change(sk); + /* Do not send POLL_HUP for half duplex close. */ + + if ((sk->sk_shutdown & SEND_SHUTDOWN) || + sk->sk_state == TCP_CLOSE) + sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_HUP); + else + sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN); + } +out: + kfree_skb(skb); +} + +static void chtls_close_con_rpl(struct sock *sk, struct sk_buff *skb) +{ + struct cpl_close_con_rpl *rpl = cplhdr(skb) + RSS_HDR; + struct chtls_sock *csk; + struct tcp_sock *tp; + + csk = rcu_dereference_sk_user_data(sk); + + if (csk_flag_nochk(csk, CSK_ABORT_RPL_PENDING)) + goto out; + + tp = tcp_sk(sk); + + tp->snd_una = ntohl(rpl->snd_nxt) - 1; /* exclude FIN */ + + switch (sk->sk_state) { + case TCP_CLOSING: + chtls_release_resources(sk); + if (csk_flag_nochk(csk, CSK_ABORT_RPL_PENDING)) + chtls_conn_done(sk); + else + chtls_timewait(sk); + break; + case TCP_LAST_ACK: + chtls_release_resources(sk); + chtls_conn_done(sk); + break; + case TCP_FIN_WAIT1: + tcp_set_state(sk, TCP_FIN_WAIT2); + sk->sk_shutdown |= SEND_SHUTDOWN; + + if (!sock_flag(sk, SOCK_DEAD)) + sk->sk_state_change(sk); + else if (tcp_sk(sk)->linger2 < 0 && + !csk_flag_nochk(csk, CSK_ABORT_SHUTDOWN)) + chtls_abort_conn(sk, skb); + break; + default: + pr_info("close_con_rpl in bad state %d\n", sk->sk_state); + } +out: + kfree_skb(skb); +} + +static struct sk_buff *get_cpl_skb(struct sk_buff *skb, + size_t len, gfp_t gfp) +{ + if (likely(!skb_is_nonlinear(skb) && !skb_cloned(skb))) { + WARN_ONCE(skb->len < len, "skb alloc error"); + __skb_trim(skb, len); + skb_get(skb); + } else { + skb = alloc_skb(len, gfp); + if (skb) + __skb_put(skb, len); + } + return skb; +} + +static void set_abort_rpl_wr(struct sk_buff *skb, unsigned int tid, + int cmd) +{ + struct cpl_abort_rpl *rpl = cplhdr(skb); + + INIT_TP_WR_CPL(rpl, CPL_ABORT_RPL, tid); + rpl->cmd = cmd; +} + +static void send_defer_abort_rpl(struct chtls_dev *cdev, struct sk_buff *skb) +{ + struct cpl_abort_req_rss *req = cplhdr(skb); + struct sk_buff *reply_skb; + + reply_skb = alloc_skb(sizeof(struct cpl_abort_rpl), + GFP_KERNEL | __GFP_NOFAIL); + __skb_put(reply_skb, sizeof(struct cpl_abort_rpl)); + set_abort_rpl_wr(reply_skb, GET_TID(req), + (req->status & CPL_ABORT_NO_RST)); + set_wr_txq(reply_skb, CPL_PRIORITY_DATA, req->status >> 1); + cxgb4_ofld_send(cdev->lldi->ports[0], reply_skb); + kfree_skb(skb); +} + +/* + * Add an skb to the deferred skb queue for processing from process context. + */ +static void t4_defer_reply(struct sk_buff *skb, struct chtls_dev *cdev, + defer_handler_t handler) +{ + DEFERRED_SKB_CB(skb)->handler = handler; + spin_lock_bh(&cdev->deferq.lock); + __skb_queue_tail(&cdev->deferq, skb); + if (skb_queue_len(&cdev->deferq) == 1) + schedule_work(&cdev->deferq_task); + spin_unlock_bh(&cdev->deferq.lock); +} + +static void send_abort_rpl(struct sock *sk, struct sk_buff *skb, + struct chtls_dev *cdev, int status, int queue) +{ + struct cpl_abort_req_rss *req = cplhdr(skb); + struct sk_buff *reply_skb; + struct chtls_sock *csk; + + csk = rcu_dereference_sk_user_data(sk); + + reply_skb = alloc_skb(sizeof(struct cpl_abort_rpl), + GFP_KERNEL); + + if (!reply_skb) { + req->status = (queue << 1); + t4_defer_reply(skb, cdev, send_defer_abort_rpl); + return; + } + + set_abort_rpl_wr(reply_skb, GET_TID(req), status); + kfree_skb(skb); + + set_wr_txq(reply_skb, CPL_PRIORITY_DATA, queue); + if (csk_conn_inline(csk)) { + struct l2t_entry *e = csk->l2t_entry; + + if (e && sk->sk_state != TCP_SYN_RECV) { + cxgb4_l2t_send(csk->egress_dev, reply_skb, e); + return; + } + } + cxgb4_ofld_send(cdev->lldi->ports[0], reply_skb); +} + +static void chtls_send_abort_rpl(struct sock *sk, struct sk_buff *skb, + struct chtls_dev *cdev, + int status, int queue) +{ + struct cpl_abort_req_rss *req = cplhdr(skb) + RSS_HDR; + struct sk_buff *reply_skb; + struct chtls_sock *csk; + unsigned int tid; + + csk = rcu_dereference_sk_user_data(sk); + tid = GET_TID(req); + + reply_skb = get_cpl_skb(skb, sizeof(struct cpl_abort_rpl), gfp_any()); + if (!reply_skb) { + req->status = (queue << 1) | status; + t4_defer_reply(skb, cdev, send_defer_abort_rpl); + return; + } + + set_abort_rpl_wr(reply_skb, tid, status); + kfree_skb(skb); + set_wr_txq(reply_skb, CPL_PRIORITY_DATA, queue); + if (csk_conn_inline(csk)) { + struct l2t_entry *e = csk->l2t_entry; + + if (e && sk->sk_state != TCP_SYN_RECV) { + cxgb4_l2t_send(csk->egress_dev, reply_skb, e); + return; + } + } + cxgb4_ofld_send(cdev->lldi->ports[0], reply_skb); +} + +/* + * This is run from a listener's backlog to abort a child connection in + * SYN_RCV state (i.e., one on the listener's SYN queue). + */ +static void bl_abort_syn_rcv(struct sock *lsk, struct sk_buff *skb) +{ + struct chtls_sock *csk; + struct sock *child; + int queue; + + child = skb->sk; + csk = rcu_dereference_sk_user_data(child); + queue = csk->txq_idx; + + skb->sk = NULL; + do_abort_syn_rcv(child, lsk); + send_abort_rpl(child, skb, BLOG_SKB_CB(skb)->cdev, + CPL_ABORT_NO_RST, queue); +} + +static int abort_syn_rcv(struct sock *sk, struct sk_buff *skb) +{ + const struct request_sock *oreq; + struct listen_ctx *listen_ctx; + struct chtls_sock *csk; + struct chtls_dev *cdev; + struct sock *psk; + void *ctx; + + csk = sk->sk_user_data; + oreq = csk->passive_reap_next; + cdev = csk->cdev; + + if (!oreq) + return -1; + + ctx = lookup_stid(cdev->tids, oreq->ts_recent); + if (!ctx) + return -1; + + listen_ctx = (struct listen_ctx *)ctx; + psk = listen_ctx->lsk; + + bh_lock_sock(psk); + if (!sock_owned_by_user(psk)) { + int queue = csk->txq_idx; + + do_abort_syn_rcv(sk, psk); + send_abort_rpl(sk, skb, cdev, CPL_ABORT_NO_RST, queue); + } else { + skb->sk = sk; + BLOG_SKB_CB(skb)->backlog_rcv = bl_abort_syn_rcv; + __sk_add_backlog(psk, skb); + } + bh_unlock_sock(psk); + return 0; +} + +static void chtls_abort_req_rss(struct sock *sk, struct sk_buff *skb) +{ + const struct cpl_abort_req_rss *req = cplhdr(skb) + RSS_HDR; + struct chtls_sock *csk = sk->sk_user_data; + int rst_status = CPL_ABORT_NO_RST; + int queue = csk->txq_idx; + + if (is_neg_adv(req->status)) { + if (sk->sk_state == TCP_SYN_RECV) + chtls_set_tcb_tflag(sk, 0, 0); + + kfree_skb(skb); + return; + } + + csk_reset_flag(csk, CSK_ABORT_REQ_RCVD); + + if (!csk_flag_nochk(csk, CSK_ABORT_SHUTDOWN) && + !csk_flag_nochk(csk, CSK_TX_DATA_SENT)) { + struct tcp_sock *tp = tcp_sk(sk); + + if (send_tx_flowc_wr(sk, 0, tp->snd_nxt, tp->rcv_nxt) < 0) + WARN_ONCE(1, "send_tx_flowc error"); + csk_set_flag(csk, CSK_TX_DATA_SENT); + } + + csk_set_flag(csk, CSK_ABORT_SHUTDOWN); + + if (!csk_flag_nochk(csk, CSK_ABORT_RPL_PENDING)) { + sk->sk_err = ETIMEDOUT; + + if (!sock_flag(sk, SOCK_DEAD)) + sk->sk_error_report(sk); + + if (sk->sk_state == TCP_SYN_RECV && !abort_syn_rcv(sk, skb)) + return; + + chtls_release_resources(sk); + chtls_conn_done(sk); + } + + chtls_send_abort_rpl(sk, skb, BLOG_SKB_CB(skb)->cdev, + rst_status, queue); +} + +static void chtls_abort_rpl_rss(struct sock *sk, struct sk_buff *skb) +{ + struct cpl_abort_rpl_rss *rpl = cplhdr(skb) + RSS_HDR; + struct chtls_sock *csk; + struct chtls_dev *cdev; + + csk = rcu_dereference_sk_user_data(sk); + cdev = csk->cdev; + + if (csk_flag_nochk(csk, CSK_ABORT_RPL_PENDING)) { + csk_reset_flag(csk, CSK_ABORT_RPL_PENDING); + if (!csk_flag_nochk(csk, CSK_ABORT_REQ_RCVD)) { + if (sk->sk_state == TCP_SYN_SENT) { + cxgb4_remove_tid(cdev->tids, + csk->port_id, + GET_TID(rpl), + sk->sk_family); + sock_put(sk); + } + chtls_release_resources(sk); + chtls_conn_done(sk); + } + } + kfree_skb(skb); +} + +static int chtls_conn_cpl(struct chtls_dev *cdev, struct sk_buff *skb) +{ + struct cpl_peer_close *req = cplhdr(skb) + RSS_HDR; + void (*fn)(struct sock *sk, struct sk_buff *skb); + unsigned int hwtid = GET_TID(req); + struct chtls_sock *csk; + struct sock *sk; + u8 opcode; + + opcode = ((const struct rss_header *)cplhdr(skb))->opcode; + + sk = lookup_tid(cdev->tids, hwtid); + if (!sk) + goto rel_skb; + + csk = sk->sk_user_data; + + switch (opcode) { + case CPL_PEER_CLOSE: + fn = chtls_peer_close; + break; + case CPL_CLOSE_CON_RPL: + fn = chtls_close_con_rpl; + break; + case CPL_ABORT_REQ_RSS: + /* + * Save the offload device in the skb, we may process this + * message after the socket has closed. + */ + BLOG_SKB_CB(skb)->cdev = csk->cdev; + fn = chtls_abort_req_rss; + break; + case CPL_ABORT_RPL_RSS: + fn = chtls_abort_rpl_rss; + break; + default: + goto rel_skb; + } + + process_cpl_msg(fn, sk, skb); + return 0; + +rel_skb: + kfree_skb(skb); + return 0; +} + +static void chtls_rx_ack(struct sock *sk, struct sk_buff *skb) +{ + struct cpl_fw4_ack *hdr = cplhdr(skb) + RSS_HDR; + struct chtls_sock *csk = sk->sk_user_data; + struct tcp_sock *tp = tcp_sk(sk); + u32 credits = hdr->credits; + u32 snd_una; + + snd_una = ntohl(hdr->snd_una); + csk->wr_credits += credits; + + if (csk->wr_unacked > csk->wr_max_credits - csk->wr_credits) + csk->wr_unacked = csk->wr_max_credits - csk->wr_credits; + + while (credits) { + struct sk_buff *pskb = csk->wr_skb_head; + u32 csum; + + if (unlikely(!pskb)) { + if (csk->wr_nondata) + csk->wr_nondata -= credits; + break; + } + csum = (__force u32)pskb->csum; + if (unlikely(credits < csum)) { + pskb->csum = (__force __wsum)(csum - credits); + break; + } + dequeue_wr(sk); + credits -= csum; + kfree_skb(pskb); + } + if (hdr->seq_vld & CPL_FW4_ACK_FLAGS_SEQVAL) { + if (unlikely(before(snd_una, tp->snd_una))) { + kfree_skb(skb); + return; + } + + if (tp->snd_una != snd_una) { + tp->snd_una = snd_una; + tp->rcv_tstamp = tcp_time_stamp(tp); + if (tp->snd_una == tp->snd_nxt && + !csk_flag_nochk(csk, CSK_TX_FAILOVER)) + csk_reset_flag(csk, CSK_TX_WAIT_IDLE); + } + } + + if (hdr->seq_vld & CPL_FW4_ACK_FLAGS_CH) { + unsigned int fclen16 = roundup(failover_flowc_wr_len, 16); + + csk->wr_credits -= fclen16; + csk_reset_flag(csk, CSK_TX_WAIT_IDLE); + csk_reset_flag(csk, CSK_TX_FAILOVER); + } + if (skb_queue_len(&csk->txq) && chtls_push_frames(csk, 0)) + sk->sk_write_space(sk); + + kfree_skb(skb); +} + +static int chtls_wr_ack(struct chtls_dev *cdev, struct sk_buff *skb) +{ + struct cpl_fw4_ack *rpl = cplhdr(skb) + RSS_HDR; + unsigned int hwtid = GET_TID(rpl); + struct sock *sk; + + sk = lookup_tid(cdev->tids, hwtid); + if (unlikely(!sk)) { + pr_err("can't find conn. for hwtid %u.\n", hwtid); + return -EINVAL; + } + process_cpl_msg(chtls_rx_ack, sk, skb); + + return 0; +} + +chtls_handler_func chtls_handlers[NUM_CPL_CMDS] = { + [CPL_PASS_OPEN_RPL] = chtls_pass_open_rpl, + [CPL_CLOSE_LISTSRV_RPL] = chtls_close_listsrv_rpl, + [CPL_PASS_ACCEPT_REQ] = chtls_pass_accept_req, + [CPL_PASS_ESTABLISH] = chtls_pass_establish, + [CPL_RX_DATA] = chtls_rx_data, + [CPL_TLS_DATA] = chtls_rx_pdu, + [CPL_RX_TLS_CMP] = chtls_rx_cmp, + [CPL_PEER_CLOSE] = chtls_conn_cpl, + [CPL_CLOSE_CON_RPL] = chtls_conn_cpl, + [CPL_ABORT_REQ_RSS] = chtls_conn_cpl, + [CPL_ABORT_RPL_RSS] = chtls_conn_cpl, + [CPL_FW4_ACK] = chtls_wr_ack, +}; diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.h b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.h new file mode 100644 index 000000000000..47ba81e42f5d --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.h @@ -0,0 +1,222 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2018 Chelsio Communications, Inc. + */ + +#ifndef __CHTLS_CM_H__ +#define __CHTLS_CM_H__ + +/* + * TCB settings + */ +/* 3:0 */ +#define TCB_ULP_TYPE_W 0 +#define TCB_ULP_TYPE_S 0 +#define TCB_ULP_TYPE_M 0xfULL +#define TCB_ULP_TYPE_V(x) ((x) << TCB_ULP_TYPE_S) + +/* 11:4 */ +#define TCB_ULP_RAW_W 0 +#define TCB_ULP_RAW_S 4 +#define TCB_ULP_RAW_M 0xffULL +#define TCB_ULP_RAW_V(x) ((x) << TCB_ULP_RAW_S) + +#define TF_TLS_KEY_SIZE_S 7 +#define TF_TLS_KEY_SIZE_V(x) ((x) << TF_TLS_KEY_SIZE_S) + +#define TF_TLS_CONTROL_S 2 +#define TF_TLS_CONTROL_V(x) ((x) << TF_TLS_CONTROL_S) + +#define TF_TLS_ACTIVE_S 1 +#define TF_TLS_ACTIVE_V(x) ((x) << TF_TLS_ACTIVE_S) + +#define TF_TLS_ENABLE_S 0 +#define TF_TLS_ENABLE_V(x) ((x) << TF_TLS_ENABLE_S) + +#define TF_RX_QUIESCE_S 15 +#define TF_RX_QUIESCE_V(x) ((x) << TF_RX_QUIESCE_S) + +/* + * Max receive window supported by HW in bytes. Only a small part of it can + * be set through option0, the rest needs to be set through RX_DATA_ACK. + */ +#define MAX_RCV_WND ((1U << 27) - 1) +#define MAX_MSS 65536 + +/* + * Min receive window. We want it to be large enough to accommodate receive + * coalescing, handle jumbo frames, and not trigger sender SWS avoidance. + */ +#define MIN_RCV_WND (24 * 1024U) +#define LOOPBACK(x) (((x) & htonl(0xff000000)) == htonl(0x7f000000)) + +/* ulp_mem_io + ulptx_idata + payload + padding */ +#define MAX_IMM_ULPTX_WR_LEN (32 + 8 + 256 + 8) + +/* for TX: a skb must have a headroom of at least TX_HEADER_LEN bytes */ +#define TX_HEADER_LEN \ + (sizeof(struct fw_ofld_tx_data_wr) + sizeof(struct sge_opaque_hdr)) +#define TX_TLSHDR_LEN \ + (sizeof(struct fw_tlstx_data_wr) + sizeof(struct cpl_tx_tls_sfo) + \ + sizeof(struct sge_opaque_hdr)) +#define TXDATA_SKB_LEN 128 + +enum { + CPL_TX_TLS_SFO_TYPE_CCS, + CPL_TX_TLS_SFO_TYPE_ALERT, + CPL_TX_TLS_SFO_TYPE_HANDSHAKE, + CPL_TX_TLS_SFO_TYPE_DATA, + CPL_TX_TLS_SFO_TYPE_HEARTBEAT, +}; + +enum { + TLS_HDR_TYPE_CCS = 20, + TLS_HDR_TYPE_ALERT, + TLS_HDR_TYPE_HANDSHAKE, + TLS_HDR_TYPE_RECORD, + TLS_HDR_TYPE_HEARTBEAT, +}; + +typedef void (*defer_handler_t)(struct chtls_dev *dev, struct sk_buff *skb); +extern struct request_sock_ops chtls_rsk_ops; +extern struct request_sock_ops chtls_rsk_opsv6; + +struct deferred_skb_cb { + defer_handler_t handler; + struct chtls_dev *dev; +}; + +#define DEFERRED_SKB_CB(skb) ((struct deferred_skb_cb *)(skb)->cb) +#define failover_flowc_wr_len offsetof(struct fw_flowc_wr, mnemval[3]) +#define WR_SKB_CB(skb) ((struct wr_skb_cb *)(skb)->cb) +#define ACCEPT_QUEUE(sk) (&inet_csk(sk)->icsk_accept_queue.rskq_accept_head) + +#define SND_WSCALE(tp) ((tp)->rx_opt.snd_wscale) +#define RCV_WSCALE(tp) ((tp)->rx_opt.rcv_wscale) +#define USER_MSS(tp) ((tp)->rx_opt.user_mss) +#define TS_RECENT_STAMP(tp) ((tp)->rx_opt.ts_recent_stamp) +#define WSCALE_OK(tp) ((tp)->rx_opt.wscale_ok) +#define TSTAMP_OK(tp) ((tp)->rx_opt.tstamp_ok) +#define SACK_OK(tp) ((tp)->rx_opt.sack_ok) +#define INC_ORPHAN_COUNT(sk) percpu_counter_inc((sk)->sk_prot->orphan_count) + +/* TLS SKB */ +#define skb_ulp_tls_inline(skb) (ULP_SKB_CB(skb)->ulp.tls.ofld) +#define skb_ulp_tls_iv_imm(skb) (ULP_SKB_CB(skb)->ulp.tls.iv) + +void chtls_defer_reply(struct sk_buff *skb, struct chtls_dev *dev, + defer_handler_t handler); + +/* + * Returns true if the socket is in one of the supplied states. + */ +static inline unsigned int sk_in_state(const struct sock *sk, + unsigned int states) +{ + return states & (1 << sk->sk_state); +} + +static void chtls_rsk_destructor(struct request_sock *req) +{ + /* do nothing */ +} + +static inline void chtls_init_rsk_ops(struct proto *chtls_tcp_prot, + struct request_sock_ops *chtls_tcp_ops, + struct proto *tcp_prot, int family) +{ + memset(chtls_tcp_ops, 0, sizeof(*chtls_tcp_ops)); + chtls_tcp_ops->family = family; + chtls_tcp_ops->obj_size = sizeof(struct tcp_request_sock); + chtls_tcp_ops->destructor = chtls_rsk_destructor; + chtls_tcp_ops->slab = tcp_prot->rsk_prot->slab; + chtls_tcp_prot->rsk_prot = chtls_tcp_ops; +} + +static inline void chtls_reqsk_free(struct request_sock *req) +{ + if (req->rsk_listener) + sock_put(req->rsk_listener); + kmem_cache_free(req->rsk_ops->slab, req); +} + +#define DECLARE_TASK_FUNC(task, task_param) \ + static void task(struct work_struct *task_param) + +static inline void sk_wakeup_sleepers(struct sock *sk, bool interruptable) +{ + struct socket_wq *wq; + + rcu_read_lock(); + wq = rcu_dereference(sk->sk_wq); + if (skwq_has_sleeper(wq)) { + if (interruptable) + wake_up_interruptible(sk_sleep(sk)); + else + wake_up_all(sk_sleep(sk)); + } + rcu_read_unlock(); +} + +static inline void chtls_set_req_port(struct request_sock *oreq, + __be16 source, __be16 dest) +{ + inet_rsk(oreq)->ir_rmt_port = source; + inet_rsk(oreq)->ir_num = ntohs(dest); +} + +static inline void chtls_set_req_addr(struct request_sock *oreq, + __be32 local_ip, __be32 peer_ip) +{ + inet_rsk(oreq)->ir_loc_addr = local_ip; + inet_rsk(oreq)->ir_rmt_addr = peer_ip; +} + +static inline void chtls_free_skb(struct sock *sk, struct sk_buff *skb) +{ + skb_dst_set(skb, NULL); + __skb_unlink(skb, &sk->sk_receive_queue); + __kfree_skb(skb); +} + +static inline void chtls_kfree_skb(struct sock *sk, struct sk_buff *skb) +{ + skb_dst_set(skb, NULL); + __skb_unlink(skb, &sk->sk_receive_queue); + kfree_skb(skb); +} + +static inline void chtls_reset_wr_list(struct chtls_sock *csk) +{ + csk->wr_skb_head = NULL; + csk->wr_skb_tail = NULL; +} + +static inline void enqueue_wr(struct chtls_sock *csk, struct sk_buff *skb) +{ + WR_SKB_CB(skb)->next_wr = NULL; + + skb_get(skb); + + if (!csk->wr_skb_head) + csk->wr_skb_head = skb; + else + WR_SKB_CB(csk->wr_skb_tail)->next_wr = skb; + csk->wr_skb_tail = skb; +} + +static inline struct sk_buff *dequeue_wr(struct sock *sk) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + struct sk_buff *skb = NULL; + + skb = csk->wr_skb_head; + + if (likely(skb)) { + /* Don't bother clearing the tail */ + csk->wr_skb_head = WR_SKB_CB(skb)->next_wr; + WR_SKB_CB(skb)->next_wr = NULL; + } + return skb; +} +#endif diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_hw.c b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_hw.c new file mode 100644 index 000000000000..f1820aca0d33 --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_hw.c @@ -0,0 +1,426 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2018 Chelsio Communications, Inc. + * + * Written by: Atul Gupta (atul.gupta@chelsio.com) + */ + +#include <linux/module.h> +#include <linux/list.h> +#include <linux/workqueue.h> +#include <linux/skbuff.h> +#include <linux/timer.h> +#include <linux/notifier.h> +#include <linux/inetdevice.h> +#include <linux/ip.h> +#include <linux/tcp.h> +#include <linux/tls.h> +#include <net/tls.h> + +#include "chtls.h" +#include "chtls_cm.h" + +static void __set_tcb_field_direct(struct chtls_sock *csk, + struct cpl_set_tcb_field *req, u16 word, + u64 mask, u64 val, u8 cookie, int no_reply) +{ + struct ulptx_idata *sc; + + INIT_TP_WR_CPL(req, CPL_SET_TCB_FIELD, csk->tid); + req->wr.wr_mid |= htonl(FW_WR_FLOWID_V(csk->tid)); + req->reply_ctrl = htons(NO_REPLY_V(no_reply) | + QUEUENO_V(csk->rss_qid)); + req->word_cookie = htons(TCB_WORD_V(word) | TCB_COOKIE_V(cookie)); + req->mask = cpu_to_be64(mask); + req->val = cpu_to_be64(val); + sc = (struct ulptx_idata *)(req + 1); + sc->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_NOOP)); + sc->len = htonl(0); +} + +static void __set_tcb_field(struct sock *sk, struct sk_buff *skb, u16 word, + u64 mask, u64 val, u8 cookie, int no_reply) +{ + struct cpl_set_tcb_field *req; + struct chtls_sock *csk; + struct ulptx_idata *sc; + unsigned int wrlen; + + wrlen = roundup(sizeof(*req) + sizeof(*sc), 16); + csk = rcu_dereference_sk_user_data(sk); + + req = (struct cpl_set_tcb_field *)__skb_put(skb, wrlen); + __set_tcb_field_direct(csk, req, word, mask, val, cookie, no_reply); + set_wr_txq(skb, CPL_PRIORITY_CONTROL, csk->port_id); +} + +/* + * Send control message to HW, message go as immediate data and packet + * is freed immediately. + */ +static int chtls_set_tcb_field(struct sock *sk, u16 word, u64 mask, u64 val) +{ + struct cpl_set_tcb_field *req; + unsigned int credits_needed; + struct chtls_sock *csk; + struct ulptx_idata *sc; + struct sk_buff *skb; + unsigned int wrlen; + int ret; + + wrlen = roundup(sizeof(*req) + sizeof(*sc), 16); + + skb = alloc_skb(wrlen, GFP_ATOMIC); + if (!skb) + return -ENOMEM; + + credits_needed = DIV_ROUND_UP(wrlen, 16); + csk = rcu_dereference_sk_user_data(sk); + + __set_tcb_field(sk, skb, word, mask, val, 0, 1); + skb_set_queue_mapping(skb, (csk->txq_idx << 1) | CPL_PRIORITY_DATA); + csk->wr_credits -= credits_needed; + csk->wr_unacked += credits_needed; + enqueue_wr(csk, skb); + ret = cxgb4_ofld_send(csk->egress_dev, skb); + if (ret < 0) + kfree_skb(skb); + return ret < 0 ? ret : 0; +} + +/* + * Set one of the t_flags bits in the TCB. + */ +int chtls_set_tcb_tflag(struct sock *sk, unsigned int bit_pos, int val) +{ + return chtls_set_tcb_field(sk, 1, 1ULL << bit_pos, + (u64)val << bit_pos); +} + +static int chtls_set_tcb_keyid(struct sock *sk, int keyid) +{ + return chtls_set_tcb_field(sk, 31, 0xFFFFFFFFULL, keyid); +} + +static int chtls_set_tcb_seqno(struct sock *sk) +{ + return chtls_set_tcb_field(sk, 28, ~0ULL, 0); +} + +static int chtls_set_tcb_quiesce(struct sock *sk, int val) +{ + return chtls_set_tcb_field(sk, 1, (1ULL << TF_RX_QUIESCE_S), + TF_RX_QUIESCE_V(val)); +} + +/* TLS Key bitmap processing */ +int chtls_init_kmap(struct chtls_dev *cdev, struct cxgb4_lld_info *lldi) +{ + unsigned int num_key_ctx, bsize; + int ksize; + + num_key_ctx = (lldi->vr->key.size / TLS_KEY_CONTEXT_SZ); + bsize = BITS_TO_LONGS(num_key_ctx); + + cdev->kmap.size = num_key_ctx; + cdev->kmap.available = bsize; + ksize = sizeof(*cdev->kmap.addr) * bsize; + cdev->kmap.addr = kvzalloc(ksize, GFP_KERNEL); + if (!cdev->kmap.addr) + return -ENOMEM; + + cdev->kmap.start = lldi->vr->key.start; + spin_lock_init(&cdev->kmap.lock); + return 0; +} + +static int get_new_keyid(struct chtls_sock *csk, u32 optname) +{ + struct net_device *dev = csk->egress_dev; + struct chtls_dev *cdev = csk->cdev; + struct chtls_hws *hws; + struct adapter *adap; + int keyid; + + adap = netdev2adap(dev); + hws = &csk->tlshws; + + spin_lock_bh(&cdev->kmap.lock); + keyid = find_first_zero_bit(cdev->kmap.addr, cdev->kmap.size); + if (keyid < cdev->kmap.size) { + __set_bit(keyid, cdev->kmap.addr); + if (optname == TLS_RX) + hws->rxkey = keyid; + else + hws->txkey = keyid; + atomic_inc(&adap->chcr_stats.tls_key); + } else { + keyid = -1; + } + spin_unlock_bh(&cdev->kmap.lock); + return keyid; +} + +void free_tls_keyid(struct sock *sk) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + struct net_device *dev = csk->egress_dev; + struct chtls_dev *cdev = csk->cdev; + struct chtls_hws *hws; + struct adapter *adap; + + if (!cdev->kmap.addr) + return; + + adap = netdev2adap(dev); + hws = &csk->tlshws; + + spin_lock_bh(&cdev->kmap.lock); + if (hws->rxkey >= 0) { + __clear_bit(hws->rxkey, cdev->kmap.addr); + atomic_dec(&adap->chcr_stats.tls_key); + hws->rxkey = -1; + } + if (hws->txkey >= 0) { + __clear_bit(hws->txkey, cdev->kmap.addr); + atomic_dec(&adap->chcr_stats.tls_key); + hws->txkey = -1; + } + spin_unlock_bh(&cdev->kmap.lock); +} + +unsigned int keyid_to_addr(int start_addr, int keyid) +{ + return (start_addr + (keyid * TLS_KEY_CONTEXT_SZ)) >> 5; +} + +static void chtls_rxkey_ivauth(struct _key_ctx *kctx) +{ + kctx->iv_to_auth = cpu_to_be64(KEYCTX_TX_WR_IV_V(6ULL) | + KEYCTX_TX_WR_AAD_V(1ULL) | + KEYCTX_TX_WR_AADST_V(5ULL) | + KEYCTX_TX_WR_CIPHER_V(14ULL) | + KEYCTX_TX_WR_CIPHERST_V(0ULL) | + KEYCTX_TX_WR_AUTH_V(14ULL) | + KEYCTX_TX_WR_AUTHST_V(16ULL) | + KEYCTX_TX_WR_AUTHIN_V(16ULL)); +} + +static int chtls_key_info(struct chtls_sock *csk, + struct _key_ctx *kctx, + u32 keylen, u32 optname, + int cipher_type) +{ + unsigned char key[AES_MAX_KEY_SIZE]; + unsigned char *key_p, *salt; + unsigned char ghash_h[AEAD_H_SIZE]; + int ck_size, key_ctx_size, kctx_mackey_size, salt_size; + struct crypto_aes_ctx aes; + int ret; + + key_ctx_size = sizeof(struct _key_ctx) + + roundup(keylen, 16) + AEAD_H_SIZE; + + /* GCM mode of AES supports 128 and 256 bit encryption, so + * prepare key context base on GCM cipher type + */ + switch (cipher_type) { + case TLS_CIPHER_AES_GCM_128: { + struct tls12_crypto_info_aes_gcm_128 *gcm_ctx_128 = + (struct tls12_crypto_info_aes_gcm_128 *) + &csk->tlshws.crypto_info; + memcpy(key, gcm_ctx_128->key, keylen); + + key_p = gcm_ctx_128->key; + salt = gcm_ctx_128->salt; + ck_size = CHCR_KEYCTX_CIPHER_KEY_SIZE_128; + salt_size = TLS_CIPHER_AES_GCM_128_SALT_SIZE; + kctx_mackey_size = CHCR_KEYCTX_MAC_KEY_SIZE_128; + break; + } + case TLS_CIPHER_AES_GCM_256: { + struct tls12_crypto_info_aes_gcm_256 *gcm_ctx_256 = + (struct tls12_crypto_info_aes_gcm_256 *) + &csk->tlshws.crypto_info; + memcpy(key, gcm_ctx_256->key, keylen); + + key_p = gcm_ctx_256->key; + salt = gcm_ctx_256->salt; + ck_size = CHCR_KEYCTX_CIPHER_KEY_SIZE_256; + salt_size = TLS_CIPHER_AES_GCM_256_SALT_SIZE; + kctx_mackey_size = CHCR_KEYCTX_MAC_KEY_SIZE_256; + break; + } + default: + pr_err("GCM: Invalid key length %d\n", keylen); + return -EINVAL; + } + + /* Calculate the H = CIPH(K, 0 repeated 16 times). + * It will go in key context + */ + ret = aes_expandkey(&aes, key, keylen); + if (ret) + return ret; + + memset(ghash_h, 0, AEAD_H_SIZE); + aes_encrypt(&aes, ghash_h, ghash_h); + memzero_explicit(&aes, sizeof(aes)); + csk->tlshws.keylen = key_ctx_size; + + /* Copy the Key context */ + if (optname == TLS_RX) { + int key_ctx; + + key_ctx = ((key_ctx_size >> 4) << 3); + kctx->ctx_hdr = FILL_KEY_CRX_HDR(ck_size, + kctx_mackey_size, + 0, 0, key_ctx); + chtls_rxkey_ivauth(kctx); + } else { + kctx->ctx_hdr = FILL_KEY_CTX_HDR(ck_size, + kctx_mackey_size, + 0, 0, key_ctx_size >> 4); + } + + memcpy(kctx->salt, salt, salt_size); + memcpy(kctx->key, key_p, keylen); + memcpy(kctx->key + keylen, ghash_h, AEAD_H_SIZE); + /* erase key info from driver */ + memset(key_p, 0, keylen); + + return 0; +} + +static void chtls_set_scmd(struct chtls_sock *csk) +{ + struct chtls_hws *hws = &csk->tlshws; + + hws->scmd.seqno_numivs = + SCMD_SEQ_NO_CTRL_V(3) | + SCMD_PROTO_VERSION_V(0) | + SCMD_ENC_DEC_CTRL_V(0) | + SCMD_CIPH_AUTH_SEQ_CTRL_V(1) | + SCMD_CIPH_MODE_V(2) | + SCMD_AUTH_MODE_V(4) | + SCMD_HMAC_CTRL_V(0) | + SCMD_IV_SIZE_V(4) | + SCMD_NUM_IVS_V(1); + + hws->scmd.ivgen_hdrlen = + SCMD_IV_GEN_CTRL_V(1) | + SCMD_KEY_CTX_INLINE_V(0) | + SCMD_TLS_FRAG_ENABLE_V(1); +} + +int chtls_setkey(struct chtls_sock *csk, u32 keylen, + u32 optname, int cipher_type) +{ + struct tls_key_req *kwr; + struct chtls_dev *cdev; + struct _key_ctx *kctx; + int wrlen, klen, len; + struct sk_buff *skb; + struct sock *sk; + int keyid; + int kaddr; + int ret; + + cdev = csk->cdev; + sk = csk->sk; + + klen = roundup((keylen + AEAD_H_SIZE) + sizeof(*kctx), 32); + wrlen = roundup(sizeof(*kwr), 16); + len = klen + wrlen; + + /* Flush out-standing data before new key takes effect */ + if (optname == TLS_TX) { + lock_sock(sk); + if (skb_queue_len(&csk->txq)) + chtls_push_frames(csk, 0); + release_sock(sk); + } + + skb = alloc_skb(len, GFP_KERNEL); + if (!skb) + return -ENOMEM; + + keyid = get_new_keyid(csk, optname); + if (keyid < 0) { + ret = -ENOSPC; + goto out_nokey; + } + + kaddr = keyid_to_addr(cdev->kmap.start, keyid); + kwr = (struct tls_key_req *)__skb_put_zero(skb, len); + kwr->wr.op_to_compl = + cpu_to_be32(FW_WR_OP_V(FW_ULPTX_WR) | FW_WR_COMPL_F | + FW_WR_ATOMIC_V(1U)); + kwr->wr.flowid_len16 = + cpu_to_be32(FW_WR_LEN16_V(DIV_ROUND_UP(len, 16) | + FW_WR_FLOWID_V(csk->tid))); + kwr->wr.protocol = 0; + kwr->wr.mfs = htons(TLS_MFS); + kwr->wr.reneg_to_write_rx = optname; + + /* ulptx command */ + kwr->req.cmd = cpu_to_be32(ULPTX_CMD_V(ULP_TX_MEM_WRITE) | + T5_ULP_MEMIO_ORDER_V(1) | + T5_ULP_MEMIO_IMM_V(1)); + kwr->req.len16 = cpu_to_be32((csk->tid << 8) | + DIV_ROUND_UP(len - sizeof(kwr->wr), 16)); + kwr->req.dlen = cpu_to_be32(ULP_MEMIO_DATA_LEN_V(klen >> 5)); + kwr->req.lock_addr = cpu_to_be32(ULP_MEMIO_ADDR_V(kaddr)); + + /* sub command */ + kwr->sc_imm.cmd_more = cpu_to_be32(ULPTX_CMD_V(ULP_TX_SC_IMM)); + kwr->sc_imm.len = cpu_to_be32(klen); + + lock_sock(sk); + /* key info */ + kctx = (struct _key_ctx *)(kwr + 1); + ret = chtls_key_info(csk, kctx, keylen, optname, cipher_type); + if (ret) + goto out_notcb; + + set_wr_txq(skb, CPL_PRIORITY_DATA, csk->tlshws.txqid); + csk->wr_credits -= DIV_ROUND_UP(len, 16); + csk->wr_unacked += DIV_ROUND_UP(len, 16); + enqueue_wr(csk, skb); + cxgb4_ofld_send(csk->egress_dev, skb); + + chtls_set_scmd(csk); + /* Clear quiesce for Rx key */ + if (optname == TLS_RX) { + ret = chtls_set_tcb_keyid(sk, keyid); + if (ret) + goto out_notcb; + ret = chtls_set_tcb_field(sk, 0, + TCB_ULP_RAW_V(TCB_ULP_RAW_M), + TCB_ULP_RAW_V((TF_TLS_KEY_SIZE_V(1) | + TF_TLS_CONTROL_V(1) | + TF_TLS_ACTIVE_V(1) | + TF_TLS_ENABLE_V(1)))); + if (ret) + goto out_notcb; + ret = chtls_set_tcb_seqno(sk); + if (ret) + goto out_notcb; + ret = chtls_set_tcb_quiesce(sk, 0); + if (ret) + goto out_notcb; + csk->tlshws.rxkey = keyid; + } else { + csk->tlshws.tx_seq_no = 0; + csk->tlshws.txkey = keyid; + } + + release_sock(sk); + return ret; +out_notcb: + release_sock(sk); + free_tls_keyid(sk); +out_nokey: + kfree_skb(skb); + return ret; +} diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c new file mode 100644 index 000000000000..2e9acae1cba3 --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c @@ -0,0 +1,1907 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2018 Chelsio Communications, Inc. + * + * Written by: Atul Gupta (atul.gupta@chelsio.com) + */ + +#include <linux/module.h> +#include <linux/list.h> +#include <linux/workqueue.h> +#include <linux/skbuff.h> +#include <linux/timer.h> +#include <linux/notifier.h> +#include <linux/inetdevice.h> +#include <linux/ip.h> +#include <linux/tcp.h> +#include <linux/sched/signal.h> +#include <net/tcp.h> +#include <net/busy_poll.h> +#include <crypto/aes.h> + +#include "chtls.h" +#include "chtls_cm.h" + +static bool is_tls_tx(struct chtls_sock *csk) +{ + return csk->tlshws.txkey >= 0; +} + +static bool is_tls_rx(struct chtls_sock *csk) +{ + return csk->tlshws.rxkey >= 0; +} + +static int data_sgl_len(const struct sk_buff *skb) +{ + unsigned int cnt; + + cnt = skb_shinfo(skb)->nr_frags; + return sgl_len(cnt) * 8; +} + +static int nos_ivs(struct sock *sk, unsigned int size) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + + return DIV_ROUND_UP(size, csk->tlshws.mfs); +} + +static int set_ivs_imm(struct sock *sk, const struct sk_buff *skb) +{ + int ivs_size = nos_ivs(sk, skb->len) * CIPHER_BLOCK_SIZE; + int hlen = TLS_WR_CPL_LEN + data_sgl_len(skb); + + if ((hlen + KEY_ON_MEM_SZ + ivs_size) < + MAX_IMM_OFLD_TX_DATA_WR_LEN) { + ULP_SKB_CB(skb)->ulp.tls.iv = 1; + return 1; + } + ULP_SKB_CB(skb)->ulp.tls.iv = 0; + return 0; +} + +static int max_ivs_size(struct sock *sk, int size) +{ + return nos_ivs(sk, size) * CIPHER_BLOCK_SIZE; +} + +static int ivs_size(struct sock *sk, const struct sk_buff *skb) +{ + return set_ivs_imm(sk, skb) ? (nos_ivs(sk, skb->len) * + CIPHER_BLOCK_SIZE) : 0; +} + +static int flowc_wr_credits(int nparams, int *flowclenp) +{ + int flowclen16, flowclen; + + flowclen = offsetof(struct fw_flowc_wr, mnemval[nparams]); + flowclen16 = DIV_ROUND_UP(flowclen, 16); + flowclen = flowclen16 * 16; + + if (flowclenp) + *flowclenp = flowclen; + + return flowclen16; +} + +static struct sk_buff *create_flowc_wr_skb(struct sock *sk, + struct fw_flowc_wr *flowc, + int flowclen) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + struct sk_buff *skb; + + skb = alloc_skb(flowclen, GFP_ATOMIC); + if (!skb) + return NULL; + + __skb_put_data(skb, flowc, flowclen); + skb_set_queue_mapping(skb, (csk->txq_idx << 1) | CPL_PRIORITY_DATA); + + return skb; +} + +static int send_flowc_wr(struct sock *sk, struct fw_flowc_wr *flowc, + int flowclen) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + struct tcp_sock *tp = tcp_sk(sk); + struct sk_buff *skb; + int flowclen16; + int ret; + + flowclen16 = flowclen / 16; + + if (csk_flag(sk, CSK_TX_DATA_SENT)) { + skb = create_flowc_wr_skb(sk, flowc, flowclen); + if (!skb) + return -ENOMEM; + + skb_entail(sk, skb, + ULPCB_FLAG_NO_HDR | ULPCB_FLAG_NO_APPEND); + return 0; + } + + ret = cxgb4_immdata_send(csk->egress_dev, + csk->txq_idx, + flowc, flowclen); + if (!ret) + return flowclen16; + skb = create_flowc_wr_skb(sk, flowc, flowclen); + if (!skb) + return -ENOMEM; + send_or_defer(sk, tp, skb, 0); + return flowclen16; +} + +static u8 tcp_state_to_flowc_state(u8 state) +{ + switch (state) { + case TCP_ESTABLISHED: + return FW_FLOWC_MNEM_TCPSTATE_ESTABLISHED; + case TCP_CLOSE_WAIT: + return FW_FLOWC_MNEM_TCPSTATE_CLOSEWAIT; + case TCP_FIN_WAIT1: + return FW_FLOWC_MNEM_TCPSTATE_FINWAIT1; + case TCP_CLOSING: + return FW_FLOWC_MNEM_TCPSTATE_CLOSING; + case TCP_LAST_ACK: + return FW_FLOWC_MNEM_TCPSTATE_LASTACK; + case TCP_FIN_WAIT2: + return FW_FLOWC_MNEM_TCPSTATE_FINWAIT2; + } + + return FW_FLOWC_MNEM_TCPSTATE_ESTABLISHED; +} + +int send_tx_flowc_wr(struct sock *sk, int compl, + u32 snd_nxt, u32 rcv_nxt) +{ + struct flowc_packed { + struct fw_flowc_wr fc; + struct fw_flowc_mnemval mnemval[FW_FLOWC_MNEM_MAX]; + } __packed sflowc; + int nparams, paramidx, flowclen16, flowclen; + struct fw_flowc_wr *flowc; + struct chtls_sock *csk; + struct tcp_sock *tp; + + csk = rcu_dereference_sk_user_data(sk); + tp = tcp_sk(sk); + memset(&sflowc, 0, sizeof(sflowc)); + flowc = &sflowc.fc; + +#define FLOWC_PARAM(__m, __v) \ + do { \ + flowc->mnemval[paramidx].mnemonic = FW_FLOWC_MNEM_##__m; \ + flowc->mnemval[paramidx].val = cpu_to_be32(__v); \ + paramidx++; \ + } while (0) + + paramidx = 0; + + FLOWC_PARAM(PFNVFN, FW_PFVF_CMD_PFN_V(csk->cdev->lldi->pf)); + FLOWC_PARAM(CH, csk->tx_chan); + FLOWC_PARAM(PORT, csk->tx_chan); + FLOWC_PARAM(IQID, csk->rss_qid); + FLOWC_PARAM(SNDNXT, tp->snd_nxt); + FLOWC_PARAM(RCVNXT, tp->rcv_nxt); + FLOWC_PARAM(SNDBUF, csk->sndbuf); + FLOWC_PARAM(MSS, tp->mss_cache); + FLOWC_PARAM(TCPSTATE, tcp_state_to_flowc_state(sk->sk_state)); + + if (SND_WSCALE(tp)) + FLOWC_PARAM(RCV_SCALE, SND_WSCALE(tp)); + + if (csk->ulp_mode == ULP_MODE_TLS) + FLOWC_PARAM(ULD_MODE, ULP_MODE_TLS); + + if (csk->tlshws.fcplenmax) + FLOWC_PARAM(TXDATAPLEN_MAX, csk->tlshws.fcplenmax); + + nparams = paramidx; +#undef FLOWC_PARAM + + flowclen16 = flowc_wr_credits(nparams, &flowclen); + flowc->op_to_nparams = + cpu_to_be32(FW_WR_OP_V(FW_FLOWC_WR) | + FW_WR_COMPL_V(compl) | + FW_FLOWC_WR_NPARAMS_V(nparams)); + flowc->flowid_len16 = cpu_to_be32(FW_WR_LEN16_V(flowclen16) | + FW_WR_FLOWID_V(csk->tid)); + + return send_flowc_wr(sk, flowc, flowclen); +} + +/* Copy IVs to WR */ +static int tls_copy_ivs(struct sock *sk, struct sk_buff *skb) + +{ + struct chtls_sock *csk; + unsigned char *iv_loc; + struct chtls_hws *hws; + unsigned char *ivs; + u16 number_of_ivs; + struct page *page; + int err = 0; + + csk = rcu_dereference_sk_user_data(sk); + hws = &csk->tlshws; + number_of_ivs = nos_ivs(sk, skb->len); + + if (number_of_ivs > MAX_IVS_PAGE) { + pr_warn("MAX IVs in PAGE exceeded %d\n", number_of_ivs); + return -ENOMEM; + } + + /* generate the IVs */ + ivs = kmalloc_array(CIPHER_BLOCK_SIZE, number_of_ivs, GFP_ATOMIC); + if (!ivs) + return -ENOMEM; + get_random_bytes(ivs, number_of_ivs * CIPHER_BLOCK_SIZE); + + if (skb_ulp_tls_iv_imm(skb)) { + /* send the IVs as immediate data in the WR */ + iv_loc = (unsigned char *)__skb_push(skb, number_of_ivs * + CIPHER_BLOCK_SIZE); + if (iv_loc) + memcpy(iv_loc, ivs, number_of_ivs * CIPHER_BLOCK_SIZE); + + hws->ivsize = number_of_ivs * CIPHER_BLOCK_SIZE; + } else { + /* Send the IVs as sgls */ + /* Already accounted IV DSGL for credits */ + skb_shinfo(skb)->nr_frags--; + page = alloc_pages(sk->sk_allocation | __GFP_COMP, 0); + if (!page) { + pr_info("%s : Page allocation for IVs failed\n", + __func__); + err = -ENOMEM; + goto out; + } + memcpy(page_address(page), ivs, number_of_ivs * + CIPHER_BLOCK_SIZE); + skb_fill_page_desc(skb, skb_shinfo(skb)->nr_frags, page, 0, + number_of_ivs * CIPHER_BLOCK_SIZE); + hws->ivsize = 0; + } +out: + kfree(ivs); + return err; +} + +/* Copy Key to WR */ +static void tls_copy_tx_key(struct sock *sk, struct sk_buff *skb) +{ + struct ulptx_sc_memrd *sc_memrd; + struct chtls_sock *csk; + struct chtls_dev *cdev; + struct ulptx_idata *sc; + struct chtls_hws *hws; + u32 immdlen; + int kaddr; + + csk = rcu_dereference_sk_user_data(sk); + hws = &csk->tlshws; + cdev = csk->cdev; + + immdlen = sizeof(*sc) + sizeof(*sc_memrd); + kaddr = keyid_to_addr(cdev->kmap.start, hws->txkey); + sc = (struct ulptx_idata *)__skb_push(skb, immdlen); + if (sc) { + sc->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_NOOP)); + sc->len = htonl(0); + sc_memrd = (struct ulptx_sc_memrd *)(sc + 1); + sc_memrd->cmd_to_len = + htonl(ULPTX_CMD_V(ULP_TX_SC_MEMRD) | + ULP_TX_SC_MORE_V(1) | + ULPTX_LEN16_V(hws->keylen >> 4)); + sc_memrd->addr = htonl(kaddr); + } +} + +static u64 tlstx_incr_seqnum(struct chtls_hws *hws) +{ + return hws->tx_seq_no++; +} + +static bool is_sg_request(const struct sk_buff *skb) +{ + return skb->peeked || + (skb->len > MAX_IMM_ULPTX_WR_LEN); +} + +/* + * Returns true if an sk_buff carries urgent data. + */ +static bool skb_urgent(struct sk_buff *skb) +{ + return ULP_SKB_CB(skb)->flags & ULPCB_FLAG_URG; +} + +/* TLS content type for CPL SFO */ +static unsigned char tls_content_type(unsigned char content_type) +{ + switch (content_type) { + case TLS_HDR_TYPE_CCS: + return CPL_TX_TLS_SFO_TYPE_CCS; + case TLS_HDR_TYPE_ALERT: + return CPL_TX_TLS_SFO_TYPE_ALERT; + case TLS_HDR_TYPE_HANDSHAKE: + return CPL_TX_TLS_SFO_TYPE_HANDSHAKE; + case TLS_HDR_TYPE_HEARTBEAT: + return CPL_TX_TLS_SFO_TYPE_HEARTBEAT; + } + return CPL_TX_TLS_SFO_TYPE_DATA; +} + +static void tls_tx_data_wr(struct sock *sk, struct sk_buff *skb, + int dlen, int tls_immd, u32 credits, + int expn, int pdus) +{ + struct fw_tlstx_data_wr *req_wr; + struct cpl_tx_tls_sfo *req_cpl; + unsigned int wr_ulp_mode_force; + struct tls_scmd *updated_scmd; + unsigned char data_type; + struct chtls_sock *csk; + struct net_device *dev; + struct chtls_hws *hws; + struct tls_scmd *scmd; + struct adapter *adap; + unsigned char *req; + int immd_len; + int iv_imm; + int len; + + csk = rcu_dereference_sk_user_data(sk); + iv_imm = skb_ulp_tls_iv_imm(skb); + dev = csk->egress_dev; + adap = netdev2adap(dev); + hws = &csk->tlshws; + scmd = &hws->scmd; + len = dlen + expn; + + dlen = (dlen < hws->mfs) ? dlen : hws->mfs; + atomic_inc(&adap->chcr_stats.tls_pdu_tx); + + updated_scmd = scmd; + updated_scmd->seqno_numivs &= 0xffffff80; + updated_scmd->seqno_numivs |= SCMD_NUM_IVS_V(pdus); + hws->scmd = *updated_scmd; + + req = (unsigned char *)__skb_push(skb, sizeof(struct cpl_tx_tls_sfo)); + req_cpl = (struct cpl_tx_tls_sfo *)req; + req = (unsigned char *)__skb_push(skb, (sizeof(struct + fw_tlstx_data_wr))); + + req_wr = (struct fw_tlstx_data_wr *)req; + immd_len = (tls_immd ? dlen : 0); + req_wr->op_to_immdlen = + htonl(FW_WR_OP_V(FW_TLSTX_DATA_WR) | + FW_TLSTX_DATA_WR_COMPL_V(1) | + FW_TLSTX_DATA_WR_IMMDLEN_V(immd_len)); + req_wr->flowid_len16 = htonl(FW_TLSTX_DATA_WR_FLOWID_V(csk->tid) | + FW_TLSTX_DATA_WR_LEN16_V(credits)); + wr_ulp_mode_force = TX_ULP_MODE_V(ULP_MODE_TLS); + + if (is_sg_request(skb)) + wr_ulp_mode_force |= FW_OFLD_TX_DATA_WR_ALIGNPLD_F | + ((tcp_sk(sk)->nonagle & TCP_NAGLE_OFF) ? 0 : + FW_OFLD_TX_DATA_WR_SHOVE_F); + + req_wr->lsodisable_to_flags = + htonl(TX_ULP_MODE_V(ULP_MODE_TLS) | + TX_URG_V(skb_urgent(skb)) | + T6_TX_FORCE_F | wr_ulp_mode_force | + TX_SHOVE_V((!csk_flag(sk, CSK_TX_MORE_DATA)) && + skb_queue_empty(&csk->txq))); + + req_wr->ctxloc_to_exp = + htonl(FW_TLSTX_DATA_WR_NUMIVS_V(pdus) | + FW_TLSTX_DATA_WR_EXP_V(expn) | + FW_TLSTX_DATA_WR_CTXLOC_V(CHTLS_KEY_CONTEXT_DDR) | + FW_TLSTX_DATA_WR_IVDSGL_V(!iv_imm) | + FW_TLSTX_DATA_WR_KEYSIZE_V(hws->keylen >> 4)); + + /* Fill in the length */ + req_wr->plen = htonl(len); + req_wr->mfs = htons(hws->mfs); + req_wr->adjustedplen_pkd = + htons(FW_TLSTX_DATA_WR_ADJUSTEDPLEN_V(hws->adjustlen)); + req_wr->expinplenmax_pkd = + htons(FW_TLSTX_DATA_WR_EXPINPLENMAX_V(hws->expansion)); + req_wr->pdusinplenmax_pkd = + FW_TLSTX_DATA_WR_PDUSINPLENMAX_V(hws->pdus); + req_wr->r10 = 0; + + data_type = tls_content_type(ULP_SKB_CB(skb)->ulp.tls.type); + req_cpl->op_to_seg_len = htonl(CPL_TX_TLS_SFO_OPCODE_V(CPL_TX_TLS_SFO) | + CPL_TX_TLS_SFO_DATA_TYPE_V(data_type) | + CPL_TX_TLS_SFO_CPL_LEN_V(2) | + CPL_TX_TLS_SFO_SEG_LEN_V(dlen)); + req_cpl->pld_len = htonl(len - expn); + + req_cpl->type_protover = htonl(CPL_TX_TLS_SFO_TYPE_V + ((data_type == CPL_TX_TLS_SFO_TYPE_HEARTBEAT) ? + TLS_HDR_TYPE_HEARTBEAT : 0) | + CPL_TX_TLS_SFO_PROTOVER_V(0)); + + /* create the s-command */ + req_cpl->r1_lo = 0; + req_cpl->seqno_numivs = cpu_to_be32(hws->scmd.seqno_numivs); + req_cpl->ivgen_hdrlen = cpu_to_be32(hws->scmd.ivgen_hdrlen); + req_cpl->scmd1 = cpu_to_be64(tlstx_incr_seqnum(hws)); +} + +/* + * Calculate the TLS data expansion size + */ +static int chtls_expansion_size(struct sock *sk, int data_len, + int fullpdu, + unsigned short *pducnt) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + struct chtls_hws *hws = &csk->tlshws; + struct tls_scmd *scmd = &hws->scmd; + int fragsize = hws->mfs; + int expnsize = 0; + int fragleft; + int fragcnt; + int expppdu; + + if (SCMD_CIPH_MODE_G(scmd->seqno_numivs) == + SCMD_CIPH_MODE_AES_GCM) { + expppdu = GCM_TAG_SIZE + AEAD_EXPLICIT_DATA_SIZE + + TLS_HEADER_LENGTH; + + if (fullpdu) { + *pducnt = data_len / (expppdu + fragsize); + if (*pducnt > 32) + *pducnt = 32; + else if (!*pducnt) + *pducnt = 1; + expnsize = (*pducnt) * expppdu; + return expnsize; + } + fragcnt = (data_len / fragsize); + expnsize = fragcnt * expppdu; + fragleft = data_len % fragsize; + if (fragleft > 0) + expnsize += expppdu; + } + return expnsize; +} + +/* WR with IV, KEY and CPL SFO added */ +static void make_tlstx_data_wr(struct sock *sk, struct sk_buff *skb, + int tls_tx_imm, int tls_len, u32 credits) +{ + unsigned short pdus_per_ulp = 0; + struct chtls_sock *csk; + struct chtls_hws *hws; + int expn_sz; + int pdus; + + csk = rcu_dereference_sk_user_data(sk); + hws = &csk->tlshws; + pdus = DIV_ROUND_UP(tls_len, hws->mfs); + expn_sz = chtls_expansion_size(sk, tls_len, 0, NULL); + if (!hws->compute) { + hws->expansion = chtls_expansion_size(sk, + hws->fcplenmax, + 1, &pdus_per_ulp); + hws->pdus = pdus_per_ulp; + hws->adjustlen = hws->pdus * + ((hws->expansion / hws->pdus) + hws->mfs); + hws->compute = 1; + } + if (tls_copy_ivs(sk, skb)) + return; + tls_copy_tx_key(sk, skb); + tls_tx_data_wr(sk, skb, tls_len, tls_tx_imm, credits, expn_sz, pdus); + hws->tx_seq_no += (pdus - 1); +} + +static void make_tx_data_wr(struct sock *sk, struct sk_buff *skb, + unsigned int immdlen, int len, + u32 credits, u32 compl) +{ + struct fw_ofld_tx_data_wr *req; + unsigned int wr_ulp_mode_force; + struct chtls_sock *csk; + unsigned int opcode; + + csk = rcu_dereference_sk_user_data(sk); + opcode = FW_OFLD_TX_DATA_WR; + + req = (struct fw_ofld_tx_data_wr *)__skb_push(skb, sizeof(*req)); + req->op_to_immdlen = htonl(WR_OP_V(opcode) | + FW_WR_COMPL_V(compl) | + FW_WR_IMMDLEN_V(immdlen)); + req->flowid_len16 = htonl(FW_WR_FLOWID_V(csk->tid) | + FW_WR_LEN16_V(credits)); + + wr_ulp_mode_force = TX_ULP_MODE_V(csk->ulp_mode); + if (is_sg_request(skb)) + wr_ulp_mode_force |= FW_OFLD_TX_DATA_WR_ALIGNPLD_F | + ((tcp_sk(sk)->nonagle & TCP_NAGLE_OFF) ? 0 : + FW_OFLD_TX_DATA_WR_SHOVE_F); + + req->tunnel_to_proxy = htonl(wr_ulp_mode_force | + TX_URG_V(skb_urgent(skb)) | + TX_SHOVE_V((!csk_flag(sk, CSK_TX_MORE_DATA)) && + skb_queue_empty(&csk->txq))); + req->plen = htonl(len); +} + +static int chtls_wr_size(struct chtls_sock *csk, const struct sk_buff *skb, + bool size) +{ + int wr_size; + + wr_size = TLS_WR_CPL_LEN; + wr_size += KEY_ON_MEM_SZ; + wr_size += ivs_size(csk->sk, skb); + + if (size) + return wr_size; + + /* frags counted for IV dsgl */ + if (!skb_ulp_tls_iv_imm(skb)) + skb_shinfo(skb)->nr_frags++; + + return wr_size; +} + +static bool is_ofld_imm(struct chtls_sock *csk, const struct sk_buff *skb) +{ + int length = skb->len; + + if (skb->peeked || skb->len > MAX_IMM_ULPTX_WR_LEN) + return false; + + if (likely(ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NEED_HDR)) { + /* Check TLS header len for Immediate */ + if (csk->ulp_mode == ULP_MODE_TLS && + skb_ulp_tls_inline(skb)) + length += chtls_wr_size(csk, skb, true); + else + length += sizeof(struct fw_ofld_tx_data_wr); + + return length <= MAX_IMM_OFLD_TX_DATA_WR_LEN; + } + return true; +} + +static unsigned int calc_tx_flits(const struct sk_buff *skb, + unsigned int immdlen) +{ + unsigned int flits, cnt; + + flits = immdlen / 8; /* headers */ + cnt = skb_shinfo(skb)->nr_frags; + if (skb_tail_pointer(skb) != skb_transport_header(skb)) + cnt++; + return flits + sgl_len(cnt); +} + +static void arp_failure_discard(void *handle, struct sk_buff *skb) +{ + kfree_skb(skb); +} + +int chtls_push_frames(struct chtls_sock *csk, int comp) +{ + struct chtls_hws *hws = &csk->tlshws; + struct tcp_sock *tp; + struct sk_buff *skb; + int total_size = 0; + struct sock *sk; + int wr_size; + + wr_size = sizeof(struct fw_ofld_tx_data_wr); + sk = csk->sk; + tp = tcp_sk(sk); + + if (unlikely(sk_in_state(sk, TCPF_SYN_SENT | TCPF_CLOSE))) + return 0; + + if (unlikely(csk_flag(sk, CSK_ABORT_SHUTDOWN))) + return 0; + + while (csk->wr_credits && (skb = skb_peek(&csk->txq)) && + (!(ULP_SKB_CB(skb)->flags & ULPCB_FLAG_HOLD) || + skb_queue_len(&csk->txq) > 1)) { + unsigned int credit_len = skb->len; + unsigned int credits_needed; + unsigned int completion = 0; + int tls_len = skb->len;/* TLS data len before IV/key */ + unsigned int immdlen; + int len = skb->len; /* length [ulp bytes] inserted by hw */ + int flowclen16 = 0; + int tls_tx_imm = 0; + + immdlen = skb->len; + if (!is_ofld_imm(csk, skb)) { + immdlen = skb_transport_offset(skb); + if (skb_ulp_tls_inline(skb)) + wr_size = chtls_wr_size(csk, skb, false); + credit_len = 8 * calc_tx_flits(skb, immdlen); + } else { + if (skb_ulp_tls_inline(skb)) { + wr_size = chtls_wr_size(csk, skb, false); + tls_tx_imm = 1; + } + } + if (likely(ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NEED_HDR)) + credit_len += wr_size; + credits_needed = DIV_ROUND_UP(credit_len, 16); + if (!csk_flag_nochk(csk, CSK_TX_DATA_SENT)) { + flowclen16 = send_tx_flowc_wr(sk, 1, tp->snd_nxt, + tp->rcv_nxt); + if (flowclen16 <= 0) + break; + csk->wr_credits -= flowclen16; + csk->wr_unacked += flowclen16; + csk->wr_nondata += flowclen16; + csk_set_flag(csk, CSK_TX_DATA_SENT); + } + + if (csk->wr_credits < credits_needed) { + if (skb_ulp_tls_inline(skb) && + !skb_ulp_tls_iv_imm(skb)) + skb_shinfo(skb)->nr_frags--; + break; + } + + __skb_unlink(skb, &csk->txq); + skb_set_queue_mapping(skb, (csk->txq_idx << 1) | + CPL_PRIORITY_DATA); + if (hws->ofld) + hws->txqid = (skb->queue_mapping >> 1); + skb->csum = (__force __wsum)(credits_needed + csk->wr_nondata); + csk->wr_credits -= credits_needed; + csk->wr_unacked += credits_needed; + csk->wr_nondata = 0; + enqueue_wr(csk, skb); + + if (likely(ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NEED_HDR)) { + if ((comp && csk->wr_unacked == credits_needed) || + (ULP_SKB_CB(skb)->flags & ULPCB_FLAG_COMPL) || + csk->wr_unacked >= csk->wr_max_credits / 2) { + completion = 1; + csk->wr_unacked = 0; + } + if (skb_ulp_tls_inline(skb)) + make_tlstx_data_wr(sk, skb, tls_tx_imm, + tls_len, credits_needed); + else + make_tx_data_wr(sk, skb, immdlen, len, + credits_needed, completion); + tp->snd_nxt += len; + tp->lsndtime = tcp_jiffies32; + if (completion) + ULP_SKB_CB(skb)->flags &= ~ULPCB_FLAG_NEED_HDR; + } else { + struct cpl_close_con_req *req = cplhdr(skb); + unsigned int cmd = CPL_OPCODE_G(ntohl + (OPCODE_TID(req))); + + if (cmd == CPL_CLOSE_CON_REQ) + csk_set_flag(csk, + CSK_CLOSE_CON_REQUESTED); + + if ((ULP_SKB_CB(skb)->flags & ULPCB_FLAG_COMPL) && + (csk->wr_unacked >= csk->wr_max_credits / 2)) { + req->wr.wr_hi |= htonl(FW_WR_COMPL_F); + csk->wr_unacked = 0; + } + } + total_size += skb->truesize; + if (ULP_SKB_CB(skb)->flags & ULPCB_FLAG_BARRIER) + csk_set_flag(csk, CSK_TX_WAIT_IDLE); + t4_set_arp_err_handler(skb, NULL, arp_failure_discard); + cxgb4_l2t_send(csk->egress_dev, skb, csk->l2t_entry); + } + sk->sk_wmem_queued -= total_size; + return total_size; +} + +static void mark_urg(struct tcp_sock *tp, int flags, + struct sk_buff *skb) +{ + if (unlikely(flags & MSG_OOB)) { + tp->snd_up = tp->write_seq; + ULP_SKB_CB(skb)->flags = ULPCB_FLAG_URG | + ULPCB_FLAG_BARRIER | + ULPCB_FLAG_NO_APPEND | + ULPCB_FLAG_NEED_HDR; + } +} + +/* + * Returns true if a connection should send more data to TCP engine + */ +static bool should_push(struct sock *sk) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + struct chtls_dev *cdev = csk->cdev; + struct tcp_sock *tp = tcp_sk(sk); + + /* + * If we've released our offload resources there's nothing to do ... + */ + if (!cdev) + return false; + + /* + * If there aren't any work requests in flight, or there isn't enough + * data in flight, or Nagle is off then send the current TX_DATA + * otherwise hold it and wait to accumulate more data. + */ + return csk->wr_credits == csk->wr_max_credits || + (tp->nonagle & TCP_NAGLE_OFF); +} + +/* + * Returns true if a TCP socket is corked. + */ +static bool corked(const struct tcp_sock *tp, int flags) +{ + return (flags & MSG_MORE) || (tp->nonagle & TCP_NAGLE_CORK); +} + +/* + * Returns true if a send should try to push new data. + */ +static bool send_should_push(struct sock *sk, int flags) +{ + return should_push(sk) && !corked(tcp_sk(sk), flags); +} + +void chtls_tcp_push(struct sock *sk, int flags) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + int qlen = skb_queue_len(&csk->txq); + + if (likely(qlen)) { + struct sk_buff *skb = skb_peek_tail(&csk->txq); + struct tcp_sock *tp = tcp_sk(sk); + + mark_urg(tp, flags, skb); + + if (!(ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NO_APPEND) && + corked(tp, flags)) { + ULP_SKB_CB(skb)->flags |= ULPCB_FLAG_HOLD; + return; + } + + ULP_SKB_CB(skb)->flags &= ~ULPCB_FLAG_HOLD; + if (qlen == 1 && + ((ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NO_APPEND) || + should_push(sk))) + chtls_push_frames(csk, 1); + } +} + +/* + * Calculate the size for a new send sk_buff. It's maximum size so we can + * pack lots of data into it, unless we plan to send it immediately, in which + * case we size it more tightly. + * + * Note: we don't bother compensating for MSS < PAGE_SIZE because it doesn't + * arise in normal cases and when it does we are just wasting memory. + */ +static int select_size(struct sock *sk, int io_len, int flags, int len) +{ + const int pgbreak = SKB_MAX_HEAD(len); + + /* + * If the data wouldn't fit in the main body anyway, put only the + * header in the main body so it can use immediate data and place all + * the payload in page fragments. + */ + if (io_len > pgbreak) + return 0; + + /* + * If we will be accumulating payload get a large main body. + */ + if (!send_should_push(sk, flags)) + return pgbreak; + + return io_len; +} + +void skb_entail(struct sock *sk, struct sk_buff *skb, int flags) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + struct tcp_sock *tp = tcp_sk(sk); + + ULP_SKB_CB(skb)->seq = tp->write_seq; + ULP_SKB_CB(skb)->flags = flags; + __skb_queue_tail(&csk->txq, skb); + sk->sk_wmem_queued += skb->truesize; + + if (TCP_PAGE(sk) && TCP_OFF(sk)) { + put_page(TCP_PAGE(sk)); + TCP_PAGE(sk) = NULL; + TCP_OFF(sk) = 0; + } +} + +static struct sk_buff *get_tx_skb(struct sock *sk, int size) +{ + struct sk_buff *skb; + + skb = alloc_skb(size + TX_HEADER_LEN, sk->sk_allocation); + if (likely(skb)) { + skb_reserve(skb, TX_HEADER_LEN); + skb_entail(sk, skb, ULPCB_FLAG_NEED_HDR); + skb_reset_transport_header(skb); + } + return skb; +} + +static struct sk_buff *get_record_skb(struct sock *sk, int size, bool zcopy) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + struct sk_buff *skb; + + skb = alloc_skb(((zcopy ? 0 : size) + TX_TLSHDR_LEN + + KEY_ON_MEM_SZ + max_ivs_size(sk, size)), + sk->sk_allocation); + if (likely(skb)) { + skb_reserve(skb, (TX_TLSHDR_LEN + + KEY_ON_MEM_SZ + max_ivs_size(sk, size))); + skb_entail(sk, skb, ULPCB_FLAG_NEED_HDR); + skb_reset_transport_header(skb); + ULP_SKB_CB(skb)->ulp.tls.ofld = 1; + ULP_SKB_CB(skb)->ulp.tls.type = csk->tlshws.type; + } + return skb; +} + +static void tx_skb_finalize(struct sk_buff *skb) +{ + struct ulp_skb_cb *cb = ULP_SKB_CB(skb); + + if (!(cb->flags & ULPCB_FLAG_NO_HDR)) + cb->flags = ULPCB_FLAG_NEED_HDR; + cb->flags |= ULPCB_FLAG_NO_APPEND; +} + +static void push_frames_if_head(struct sock *sk) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + + if (skb_queue_len(&csk->txq) == 1) + chtls_push_frames(csk, 1); +} + +static int chtls_skb_copy_to_page_nocache(struct sock *sk, + struct iov_iter *from, + struct sk_buff *skb, + struct page *page, + int off, int copy) +{ + int err; + + err = skb_do_copy_data_nocache(sk, skb, from, page_address(page) + + off, copy, skb->len); + if (err) + return err; + + skb->len += copy; + skb->data_len += copy; + skb->truesize += copy; + sk->sk_wmem_queued += copy; + return 0; +} + +static int csk_mem_free(struct chtls_dev *cdev, struct sock *sk) +{ + return (cdev->max_host_sndbuf - sk->sk_wmem_queued); +} + +static int csk_wait_memory(struct chtls_dev *cdev, + struct sock *sk, long *timeo_p) +{ + DEFINE_WAIT_FUNC(wait, woken_wake_function); + int err = 0; + long current_timeo; + long vm_wait = 0; + bool noblock; + + current_timeo = *timeo_p; + noblock = (*timeo_p ? false : true); + if (csk_mem_free(cdev, sk)) { + current_timeo = (prandom_u32() % (HZ / 5)) + 2; + vm_wait = (prandom_u32() % (HZ / 5)) + 2; + } + + add_wait_queue(sk_sleep(sk), &wait); + while (1) { + sk_set_bit(SOCKWQ_ASYNC_NOSPACE, sk); + + if (sk->sk_err || (sk->sk_shutdown & SEND_SHUTDOWN)) + goto do_error; + if (!*timeo_p) { + if (noblock) + set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); + goto do_nonblock; + } + if (signal_pending(current)) + goto do_interrupted; + sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk); + if (csk_mem_free(cdev, sk) && !vm_wait) + break; + + set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); + sk->sk_write_pending++; + sk_wait_event(sk, ¤t_timeo, sk->sk_err || + (sk->sk_shutdown & SEND_SHUTDOWN) || + (csk_mem_free(cdev, sk) && !vm_wait), &wait); + sk->sk_write_pending--; + + if (vm_wait) { + vm_wait -= current_timeo; + current_timeo = *timeo_p; + if (current_timeo != MAX_SCHEDULE_TIMEOUT) { + current_timeo -= vm_wait; + if (current_timeo < 0) + current_timeo = 0; + } + vm_wait = 0; + } + *timeo_p = current_timeo; + } +do_rm_wq: + remove_wait_queue(sk_sleep(sk), &wait); + return err; +do_error: + err = -EPIPE; + goto do_rm_wq; +do_nonblock: + err = -EAGAIN; + goto do_rm_wq; +do_interrupted: + err = sock_intr_errno(*timeo_p); + goto do_rm_wq; +} + +static int chtls_proccess_cmsg(struct sock *sk, struct msghdr *msg, + unsigned char *record_type) +{ + struct cmsghdr *cmsg; + int rc = -EINVAL; + + for_each_cmsghdr(cmsg, msg) { + if (!CMSG_OK(msg, cmsg)) + return -EINVAL; + if (cmsg->cmsg_level != SOL_TLS) + continue; + + switch (cmsg->cmsg_type) { + case TLS_SET_RECORD_TYPE: + if (cmsg->cmsg_len < CMSG_LEN(sizeof(*record_type))) + return -EINVAL; + + if (msg->msg_flags & MSG_MORE) + return -EINVAL; + + *record_type = *(unsigned char *)CMSG_DATA(cmsg); + rc = 0; + break; + default: + return -EINVAL; + } + } + + return rc; +} + +int chtls_sendmsg(struct sock *sk, struct msghdr *msg, size_t size) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + struct chtls_dev *cdev = csk->cdev; + struct tcp_sock *tp = tcp_sk(sk); + struct sk_buff *skb; + int mss, flags, err; + int recordsz = 0; + int copied = 0; + long timeo; + + lock_sock(sk); + flags = msg->msg_flags; + timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT); + + if (!sk_in_state(sk, TCPF_ESTABLISHED | TCPF_CLOSE_WAIT)) { + err = sk_stream_wait_connect(sk, &timeo); + if (err) + goto out_err; + } + + sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk); + err = -EPIPE; + if (sk->sk_err || (sk->sk_shutdown & SEND_SHUTDOWN)) + goto out_err; + + mss = csk->mss; + csk_set_flag(csk, CSK_TX_MORE_DATA); + + while (msg_data_left(msg)) { + int copy = 0; + + skb = skb_peek_tail(&csk->txq); + if (skb) { + copy = mss - skb->len; + skb->ip_summed = CHECKSUM_UNNECESSARY; + } + if (!csk_mem_free(cdev, sk)) + goto wait_for_sndbuf; + + if (is_tls_tx(csk) && !csk->tlshws.txleft) { + unsigned char record_type = TLS_RECORD_TYPE_DATA; + + if (unlikely(msg->msg_controllen)) { + err = chtls_proccess_cmsg(sk, msg, + &record_type); + if (err) + goto out_err; + + /* Avoid appending tls handshake, alert to tls data */ + if (skb) + tx_skb_finalize(skb); + } + + recordsz = size; + csk->tlshws.txleft = recordsz; + csk->tlshws.type = record_type; + } + + if (!skb || (ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NO_APPEND) || + copy <= 0) { +new_buf: + if (skb) { + tx_skb_finalize(skb); + push_frames_if_head(sk); + } + + if (is_tls_tx(csk)) { + skb = get_record_skb(sk, + select_size(sk, + recordsz, + flags, + TX_TLSHDR_LEN), + false); + } else { + skb = get_tx_skb(sk, + select_size(sk, size, flags, + TX_HEADER_LEN)); + } + if (unlikely(!skb)) + goto wait_for_memory; + + skb->ip_summed = CHECKSUM_UNNECESSARY; + copy = mss; + } + if (copy > size) + copy = size; + + if (skb_tailroom(skb) > 0) { + copy = min(copy, skb_tailroom(skb)); + if (is_tls_tx(csk)) + copy = min_t(int, copy, csk->tlshws.txleft); + err = skb_add_data_nocache(sk, skb, + &msg->msg_iter, copy); + if (err) + goto do_fault; + } else { + int i = skb_shinfo(skb)->nr_frags; + struct page *page = TCP_PAGE(sk); + int pg_size = PAGE_SIZE; + int off = TCP_OFF(sk); + bool merge; + + if (page) + pg_size = page_size(page); + if (off < pg_size && + skb_can_coalesce(skb, i, page, off)) { + merge = true; + goto copy; + } + merge = false; + if (i == (is_tls_tx(csk) ? (MAX_SKB_FRAGS - 1) : + MAX_SKB_FRAGS)) + goto new_buf; + + if (page && off == pg_size) { + put_page(page); + TCP_PAGE(sk) = page = NULL; + pg_size = PAGE_SIZE; + } + + if (!page) { + gfp_t gfp = sk->sk_allocation; + int order = cdev->send_page_order; + + if (order) { + page = alloc_pages(gfp | __GFP_COMP | + __GFP_NOWARN | + __GFP_NORETRY, + order); + if (page) + pg_size <<= order; + } + if (!page) { + page = alloc_page(gfp); + pg_size = PAGE_SIZE; + } + if (!page) + goto wait_for_memory; + off = 0; + } +copy: + if (copy > pg_size - off) + copy = pg_size - off; + if (is_tls_tx(csk)) + copy = min_t(int, copy, csk->tlshws.txleft); + + err = chtls_skb_copy_to_page_nocache(sk, &msg->msg_iter, + skb, page, + off, copy); + if (unlikely(err)) { + if (!TCP_PAGE(sk)) { + TCP_PAGE(sk) = page; + TCP_OFF(sk) = 0; + } + goto do_fault; + } + /* Update the skb. */ + if (merge) { + skb_frag_size_add( + &skb_shinfo(skb)->frags[i - 1], + copy); + } else { + skb_fill_page_desc(skb, i, page, off, copy); + if (off + copy < pg_size) { + /* space left keep page */ + get_page(page); + TCP_PAGE(sk) = page; + } else { + TCP_PAGE(sk) = NULL; + } + } + TCP_OFF(sk) = off + copy; + } + if (unlikely(skb->len == mss)) + tx_skb_finalize(skb); + tp->write_seq += copy; + copied += copy; + size -= copy; + + if (is_tls_tx(csk)) + csk->tlshws.txleft -= copy; + + if (corked(tp, flags) && + (sk_stream_wspace(sk) < sk_stream_min_wspace(sk))) + ULP_SKB_CB(skb)->flags |= ULPCB_FLAG_NO_APPEND; + + if (size == 0) + goto out; + + if (ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NO_APPEND) + push_frames_if_head(sk); + continue; +wait_for_sndbuf: + set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); +wait_for_memory: + err = csk_wait_memory(cdev, sk, &timeo); + if (err) + goto do_error; + } +out: + csk_reset_flag(csk, CSK_TX_MORE_DATA); + if (copied) + chtls_tcp_push(sk, flags); +done: + release_sock(sk); + return copied; +do_fault: + if (!skb->len) { + __skb_unlink(skb, &csk->txq); + sk->sk_wmem_queued -= skb->truesize; + __kfree_skb(skb); + } +do_error: + if (copied) + goto out; +out_err: + if (csk_conn_inline(csk)) + csk_reset_flag(csk, CSK_TX_MORE_DATA); + copied = sk_stream_error(sk, flags, err); + goto done; +} + +int chtls_sendpage(struct sock *sk, struct page *page, + int offset, size_t size, int flags) +{ + struct chtls_sock *csk; + struct chtls_dev *cdev; + int mss, err, copied; + struct tcp_sock *tp; + long timeo; + + tp = tcp_sk(sk); + copied = 0; + csk = rcu_dereference_sk_user_data(sk); + cdev = csk->cdev; + timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT); + + err = sk_stream_wait_connect(sk, &timeo); + if (!sk_in_state(sk, TCPF_ESTABLISHED | TCPF_CLOSE_WAIT) && + err != 0) + goto out_err; + + mss = csk->mss; + csk_set_flag(csk, CSK_TX_MORE_DATA); + + while (size > 0) { + struct sk_buff *skb = skb_peek_tail(&csk->txq); + int copy, i; + + if (!skb || (ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NO_APPEND) || + (copy = mss - skb->len) <= 0) { +new_buf: + if (!csk_mem_free(cdev, sk)) + goto wait_for_sndbuf; + + if (is_tls_tx(csk)) { + skb = get_record_skb(sk, + select_size(sk, size, + flags, + TX_TLSHDR_LEN), + true); + } else { + skb = get_tx_skb(sk, 0); + } + if (!skb) + goto wait_for_memory; + copy = mss; + } + if (copy > size) + copy = size; + + i = skb_shinfo(skb)->nr_frags; + if (skb_can_coalesce(skb, i, page, offset)) { + skb_frag_size_add(&skb_shinfo(skb)->frags[i - 1], copy); + } else if (i < MAX_SKB_FRAGS) { + get_page(page); + skb_fill_page_desc(skb, i, page, offset, copy); + } else { + tx_skb_finalize(skb); + push_frames_if_head(sk); + goto new_buf; + } + + skb->len += copy; + if (skb->len == mss) + tx_skb_finalize(skb); + skb->data_len += copy; + skb->truesize += copy; + sk->sk_wmem_queued += copy; + tp->write_seq += copy; + copied += copy; + offset += copy; + size -= copy; + + if (corked(tp, flags) && + (sk_stream_wspace(sk) < sk_stream_min_wspace(sk))) + ULP_SKB_CB(skb)->flags |= ULPCB_FLAG_NO_APPEND; + + if (!size) + break; + + if (unlikely(ULP_SKB_CB(skb)->flags & ULPCB_FLAG_NO_APPEND)) + push_frames_if_head(sk); + continue; +wait_for_sndbuf: + set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); +wait_for_memory: + err = csk_wait_memory(cdev, sk, &timeo); + if (err) + goto do_error; + } +out: + csk_reset_flag(csk, CSK_TX_MORE_DATA); + if (copied) + chtls_tcp_push(sk, flags); +done: + release_sock(sk); + return copied; + +do_error: + if (copied) + goto out; + +out_err: + if (csk_conn_inline(csk)) + csk_reset_flag(csk, CSK_TX_MORE_DATA); + copied = sk_stream_error(sk, flags, err); + goto done; +} + +static void chtls_select_window(struct sock *sk) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + struct tcp_sock *tp = tcp_sk(sk); + unsigned int wnd = tp->rcv_wnd; + + wnd = max_t(unsigned int, wnd, tcp_full_space(sk)); + wnd = max_t(unsigned int, MIN_RCV_WND, wnd); + + if (wnd > MAX_RCV_WND) + wnd = MAX_RCV_WND; + +/* + * Check if we need to grow the receive window in response to an increase in + * the socket's receive buffer size. Some applications increase the buffer + * size dynamically and rely on the window to grow accordingly. + */ + + if (wnd > tp->rcv_wnd) { + tp->rcv_wup -= wnd - tp->rcv_wnd; + tp->rcv_wnd = wnd; + /* Mark the receive window as updated */ + csk_reset_flag(csk, CSK_UPDATE_RCV_WND); + } +} + +/* + * Send RX credits through an RX_DATA_ACK CPL message. We are permitted + * to return without sending the message in case we cannot allocate + * an sk_buff. Returns the number of credits sent. + */ +static u32 send_rx_credits(struct chtls_sock *csk, u32 credits) +{ + struct cpl_rx_data_ack *req; + struct sk_buff *skb; + + skb = alloc_skb(sizeof(*req), GFP_ATOMIC); + if (!skb) + return 0; + __skb_put(skb, sizeof(*req)); + req = (struct cpl_rx_data_ack *)skb->head; + + set_wr_txq(skb, CPL_PRIORITY_ACK, csk->port_id); + INIT_TP_WR(req, csk->tid); + OPCODE_TID(req) = cpu_to_be32(MK_OPCODE_TID(CPL_RX_DATA_ACK, + csk->tid)); + req->credit_dack = cpu_to_be32(RX_CREDITS_V(credits) | + RX_FORCE_ACK_F); + cxgb4_ofld_send(csk->cdev->ports[csk->port_id], skb); + return credits; +} + +#define CREDIT_RETURN_STATE (TCPF_ESTABLISHED | \ + TCPF_FIN_WAIT1 | \ + TCPF_FIN_WAIT2) + +/* + * Called after some received data has been read. It returns RX credits + * to the HW for the amount of data processed. + */ +static void chtls_cleanup_rbuf(struct sock *sk, int copied) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + struct tcp_sock *tp; + int must_send; + u32 credits; + u32 thres; + + thres = 15 * 1024; + + if (!sk_in_state(sk, CREDIT_RETURN_STATE)) + return; + + chtls_select_window(sk); + tp = tcp_sk(sk); + credits = tp->copied_seq - tp->rcv_wup; + if (unlikely(!credits)) + return; + +/* + * For coalescing to work effectively ensure the receive window has + * at least 16KB left. + */ + must_send = credits + 16384 >= tp->rcv_wnd; + + if (must_send || credits >= thres) + tp->rcv_wup += send_rx_credits(csk, credits); +} + +static int chtls_pt_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, + int nonblock, int flags, int *addr_len) +{ + struct chtls_sock *csk = rcu_dereference_sk_user_data(sk); + struct chtls_hws *hws = &csk->tlshws; + struct net_device *dev = csk->egress_dev; + struct adapter *adap = netdev2adap(dev); + struct tcp_sock *tp = tcp_sk(sk); + unsigned long avail; + int buffers_freed; + int copied = 0; + int target; + long timeo; + + buffers_freed = 0; + + timeo = sock_rcvtimeo(sk, nonblock); + target = sock_rcvlowat(sk, flags & MSG_WAITALL, len); + + if (unlikely(csk_flag(sk, CSK_UPDATE_RCV_WND))) + chtls_cleanup_rbuf(sk, copied); + + do { + struct sk_buff *skb; + u32 offset = 0; + + if (unlikely(tp->urg_data && + tp->urg_seq == tp->copied_seq)) { + if (copied) + break; + if (signal_pending(current)) { + copied = timeo ? sock_intr_errno(timeo) : + -EAGAIN; + break; + } + } + skb = skb_peek(&sk->sk_receive_queue); + if (skb) + goto found_ok_skb; + if (csk->wr_credits && + skb_queue_len(&csk->txq) && + chtls_push_frames(csk, csk->wr_credits == + csk->wr_max_credits)) + sk->sk_write_space(sk); + + if (copied >= target && !READ_ONCE(sk->sk_backlog.tail)) + break; + + if (copied) { + if (sk->sk_err || sk->sk_state == TCP_CLOSE || + (sk->sk_shutdown & RCV_SHUTDOWN) || + signal_pending(current)) + break; + + if (!timeo) + break; + } else { + if (sock_flag(sk, SOCK_DONE)) + break; + if (sk->sk_err) { + copied = sock_error(sk); + break; + } + if (sk->sk_shutdown & RCV_SHUTDOWN) + break; + if (sk->sk_state == TCP_CLOSE) { + copied = -ENOTCONN; + break; + } + if (!timeo) { + copied = -EAGAIN; + break; + } + if (signal_pending(current)) { + copied = sock_intr_errno(timeo); + break; + } + } + if (READ_ONCE(sk->sk_backlog.tail)) { + release_sock(sk); + lock_sock(sk); + chtls_cleanup_rbuf(sk, copied); + continue; + } + + if (copied >= target) + break; + chtls_cleanup_rbuf(sk, copied); + sk_wait_data(sk, &timeo, NULL); + continue; +found_ok_skb: + if (!skb->len) { + skb_dst_set(skb, NULL); + __skb_unlink(skb, &sk->sk_receive_queue); + kfree_skb(skb); + + if (!copied && !timeo) { + copied = -EAGAIN; + break; + } + + if (copied < target) { + release_sock(sk); + lock_sock(sk); + continue; + } + break; + } + offset = hws->copied_seq; + avail = skb->len - offset; + if (len < avail) + avail = len; + + if (unlikely(tp->urg_data)) { + u32 urg_offset = tp->urg_seq - tp->copied_seq; + + if (urg_offset < avail) { + if (urg_offset) { + avail = urg_offset; + } else if (!sock_flag(sk, SOCK_URGINLINE)) { + /* First byte is urgent, skip */ + tp->copied_seq++; + offset++; + avail--; + if (!avail) + goto skip_copy; + } + } + } + /* Set record type if not already done. For a non-data record, + * do not proceed if record type could not be copied. + */ + if (ULP_SKB_CB(skb)->flags & ULPCB_FLAG_TLS_HDR) { + struct tls_hdr *thdr = (struct tls_hdr *)skb->data; + int cerr = 0; + + cerr = put_cmsg(msg, SOL_TLS, TLS_GET_RECORD_TYPE, + sizeof(thdr->type), &thdr->type); + + if (cerr && thdr->type != TLS_RECORD_TYPE_DATA) + return -EIO; + /* don't send tls header, skip copy */ + goto skip_copy; + } + + if (skb_copy_datagram_msg(skb, offset, msg, avail)) { + if (!copied) { + copied = -EFAULT; + break; + } + } + + copied += avail; + len -= avail; + hws->copied_seq += avail; +skip_copy: + if (tp->urg_data && after(tp->copied_seq, tp->urg_seq)) + tp->urg_data = 0; + + if ((avail + offset) >= skb->len) { + if (ULP_SKB_CB(skb)->flags & ULPCB_FLAG_TLS_HDR) { + tp->copied_seq += skb->len; + hws->rcvpld = skb->hdr_len; + } else { + atomic_inc(&adap->chcr_stats.tls_pdu_rx); + tp->copied_seq += hws->rcvpld; + } + chtls_free_skb(sk, skb); + buffers_freed++; + hws->copied_seq = 0; + if (copied >= target && + !skb_peek(&sk->sk_receive_queue)) + break; + } + } while (len > 0); + + if (buffers_freed) + chtls_cleanup_rbuf(sk, copied); + release_sock(sk); + return copied; +} + +/* + * Peek at data in a socket's receive buffer. + */ +static int peekmsg(struct sock *sk, struct msghdr *msg, + size_t len, int nonblock, int flags) +{ + struct tcp_sock *tp = tcp_sk(sk); + u32 peek_seq, offset; + struct sk_buff *skb; + int copied = 0; + size_t avail; /* amount of available data in current skb */ + long timeo; + + lock_sock(sk); + timeo = sock_rcvtimeo(sk, nonblock); + peek_seq = tp->copied_seq; + + do { + if (unlikely(tp->urg_data && tp->urg_seq == peek_seq)) { + if (copied) + break; + if (signal_pending(current)) { + copied = timeo ? sock_intr_errno(timeo) : + -EAGAIN; + break; + } + } + + skb_queue_walk(&sk->sk_receive_queue, skb) { + offset = peek_seq - ULP_SKB_CB(skb)->seq; + if (offset < skb->len) + goto found_ok_skb; + } + + /* empty receive queue */ + if (copied) + break; + if (sock_flag(sk, SOCK_DONE)) + break; + if (sk->sk_err) { + copied = sock_error(sk); + break; + } + if (sk->sk_shutdown & RCV_SHUTDOWN) + break; + if (sk->sk_state == TCP_CLOSE) { + copied = -ENOTCONN; + break; + } + if (!timeo) { + copied = -EAGAIN; + break; + } + if (signal_pending(current)) { + copied = sock_intr_errno(timeo); + break; + } + + if (READ_ONCE(sk->sk_backlog.tail)) { + /* Do not sleep, just process backlog. */ + release_sock(sk); + lock_sock(sk); + } else { + sk_wait_data(sk, &timeo, NULL); + } + + if (unlikely(peek_seq != tp->copied_seq)) { + if (net_ratelimit()) + pr_info("TCP(%s:%d), race in MSG_PEEK.\n", + current->comm, current->pid); + peek_seq = tp->copied_seq; + } + continue; + +found_ok_skb: + avail = skb->len - offset; + if (len < avail) + avail = len; + /* + * Do we have urgent data here? We need to skip over the + * urgent byte. + */ + if (unlikely(tp->urg_data)) { + u32 urg_offset = tp->urg_seq - peek_seq; + + if (urg_offset < avail) { + /* + * The amount of data we are preparing to copy + * contains urgent data. + */ + if (!urg_offset) { /* First byte is urgent */ + if (!sock_flag(sk, SOCK_URGINLINE)) { + peek_seq++; + offset++; + avail--; + } + if (!avail) + continue; + } else { + /* stop short of the urgent data */ + avail = urg_offset; + } + } + } + + /* + * If MSG_TRUNC is specified the data is discarded. + */ + if (likely(!(flags & MSG_TRUNC))) + if (skb_copy_datagram_msg(skb, offset, msg, len)) { + if (!copied) { + copied = -EFAULT; + break; + } + } + peek_seq += avail; + copied += avail; + len -= avail; + } while (len > 0); + + release_sock(sk); + return copied; +} + +int chtls_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, + int nonblock, int flags, int *addr_len) +{ + struct tcp_sock *tp = tcp_sk(sk); + struct chtls_sock *csk; + unsigned long avail; /* amount of available data in current skb */ + int buffers_freed; + int copied = 0; + long timeo; + int target; /* Read at least this many bytes */ + + buffers_freed = 0; + + if (unlikely(flags & MSG_OOB)) + return tcp_prot.recvmsg(sk, msg, len, nonblock, flags, + addr_len); + + if (unlikely(flags & MSG_PEEK)) + return peekmsg(sk, msg, len, nonblock, flags); + + if (sk_can_busy_loop(sk) && + skb_queue_empty_lockless(&sk->sk_receive_queue) && + sk->sk_state == TCP_ESTABLISHED) + sk_busy_loop(sk, nonblock); + + lock_sock(sk); + csk = rcu_dereference_sk_user_data(sk); + + if (is_tls_rx(csk)) + return chtls_pt_recvmsg(sk, msg, len, nonblock, + flags, addr_len); + + timeo = sock_rcvtimeo(sk, nonblock); + target = sock_rcvlowat(sk, flags & MSG_WAITALL, len); + + if (unlikely(csk_flag(sk, CSK_UPDATE_RCV_WND))) + chtls_cleanup_rbuf(sk, copied); + + do { + struct sk_buff *skb; + u32 offset; + + if (unlikely(tp->urg_data && tp->urg_seq == tp->copied_seq)) { + if (copied) + break; + if (signal_pending(current)) { + copied = timeo ? sock_intr_errno(timeo) : + -EAGAIN; + break; + } + } + + skb = skb_peek(&sk->sk_receive_queue); + if (skb) + goto found_ok_skb; + + if (csk->wr_credits && + skb_queue_len(&csk->txq) && + chtls_push_frames(csk, csk->wr_credits == + csk->wr_max_credits)) + sk->sk_write_space(sk); + + if (copied >= target && !READ_ONCE(sk->sk_backlog.tail)) + break; + + if (copied) { + if (sk->sk_err || sk->sk_state == TCP_CLOSE || + (sk->sk_shutdown & RCV_SHUTDOWN) || + signal_pending(current)) + break; + } else { + if (sock_flag(sk, SOCK_DONE)) + break; + if (sk->sk_err) { + copied = sock_error(sk); + break; + } + if (sk->sk_shutdown & RCV_SHUTDOWN) + break; + if (sk->sk_state == TCP_CLOSE) { + copied = -ENOTCONN; + break; + } + if (!timeo) { + copied = -EAGAIN; + break; + } + if (signal_pending(current)) { + copied = sock_intr_errno(timeo); + break; + } + } + + if (READ_ONCE(sk->sk_backlog.tail)) { + release_sock(sk); + lock_sock(sk); + chtls_cleanup_rbuf(sk, copied); + continue; + } + + if (copied >= target) + break; + chtls_cleanup_rbuf(sk, copied); + sk_wait_data(sk, &timeo, NULL); + continue; + +found_ok_skb: + if (!skb->len) { + chtls_kfree_skb(sk, skb); + if (!copied && !timeo) { + copied = -EAGAIN; + break; + } + + if (copied < target) + continue; + + break; + } + + offset = tp->copied_seq - ULP_SKB_CB(skb)->seq; + avail = skb->len - offset; + if (len < avail) + avail = len; + + if (unlikely(tp->urg_data)) { + u32 urg_offset = tp->urg_seq - tp->copied_seq; + + if (urg_offset < avail) { + if (urg_offset) { + avail = urg_offset; + } else if (!sock_flag(sk, SOCK_URGINLINE)) { + tp->copied_seq++; + offset++; + avail--; + if (!avail) + goto skip_copy; + } + } + } + + if (likely(!(flags & MSG_TRUNC))) { + if (skb_copy_datagram_msg(skb, offset, + msg, avail)) { + if (!copied) { + copied = -EFAULT; + break; + } + } + } + + tp->copied_seq += avail; + copied += avail; + len -= avail; + +skip_copy: + if (tp->urg_data && after(tp->copied_seq, tp->urg_seq)) + tp->urg_data = 0; + + if (avail + offset >= skb->len) { + chtls_free_skb(sk, skb); + buffers_freed++; + + if (copied >= target && + !skb_peek(&sk->sk_receive_queue)) + break; + } + } while (len > 0); + + if (buffers_freed) + chtls_cleanup_rbuf(sk, copied); + + release_sock(sk); + return copied; +} diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_main.c b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_main.c new file mode 100644 index 000000000000..9098b3eed4da --- /dev/null +++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_main.c @@ -0,0 +1,641 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2018 Chelsio Communications, Inc. + * + * Written by: Atul Gupta (atul.gupta@chelsio.com) + */ +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/skbuff.h> +#include <linux/socket.h> +#include <linux/hash.h> +#include <linux/in.h> +#include <linux/net.h> +#include <linux/ip.h> +#include <linux/tcp.h> +#include <net/ipv6.h> +#include <net/transp_v6.h> +#include <net/tcp.h> +#include <net/tls.h> + +#include "chtls.h" +#include "chtls_cm.h" + +#define DRV_NAME "chtls" + +/* + * chtls device management + * maintains a list of the chtls devices + */ +static LIST_HEAD(cdev_list); +static DEFINE_MUTEX(cdev_mutex); + +static DEFINE_MUTEX(notify_mutex); +static RAW_NOTIFIER_HEAD(listen_notify_list); +static struct proto chtls_cpl_prot, chtls_cpl_protv6; +struct request_sock_ops chtls_rsk_ops, chtls_rsk_opsv6; +static uint send_page_order = (14 - PAGE_SHIFT < 0) ? 0 : 14 - PAGE_SHIFT; + +static void register_listen_notifier(struct notifier_block *nb) +{ + mutex_lock(¬ify_mutex); + raw_notifier_chain_register(&listen_notify_list, nb); + mutex_unlock(¬ify_mutex); +} + +static void unregister_listen_notifier(struct notifier_block *nb) +{ + mutex_lock(¬ify_mutex); + raw_notifier_chain_unregister(&listen_notify_list, nb); + mutex_unlock(¬ify_mutex); +} + +static int listen_notify_handler(struct notifier_block *this, + unsigned long event, void *data) +{ + struct chtls_listen *clisten; + int ret = NOTIFY_DONE; + + clisten = (struct chtls_listen *)data; + + switch (event) { + case CHTLS_LISTEN_START: + ret = chtls_listen_start(clisten->cdev, clisten->sk); + kfree(clisten); + break; + case CHTLS_LISTEN_STOP: + chtls_listen_stop(clisten->cdev, clisten->sk); + kfree(clisten); + break; + } + return ret; +} + +static struct notifier_block listen_notifier = { + .notifier_call = listen_notify_handler +}; + +static int listen_backlog_rcv(struct sock *sk, struct sk_buff *skb) +{ + if (likely(skb_transport_header(skb) != skb_network_header(skb))) + return tcp_v4_do_rcv(sk, skb); + BLOG_SKB_CB(skb)->backlog_rcv(sk, skb); + return 0; +} + +static int chtls_start_listen(struct chtls_dev *cdev, struct sock *sk) +{ + struct chtls_listen *clisten; + + if (sk->sk_protocol != IPPROTO_TCP) + return -EPROTONOSUPPORT; + + if (sk->sk_family == PF_INET && + LOOPBACK(inet_sk(sk)->inet_rcv_saddr)) + return -EADDRNOTAVAIL; + + sk->sk_backlog_rcv = listen_backlog_rcv; + clisten = kmalloc(sizeof(*clisten), GFP_KERNEL); + if (!clisten) + return -ENOMEM; + clisten->cdev = cdev; + clisten->sk = sk; + mutex_lock(¬ify_mutex); + raw_notifier_call_chain(&listen_notify_list, + CHTLS_LISTEN_START, clisten); + mutex_unlock(¬ify_mutex); + return 0; +} + +static void chtls_stop_listen(struct chtls_dev *cdev, struct sock *sk) +{ + struct chtls_listen *clisten; + + if (sk->sk_protocol != IPPROTO_TCP) + return; + + clisten = kmalloc(sizeof(*clisten), GFP_KERNEL); + if (!clisten) + return; + clisten->cdev = cdev; + clisten->sk = sk; + mutex_lock(¬ify_mutex); + raw_notifier_call_chain(&listen_notify_list, + CHTLS_LISTEN_STOP, clisten); + mutex_unlock(¬ify_mutex); +} + +static int chtls_inline_feature(struct tls_toe_device *dev) +{ + struct net_device *netdev; + struct chtls_dev *cdev; + int i; + + cdev = to_chtls_dev(dev); + + for (i = 0; i < cdev->lldi->nports; i++) { + netdev = cdev->ports[i]; + if (netdev->features & NETIF_F_HW_TLS_RECORD) + return 1; + } + return 0; +} + +static int chtls_create_hash(struct tls_toe_device *dev, struct sock *sk) +{ + struct chtls_dev *cdev = to_chtls_dev(dev); + + if (sk->sk_state == TCP_LISTEN) + return chtls_start_listen(cdev, sk); + return 0; +} + +static void chtls_destroy_hash(struct tls_toe_device *dev, struct sock *sk) +{ + struct chtls_dev *cdev = to_chtls_dev(dev); + + if (sk->sk_state == TCP_LISTEN) + chtls_stop_listen(cdev, sk); +} + +static void chtls_free_uld(struct chtls_dev *cdev) +{ + int i; + + tls_toe_unregister_device(&cdev->tlsdev); + kvfree(cdev->kmap.addr); + idr_destroy(&cdev->hwtid_idr); + for (i = 0; i < (1 << RSPQ_HASH_BITS); i++) + kfree_skb(cdev->rspq_skb_cache[i]); + kfree(cdev->lldi); + kfree_skb(cdev->askb); + kfree(cdev); +} + +static inline void chtls_dev_release(struct kref *kref) +{ + struct tls_toe_device *dev; + struct chtls_dev *cdev; + struct adapter *adap; + + dev = container_of(kref, struct tls_toe_device, kref); + cdev = to_chtls_dev(dev); + + /* Reset tls rx/tx stats */ + adap = pci_get_drvdata(cdev->pdev); + atomic_set(&adap->chcr_stats.tls_pdu_tx, 0); + atomic_set(&adap->chcr_stats.tls_pdu_rx, 0); + + chtls_free_uld(cdev); +} + +static void chtls_register_dev(struct chtls_dev *cdev) +{ + struct tls_toe_device *tlsdev = &cdev->tlsdev; + + strlcpy(tlsdev->name, "chtls", TLS_TOE_DEVICE_NAME_MAX); + strlcat(tlsdev->name, cdev->lldi->ports[0]->name, + TLS_TOE_DEVICE_NAME_MAX); + tlsdev->feature = chtls_inline_feature; + tlsdev->hash = chtls_create_hash; + tlsdev->unhash = chtls_destroy_hash; + tlsdev->release = chtls_dev_release; + kref_init(&tlsdev->kref); + tls_toe_register_device(tlsdev); + cdev->cdev_state = CHTLS_CDEV_STATE_UP; +} + +static void process_deferq(struct work_struct *task_param) +{ + struct chtls_dev *cdev = container_of(task_param, + struct chtls_dev, deferq_task); + struct sk_buff *skb; + + spin_lock_bh(&cdev->deferq.lock); + while ((skb = __skb_dequeue(&cdev->deferq)) != NULL) { + spin_unlock_bh(&cdev->deferq.lock); + DEFERRED_SKB_CB(skb)->handler(cdev, skb); + spin_lock_bh(&cdev->deferq.lock); + } + spin_unlock_bh(&cdev->deferq.lock); +} + +static int chtls_get_skb(struct chtls_dev *cdev) +{ + cdev->askb = alloc_skb(sizeof(struct tcphdr), GFP_KERNEL); + if (!cdev->askb) + return -ENOMEM; + + skb_put(cdev->askb, sizeof(struct tcphdr)); + skb_reset_transport_header(cdev->askb); + memset(cdev->askb->data, 0, cdev->askb->len); + return 0; +} + +static void *chtls_uld_add(const struct cxgb4_lld_info *info) +{ + struct cxgb4_lld_info *lldi; + struct chtls_dev *cdev; + int i, j; + + cdev = kzalloc(sizeof(*cdev), GFP_KERNEL); + if (!cdev) + goto out; + + lldi = kzalloc(sizeof(*lldi), GFP_KERNEL); + if (!lldi) + goto out_lldi; + + if (chtls_get_skb(cdev)) + goto out_skb; + + *lldi = *info; + cdev->lldi = lldi; + cdev->pdev = lldi->pdev; + cdev->tids = lldi->tids; + cdev->ports = lldi->ports; + cdev->mtus = lldi->mtus; + cdev->tids = lldi->tids; + cdev->pfvf = FW_VIID_PFN_G(cxgb4_port_viid(lldi->ports[0])) + << FW_VIID_PFN_S; + + for (i = 0; i < (1 << RSPQ_HASH_BITS); i++) { + unsigned int size = 64 - sizeof(struct rsp_ctrl) - 8; + + cdev->rspq_skb_cache[i] = __alloc_skb(size, + gfp_any(), 0, + lldi->nodeid); + if (unlikely(!cdev->rspq_skb_cache[i])) + goto out_rspq_skb; + } + + idr_init(&cdev->hwtid_idr); + INIT_WORK(&cdev->deferq_task, process_deferq); + spin_lock_init(&cdev->listen_lock); + spin_lock_init(&cdev->idr_lock); + cdev->send_page_order = min_t(uint, get_order(32768), + send_page_order); + cdev->max_host_sndbuf = 48 * 1024; + + if (lldi->vr->key.size) + if (chtls_init_kmap(cdev, lldi)) + goto out_rspq_skb; + + mutex_lock(&cdev_mutex); + list_add_tail(&cdev->list, &cdev_list); + mutex_unlock(&cdev_mutex); + + return cdev; +out_rspq_skb: + for (j = 0; j < i; j++) + kfree_skb(cdev->rspq_skb_cache[j]); + kfree_skb(cdev->askb); +out_skb: + kfree(lldi); +out_lldi: + kfree(cdev); +out: + return NULL; +} + +static void chtls_free_all_uld(void) +{ + struct chtls_dev *cdev, *tmp; + + mutex_lock(&cdev_mutex); + list_for_each_entry_safe(cdev, tmp, &cdev_list, list) { + if (cdev->cdev_state == CHTLS_CDEV_STATE_UP) { + list_del(&cdev->list); + kref_put(&cdev->tlsdev.kref, cdev->tlsdev.release); + } + } + mutex_unlock(&cdev_mutex); +} + +static int chtls_uld_state_change(void *handle, enum cxgb4_state new_state) +{ + struct chtls_dev *cdev = handle; + + switch (new_state) { + case CXGB4_STATE_UP: + chtls_register_dev(cdev); + break; + case CXGB4_STATE_DOWN: + break; + case CXGB4_STATE_START_RECOVERY: + break; + case CXGB4_STATE_DETACH: + mutex_lock(&cdev_mutex); + list_del(&cdev->list); + mutex_unlock(&cdev_mutex); + kref_put(&cdev->tlsdev.kref, cdev->tlsdev.release); + break; + default: + break; + } + return 0; +} + +static struct sk_buff *copy_gl_to_skb_pkt(const struct pkt_gl *gl, + const __be64 *rsp, + u32 pktshift) +{ + struct sk_buff *skb; + + /* Allocate space for cpl_pass_accpet_req which will be synthesized by + * driver. Once driver synthesizes cpl_pass_accpet_req the skb will go + * through the regular cpl_pass_accept_req processing in TOM. + */ + skb = alloc_skb(gl->tot_len + sizeof(struct cpl_pass_accept_req) + - pktshift, GFP_ATOMIC); + if (unlikely(!skb)) + return NULL; + __skb_put(skb, gl->tot_len + sizeof(struct cpl_pass_accept_req) + - pktshift); + /* For now we will copy cpl_rx_pkt in the skb */ + skb_copy_to_linear_data(skb, rsp, sizeof(struct cpl_rx_pkt)); + skb_copy_to_linear_data_offset(skb, sizeof(struct cpl_pass_accept_req) + , gl->va + pktshift, + gl->tot_len - pktshift); + + return skb; +} + +static int chtls_recv_packet(struct chtls_dev *cdev, + const struct pkt_gl *gl, const __be64 *rsp) +{ + unsigned int opcode = *(u8 *)rsp; + struct sk_buff *skb; + int ret; + + skb = copy_gl_to_skb_pkt(gl, rsp, cdev->lldi->sge_pktshift); + if (!skb) + return -ENOMEM; + + ret = chtls_handlers[opcode](cdev, skb); + if (ret & CPL_RET_BUF_DONE) + kfree_skb(skb); + + return 0; +} + +static int chtls_recv_rsp(struct chtls_dev *cdev, const __be64 *rsp) +{ + unsigned long rspq_bin; + unsigned int opcode; + struct sk_buff *skb; + unsigned int len; + int ret; + + len = 64 - sizeof(struct rsp_ctrl) - 8; + opcode = *(u8 *)rsp; + + rspq_bin = hash_ptr((void *)rsp, RSPQ_HASH_BITS); + skb = cdev->rspq_skb_cache[rspq_bin]; + if (skb && !skb_is_nonlinear(skb) && + !skb_shared(skb) && !skb_cloned(skb)) { + refcount_inc(&skb->users); + if (refcount_read(&skb->users) == 2) { + __skb_trim(skb, 0); + if (skb_tailroom(skb) >= len) + goto copy_out; + } + refcount_dec(&skb->users); + } + skb = alloc_skb(len, GFP_ATOMIC); + if (unlikely(!skb)) + return -ENOMEM; + +copy_out: + __skb_put(skb, len); + skb_copy_to_linear_data(skb, rsp, len); + skb_reset_network_header(skb); + skb_reset_transport_header(skb); + ret = chtls_handlers[opcode](cdev, skb); + + if (ret & CPL_RET_BUF_DONE) + kfree_skb(skb); + return 0; +} + +static void chtls_recv(struct chtls_dev *cdev, + struct sk_buff **skbs, const __be64 *rsp) +{ + struct sk_buff *skb = *skbs; + unsigned int opcode; + int ret; + + opcode = *(u8 *)rsp; + + __skb_push(skb, sizeof(struct rss_header)); + skb_copy_to_linear_data(skb, rsp, sizeof(struct rss_header)); + + ret = chtls_handlers[opcode](cdev, skb); + if (ret & CPL_RET_BUF_DONE) + kfree_skb(skb); +} + +static int chtls_uld_rx_handler(void *handle, const __be64 *rsp, + const struct pkt_gl *gl) +{ + struct chtls_dev *cdev = handle; + unsigned int opcode; + struct sk_buff *skb; + + opcode = *(u8 *)rsp; + + if (unlikely(opcode == CPL_RX_PKT)) { + if (chtls_recv_packet(cdev, gl, rsp) < 0) + goto nomem; + return 0; + } + + if (!gl) + return chtls_recv_rsp(cdev, rsp); + +#define RX_PULL_LEN 128 + skb = cxgb4_pktgl_to_skb(gl, RX_PULL_LEN, RX_PULL_LEN); + if (unlikely(!skb)) + goto nomem; + chtls_recv(cdev, &skb, rsp); + return 0; + +nomem: + return -ENOMEM; +} + +static int do_chtls_getsockopt(struct sock *sk, char __user *optval, + int __user *optlen) +{ + struct tls_crypto_info crypto_info = { 0 }; + + crypto_info.version = TLS_1_2_VERSION; + if (copy_to_user(optval, &crypto_info, sizeof(struct tls_crypto_info))) + return -EFAULT; + return 0; +} + +static int chtls_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen) +{ + struct tls_context *ctx = tls_get_ctx(sk); + + if (level != SOL_TLS) + return ctx->sk_proto->getsockopt(sk, level, + optname, optval, optlen); + + return do_chtls_getsockopt(sk, optval, optlen); +} + +static int do_chtls_setsockopt(struct sock *sk, int optname, + sockptr_t optval, unsigned int optlen) +{ + struct tls_crypto_info *crypto_info, tmp_crypto_info; + struct chtls_sock *csk; + int keylen; + int cipher_type; + int rc = 0; + + csk = rcu_dereference_sk_user_data(sk); + + if (sockptr_is_null(optval) || optlen < sizeof(*crypto_info)) { + rc = -EINVAL; + goto out; + } + + rc = copy_from_sockptr(&tmp_crypto_info, optval, sizeof(*crypto_info)); + if (rc) { + rc = -EFAULT; + goto out; + } + + /* check version */ + if (tmp_crypto_info.version != TLS_1_2_VERSION) { + rc = -ENOTSUPP; + goto out; + } + + crypto_info = (struct tls_crypto_info *)&csk->tlshws.crypto_info; + + /* GCM mode of AES supports 128 and 256 bit encryption, so + * copy keys from user based on GCM cipher type. + */ + switch (tmp_crypto_info.cipher_type) { + case TLS_CIPHER_AES_GCM_128: { + /* Obtain version and type from previous copy */ + crypto_info[0] = tmp_crypto_info; + /* Now copy the following data */ + rc = copy_from_sockptr_offset((char *)crypto_info + + sizeof(*crypto_info), + optval, sizeof(*crypto_info), + sizeof(struct tls12_crypto_info_aes_gcm_128) + - sizeof(*crypto_info)); + + if (rc) { + rc = -EFAULT; + goto out; + } + + keylen = TLS_CIPHER_AES_GCM_128_KEY_SIZE; + cipher_type = TLS_CIPHER_AES_GCM_128; + break; + } + case TLS_CIPHER_AES_GCM_256: { + crypto_info[0] = tmp_crypto_info; + rc = copy_from_sockptr_offset((char *)crypto_info + + sizeof(*crypto_info), + optval, sizeof(*crypto_info), + sizeof(struct tls12_crypto_info_aes_gcm_256) + - sizeof(*crypto_info)); + + if (rc) { + rc = -EFAULT; + goto out; + } + + keylen = TLS_CIPHER_AES_GCM_256_KEY_SIZE; + cipher_type = TLS_CIPHER_AES_GCM_256; + break; + } + default: + rc = -EINVAL; + goto out; + } + rc = chtls_setkey(csk, keylen, optname, cipher_type); +out: + return rc; +} + +static int chtls_setsockopt(struct sock *sk, int level, int optname, + sockptr_t optval, unsigned int optlen) +{ + struct tls_context *ctx = tls_get_ctx(sk); + + if (level != SOL_TLS) + return ctx->sk_proto->setsockopt(sk, level, + optname, optval, optlen); + + return do_chtls_setsockopt(sk, optname, optval, optlen); +} + +static struct cxgb4_uld_info chtls_uld_info = { + .name = DRV_NAME, + .nrxq = MAX_ULD_QSETS, + .ntxq = MAX_ULD_QSETS, + .rxq_size = 1024, + .add = chtls_uld_add, + .state_change = chtls_uld_state_change, + .rx_handler = chtls_uld_rx_handler, +}; + +void chtls_install_cpl_ops(struct sock *sk) +{ + if (sk->sk_family == AF_INET) + sk->sk_prot = &chtls_cpl_prot; + else + sk->sk_prot = &chtls_cpl_protv6; +} + +static void __init chtls_init_ulp_ops(void) +{ + chtls_cpl_prot = tcp_prot; + chtls_init_rsk_ops(&chtls_cpl_prot, &chtls_rsk_ops, + &tcp_prot, PF_INET); + chtls_cpl_prot.close = chtls_close; + chtls_cpl_prot.disconnect = chtls_disconnect; + chtls_cpl_prot.destroy = chtls_destroy_sock; + chtls_cpl_prot.shutdown = chtls_shutdown; + chtls_cpl_prot.sendmsg = chtls_sendmsg; + chtls_cpl_prot.sendpage = chtls_sendpage; + chtls_cpl_prot.recvmsg = chtls_recvmsg; + chtls_cpl_prot.setsockopt = chtls_setsockopt; + chtls_cpl_prot.getsockopt = chtls_getsockopt; +#if IS_ENABLED(CONFIG_IPV6) + chtls_cpl_protv6 = chtls_cpl_prot; + chtls_init_rsk_ops(&chtls_cpl_protv6, &chtls_rsk_opsv6, + &tcpv6_prot, PF_INET6); +#endif +} + +static int __init chtls_register(void) +{ + chtls_init_ulp_ops(); + register_listen_notifier(&listen_notifier); + cxgb4_register_uld(CXGB4_ULD_TLS, &chtls_uld_info); + return 0; +} + +static void __exit chtls_unregister(void) +{ + unregister_listen_notifier(&listen_notifier); + chtls_free_all_uld(); + cxgb4_unregister_uld(CXGB4_ULD_TLS); +} + +module_init(chtls_register); +module_exit(chtls_unregister); + +MODULE_DESCRIPTION("Chelsio TLS Inline driver"); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Chelsio Communications"); +MODULE_VERSION(CHTLS_DRV_VERSION); diff --git a/drivers/net/ethernet/cirrus/cs89x0.h b/drivers/net/ethernet/cirrus/cs89x0.h index 91423b70bb45..210f9ec9af4b 100644 --- a/drivers/net/ethernet/cirrus/cs89x0.h +++ b/drivers/net/ethernet/cirrus/cs89x0.h @@ -459,7 +459,3 @@ #define PNP_CNF_INT 0x70 #define PNP_CNF_DMA 0x74 #define PNP_CNF_MEM 0x48 - -#define BIT0 1 -#define BIT15 0x8000 - diff --git a/drivers/net/ethernet/cisco/enic/enic.h b/drivers/net/ethernet/cisco/enic/enic.h index 18f3aeb88f22..c67a16a48d62 100644 --- a/drivers/net/ethernet/cisco/enic/enic.h +++ b/drivers/net/ethernet/cisco/enic/enic.h @@ -169,6 +169,7 @@ struct enic { u16 num_vfs; #endif spinlock_t enic_api_lock; + bool enic_api_busy; struct enic_port_profile *pp; /* work queue cache line section */ diff --git a/drivers/net/ethernet/cisco/enic/enic_api.c b/drivers/net/ethernet/cisco/enic/enic_api.c index b161f24522b8..3bdc74fba1e3 100644 --- a/drivers/net/ethernet/cisco/enic/enic_api.c +++ b/drivers/net/ethernet/cisco/enic/enic_api.c @@ -1,4 +1,4 @@ -/** +/* * Copyright 2013 Cisco Systems, Inc. All rights reserved. * * This program is free software; you may redistribute it and/or modify @@ -34,6 +34,12 @@ int enic_api_devcmd_proxy_by_index(struct net_device *netdev, int vf, struct vnic_dev *vdev = enic->vdev; spin_lock(&enic->enic_api_lock); + while (enic->enic_api_busy) { + spin_unlock(&enic->enic_api_lock); + cpu_relax(); + spin_lock(&enic->enic_api_lock); + } + spin_lock_bh(&enic->devcmd_lock); vnic_dev_cmd_proxy_by_index_start(vdev, vf); diff --git a/drivers/net/ethernet/cisco/enic/enic_ethtool.c b/drivers/net/ethernet/cisco/enic/enic_ethtool.c index 4d8e0aa447fb..a4dd52bba2c3 100644 --- a/drivers/net/ethernet/cisco/enic/enic_ethtool.c +++ b/drivers/net/ethernet/cisco/enic/enic_ethtool.c @@ -1,4 +1,4 @@ -/** +/* * Copyright 2013 Cisco Systems, Inc. All rights reserved. * * This program is free software; you may redistribute it and/or modify diff --git a/drivers/net/ethernet/cisco/enic/enic_main.c b/drivers/net/ethernet/cisco/enic/enic_main.c index 552d89fdf54a..fb269d587b74 100644 --- a/drivers/net/ethernet/cisco/enic/enic_main.c +++ b/drivers/net/ethernet/cisco/enic/enic_main.c @@ -326,11 +326,11 @@ static void enic_free_wq_buf(struct vnic_wq *wq, struct vnic_wq_buf *buf) struct enic *enic = vnic_dev_priv(wq->vdev); if (buf->sop) - pci_unmap_single(enic->pdev, buf->dma_addr, - buf->len, PCI_DMA_TODEVICE); + dma_unmap_single(&enic->pdev->dev, buf->dma_addr, buf->len, + DMA_TO_DEVICE); else - pci_unmap_page(enic->pdev, buf->dma_addr, - buf->len, PCI_DMA_TODEVICE); + dma_unmap_page(&enic->pdev->dev, buf->dma_addr, buf->len, + DMA_TO_DEVICE); if (buf->os_buf) dev_kfree_skb_any(buf->os_buf); @@ -574,8 +574,8 @@ static int enic_queue_wq_skb_vlan(struct enic *enic, struct vnic_wq *wq, dma_addr_t dma_addr; int err = 0; - dma_addr = pci_map_single(enic->pdev, skb->data, head_len, - PCI_DMA_TODEVICE); + dma_addr = dma_map_single(&enic->pdev->dev, skb->data, head_len, + DMA_TO_DEVICE); if (unlikely(enic_dma_map_check(enic, dma_addr))) return -ENOMEM; @@ -605,8 +605,8 @@ static int enic_queue_wq_skb_csum_l4(struct enic *enic, struct vnic_wq *wq, dma_addr_t dma_addr; int err = 0; - dma_addr = pci_map_single(enic->pdev, skb->data, head_len, - PCI_DMA_TODEVICE); + dma_addr = dma_map_single(&enic->pdev->dev, skb->data, head_len, + DMA_TO_DEVICE); if (unlikely(enic_dma_map_check(enic, dma_addr))) return -ENOMEM; @@ -693,8 +693,9 @@ static int enic_queue_wq_skb_tso(struct enic *enic, struct vnic_wq *wq, */ while (frag_len_left) { len = min(frag_len_left, (unsigned int)WQ_ENET_MAX_DESC_LEN); - dma_addr = pci_map_single(enic->pdev, skb->data + offset, len, - PCI_DMA_TODEVICE); + dma_addr = dma_map_single(&enic->pdev->dev, + skb->data + offset, len, + DMA_TO_DEVICE); if (unlikely(enic_dma_map_check(enic, dma_addr))) return -ENOMEM; enic_queue_wq_desc_tso(wq, skb, dma_addr, len, mss, hdr_len, @@ -752,8 +753,8 @@ static inline int enic_queue_wq_skb_encap(struct enic *enic, struct vnic_wq *wq, dma_addr_t dma_addr; int err = 0; - dma_addr = pci_map_single(enic->pdev, skb->data, head_len, - PCI_DMA_TODEVICE); + dma_addr = dma_map_single(&enic->pdev->dev, skb->data, head_len, + DMA_TO_DEVICE); if (unlikely(enic_dma_map_check(enic, dma_addr))) return -ENOMEM; @@ -1222,8 +1223,8 @@ static void enic_free_rq_buf(struct vnic_rq *rq, struct vnic_rq_buf *buf) if (!buf->os_buf) return; - pci_unmap_single(enic->pdev, buf->dma_addr, - buf->len, PCI_DMA_FROMDEVICE); + dma_unmap_single(&enic->pdev->dev, buf->dma_addr, buf->len, + DMA_FROM_DEVICE); dev_kfree_skb_any(buf->os_buf); buf->os_buf = NULL; } @@ -1248,8 +1249,8 @@ static int enic_rq_alloc_buf(struct vnic_rq *rq) if (!skb) return -ENOMEM; - dma_addr = pci_map_single(enic->pdev, skb->data, len, - PCI_DMA_FROMDEVICE); + dma_addr = dma_map_single(&enic->pdev->dev, skb->data, len, + DMA_FROM_DEVICE); if (unlikely(enic_dma_map_check(enic, dma_addr))) { dev_kfree_skb(skb); return -ENOMEM; @@ -1281,8 +1282,8 @@ static bool enic_rxcopybreak(struct net_device *netdev, struct sk_buff **skb, new_skb = netdev_alloc_skb_ip_align(netdev, len); if (!new_skb) return false; - pci_dma_sync_single_for_cpu(enic->pdev, buf->dma_addr, len, - DMA_FROM_DEVICE); + dma_sync_single_for_cpu(&enic->pdev->dev, buf->dma_addr, len, + DMA_FROM_DEVICE); memcpy(new_skb->data, (*skb)->data, len); *skb = new_skb; @@ -1331,8 +1332,8 @@ static void enic_rq_indicate_buf(struct vnic_rq *rq, enic->rq_truncated_pkts++; } - pci_unmap_single(enic->pdev, buf->dma_addr, buf->len, - PCI_DMA_FROMDEVICE); + dma_unmap_single(&enic->pdev->dev, buf->dma_addr, buf->len, + DMA_FROM_DEVICE); dev_kfree_skb_any(skb); buf->os_buf = NULL; @@ -1346,8 +1347,8 @@ static void enic_rq_indicate_buf(struct vnic_rq *rq, if (!enic_rxcopybreak(netdev, &skb, buf, bytes_written)) { buf->os_buf = NULL; - pci_unmap_single(enic->pdev, buf->dma_addr, buf->len, - PCI_DMA_FROMDEVICE); + dma_unmap_single(&enic->pdev->dev, buf->dma_addr, + buf->len, DMA_FROM_DEVICE); } prefetch(skb->data - NET_IP_ALIGN); @@ -1420,8 +1421,8 @@ static void enic_rq_indicate_buf(struct vnic_rq *rq, /* Buffer overflow */ - pci_unmap_single(enic->pdev, buf->dma_addr, buf->len, - PCI_DMA_FROMDEVICE); + dma_unmap_single(&enic->pdev->dev, buf->dma_addr, buf->len, + DMA_FROM_DEVICE); dev_kfree_skb_any(skb); buf->os_buf = NULL; } @@ -2106,8 +2107,6 @@ static int enic_dev_wait(struct vnic_dev *vdev, int done; int err; - BUG_ON(in_interrupt()); - err = start(vdev, arg); if (err) return err; @@ -2178,9 +2177,9 @@ int __enic_set_rsskey(struct enic *enic) dma_addr_t rss_key_buf_pa; int i, kidx, bidx, err; - rss_key_buf_va = pci_zalloc_consistent(enic->pdev, - sizeof(union vnic_rss_key), - &rss_key_buf_pa); + rss_key_buf_va = dma_alloc_coherent(&enic->pdev->dev, + sizeof(union vnic_rss_key), + &rss_key_buf_pa, GFP_ATOMIC); if (!rss_key_buf_va) return -ENOMEM; @@ -2195,8 +2194,8 @@ int __enic_set_rsskey(struct enic *enic) sizeof(union vnic_rss_key)); spin_unlock_bh(&enic->devcmd_lock); - pci_free_consistent(enic->pdev, sizeof(union vnic_rss_key), - rss_key_buf_va, rss_key_buf_pa); + dma_free_coherent(&enic->pdev->dev, sizeof(union vnic_rss_key), + rss_key_buf_va, rss_key_buf_pa); return err; } @@ -2215,8 +2214,9 @@ static int enic_set_rsscpu(struct enic *enic, u8 rss_hash_bits) unsigned int i; int err; - rss_cpu_buf_va = pci_alloc_consistent(enic->pdev, - sizeof(union vnic_rss_cpu), &rss_cpu_buf_pa); + rss_cpu_buf_va = dma_alloc_coherent(&enic->pdev->dev, + sizeof(union vnic_rss_cpu), + &rss_cpu_buf_pa, GFP_ATOMIC); if (!rss_cpu_buf_va) return -ENOMEM; @@ -2229,8 +2229,8 @@ static int enic_set_rsscpu(struct enic *enic, u8 rss_hash_bits) sizeof(union vnic_rss_cpu)); spin_unlock_bh(&enic->devcmd_lock); - pci_free_consistent(enic->pdev, sizeof(union vnic_rss_cpu), - rss_cpu_buf_va, rss_cpu_buf_pa); + dma_free_coherent(&enic->pdev->dev, sizeof(union vnic_rss_cpu), + rss_cpu_buf_va, rss_cpu_buf_pa); return err; } @@ -2295,6 +2295,13 @@ static int enic_set_rss_nic_cfg(struct enic *enic) rss_hash_bits, rss_base_cpu, rss_enable); } +static void enic_set_api_busy(struct enic *enic, bool busy) +{ + spin_lock(&enic->enic_api_lock); + enic->enic_api_busy = busy; + spin_unlock(&enic->enic_api_lock); +} + static void enic_reset(struct work_struct *work) { struct enic *enic = container_of(work, struct enic, reset); @@ -2304,7 +2311,9 @@ static void enic_reset(struct work_struct *work) rtnl_lock(); - spin_lock(&enic->enic_api_lock); + /* Stop any activity from infiniband */ + enic_set_api_busy(enic, true); + enic_stop(enic->netdev); enic_dev_soft_reset(enic); enic_reset_addr_lists(enic); @@ -2312,7 +2321,10 @@ static void enic_reset(struct work_struct *work) enic_set_rss_nic_cfg(enic); enic_dev_set_ig_vlan_rewrite_mode(enic); enic_open(enic->netdev); - spin_unlock(&enic->enic_api_lock); + + /* Allow infiniband to fiddle with the device again */ + enic_set_api_busy(enic, false); + call_netdevice_notifiers(NETDEV_REBOOT, enic->netdev); rtnl_unlock(); @@ -2324,7 +2336,9 @@ static void enic_tx_hang_reset(struct work_struct *work) rtnl_lock(); - spin_lock(&enic->enic_api_lock); + /* Stop any activity from infiniband */ + enic_set_api_busy(enic, true); + enic_dev_hang_notify(enic); enic_stop(enic->netdev); enic_dev_hang_reset(enic); @@ -2333,7 +2347,10 @@ static void enic_tx_hang_reset(struct work_struct *work) enic_set_rss_nic_cfg(enic); enic_dev_set_ig_vlan_rewrite_mode(enic); enic_open(enic->netdev); - spin_unlock(&enic->enic_api_lock); + + /* Allow infiniband to fiddle with the device again */ + enic_set_api_busy(enic, false); + call_netdevice_notifiers(NETDEV_REBOOT, enic->netdev); rtnl_unlock(); @@ -2527,13 +2544,15 @@ static void enic_dev_deinit(struct enic *enic) { unsigned int i; - for (i = 0; i < enic->rq_count; i++) { - napi_hash_del(&enic->napi[i]); - netif_napi_del(&enic->napi[i]); - } + for (i = 0; i < enic->rq_count; i++) + __netif_napi_del(&enic->napi[i]); + if (vnic_dev_get_intr_mode(enic->vdev) == VNIC_DEV_INTR_MODE_MSIX) for (i = 0; i < enic->wq_count; i++) - netif_napi_del(&enic->napi[enic_cq_wq(enic, i)]); + __netif_napi_del(&enic->napi[enic_cq_wq(enic, i)]); + + /* observe RCU grace period after __netif_napi_del() calls */ + synchronize_net(); enic_free_vnic_resources(enic); enic_clear_intr_mode(enic); @@ -2699,21 +2718,21 @@ static int enic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) * fail to 32-bit. */ - err = pci_set_dma_mask(pdev, DMA_BIT_MASK(47)); + err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(47)); if (err) { - err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32)); + err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)); if (err) { dev_err(dev, "No usable DMA configuration, aborting\n"); goto err_out_release_regions; } - err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32)); + err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32)); if (err) { dev_err(dev, "Unable to obtain %u-bit DMA " "for consistent allocations, aborting\n", 32); goto err_out_release_regions; } } else { - err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(47)); + err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(47)); if (err) { dev_err(dev, "Unable to obtain %u-bit DMA " "for consistent allocations, aborting\n", 47); diff --git a/drivers/net/ethernet/cisco/enic/vnic_dev.c b/drivers/net/ethernet/cisco/enic/vnic_dev.c index 901e44b0b795..45015931b335 100644 --- a/drivers/net/ethernet/cisco/enic/vnic_dev.c +++ b/drivers/net/ethernet/cisco/enic/vnic_dev.c @@ -193,9 +193,10 @@ int vnic_dev_alloc_desc_ring(struct vnic_dev *vdev, struct vnic_dev_ring *ring, { vnic_dev_desc_ring_size(ring, desc_count, desc_size); - ring->descs_unaligned = pci_alloc_consistent(vdev->pdev, - ring->size_unaligned, - &ring->base_addr_unaligned); + ring->descs_unaligned = dma_alloc_coherent(&vdev->pdev->dev, + ring->size_unaligned, + &ring->base_addr_unaligned, + GFP_KERNEL); if (!ring->descs_unaligned) { vdev_err(vdev, "Failed to allocate ring (size=%d), aborting\n", @@ -218,10 +219,9 @@ int vnic_dev_alloc_desc_ring(struct vnic_dev *vdev, struct vnic_dev_ring *ring, void vnic_dev_free_desc_ring(struct vnic_dev *vdev, struct vnic_dev_ring *ring) { if (ring->descs) { - pci_free_consistent(vdev->pdev, - ring->size_unaligned, - ring->descs_unaligned, - ring->base_addr_unaligned); + dma_free_coherent(&vdev->pdev->dev, ring->size_unaligned, + ring->descs_unaligned, + ring->base_addr_unaligned); ring->descs = NULL; } } @@ -551,9 +551,9 @@ int vnic_dev_fw_info(struct vnic_dev *vdev, int err = 0; if (!vdev->fw_info) { - vdev->fw_info = pci_zalloc_consistent(vdev->pdev, - sizeof(struct vnic_devcmd_fw_info), - &vdev->fw_info_pa); + vdev->fw_info = dma_alloc_coherent(&vdev->pdev->dev, + sizeof(struct vnic_devcmd_fw_info), + &vdev->fw_info_pa, GFP_ATOMIC); if (!vdev->fw_info) return -ENOMEM; @@ -603,8 +603,9 @@ int vnic_dev_stats_dump(struct vnic_dev *vdev, struct vnic_stats **stats) int wait = 1000; if (!vdev->stats) { - vdev->stats = pci_alloc_consistent(vdev->pdev, - sizeof(struct vnic_stats), &vdev->stats_pa); + vdev->stats = dma_alloc_coherent(&vdev->pdev->dev, + sizeof(struct vnic_stats), + &vdev->stats_pa, GFP_ATOMIC); if (!vdev->stats) return -ENOMEM; } @@ -852,9 +853,9 @@ int vnic_dev_notify_set(struct vnic_dev *vdev, u16 intr) return -EINVAL; } - notify_addr = pci_alloc_consistent(vdev->pdev, - sizeof(struct vnic_devcmd_notify), - ¬ify_pa); + notify_addr = dma_alloc_coherent(&vdev->pdev->dev, + sizeof(struct vnic_devcmd_notify), + ¬ify_pa, GFP_ATOMIC); if (!notify_addr) return -ENOMEM; @@ -882,10 +883,9 @@ static int vnic_dev_notify_unsetcmd(struct vnic_dev *vdev) int vnic_dev_notify_unset(struct vnic_dev *vdev) { if (vdev->notify) { - pci_free_consistent(vdev->pdev, - sizeof(struct vnic_devcmd_notify), - vdev->notify, - vdev->notify_pa); + dma_free_coherent(&vdev->pdev->dev, + sizeof(struct vnic_devcmd_notify), + vdev->notify, vdev->notify_pa); } return vnic_dev_notify_unsetcmd(vdev); @@ -1046,18 +1046,17 @@ void vnic_dev_unregister(struct vnic_dev *vdev) { if (vdev) { if (vdev->notify) - pci_free_consistent(vdev->pdev, - sizeof(struct vnic_devcmd_notify), - vdev->notify, - vdev->notify_pa); + dma_free_coherent(&vdev->pdev->dev, + sizeof(struct vnic_devcmd_notify), + vdev->notify, vdev->notify_pa); if (vdev->stats) - pci_free_consistent(vdev->pdev, - sizeof(struct vnic_stats), - vdev->stats, vdev->stats_pa); + dma_free_coherent(&vdev->pdev->dev, + sizeof(struct vnic_stats), + vdev->stats, vdev->stats_pa); if (vdev->fw_info) - pci_free_consistent(vdev->pdev, - sizeof(struct vnic_devcmd_fw_info), - vdev->fw_info, vdev->fw_info_pa); + dma_free_coherent(&vdev->pdev->dev, + sizeof(struct vnic_devcmd_fw_info), + vdev->fw_info, vdev->fw_info_pa); if (vdev->devcmd2) vnic_dev_deinit_devcmd2(vdev); @@ -1127,7 +1126,7 @@ int vnic_dev_init_prov2(struct vnic_dev *vdev, u8 *buf, u32 len) void *prov_buf; int ret; - prov_buf = pci_alloc_consistent(vdev->pdev, len, &prov_pa); + prov_buf = dma_alloc_coherent(&vdev->pdev->dev, len, &prov_pa, GFP_ATOMIC); if (!prov_buf) return -ENOMEM; @@ -1137,7 +1136,7 @@ int vnic_dev_init_prov2(struct vnic_dev *vdev, u8 *buf, u32 len) ret = vnic_dev_cmd(vdev, CMD_INIT_PROV_INFO2, &a0, &a1, wait); - pci_free_consistent(vdev->pdev, len, prov_buf, prov_pa); + dma_free_coherent(&vdev->pdev->dev, len, prov_buf, prov_pa); return ret; } @@ -1217,7 +1216,8 @@ int vnic_dev_classifier(struct vnic_dev *vdev, u8 cmd, u16 *entry, tlv_size = sizeof(struct filter) + sizeof(struct filter_action) + 2 * sizeof(struct filter_tlv); - tlv_va = pci_alloc_consistent(vdev->pdev, tlv_size, &tlv_pa); + tlv_va = dma_alloc_coherent(&vdev->pdev->dev, tlv_size, + &tlv_pa, GFP_ATOMIC); if (!tlv_va) return -ENOMEM; tlv = tlv_va; @@ -1240,7 +1240,7 @@ int vnic_dev_classifier(struct vnic_dev *vdev, u8 cmd, u16 *entry, ret = vnic_dev_cmd(vdev, CMD_ADD_FILTER, &a0, &a1, wait); *entry = (u16)a0; - pci_free_consistent(vdev->pdev, tlv_size, tlv_va, tlv_pa); + dma_free_coherent(&vdev->pdev->dev, tlv_size, tlv_va, tlv_pa); } else if (cmd == CLSF_DEL) { a0 = *entry; ret = vnic_dev_cmd(vdev, CMD_DEL_FILTER, &a0, &a1, wait); diff --git a/drivers/net/ethernet/cortina/gemini.c b/drivers/net/ethernet/cortina/gemini.c index ffec0f3dd957..8df6f081f244 100644 --- a/drivers/net/ethernet/cortina/gemini.c +++ b/drivers/net/ethernet/cortina/gemini.c @@ -85,6 +85,8 @@ MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)"); /** * struct gmac_queue_page - page buffer per-page info + * @page: the page struct + * @mapping: the dma address handle */ struct gmac_queue_page { struct page *page; @@ -509,7 +511,6 @@ static int gmac_init(struct net_device *netdev) .rel_threshold = 0, } }; union gmac_config0 tmp; - u32 val; config0.bits.max_len = gmac_pick_rx_max_len(netdev->mtu); tmp.bits32 = readl(port->gmac_base + GMAC_CONFIG0); @@ -519,7 +520,7 @@ static int gmac_init(struct net_device *netdev) writel(config2.bits32, port->gmac_base + GMAC_CONFIG2); writel(config3.bits32, port->gmac_base + GMAC_CONFIG3); - val = readl(port->dma_base + GMAC_AHB_WEIGHT_REG); + readl(port->dma_base + GMAC_AHB_WEIGHT_REG); writel(ahb_weight.bits32, port->dma_base + GMAC_AHB_WEIGHT_REG); writel(hw_weigh.bits32, @@ -539,12 +540,6 @@ static int gmac_init(struct net_device *netdev) return 0; } -static void gmac_uninit(struct net_device *netdev) -{ - if (netdev->phydev) - phy_disconnect(netdev->phydev); -} - static int gmac_setup_txqs(struct net_device *netdev) { struct gemini_ethernet_port *port = netdev_priv(netdev); @@ -1768,15 +1763,6 @@ static int gmac_open(struct net_device *netdev) struct gemini_ethernet_port *port = netdev_priv(netdev); int err; - if (!netdev->phydev) { - err = gmac_setup_phy(netdev); - if (err) { - netif_err(port, ifup, netdev, - "PHY init failed: %d\n", err); - return err; - } - } - err = request_irq(netdev->irq, gmac_irq, IRQF_SHARED, netdev->name, netdev); if (err) { @@ -2122,9 +2108,8 @@ static void gmac_get_ringparam(struct net_device *netdev, struct ethtool_ringparam *rp) { struct gemini_ethernet_port *port = netdev_priv(netdev); - union gmac_config0 config0; - config0.bits32 = readl(port->gmac_base + GMAC_CONFIG0); + readl(port->gmac_base + GMAC_CONFIG0); rp->rx_max_pending = 1 << 15; rp->rx_mini_max_pending = 0; @@ -2209,7 +2194,6 @@ static void gmac_get_drvinfo(struct net_device *netdev, static const struct net_device_ops gmac_351x_ops = { .ndo_init = gmac_init, - .ndo_uninit = gmac_uninit, .ndo_open = gmac_open, .ndo_stop = gmac_stop, .ndo_start_xmit = gmac_start_xmit, @@ -2295,8 +2279,10 @@ static irqreturn_t gemini_port_irq(int irq, void *data) static void gemini_port_remove(struct gemini_ethernet_port *port) { - if (port->netdev) + if (port->netdev) { + phy_disconnect(port->netdev->phydev); unregister_netdev(port->netdev); + } clk_disable_unprepare(port->pclk); geth_cleanup_freeq(port->geth); } @@ -2505,6 +2491,13 @@ static int gemini_ethernet_port_probe(struct platform_device *pdev) if (ret) goto unprepare; + ret = gmac_setup_phy(netdev); + if (ret) { + netdev_err(netdev, + "PHY init failed\n"); + goto unprepare; + } + ret = register_netdev(netdev); if (ret) goto unprepare; @@ -2513,10 +2506,6 @@ static int gemini_ethernet_port_probe(struct platform_device *pdev) "irq %d, DMA @ 0x%pap, GMAC @ 0x%pap\n", port->irq, &dmares->start, &gmacres->start); - ret = gmac_setup_phy(netdev); - if (ret) - netdev_info(netdev, - "PHY init failed, deferring to ifup time\n"); return 0; unprepare: @@ -2529,6 +2518,7 @@ static int gemini_ethernet_port_remove(struct platform_device *pdev) struct gemini_ethernet_port *port = platform_get_drvdata(pdev); gemini_port_remove(port); + return 0; } diff --git a/drivers/net/ethernet/dec/tulip/de2104x.c b/drivers/net/ethernet/dec/tulip/de2104x.c index 2610efe4f873..d9f6c19940ef 100644 --- a/drivers/net/ethernet/dec/tulip/de2104x.c +++ b/drivers/net/ethernet/dec/tulip/de2104x.c @@ -443,21 +443,23 @@ static void de_rx (struct de_private *de) } if (!copying_skb) { - pci_unmap_single(de->pdev, mapping, - buflen, PCI_DMA_FROMDEVICE); + dma_unmap_single(&de->pdev->dev, mapping, buflen, + DMA_FROM_DEVICE); skb_put(skb, len); mapping = de->rx_skb[rx_tail].mapping = - pci_map_single(de->pdev, copy_skb->data, - buflen, PCI_DMA_FROMDEVICE); + dma_map_single(&de->pdev->dev, copy_skb->data, + buflen, DMA_FROM_DEVICE); de->rx_skb[rx_tail].skb = copy_skb; } else { - pci_dma_sync_single_for_cpu(de->pdev, mapping, len, PCI_DMA_FROMDEVICE); + dma_sync_single_for_cpu(&de->pdev->dev, mapping, len, + DMA_FROM_DEVICE); skb_reserve(copy_skb, RX_OFFSET); skb_copy_from_linear_data(skb, skb_put(copy_skb, len), len); - pci_dma_sync_single_for_device(de->pdev, mapping, len, PCI_DMA_FROMDEVICE); + dma_sync_single_for_device(&de->pdev->dev, mapping, + len, DMA_FROM_DEVICE); /* We'll reuse the original ring buffer. */ skb = copy_skb; @@ -554,13 +556,15 @@ static void de_tx (struct de_private *de) goto next; if (unlikely(skb == DE_SETUP_SKB)) { - pci_unmap_single(de->pdev, de->tx_skb[tx_tail].mapping, - sizeof(de->setup_frame), PCI_DMA_TODEVICE); + dma_unmap_single(&de->pdev->dev, + de->tx_skb[tx_tail].mapping, + sizeof(de->setup_frame), + DMA_TO_DEVICE); goto next; } - pci_unmap_single(de->pdev, de->tx_skb[tx_tail].mapping, - skb->len, PCI_DMA_TODEVICE); + dma_unmap_single(&de->pdev->dev, de->tx_skb[tx_tail].mapping, + skb->len, DMA_TO_DEVICE); if (status & LastFrag) { if (status & TxError) { @@ -620,7 +624,8 @@ static netdev_tx_t de_start_xmit (struct sk_buff *skb, txd = &de->tx_ring[entry]; len = skb->len; - mapping = pci_map_single(de->pdev, skb->data, len, PCI_DMA_TODEVICE); + mapping = dma_map_single(&de->pdev->dev, skb->data, len, + DMA_TO_DEVICE); if (entry == (DE_TX_RING_SIZE - 1)) flags |= RingEnd; if (!tx_free || (tx_free == (DE_TX_RING_SIZE / 2))) @@ -763,8 +768,8 @@ static void __de_set_rx_mode (struct net_device *dev) de->tx_skb[entry].skb = DE_SETUP_SKB; de->tx_skb[entry].mapping = mapping = - pci_map_single (de->pdev, de->setup_frame, - sizeof (de->setup_frame), PCI_DMA_TODEVICE); + dma_map_single(&de->pdev->dev, de->setup_frame, + sizeof(de->setup_frame), DMA_TO_DEVICE); /* Put the setup frame on the Tx list. */ txd = &de->tx_ring[entry]; @@ -1279,8 +1284,10 @@ static int de_refill_rx (struct de_private *de) if (!skb) goto err_out; - de->rx_skb[i].mapping = pci_map_single(de->pdev, - skb->data, de->rx_buf_sz, PCI_DMA_FROMDEVICE); + de->rx_skb[i].mapping = dma_map_single(&de->pdev->dev, + skb->data, + de->rx_buf_sz, + DMA_FROM_DEVICE); de->rx_skb[i].skb = skb; de->rx_ring[i].opts1 = cpu_to_le32(DescOwn); @@ -1313,7 +1320,8 @@ static int de_init_rings (struct de_private *de) static int de_alloc_rings (struct de_private *de) { - de->rx_ring = pci_alloc_consistent(de->pdev, DE_RING_BYTES, &de->ring_dma); + de->rx_ring = dma_alloc_coherent(&de->pdev->dev, DE_RING_BYTES, + &de->ring_dma, GFP_KERNEL); if (!de->rx_ring) return -ENOMEM; de->tx_ring = &de->rx_ring[DE_RX_RING_SIZE]; @@ -1333,8 +1341,9 @@ static void de_clean_rings (struct de_private *de) for (i = 0; i < DE_RX_RING_SIZE; i++) { if (de->rx_skb[i].skb) { - pci_unmap_single(de->pdev, de->rx_skb[i].mapping, - de->rx_buf_sz, PCI_DMA_FROMDEVICE); + dma_unmap_single(&de->pdev->dev, + de->rx_skb[i].mapping, de->rx_buf_sz, + DMA_FROM_DEVICE); dev_kfree_skb(de->rx_skb[i].skb); } } @@ -1344,15 +1353,15 @@ static void de_clean_rings (struct de_private *de) if ((skb) && (skb != DE_DUMMY_SKB)) { if (skb != DE_SETUP_SKB) { de->dev->stats.tx_dropped++; - pci_unmap_single(de->pdev, - de->tx_skb[i].mapping, - skb->len, PCI_DMA_TODEVICE); + dma_unmap_single(&de->pdev->dev, + de->tx_skb[i].mapping, + skb->len, DMA_TO_DEVICE); dev_kfree_skb(skb); } else { - pci_unmap_single(de->pdev, - de->tx_skb[i].mapping, - sizeof(de->setup_frame), - PCI_DMA_TODEVICE); + dma_unmap_single(&de->pdev->dev, + de->tx_skb[i].mapping, + sizeof(de->setup_frame), + DMA_TO_DEVICE); } } } @@ -1364,7 +1373,8 @@ static void de_clean_rings (struct de_private *de) static void de_free_rings (struct de_private *de) { de_clean_rings(de); - pci_free_consistent(de->pdev, DE_RING_BYTES, de->rx_ring, de->ring_dma); + dma_free_coherent(&de->pdev->dev, DE_RING_BYTES, de->rx_ring, + de->ring_dma); de->rx_ring = NULL; de->tx_ring = NULL; } diff --git a/drivers/net/ethernet/dec/tulip/de4x5.c b/drivers/net/ethernet/dec/tulip/de4x5.c index f9dd1aa9f2da..683e328b5461 100644 --- a/drivers/net/ethernet/dec/tulip/de4x5.c +++ b/drivers/net/ethernet/dec/tulip/de4x5.c @@ -4925,11 +4925,11 @@ mii_get_oui(u_char phyaddr, u_long ioaddr) u_char breg[2]; } a; int i, r2, r3, ret=0;*/ - int r2, r3; + int r2; /* Read r2 and r3 */ r2 = mii_rd(MII_ID0, phyaddr, ioaddr); - r3 = mii_rd(MII_ID1, phyaddr, ioaddr); + mii_rd(MII_ID1, phyaddr, ioaddr); /* SEEQ and Cypress way * / / * Shuffle r2 and r3 * / a.reg=0; diff --git a/drivers/net/ethernet/dec/tulip/dmfe.c b/drivers/net/ethernet/dec/tulip/dmfe.c index c3b4abff48b5..87a27fe2992d 100644 --- a/drivers/net/ethernet/dec/tulip/dmfe.c +++ b/drivers/net/ethernet/dec/tulip/dmfe.c @@ -380,7 +380,7 @@ static int dmfe_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) return -ENOMEM; SET_NETDEV_DEV(dev, &pdev->dev); - if (pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) { + if (dma_set_mask(&pdev->dev, DMA_BIT_MASK(32))) { pr_warn("32-bit PCI DMA not available\n"); err = -ENODEV; goto err_out_free; @@ -422,15 +422,17 @@ static int dmfe_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) db = netdev_priv(dev); /* Allocate Tx/Rx descriptor memory */ - db->desc_pool_ptr = pci_alloc_consistent(pdev, sizeof(struct tx_desc) * - DESC_ALL_CNT + 0x20, &db->desc_pool_dma_ptr); + db->desc_pool_ptr = dma_alloc_coherent(&pdev->dev, + sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20, + &db->desc_pool_dma_ptr, GFP_KERNEL); if (!db->desc_pool_ptr) { err = -ENOMEM; goto err_out_res; } - db->buf_pool_ptr = pci_alloc_consistent(pdev, TX_BUF_ALLOC * - TX_DESC_CNT + 4, &db->buf_pool_dma_ptr); + db->buf_pool_ptr = dma_alloc_coherent(&pdev->dev, + TX_BUF_ALLOC * TX_DESC_CNT + 4, + &db->buf_pool_dma_ptr, GFP_KERNEL); if (!db->buf_pool_ptr) { err = -ENOMEM; goto err_out_free_desc; @@ -492,11 +494,12 @@ static int dmfe_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) err_out_unmap: pci_iounmap(pdev, db->ioaddr); err_out_free_buf: - pci_free_consistent(pdev, TX_BUF_ALLOC * TX_DESC_CNT + 4, - db->buf_pool_ptr, db->buf_pool_dma_ptr); + dma_free_coherent(&pdev->dev, TX_BUF_ALLOC * TX_DESC_CNT + 4, + db->buf_pool_ptr, db->buf_pool_dma_ptr); err_out_free_desc: - pci_free_consistent(pdev, sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20, - db->desc_pool_ptr, db->desc_pool_dma_ptr); + dma_free_coherent(&pdev->dev, + sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20, + db->desc_pool_ptr, db->desc_pool_dma_ptr); err_out_res: pci_release_regions(pdev); err_out_disable: @@ -519,11 +522,12 @@ static void dmfe_remove_one(struct pci_dev *pdev) unregister_netdev(dev); pci_iounmap(db->pdev, db->ioaddr); - pci_free_consistent(db->pdev, sizeof(struct tx_desc) * - DESC_ALL_CNT + 0x20, db->desc_pool_ptr, - db->desc_pool_dma_ptr); - pci_free_consistent(db->pdev, TX_BUF_ALLOC * TX_DESC_CNT + 4, - db->buf_pool_ptr, db->buf_pool_dma_ptr); + dma_free_coherent(&db->pdev->dev, + sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20, + db->desc_pool_ptr, db->desc_pool_dma_ptr); + dma_free_coherent(&db->pdev->dev, + TX_BUF_ALLOC * TX_DESC_CNT + 4, + db->buf_pool_ptr, db->buf_pool_dma_ptr); pci_release_regions(pdev); free_netdev(dev); /* free board information */ } @@ -955,8 +959,8 @@ static void dmfe_rx_packet(struct net_device *dev, struct dmfe_board_info *db) db->rx_avail_cnt--; db->interval_rx_cnt++; - pci_unmap_single(db->pdev, le32_to_cpu(rxptr->rdes2), - RX_ALLOC_SIZE, PCI_DMA_FROMDEVICE); + dma_unmap_single(&db->pdev->dev, le32_to_cpu(rxptr->rdes2), + RX_ALLOC_SIZE, DMA_FROM_DEVICE); if ( (rdes0 & 0x300) != 0x300) { /* A packet without First/Last flag */ @@ -1329,8 +1333,8 @@ static void dmfe_reuse_skb(struct dmfe_board_info *db, struct sk_buff * skb) if (!(rxptr->rdes0 & cpu_to_le32(0x80000000))) { rxptr->rx_skb_ptr = skb; - rxptr->rdes2 = cpu_to_le32( pci_map_single(db->pdev, - skb->data, RX_ALLOC_SIZE, PCI_DMA_FROMDEVICE) ); + rxptr->rdes2 = cpu_to_le32(dma_map_single(&db->pdev->dev, skb->data, + RX_ALLOC_SIZE, DMA_FROM_DEVICE)); wmb(); rxptr->rdes0 = cpu_to_le32(0x80000000); db->rx_avail_cnt++; @@ -1544,8 +1548,8 @@ static void allocate_rx_buffer(struct net_device *dev) if ( ( skb = netdev_alloc_skb(dev, RX_ALLOC_SIZE) ) == NULL ) break; rxptr->rx_skb_ptr = skb; /* FIXME (?) */ - rxptr->rdes2 = cpu_to_le32( pci_map_single(db->pdev, skb->data, - RX_ALLOC_SIZE, PCI_DMA_FROMDEVICE) ); + rxptr->rdes2 = cpu_to_le32(dma_map_single(&db->pdev->dev, skb->data, + RX_ALLOC_SIZE, DMA_FROM_DEVICE)); wmb(); rxptr->rdes0 = cpu_to_le32(0x80000000); rxptr = rxptr->next_rx_desc; diff --git a/drivers/net/ethernet/dec/tulip/interrupt.c b/drivers/net/ethernet/dec/tulip/interrupt.c index c1ca0765d56d..54560f9a1651 100644 --- a/drivers/net/ethernet/dec/tulip/interrupt.c +++ b/drivers/net/ethernet/dec/tulip/interrupt.c @@ -74,8 +74,8 @@ int tulip_refill_rx(struct net_device *dev) if (skb == NULL) break; - mapping = pci_map_single(tp->pdev, skb->data, PKT_BUF_SZ, - PCI_DMA_FROMDEVICE); + mapping = dma_map_single(&tp->pdev->dev, skb->data, + PKT_BUF_SZ, DMA_FROM_DEVICE); if (dma_mapping_error(&tp->pdev->dev, mapping)) { dev_kfree_skb(skb); tp->rx_buffers[entry].skb = NULL; @@ -210,9 +210,10 @@ int tulip_poll(struct napi_struct *napi, int budget) if (pkt_len < tulip_rx_copybreak && (skb = netdev_alloc_skb(dev, pkt_len + 2)) != NULL) { skb_reserve(skb, 2); /* 16 byte align the IP header */ - pci_dma_sync_single_for_cpu(tp->pdev, - tp->rx_buffers[entry].mapping, - pkt_len, PCI_DMA_FROMDEVICE); + dma_sync_single_for_cpu(&tp->pdev->dev, + tp->rx_buffers[entry].mapping, + pkt_len, + DMA_FROM_DEVICE); #if ! defined(__alpha__) skb_copy_to_linear_data(skb, tp->rx_buffers[entry].skb->data, pkt_len); @@ -222,9 +223,10 @@ int tulip_poll(struct napi_struct *napi, int budget) tp->rx_buffers[entry].skb->data, pkt_len); #endif - pci_dma_sync_single_for_device(tp->pdev, - tp->rx_buffers[entry].mapping, - pkt_len, PCI_DMA_FROMDEVICE); + dma_sync_single_for_device(&tp->pdev->dev, + tp->rx_buffers[entry].mapping, + pkt_len, + DMA_FROM_DEVICE); } else { /* Pass up the skb already on the Rx ring. */ char *temp = skb_put(skb = tp->rx_buffers[entry].skb, pkt_len); @@ -240,8 +242,10 @@ int tulip_poll(struct napi_struct *napi, int budget) } #endif - pci_unmap_single(tp->pdev, tp->rx_buffers[entry].mapping, - PKT_BUF_SZ, PCI_DMA_FROMDEVICE); + dma_unmap_single(&tp->pdev->dev, + tp->rx_buffers[entry].mapping, + PKT_BUF_SZ, + DMA_FROM_DEVICE); tp->rx_buffers[entry].skb = NULL; tp->rx_buffers[entry].mapping = 0; @@ -436,9 +440,10 @@ static int tulip_rx(struct net_device *dev) if (pkt_len < tulip_rx_copybreak && (skb = netdev_alloc_skb(dev, pkt_len + 2)) != NULL) { skb_reserve(skb, 2); /* 16 byte align the IP header */ - pci_dma_sync_single_for_cpu(tp->pdev, - tp->rx_buffers[entry].mapping, - pkt_len, PCI_DMA_FROMDEVICE); + dma_sync_single_for_cpu(&tp->pdev->dev, + tp->rx_buffers[entry].mapping, + pkt_len, + DMA_FROM_DEVICE); #if ! defined(__alpha__) skb_copy_to_linear_data(skb, tp->rx_buffers[entry].skb->data, pkt_len); @@ -448,9 +453,10 @@ static int tulip_rx(struct net_device *dev) tp->rx_buffers[entry].skb->data, pkt_len); #endif - pci_dma_sync_single_for_device(tp->pdev, - tp->rx_buffers[entry].mapping, - pkt_len, PCI_DMA_FROMDEVICE); + dma_sync_single_for_device(&tp->pdev->dev, + tp->rx_buffers[entry].mapping, + pkt_len, + DMA_FROM_DEVICE); } else { /* Pass up the skb already on the Rx ring. */ char *temp = skb_put(skb = tp->rx_buffers[entry].skb, pkt_len); @@ -466,8 +472,9 @@ static int tulip_rx(struct net_device *dev) } #endif - pci_unmap_single(tp->pdev, tp->rx_buffers[entry].mapping, - PKT_BUF_SZ, PCI_DMA_FROMDEVICE); + dma_unmap_single(&tp->pdev->dev, + tp->rx_buffers[entry].mapping, + PKT_BUF_SZ, DMA_FROM_DEVICE); tp->rx_buffers[entry].skb = NULL; tp->rx_buffers[entry].mapping = 0; @@ -597,10 +604,10 @@ irqreturn_t tulip_interrupt(int irq, void *dev_instance) if (tp->tx_buffers[entry].skb == NULL) { /* test because dummy frames not mapped */ if (tp->tx_buffers[entry].mapping) - pci_unmap_single(tp->pdev, - tp->tx_buffers[entry].mapping, - sizeof(tp->setup_frame), - PCI_DMA_TODEVICE); + dma_unmap_single(&tp->pdev->dev, + tp->tx_buffers[entry].mapping, + sizeof(tp->setup_frame), + DMA_TO_DEVICE); continue; } @@ -629,9 +636,10 @@ irqreturn_t tulip_interrupt(int irq, void *dev_instance) dev->stats.tx_packets++; } - pci_unmap_single(tp->pdev, tp->tx_buffers[entry].mapping, + dma_unmap_single(&tp->pdev->dev, + tp->tx_buffers[entry].mapping, tp->tx_buffers[entry].skb->len, - PCI_DMA_TODEVICE); + DMA_TO_DEVICE); /* Free the original skb. */ dev_kfree_skb_irq(tp->tx_buffers[entry].skb); diff --git a/drivers/net/ethernet/dec/tulip/media.c b/drivers/net/ethernet/dec/tulip/media.c index dcf21a36a9cf..011604787b8e 100644 --- a/drivers/net/ethernet/dec/tulip/media.c +++ b/drivers/net/ethernet/dec/tulip/media.c @@ -319,13 +319,8 @@ void tulip_select_media(struct net_device *dev, int startup) break; } case 5: case 6: { - u16 setup[5]; - new_csr6 = 0; /* FIXME */ - for (i = 0; i < 5; i++) - setup[i] = get_u16(&p[i*2 + 1]); - if (startup && mtable->has_reset) { struct medialeaf *rleaf = &mtable->mleaf[mtable->has_reset]; unsigned char *rst = rleaf->leafdata; diff --git a/drivers/net/ethernet/dec/tulip/tulip_core.c b/drivers/net/ethernet/dec/tulip/tulip_core.c index 3a8659c5da06..e7b0d7de40fd 100644 --- a/drivers/net/ethernet/dec/tulip/tulip_core.c +++ b/drivers/net/ethernet/dec/tulip/tulip_core.c @@ -350,9 +350,9 @@ static void tulip_up(struct net_device *dev) *setup_frm++ = eaddrs[1]; *setup_frm++ = eaddrs[1]; *setup_frm++ = eaddrs[2]; *setup_frm++ = eaddrs[2]; - mapping = pci_map_single(tp->pdev, tp->setup_frame, + mapping = dma_map_single(&tp->pdev->dev, tp->setup_frame, sizeof(tp->setup_frame), - PCI_DMA_TODEVICE); + DMA_TO_DEVICE); tp->tx_buffers[tp->cur_tx].skb = NULL; tp->tx_buffers[tp->cur_tx].mapping = mapping; @@ -630,8 +630,8 @@ static void tulip_init_ring(struct net_device *dev) tp->rx_buffers[i].skb = skb; if (skb == NULL) break; - mapping = pci_map_single(tp->pdev, skb->data, - PKT_BUF_SZ, PCI_DMA_FROMDEVICE); + mapping = dma_map_single(&tp->pdev->dev, skb->data, + PKT_BUF_SZ, DMA_FROM_DEVICE); tp->rx_buffers[i].mapping = mapping; tp->rx_ring[i].status = cpu_to_le32(DescOwned); /* Owned by Tulip chip */ tp->rx_ring[i].buffer1 = cpu_to_le32(mapping); @@ -664,8 +664,8 @@ tulip_start_xmit(struct sk_buff *skb, struct net_device *dev) entry = tp->cur_tx % TX_RING_SIZE; tp->tx_buffers[entry].skb = skb; - mapping = pci_map_single(tp->pdev, skb->data, - skb->len, PCI_DMA_TODEVICE); + mapping = dma_map_single(&tp->pdev->dev, skb->data, skb->len, + DMA_TO_DEVICE); tp->tx_buffers[entry].mapping = mapping; tp->tx_ring[entry].buffer1 = cpu_to_le32(mapping); @@ -716,16 +716,17 @@ static void tulip_clean_tx_ring(struct tulip_private *tp) if (tp->tx_buffers[entry].skb == NULL) { /* test because dummy frames not mapped */ if (tp->tx_buffers[entry].mapping) - pci_unmap_single(tp->pdev, - tp->tx_buffers[entry].mapping, - sizeof(tp->setup_frame), - PCI_DMA_TODEVICE); + dma_unmap_single(&tp->pdev->dev, + tp->tx_buffers[entry].mapping, + sizeof(tp->setup_frame), + DMA_TO_DEVICE); continue; } - pci_unmap_single(tp->pdev, tp->tx_buffers[entry].mapping, - tp->tx_buffers[entry].skb->len, - PCI_DMA_TODEVICE); + dma_unmap_single(&tp->pdev->dev, + tp->tx_buffers[entry].mapping, + tp->tx_buffers[entry].skb->len, + DMA_TO_DEVICE); /* Free the original skb. */ dev_kfree_skb_irq(tp->tx_buffers[entry].skb); @@ -795,8 +796,8 @@ static void tulip_free_ring (struct net_device *dev) /* An invalid address. */ tp->rx_ring[i].buffer1 = cpu_to_le32(0xBADF00D0); if (skb) { - pci_unmap_single(tp->pdev, mapping, PKT_BUF_SZ, - PCI_DMA_FROMDEVICE); + dma_unmap_single(&tp->pdev->dev, mapping, PKT_BUF_SZ, + DMA_FROM_DEVICE); dev_kfree_skb (skb); } } @@ -805,8 +806,9 @@ static void tulip_free_ring (struct net_device *dev) struct sk_buff *skb = tp->tx_buffers[i].skb; if (skb != NULL) { - pci_unmap_single(tp->pdev, tp->tx_buffers[i].mapping, - skb->len, PCI_DMA_TODEVICE); + dma_unmap_single(&tp->pdev->dev, + tp->tx_buffers[i].mapping, skb->len, + DMA_TO_DEVICE); dev_kfree_skb (skb); } tp->tx_buffers[i].skb = NULL; @@ -1149,9 +1151,10 @@ static void set_rx_mode(struct net_device *dev) tp->tx_buffers[entry].skb = NULL; tp->tx_buffers[entry].mapping = - pci_map_single(tp->pdev, tp->setup_frame, + dma_map_single(&tp->pdev->dev, + tp->setup_frame, sizeof(tp->setup_frame), - PCI_DMA_TODEVICE); + DMA_TO_DEVICE); /* Put the setup frame on the Tx list. */ if (entry == TX_RING_SIZE-1) tx_flags |= DESC_RING_WRAP; /* Wrap ring. */ @@ -1422,10 +1425,10 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) tp = netdev_priv(dev); tp->dev = dev; - tp->rx_ring = pci_alloc_consistent(pdev, - sizeof(struct tulip_rx_desc) * RX_RING_SIZE + - sizeof(struct tulip_tx_desc) * TX_RING_SIZE, - &tp->rx_ring_dma); + tp->rx_ring = dma_alloc_coherent(&pdev->dev, + sizeof(struct tulip_rx_desc) * RX_RING_SIZE + + sizeof(struct tulip_tx_desc) * TX_RING_SIZE, + &tp->rx_ring_dma, GFP_KERNEL); if (!tp->rx_ring) goto err_out_mtable; tp->tx_ring = (struct tulip_tx_desc *)(tp->rx_ring + RX_RING_SIZE); @@ -1757,10 +1760,10 @@ static int tulip_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) return 0; err_out_free_ring: - pci_free_consistent (pdev, - sizeof (struct tulip_rx_desc) * RX_RING_SIZE + - sizeof (struct tulip_tx_desc) * TX_RING_SIZE, - tp->rx_ring, tp->rx_ring_dma); + dma_free_coherent(&pdev->dev, + sizeof(struct tulip_rx_desc) * RX_RING_SIZE + + sizeof(struct tulip_tx_desc) * TX_RING_SIZE, + tp->rx_ring, tp->rx_ring_dma); err_out_mtable: kfree (tp->mtable); @@ -1878,10 +1881,10 @@ static void tulip_remove_one(struct pci_dev *pdev) tp = netdev_priv(dev); unregister_netdev(dev); - pci_free_consistent (pdev, - sizeof (struct tulip_rx_desc) * RX_RING_SIZE + - sizeof (struct tulip_tx_desc) * TX_RING_SIZE, - tp->rx_ring, tp->rx_ring_dma); + dma_free_coherent(&pdev->dev, + sizeof(struct tulip_rx_desc) * RX_RING_SIZE + + sizeof(struct tulip_tx_desc) * TX_RING_SIZE, + tp->rx_ring, tp->rx_ring_dma); kfree (tp->mtable); pci_iounmap(pdev, tp->base_addr); free_netdev (dev); diff --git a/drivers/net/ethernet/dec/tulip/uli526x.c b/drivers/net/ethernet/dec/tulip/uli526x.c index f942399f0f32..13e73ed15ef0 100644 --- a/drivers/net/ethernet/dec/tulip/uli526x.c +++ b/drivers/net/ethernet/dec/tulip/uli526x.c @@ -282,7 +282,7 @@ static int uli526x_init_one(struct pci_dev *pdev, return -ENOMEM; SET_NETDEV_DEV(dev, &pdev->dev); - if (pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) { + if (dma_set_mask(&pdev->dev, DMA_BIT_MASK(32))) { pr_warn("32-bit PCI DMA not available\n"); err = -ENODEV; goto err_out_free; @@ -317,11 +317,15 @@ static int uli526x_init_one(struct pci_dev *pdev, /* Allocate Tx/Rx descriptor memory */ err = -ENOMEM; - db->desc_pool_ptr = pci_alloc_consistent(pdev, sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20, &db->desc_pool_dma_ptr); + db->desc_pool_ptr = dma_alloc_coherent(&pdev->dev, + sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20, + &db->desc_pool_dma_ptr, GFP_KERNEL); if (!db->desc_pool_ptr) goto err_out_release; - db->buf_pool_ptr = pci_alloc_consistent(pdev, TX_BUF_ALLOC * TX_DESC_CNT + 4, &db->buf_pool_dma_ptr); + db->buf_pool_ptr = dma_alloc_coherent(&pdev->dev, + TX_BUF_ALLOC * TX_DESC_CNT + 4, + &db->buf_pool_dma_ptr, GFP_KERNEL); if (!db->buf_pool_ptr) goto err_out_free_tx_desc; @@ -401,11 +405,12 @@ static int uli526x_init_one(struct pci_dev *pdev, err_out_unmap: pci_iounmap(pdev, db->ioaddr); err_out_free_tx_buf: - pci_free_consistent(pdev, TX_BUF_ALLOC * TX_DESC_CNT + 4, - db->buf_pool_ptr, db->buf_pool_dma_ptr); + dma_free_coherent(&pdev->dev, TX_BUF_ALLOC * TX_DESC_CNT + 4, + db->buf_pool_ptr, db->buf_pool_dma_ptr); err_out_free_tx_desc: - pci_free_consistent(pdev, sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20, - db->desc_pool_ptr, db->desc_pool_dma_ptr); + dma_free_coherent(&pdev->dev, + sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20, + db->desc_pool_ptr, db->desc_pool_dma_ptr); err_out_release: pci_release_regions(pdev); err_out_disable: @@ -424,11 +429,11 @@ static void uli526x_remove_one(struct pci_dev *pdev) unregister_netdev(dev); pci_iounmap(pdev, db->ioaddr); - pci_free_consistent(db->pdev, sizeof(struct tx_desc) * - DESC_ALL_CNT + 0x20, db->desc_pool_ptr, - db->desc_pool_dma_ptr); - pci_free_consistent(db->pdev, TX_BUF_ALLOC * TX_DESC_CNT + 4, - db->buf_pool_ptr, db->buf_pool_dma_ptr); + dma_free_coherent(&db->pdev->dev, + sizeof(struct tx_desc) * DESC_ALL_CNT + 0x20, + db->desc_pool_ptr, db->desc_pool_dma_ptr); + dma_free_coherent(&db->pdev->dev, TX_BUF_ALLOC * TX_DESC_CNT + 4, + db->buf_pool_ptr, db->buf_pool_dma_ptr); pci_release_regions(pdev); pci_disable_device(pdev); free_netdev(dev); @@ -810,7 +815,8 @@ static void uli526x_rx_packet(struct net_device *dev, struct uli526x_board_info db->rx_avail_cnt--; db->interval_rx_cnt++; - pci_unmap_single(db->pdev, le32_to_cpu(rxptr->rdes2), RX_ALLOC_SIZE, PCI_DMA_FROMDEVICE); + dma_unmap_single(&db->pdev->dev, le32_to_cpu(rxptr->rdes2), + RX_ALLOC_SIZE, DMA_FROM_DEVICE); if ( (rdes0 & 0x300) != 0x300) { /* A packet without First/Last flag */ /* reuse this SKB */ @@ -1234,10 +1240,8 @@ static void uli526x_reuse_skb(struct uli526x_board_info *db, struct sk_buff * sk if (!(rxptr->rdes0 & cpu_to_le32(0x80000000))) { rxptr->rx_skb_ptr = skb; - rxptr->rdes2 = cpu_to_le32(pci_map_single(db->pdev, - skb_tail_pointer(skb), - RX_ALLOC_SIZE, - PCI_DMA_FROMDEVICE)); + rxptr->rdes2 = cpu_to_le32(dma_map_single(&db->pdev->dev, skb_tail_pointer(skb), + RX_ALLOC_SIZE, DMA_FROM_DEVICE)); wmb(); rxptr->rdes0 = cpu_to_le32(0x80000000); db->rx_avail_cnt++; @@ -1409,10 +1413,8 @@ static void allocate_rx_buffer(struct net_device *dev) if (skb == NULL) break; rxptr->rx_skb_ptr = skb; /* FIXME (?) */ - rxptr->rdes2 = cpu_to_le32(pci_map_single(db->pdev, - skb_tail_pointer(skb), - RX_ALLOC_SIZE, - PCI_DMA_FROMDEVICE)); + rxptr->rdes2 = cpu_to_le32(dma_map_single(&db->pdev->dev, skb_tail_pointer(skb), + RX_ALLOC_SIZE, DMA_FROM_DEVICE)); wmb(); rxptr->rdes0 = cpu_to_le32(0x80000000); rxptr = rxptr->next_rx_desc; diff --git a/drivers/net/ethernet/dec/tulip/winbond-840.c b/drivers/net/ethernet/dec/tulip/winbond-840.c index 5a43be327f58..89cbdc1f4857 100644 --- a/drivers/net/ethernet/dec/tulip/winbond-840.c +++ b/drivers/net/ethernet/dec/tulip/winbond-840.c @@ -364,7 +364,7 @@ static int w840_probe1(struct pci_dev *pdev, const struct pci_device_id *ent) irq = pdev->irq; - if (pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) { + if (dma_set_mask(&pdev->dev, DMA_BIT_MASK(32))) { pr_warn("Device %s disabled due to DMA limitations\n", pci_name(pdev)); return -EIO; @@ -630,9 +630,10 @@ static int netdev_open(struct net_device *dev) goto out_err; if (debug > 1) - netdev_dbg(dev, "w89c840_open() irq %d\n", irq); + netdev_dbg(dev, "%s() irq %d\n", __func__, irq); - if((i=alloc_ringdesc(dev))) + i = alloc_ringdesc(dev); + if (i) goto out_err; spin_lock_irq(&np->lock); @@ -642,7 +643,7 @@ static int netdev_open(struct net_device *dev) netif_start_queue(dev); if (debug > 2) - netdev_dbg(dev, "Done netdev_open()\n"); + netdev_dbg(dev, "Done %s()\n", __func__); /* Set the timer to check for link beat. */ timer_setup(&np->timer, netdev_timer, 0); @@ -802,8 +803,9 @@ static void init_rxtx_rings(struct net_device *dev) np->rx_skbuff[i] = skb; if (skb == NULL) break; - np->rx_addr[i] = pci_map_single(np->pci_dev,skb->data, - np->rx_buf_sz,PCI_DMA_FROMDEVICE); + np->rx_addr[i] = dma_map_single(&np->pci_dev->dev, skb->data, + np->rx_buf_sz, + DMA_FROM_DEVICE); np->rx_ring[i].buffer1 = np->rx_addr[i]; np->rx_ring[i].status = DescOwned; @@ -833,20 +835,17 @@ static void free_rxtx_rings(struct netdev_private* np) for (i = 0; i < RX_RING_SIZE; i++) { np->rx_ring[i].status = 0; if (np->rx_skbuff[i]) { - pci_unmap_single(np->pci_dev, - np->rx_addr[i], - np->rx_skbuff[i]->len, - PCI_DMA_FROMDEVICE); + dma_unmap_single(&np->pci_dev->dev, np->rx_addr[i], + np->rx_skbuff[i]->len, + DMA_FROM_DEVICE); dev_kfree_skb(np->rx_skbuff[i]); } np->rx_skbuff[i] = NULL; } for (i = 0; i < TX_RING_SIZE; i++) { if (np->tx_skbuff[i]) { - pci_unmap_single(np->pci_dev, - np->tx_addr[i], - np->tx_skbuff[i]->len, - PCI_DMA_TODEVICE); + dma_unmap_single(&np->pci_dev->dev, np->tx_addr[i], + np->tx_skbuff[i]->len, DMA_TO_DEVICE); dev_kfree_skb(np->tx_skbuff[i]); } np->tx_skbuff[i] = NULL; @@ -964,10 +963,10 @@ static int alloc_ringdesc(struct net_device *dev) np->rx_buf_sz = (dev->mtu <= 1500 ? PKT_BUF_SZ : dev->mtu + 32); - np->rx_ring = pci_alloc_consistent(np->pci_dev, - sizeof(struct w840_rx_desc)*RX_RING_SIZE + - sizeof(struct w840_tx_desc)*TX_RING_SIZE, - &np->ring_dma_addr); + np->rx_ring = dma_alloc_coherent(&np->pci_dev->dev, + sizeof(struct w840_rx_desc) * RX_RING_SIZE + + sizeof(struct w840_tx_desc) * TX_RING_SIZE, + &np->ring_dma_addr, GFP_KERNEL); if(!np->rx_ring) return -ENOMEM; init_rxtx_rings(dev); @@ -976,10 +975,10 @@ static int alloc_ringdesc(struct net_device *dev) static void free_ringdesc(struct netdev_private *np) { - pci_free_consistent(np->pci_dev, - sizeof(struct w840_rx_desc)*RX_RING_SIZE + - sizeof(struct w840_tx_desc)*TX_RING_SIZE, - np->rx_ring, np->ring_dma_addr); + dma_free_coherent(&np->pci_dev->dev, + sizeof(struct w840_rx_desc) * RX_RING_SIZE + + sizeof(struct w840_tx_desc) * TX_RING_SIZE, + np->rx_ring, np->ring_dma_addr); } @@ -994,8 +993,8 @@ static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev) /* Calculate the next Tx descriptor entry. */ entry = np->cur_tx % TX_RING_SIZE; - np->tx_addr[entry] = pci_map_single(np->pci_dev, - skb->data,skb->len, PCI_DMA_TODEVICE); + np->tx_addr[entry] = dma_map_single(&np->pci_dev->dev, skb->data, + skb->len, DMA_TO_DEVICE); np->tx_skbuff[entry] = skb; np->tx_ring[entry].buffer1 = np->tx_addr[entry]; @@ -1078,9 +1077,8 @@ static void netdev_tx_done(struct net_device *dev) np->stats.tx_packets++; } /* Free the original skb. */ - pci_unmap_single(np->pci_dev,np->tx_addr[entry], - np->tx_skbuff[entry]->len, - PCI_DMA_TODEVICE); + dma_unmap_single(&np->pci_dev->dev, np->tx_addr[entry], + np->tx_skbuff[entry]->len, DMA_TO_DEVICE); np->tx_q_bytes -= np->tx_skbuff[entry]->len; dev_kfree_skb_irq(np->tx_skbuff[entry]); np->tx_skbuff[entry] = NULL; @@ -1217,18 +1215,21 @@ static int netdev_rx(struct net_device *dev) if (pkt_len < rx_copybreak && (skb = netdev_alloc_skb(dev, pkt_len + 2)) != NULL) { skb_reserve(skb, 2); /* 16 byte align the IP header */ - pci_dma_sync_single_for_cpu(np->pci_dev,np->rx_addr[entry], - np->rx_skbuff[entry]->len, - PCI_DMA_FROMDEVICE); + dma_sync_single_for_cpu(&np->pci_dev->dev, + np->rx_addr[entry], + np->rx_skbuff[entry]->len, + DMA_FROM_DEVICE); skb_copy_to_linear_data(skb, np->rx_skbuff[entry]->data, pkt_len); skb_put(skb, pkt_len); - pci_dma_sync_single_for_device(np->pci_dev,np->rx_addr[entry], - np->rx_skbuff[entry]->len, - PCI_DMA_FROMDEVICE); + dma_sync_single_for_device(&np->pci_dev->dev, + np->rx_addr[entry], + np->rx_skbuff[entry]->len, + DMA_FROM_DEVICE); } else { - pci_unmap_single(np->pci_dev,np->rx_addr[entry], - np->rx_skbuff[entry]->len, - PCI_DMA_FROMDEVICE); + dma_unmap_single(&np->pci_dev->dev, + np->rx_addr[entry], + np->rx_skbuff[entry]->len, + DMA_FROM_DEVICE); skb_put(skb = np->rx_skbuff[entry], pkt_len); np->rx_skbuff[entry] = NULL; } @@ -1258,9 +1259,10 @@ static int netdev_rx(struct net_device *dev) np->rx_skbuff[entry] = skb; if (skb == NULL) break; /* Better luck next round. */ - np->rx_addr[entry] = pci_map_single(np->pci_dev, - skb->data, - np->rx_buf_sz, PCI_DMA_FROMDEVICE); + np->rx_addr[entry] = dma_map_single(&np->pci_dev->dev, + skb->data, + np->rx_buf_sz, + DMA_FROM_DEVICE); np->rx_ring[entry].buffer1 = np->rx_addr[entry]; } wmb(); diff --git a/drivers/net/ethernet/dlink/dl2k.c b/drivers/net/ethernet/dlink/dl2k.c index be6d8a9ada27..734acb834c98 100644 --- a/drivers/net/ethernet/dlink/dl2k.c +++ b/drivers/net/ethernet/dlink/dl2k.c @@ -7,7 +7,6 @@ */ -#define DRV_NAME "DL2000/TC902x-based linux driver" #include "dl2k.h" #include <linux/dma-mapping.h> @@ -223,13 +222,15 @@ rio_probe1 (struct pci_dev *pdev, const struct pci_device_id *ent) pci_set_drvdata (pdev, dev); - ring_space = pci_alloc_consistent (pdev, TX_TOTAL_SIZE, &ring_dma); + ring_space = dma_alloc_coherent(&pdev->dev, TX_TOTAL_SIZE, &ring_dma, + GFP_KERNEL); if (!ring_space) goto err_out_iounmap; np->tx_ring = ring_space; np->tx_ring_dma = ring_dma; - ring_space = pci_alloc_consistent (pdev, RX_TOTAL_SIZE, &ring_dma); + ring_space = dma_alloc_coherent(&pdev->dev, RX_TOTAL_SIZE, &ring_dma, + GFP_KERNEL); if (!ring_space) goto err_out_unmap_tx; np->rx_ring = ring_space; @@ -280,9 +281,11 @@ rio_probe1 (struct pci_dev *pdev, const struct pci_device_id *ent) return 0; err_out_unmap_rx: - pci_free_consistent (pdev, RX_TOTAL_SIZE, np->rx_ring, np->rx_ring_dma); + dma_free_coherent(&pdev->dev, RX_TOTAL_SIZE, np->rx_ring, + np->rx_ring_dma); err_out_unmap_tx: - pci_free_consistent (pdev, TX_TOTAL_SIZE, np->tx_ring, np->tx_ring_dma); + dma_free_coherent(&pdev->dev, TX_TOTAL_SIZE, np->tx_ring, + np->tx_ring_dma); err_out_iounmap: #ifdef MEM_MAPPING pci_iounmap(pdev, np->ioaddr); @@ -436,8 +439,9 @@ static void free_list(struct net_device *dev) for (i = 0; i < RX_RING_SIZE; i++) { skb = np->rx_skbuff[i]; if (skb) { - pci_unmap_single(np->pdev, desc_to_dma(&np->rx_ring[i]), - skb->len, PCI_DMA_FROMDEVICE); + dma_unmap_single(&np->pdev->dev, + desc_to_dma(&np->rx_ring[i]), + skb->len, DMA_FROM_DEVICE); dev_kfree_skb(skb); np->rx_skbuff[i] = NULL; } @@ -447,8 +451,9 @@ static void free_list(struct net_device *dev) for (i = 0; i < TX_RING_SIZE; i++) { skb = np->tx_skbuff[i]; if (skb) { - pci_unmap_single(np->pdev, desc_to_dma(&np->tx_ring[i]), - skb->len, PCI_DMA_TODEVICE); + dma_unmap_single(&np->pdev->dev, + desc_to_dma(&np->tx_ring[i]), + skb->len, DMA_TO_DEVICE); dev_kfree_skb(skb); np->tx_skbuff[i] = NULL; } @@ -505,9 +510,8 @@ static int alloc_list(struct net_device *dev) sizeof(struct netdev_desc)); /* Rubicon now supports 40 bits of addressing space. */ np->rx_ring[i].fraginfo = - cpu_to_le64(pci_map_single( - np->pdev, skb->data, np->rx_buf_sz, - PCI_DMA_FROMDEVICE)); + cpu_to_le64(dma_map_single(&np->pdev->dev, skb->data, + np->rx_buf_sz, DMA_FROM_DEVICE)); np->rx_ring[i].fraginfo |= cpu_to_le64((u64)np->rx_buf_sz << 48); } @@ -673,9 +677,8 @@ rio_timer (struct timer_list *t) } np->rx_skbuff[entry] = skb; np->rx_ring[entry].fraginfo = - cpu_to_le64 (pci_map_single - (np->pdev, skb->data, np->rx_buf_sz, - PCI_DMA_FROMDEVICE)); + cpu_to_le64 (dma_map_single(&np->pdev->dev, skb->data, + np->rx_buf_sz, DMA_FROM_DEVICE)); } np->rx_ring[entry].fraginfo |= cpu_to_le64((u64)np->rx_buf_sz << 48); @@ -729,9 +732,8 @@ start_xmit (struct sk_buff *skb, struct net_device *dev) ((u64)np->vlan << 32) | ((u64)skb->priority << 45); } - txdesc->fraginfo = cpu_to_le64 (pci_map_single (np->pdev, skb->data, - skb->len, - PCI_DMA_TODEVICE)); + txdesc->fraginfo = cpu_to_le64 (dma_map_single(&np->pdev->dev, skb->data, + skb->len, DMA_TO_DEVICE)); txdesc->fraginfo |= cpu_to_le64((u64)skb->len << 48); /* DL2K bug: DMA fails to get next descriptor ptr in 10Mbps mode @@ -828,9 +830,9 @@ rio_free_tx (struct net_device *dev, int irq) if (!(np->tx_ring[entry].status & cpu_to_le64(TFDDone))) break; skb = np->tx_skbuff[entry]; - pci_unmap_single (np->pdev, - desc_to_dma(&np->tx_ring[entry]), - skb->len, PCI_DMA_TODEVICE); + dma_unmap_single(&np->pdev->dev, + desc_to_dma(&np->tx_ring[entry]), skb->len, + DMA_TO_DEVICE); if (irq) dev_consume_skb_irq(skb); else @@ -950,25 +952,25 @@ receive_packet (struct net_device *dev) /* Small skbuffs for short packets */ if (pkt_len > copy_thresh) { - pci_unmap_single (np->pdev, - desc_to_dma(desc), - np->rx_buf_sz, - PCI_DMA_FROMDEVICE); + dma_unmap_single(&np->pdev->dev, + desc_to_dma(desc), + np->rx_buf_sz, + DMA_FROM_DEVICE); skb_put (skb = np->rx_skbuff[entry], pkt_len); np->rx_skbuff[entry] = NULL; } else if ((skb = netdev_alloc_skb_ip_align(dev, pkt_len))) { - pci_dma_sync_single_for_cpu(np->pdev, - desc_to_dma(desc), - np->rx_buf_sz, - PCI_DMA_FROMDEVICE); + dma_sync_single_for_cpu(&np->pdev->dev, + desc_to_dma(desc), + np->rx_buf_sz, + DMA_FROM_DEVICE); skb_copy_to_linear_data (skb, np->rx_skbuff[entry]->data, pkt_len); skb_put (skb, pkt_len); - pci_dma_sync_single_for_device(np->pdev, - desc_to_dma(desc), - np->rx_buf_sz, - PCI_DMA_FROMDEVICE); + dma_sync_single_for_device(&np->pdev->dev, + desc_to_dma(desc), + np->rx_buf_sz, + DMA_FROM_DEVICE); } skb->protocol = eth_type_trans (skb, dev); #if 0 @@ -1001,9 +1003,8 @@ receive_packet (struct net_device *dev) } np->rx_skbuff[entry] = skb; np->rx_ring[entry].fraginfo = - cpu_to_le64 (pci_map_single - (np->pdev, skb->data, np->rx_buf_sz, - PCI_DMA_FROMDEVICE)); + cpu_to_le64(dma_map_single(&np->pdev->dev, skb->data, + np->rx_buf_sz, DMA_FROM_DEVICE)); } np->rx_ring[entry].fraginfo |= cpu_to_le64((u64)np->rx_buf_sz << 48); @@ -1797,10 +1798,10 @@ rio_remove1 (struct pci_dev *pdev) struct netdev_private *np = netdev_priv(dev); unregister_netdev (dev); - pci_free_consistent (pdev, RX_TOTAL_SIZE, np->rx_ring, - np->rx_ring_dma); - pci_free_consistent (pdev, TX_TOTAL_SIZE, np->tx_ring, - np->tx_ring_dma); + dma_free_coherent(&pdev->dev, RX_TOTAL_SIZE, np->rx_ring, + np->rx_ring_dma); + dma_free_coherent(&pdev->dev, TX_TOTAL_SIZE, np->tx_ring, + np->tx_ring_dma); #ifdef MEM_MAPPING pci_iounmap(pdev, np->ioaddr); #endif diff --git a/drivers/net/ethernet/dlink/sundance.c b/drivers/net/ethernet/dlink/sundance.c index b3f8597e77aa..e3a8858915b3 100644 --- a/drivers/net/ethernet/dlink/sundance.c +++ b/drivers/net/ethernet/dlink/sundance.c @@ -367,6 +367,7 @@ struct netdev_private { dma_addr_t tx_ring_dma; dma_addr_t rx_ring_dma; struct timer_list timer; /* Media monitoring timer. */ + struct net_device *ndev; /* backpointer */ /* ethtool extra stats */ struct { u64 tx_multiple_collisions; @@ -429,8 +430,8 @@ static void init_ring(struct net_device *dev); static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev); static int reset_tx (struct net_device *dev); static irqreturn_t intr_handler(int irq, void *dev_instance); -static void rx_poll(unsigned long data); -static void tx_poll(unsigned long data); +static void rx_poll(struct tasklet_struct *t); +static void tx_poll(struct tasklet_struct *t); static void refill_rx (struct net_device *dev); static void netdev_error(struct net_device *dev, int intr_status); static void netdev_error(struct net_device *dev, int intr_status); @@ -531,14 +532,15 @@ static int sundance_probe1(struct pci_dev *pdev, cpu_to_le16(eeprom_read(ioaddr, i + EEPROM_SA_OFFSET)); np = netdev_priv(dev); + np->ndev = dev; np->base = ioaddr; np->pci_dev = pdev; np->chip_id = chip_idx; np->msg_enable = (1 << debug) - 1; spin_lock_init(&np->lock); spin_lock_init(&np->statlock); - tasklet_init(&np->rx_tasklet, rx_poll, (unsigned long)dev); - tasklet_init(&np->tx_tasklet, tx_poll, (unsigned long)dev); + tasklet_setup(&np->rx_tasklet, rx_poll); + tasklet_setup(&np->tx_tasklet, tx_poll); ring_space = dma_alloc_coherent(&pdev->dev, TX_TOTAL_SIZE, &ring_dma, GFP_KERNEL); @@ -1054,10 +1056,9 @@ static void init_ring(struct net_device *dev) } } -static void tx_poll (unsigned long data) +static void tx_poll(struct tasklet_struct *t) { - struct net_device *dev = (struct net_device *)data; - struct netdev_private *np = netdev_priv(dev); + struct netdev_private *np = from_tasklet(np, t, tx_tasklet); unsigned head = np->cur_task % TX_RING_SIZE; struct netdev_desc *txdesc = &np->tx_ring[(np->cur_tx - 1) % TX_RING_SIZE]; @@ -1312,10 +1313,10 @@ static irqreturn_t intr_handler(int irq, void *dev_instance) return IRQ_RETVAL(handled); } -static void rx_poll(unsigned long data) +static void rx_poll(struct tasklet_struct *t) { - struct net_device *dev = (struct net_device *)data; - struct netdev_private *np = netdev_priv(dev); + struct netdev_private *np = from_tasklet(np, t, rx_tasklet); + struct net_device *dev = np->ndev; int entry = np->cur_rx % RX_RING_SIZE; int boguscnt = np->budget; void __iomem *ioaddr = np->base; diff --git a/drivers/net/ethernet/dnet.c b/drivers/net/ethernet/dnet.c index db98274501a0..48c6eb142dcc 100644 --- a/drivers/net/ethernet/dnet.c +++ b/drivers/net/ethernet/dnet.c @@ -507,23 +507,20 @@ static netdev_tx_t dnet_start_xmit(struct sk_buff *skb, struct net_device *dev) { struct dnet *bp = netdev_priv(dev); - u32 tx_status, irq_enable; - unsigned int len, i, tx_cmd, wrsz; + unsigned int i, tx_cmd, wrsz; unsigned long flags; unsigned int *bufp; + u32 irq_enable; - tx_status = dnet_readl(bp, TX_STATUS); + dnet_readl(bp, TX_STATUS); pr_debug("start_xmit: len %u head %p data %p\n", skb->len, skb->head, skb->data); dnet_print_skb(skb); - /* frame size (words) */ - len = (skb->len + 3) >> 2; - spin_lock_irqsave(&bp->lock, flags); - tx_status = dnet_readl(bp, TX_STATUS); + dnet_readl(bp, TX_STATUS); bufp = (unsigned int *)(((unsigned long) skb->data) & ~0x3UL); wrsz = (u32) skb->len + 3; @@ -545,7 +542,7 @@ static netdev_tx_t dnet_start_xmit(struct sk_buff *skb, struct net_device *dev) if (dnet_readl(bp, TX_FIFO_WCNT) > DNET_FIFO_TX_DATA_AF_TH) { netif_stop_queue(dev); - tx_status = dnet_readl(bp, INTR_SRC); + dnet_readl(bp, INTR_SRC); irq_enable = dnet_readl(bp, INTR_ENB); irq_enable |= DNET_INTR_ENB_TX_FIFOAE; dnet_writel(bp, irq_enable, INTR_ENB); diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c index a817ca661c1f..0981fe9652e5 100644 --- a/drivers/net/ethernet/ethoc.c +++ b/drivers/net/ethernet/ethoc.c @@ -177,6 +177,7 @@ MODULE_PARM_DESC(buffer_size, "DMA buffer allocation size"); * struct ethoc - driver-private device structure * @iobase: pointer to I/O memory region * @membase: pointer to buffer memory region + * @big_endian: just big or little (endian) * @num_bd: number of buffer descriptors * @num_tx: number of send buffers * @cur_tx: last send buffer written @@ -189,7 +190,10 @@ MODULE_PARM_DESC(buffer_size, "DMA buffer allocation size"); * @msg_enable: device state flags * @lock: device lock * @mdio: MDIO bus for PHY access + * @clk: clock * @phy_id: address of attached PHY + * @old_link: previous link info + * @old_duplex: previous duplex info */ struct ethoc { void __iomem *iobase; @@ -1015,7 +1019,7 @@ static const struct net_device_ops ethoc_netdev_ops = { /** * ethoc_probe - initialize OpenCores ethernet MAC - * pdev: platform device + * @pdev: platform device */ static int ethoc_probe(struct platform_device *pdev) { diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c index fdff3b4723ba..06cc863f4dd6 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c @@ -87,7 +87,7 @@ MODULE_PARM_DESC(tx_timeout, "The Tx timeout in ms"); #define DPAA_MSG_DEFAULT (NETIF_MSG_DRV | NETIF_MSG_PROBE | \ NETIF_MSG_LINK | NETIF_MSG_IFUP | \ - NETIF_MSG_IFDOWN) + NETIF_MSG_IFDOWN | NETIF_MSG_HW) #define DPAA_INGRESS_CS_THRESHOLD 0x10000000 /* Ingress congestion threshold on FMan ports diff --git a/drivers/net/ethernet/freescale/dpaa2/Kconfig b/drivers/net/ethernet/freescale/dpaa2/Kconfig index feea797cde02..cfd369cf4c8c 100644 --- a/drivers/net/ethernet/freescale/dpaa2/Kconfig +++ b/drivers/net/ethernet/freescale/dpaa2/Kconfig @@ -3,6 +3,7 @@ config FSL_DPAA2_ETH tristate "Freescale DPAA2 Ethernet" depends on FSL_MC_BUS && FSL_MC_DPIO select PHYLINK + select PCS_LYNX help This is the DPAA2 Ethernet driver supporting Freescale SoCs with DPAA2 (DataPath Acceleration Architecture v2). diff --git a/drivers/net/ethernet/freescale/dpaa2/Makefile b/drivers/net/ethernet/freescale/dpaa2/Makefile index 6e7f33c956bf..146cb3540e61 100644 --- a/drivers/net/ethernet/freescale/dpaa2/Makefile +++ b/drivers/net/ethernet/freescale/dpaa2/Makefile @@ -6,7 +6,7 @@ obj-$(CONFIG_FSL_DPAA2_ETH) += fsl-dpaa2-eth.o obj-$(CONFIG_FSL_DPAA2_PTP_CLOCK) += fsl-dpaa2-ptp.o -fsl-dpaa2-eth-objs := dpaa2-eth.o dpaa2-ethtool.o dpni.o dpaa2-mac.o dpmac.o +fsl-dpaa2-eth-objs := dpaa2-eth.o dpaa2-ethtool.o dpni.o dpaa2-mac.o dpmac.o dpaa2-eth-devlink.o fsl-dpaa2-eth-${CONFIG_FSL_DPAA2_ETH_DCB} += dpaa2-eth-dcb.o fsl-dpaa2-eth-${CONFIG_DEBUG_FS} += dpaa2-eth-debugfs.o fsl-dpaa2-ptp-objs := dpaa2-ptp.o dprtc.o diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-dcb.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-dcb.c index 83dee575c2fa..84de0644168d 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-dcb.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-dcb.c @@ -17,12 +17,12 @@ static int dpaa2_eth_dcbnl_ieee_getpfc(struct net_device *net_dev, return 0; } -static inline bool is_prio_enabled(u8 pfc_en, u8 tc) +static inline bool dpaa2_eth_is_prio_enabled(u8 pfc_en, u8 tc) { return !!(pfc_en & (1 << tc)); } -static int set_pfc_cn(struct dpaa2_eth_priv *priv, u8 pfc_en) +static int dpaa2_eth_set_pfc_cn(struct dpaa2_eth_priv *priv, u8 pfc_en) { struct dpni_congestion_notification_cfg cfg = {0}; int i, err; @@ -33,7 +33,7 @@ static int set_pfc_cn(struct dpaa2_eth_priv *priv, u8 pfc_en) cfg.message_ctx = 0ULL; for (i = 0; i < dpaa2_eth_tc_count(priv); i++) { - if (is_prio_enabled(pfc_en, i)) { + if (dpaa2_eth_is_prio_enabled(pfc_en, i)) { cfg.threshold_entry = DPAA2_ETH_CN_THRESH_ENTRY(priv); cfg.threshold_exit = DPAA2_ETH_CN_THRESH_EXIT(priv); } else { @@ -93,7 +93,7 @@ static int dpaa2_eth_dcbnl_ieee_setpfc(struct net_device *net_dev, } /* Configure congestion notifications for the enabled priorities */ - err = set_pfc_cn(priv, pfc->pfc_en); + err = dpaa2_eth_set_pfc_cn(priv, pfc->pfc_en); if (err) return err; diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-debugfs.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-debugfs.c index 56d9927fbfda..b87db0846e10 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-debugfs.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-debugfs.c @@ -42,24 +42,7 @@ static int dpaa2_dbg_cpu_show(struct seq_file *file, void *offset) return 0; } -static int dpaa2_dbg_cpu_open(struct inode *inode, struct file *file) -{ - int err; - struct dpaa2_eth_priv *priv = (struct dpaa2_eth_priv *)inode->i_private; - - err = single_open(file, dpaa2_dbg_cpu_show, priv); - if (err < 0) - netdev_err(priv->net_dev, "single_open() failed\n"); - - return err; -} - -static const struct file_operations dpaa2_dbg_cpu_ops = { - .open = dpaa2_dbg_cpu_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; +DEFINE_SHOW_ATTRIBUTE(dpaa2_dbg_cpu); static char *fq_type_to_str(struct dpaa2_eth_fq *fq) { @@ -106,24 +89,7 @@ static int dpaa2_dbg_fqs_show(struct seq_file *file, void *offset) return 0; } -static int dpaa2_dbg_fqs_open(struct inode *inode, struct file *file) -{ - int err; - struct dpaa2_eth_priv *priv = (struct dpaa2_eth_priv *)inode->i_private; - - err = single_open(file, dpaa2_dbg_fqs_show, priv); - if (err < 0) - netdev_err(priv->net_dev, "single_open() failed\n"); - - return err; -} - -static const struct file_operations dpaa2_dbg_fq_ops = { - .open = dpaa2_dbg_fqs_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; +DEFINE_SHOW_ATTRIBUTE(dpaa2_dbg_fqs); static int dpaa2_dbg_ch_show(struct seq_file *file, void *offset) { @@ -151,24 +117,7 @@ static int dpaa2_dbg_ch_show(struct seq_file *file, void *offset) return 0; } -static int dpaa2_dbg_ch_open(struct inode *inode, struct file *file) -{ - int err; - struct dpaa2_eth_priv *priv = (struct dpaa2_eth_priv *)inode->i_private; - - err = single_open(file, dpaa2_dbg_ch_show, priv); - if (err < 0) - netdev_err(priv->net_dev, "single_open() failed\n"); - - return err; -} - -static const struct file_operations dpaa2_dbg_ch_ops = { - .open = dpaa2_dbg_ch_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; +DEFINE_SHOW_ATTRIBUTE(dpaa2_dbg_ch); void dpaa2_dbg_add(struct dpaa2_eth_priv *priv) { @@ -179,13 +128,13 @@ void dpaa2_dbg_add(struct dpaa2_eth_priv *priv) priv->dbg.dir = dir; /* per-cpu stats file */ - debugfs_create_file("cpu_stats", 0444, dir, priv, &dpaa2_dbg_cpu_ops); + debugfs_create_file("cpu_stats", 0444, dir, priv, &dpaa2_dbg_cpu_fops); /* per-fq stats file */ - debugfs_create_file("fq_stats", 0444, dir, priv, &dpaa2_dbg_fq_ops); + debugfs_create_file("fq_stats", 0444, dir, priv, &dpaa2_dbg_fqs_fops); /* per-fq stats file */ - debugfs_create_file("ch_stats", 0444, dir, priv, &dpaa2_dbg_ch_ops); + debugfs_create_file("ch_stats", 0444, dir, priv, &dpaa2_dbg_ch_fops); } void dpaa2_dbg_remove(struct dpaa2_eth_priv *priv) diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-devlink.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-devlink.c new file mode 100644 index 000000000000..833696245565 --- /dev/null +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-devlink.c @@ -0,0 +1,309 @@ +// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) +#include "dpaa2-eth.h" +/* Copyright 2020 NXP + */ + +#define DPAA2_ETH_TRAP_DROP(_id, _group_id) \ + DEVLINK_TRAP_GENERIC(DROP, DROP, _id, \ + DEVLINK_TRAP_GROUP_GENERIC_ID_##_group_id, 0) + +static const struct devlink_trap_group dpaa2_eth_trap_groups_arr[] = { + DEVLINK_TRAP_GROUP_GENERIC(PARSER_ERROR_DROPS, 0), +}; + +static const struct devlink_trap dpaa2_eth_traps_arr[] = { + DPAA2_ETH_TRAP_DROP(VXLAN_PARSING, PARSER_ERROR_DROPS), + DPAA2_ETH_TRAP_DROP(LLC_SNAP_PARSING, PARSER_ERROR_DROPS), + DPAA2_ETH_TRAP_DROP(VLAN_PARSING, PARSER_ERROR_DROPS), + DPAA2_ETH_TRAP_DROP(PPPOE_PPP_PARSING, PARSER_ERROR_DROPS), + DPAA2_ETH_TRAP_DROP(MPLS_PARSING, PARSER_ERROR_DROPS), + DPAA2_ETH_TRAP_DROP(ARP_PARSING, PARSER_ERROR_DROPS), + DPAA2_ETH_TRAP_DROP(IP_1_PARSING, PARSER_ERROR_DROPS), + DPAA2_ETH_TRAP_DROP(IP_N_PARSING, PARSER_ERROR_DROPS), + DPAA2_ETH_TRAP_DROP(GRE_PARSING, PARSER_ERROR_DROPS), + DPAA2_ETH_TRAP_DROP(UDP_PARSING, PARSER_ERROR_DROPS), + DPAA2_ETH_TRAP_DROP(TCP_PARSING, PARSER_ERROR_DROPS), + DPAA2_ETH_TRAP_DROP(IPSEC_PARSING, PARSER_ERROR_DROPS), + DPAA2_ETH_TRAP_DROP(SCTP_PARSING, PARSER_ERROR_DROPS), + DPAA2_ETH_TRAP_DROP(DCCP_PARSING, PARSER_ERROR_DROPS), + DPAA2_ETH_TRAP_DROP(GTP_PARSING, PARSER_ERROR_DROPS), + DPAA2_ETH_TRAP_DROP(ESP_PARSING, PARSER_ERROR_DROPS), +}; + +static int dpaa2_eth_dl_info_get(struct devlink *devlink, + struct devlink_info_req *req, + struct netlink_ext_ack *extack) +{ + struct dpaa2_eth_devlink_priv *dl_priv = devlink_priv(devlink); + struct dpaa2_eth_priv *priv = dl_priv->dpaa2_priv; + char buf[10]; + int err; + + err = devlink_info_driver_name_put(req, KBUILD_MODNAME); + if (err) + return err; + + scnprintf(buf, 10, "%d.%d", priv->dpni_ver_major, priv->dpni_ver_minor); + err = devlink_info_version_running_put(req, "dpni", buf); + if (err) + return err; + + return 0; +} + +static struct dpaa2_eth_trap_item * +dpaa2_eth_dl_trap_item_lookup(struct dpaa2_eth_priv *priv, u16 trap_id) +{ + struct dpaa2_eth_trap_data *dpaa2_eth_trap_data = priv->trap_data; + int i; + + for (i = 0; i < ARRAY_SIZE(dpaa2_eth_traps_arr); i++) { + if (dpaa2_eth_traps_arr[i].id == trap_id) + return &dpaa2_eth_trap_data->trap_items_arr[i]; + } + + return NULL; +} + +struct dpaa2_eth_trap_item *dpaa2_eth_dl_get_trap(struct dpaa2_eth_priv *priv, + struct dpaa2_fapr *fapr) +{ + struct dpaa2_faf_error_bit { + int position; + enum devlink_trap_generic_id trap_id; + } faf_bits[] = { + { .position = 5, .trap_id = DEVLINK_TRAP_GENERIC_ID_VXLAN_PARSING }, + { .position = 20, .trap_id = DEVLINK_TRAP_GENERIC_ID_LLC_SNAP_PARSING }, + { .position = 24, .trap_id = DEVLINK_TRAP_GENERIC_ID_VLAN_PARSING }, + { .position = 26, .trap_id = DEVLINK_TRAP_GENERIC_ID_PPPOE_PPP_PARSING }, + { .position = 29, .trap_id = DEVLINK_TRAP_GENERIC_ID_MPLS_PARSING }, + { .position = 31, .trap_id = DEVLINK_TRAP_GENERIC_ID_ARP_PARSING }, + { .position = 52, .trap_id = DEVLINK_TRAP_GENERIC_ID_IP_1_PARSING }, + { .position = 61, .trap_id = DEVLINK_TRAP_GENERIC_ID_IP_N_PARSING }, + { .position = 67, .trap_id = DEVLINK_TRAP_GENERIC_ID_GRE_PARSING }, + { .position = 71, .trap_id = DEVLINK_TRAP_GENERIC_ID_UDP_PARSING }, + { .position = 76, .trap_id = DEVLINK_TRAP_GENERIC_ID_TCP_PARSING }, + { .position = 80, .trap_id = DEVLINK_TRAP_GENERIC_ID_IPSEC_PARSING }, + { .position = 82, .trap_id = DEVLINK_TRAP_GENERIC_ID_SCTP_PARSING }, + { .position = 84, .trap_id = DEVLINK_TRAP_GENERIC_ID_DCCP_PARSING }, + { .position = 88, .trap_id = DEVLINK_TRAP_GENERIC_ID_GTP_PARSING }, + { .position = 90, .trap_id = DEVLINK_TRAP_GENERIC_ID_ESP_PARSING }, + }; + u64 faf_word; + u64 mask; + int i; + + for (i = 0; i < ARRAY_SIZE(faf_bits); i++) { + if (faf_bits[i].position < 32) { + /* Low part of FAF. + * position ranges from 31 to 0, mask from 0 to 31. + */ + mask = 1ull << (31 - faf_bits[i].position); + faf_word = __le32_to_cpu(fapr->faf_lo); + } else { + /* High part of FAF. + * position ranges from 95 to 32, mask from 0 to 63. + */ + mask = 1ull << (63 - (faf_bits[i].position - 32)); + faf_word = __le64_to_cpu(fapr->faf_hi); + } + if (faf_word & mask) + return dpaa2_eth_dl_trap_item_lookup(priv, faf_bits[i].trap_id); + } + return NULL; +} + +static int dpaa2_eth_dl_trap_init(struct devlink *devlink, + const struct devlink_trap *trap, + void *trap_ctx) +{ + struct dpaa2_eth_devlink_priv *dl_priv = devlink_priv(devlink); + struct dpaa2_eth_priv *priv = dl_priv->dpaa2_priv; + struct dpaa2_eth_trap_item *dpaa2_eth_trap_item; + + dpaa2_eth_trap_item = dpaa2_eth_dl_trap_item_lookup(priv, trap->id); + if (WARN_ON(!dpaa2_eth_trap_item)) + return -ENOENT; + + dpaa2_eth_trap_item->trap_ctx = trap_ctx; + + return 0; +} + +static int dpaa2_eth_dl_trap_action_set(struct devlink *devlink, + const struct devlink_trap *trap, + enum devlink_trap_action action, + struct netlink_ext_ack *extack) +{ + /* No support for changing the action of an independent packet trap, + * only per trap group - parser error drops + */ + NL_SET_ERR_MSG_MOD(extack, + "Cannot change trap action independently of group"); + return -EOPNOTSUPP; +} + +static int dpaa2_eth_dl_trap_group_action_set(struct devlink *devlink, + const struct devlink_trap_group *group, + enum devlink_trap_action action, + struct netlink_ext_ack *extack) +{ + struct dpaa2_eth_devlink_priv *dl_priv = devlink_priv(devlink); + struct dpaa2_eth_priv *priv = dl_priv->dpaa2_priv; + struct net_device *net_dev = priv->net_dev; + struct device *dev = net_dev->dev.parent; + struct dpni_error_cfg err_cfg = {0}; + int err; + + if (group->id != DEVLINK_TRAP_GROUP_GENERIC_ID_PARSER_ERROR_DROPS) + return -EOPNOTSUPP; + + /* Configure handling of frames marked as errors from the parser */ + err_cfg.errors = DPAA2_FAS_RX_ERR_MASK; + err_cfg.set_frame_annotation = 1; + + switch (action) { + case DEVLINK_TRAP_ACTION_DROP: + err_cfg.error_action = DPNI_ERROR_ACTION_DISCARD; + break; + case DEVLINK_TRAP_ACTION_TRAP: + err_cfg.error_action = DPNI_ERROR_ACTION_SEND_TO_ERROR_QUEUE; + break; + default: + return -EOPNOTSUPP; + } + + err = dpni_set_errors_behavior(priv->mc_io, 0, priv->mc_token, &err_cfg); + if (err) { + dev_err(dev, "dpni_set_errors_behavior failed\n"); + return err; + } + + return 0; +} + +static const struct devlink_ops dpaa2_eth_devlink_ops = { + .info_get = dpaa2_eth_dl_info_get, + .trap_init = dpaa2_eth_dl_trap_init, + .trap_action_set = dpaa2_eth_dl_trap_action_set, + .trap_group_action_set = dpaa2_eth_dl_trap_group_action_set, +}; + +int dpaa2_eth_dl_register(struct dpaa2_eth_priv *priv) +{ + struct net_device *net_dev = priv->net_dev; + struct device *dev = net_dev->dev.parent; + struct dpaa2_eth_devlink_priv *dl_priv; + int err; + + priv->devlink = devlink_alloc(&dpaa2_eth_devlink_ops, sizeof(*dl_priv)); + if (!priv->devlink) { + dev_err(dev, "devlink_alloc failed\n"); + return -ENOMEM; + } + dl_priv = devlink_priv(priv->devlink); + dl_priv->dpaa2_priv = priv; + + err = devlink_register(priv->devlink, dev); + if (err) { + dev_err(dev, "devlink_register() = %d\n", err); + goto devlink_free; + } + + return 0; + +devlink_free: + devlink_free(priv->devlink); + + return err; +} + +void dpaa2_eth_dl_unregister(struct dpaa2_eth_priv *priv) +{ + devlink_unregister(priv->devlink); + devlink_free(priv->devlink); +} + +int dpaa2_eth_dl_port_add(struct dpaa2_eth_priv *priv) +{ + struct devlink_port *devlink_port = &priv->devlink_port; + struct devlink_port_attrs attrs = {}; + int err; + + attrs.flavour = DEVLINK_PORT_FLAVOUR_PHYSICAL; + devlink_port_attrs_set(devlink_port, &attrs); + + err = devlink_port_register(priv->devlink, devlink_port, 0); + if (err) + return err; + + devlink_port_type_eth_set(devlink_port, priv->net_dev); + + return 0; +} + +void dpaa2_eth_dl_port_del(struct dpaa2_eth_priv *priv) +{ + struct devlink_port *devlink_port = &priv->devlink_port; + + devlink_port_type_clear(devlink_port); + devlink_port_unregister(devlink_port); +} + +int dpaa2_eth_dl_traps_register(struct dpaa2_eth_priv *priv) +{ + struct dpaa2_eth_trap_data *dpaa2_eth_trap_data; + struct net_device *net_dev = priv->net_dev; + struct device *dev = net_dev->dev.parent; + int err; + + dpaa2_eth_trap_data = kzalloc(sizeof(*dpaa2_eth_trap_data), GFP_KERNEL); + if (!dpaa2_eth_trap_data) + return -ENOMEM; + priv->trap_data = dpaa2_eth_trap_data; + + dpaa2_eth_trap_data->trap_items_arr = kcalloc(ARRAY_SIZE(dpaa2_eth_traps_arr), + sizeof(struct dpaa2_eth_trap_item), + GFP_KERNEL); + if (!dpaa2_eth_trap_data->trap_items_arr) { + err = -ENOMEM; + goto trap_data_free; + } + + err = devlink_trap_groups_register(priv->devlink, dpaa2_eth_trap_groups_arr, + ARRAY_SIZE(dpaa2_eth_trap_groups_arr)); + if (err) { + dev_err(dev, "devlink_trap_groups_register() = %d\n", err); + goto trap_items_arr_free; + } + + err = devlink_traps_register(priv->devlink, dpaa2_eth_traps_arr, + ARRAY_SIZE(dpaa2_eth_traps_arr), priv); + if (err) { + dev_err(dev, "devlink_traps_register() = %d\n", err); + goto trap_groups_unregiser; + } + + return 0; + +trap_groups_unregiser: + devlink_trap_groups_unregister(priv->devlink, dpaa2_eth_trap_groups_arr, + ARRAY_SIZE(dpaa2_eth_trap_groups_arr)); +trap_items_arr_free: + kfree(dpaa2_eth_trap_data->trap_items_arr); +trap_data_free: + kfree(dpaa2_eth_trap_data); + priv->trap_data = NULL; + + return err; +} + +void dpaa2_eth_dl_traps_unregister(struct dpaa2_eth_priv *priv) +{ + devlink_traps_unregister(priv->devlink, dpaa2_eth_traps_arr, + ARRAY_SIZE(dpaa2_eth_traps_arr)); + devlink_trap_groups_unregister(priv->devlink, dpaa2_eth_trap_groups_arr, + ARRAY_SIZE(dpaa2_eth_trap_groups_arr)); + kfree(priv->trap_data->trap_items_arr); + kfree(priv->trap_data); +} diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c index cf5383bb8331..cf9400a9886d 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c @@ -11,10 +11,11 @@ #include <linux/msi.h> #include <linux/kthread.h> #include <linux/iommu.h> -#include <linux/net_tstamp.h> #include <linux/fsl/mc.h> #include <linux/bpf.h> #include <linux/bpf_trace.h> +#include <linux/fsl/ptp_qoriq.h> +#include <linux/ptp_classify.h> #include <net/pkt_cls.h> #include <net/sock.h> @@ -30,6 +31,9 @@ MODULE_LICENSE("Dual BSD/GPL"); MODULE_AUTHOR("Freescale Semiconductor, Inc"); MODULE_DESCRIPTION("Freescale DPAA2 Ethernet Driver"); +struct ptp_qoriq *dpaa2_ptp; +EXPORT_SYMBOL(dpaa2_ptp); + static void *dpaa2_iova_to_virt(struct iommu_domain *domain, dma_addr_t iova_addr) { @@ -40,9 +44,9 @@ static void *dpaa2_iova_to_virt(struct iommu_domain *domain, return phys_to_virt(phys_addr); } -static void validate_rx_csum(struct dpaa2_eth_priv *priv, - u32 fd_status, - struct sk_buff *skb) +static void dpaa2_eth_validate_rx_csum(struct dpaa2_eth_priv *priv, + u32 fd_status, + struct sk_buff *skb) { skb_checksum_none_assert(skb); @@ -62,9 +66,9 @@ static void validate_rx_csum(struct dpaa2_eth_priv *priv, /* Free a received FD. * Not to be used for Tx conf FDs or on any other paths. */ -static void free_rx_fd(struct dpaa2_eth_priv *priv, - const struct dpaa2_fd *fd, - void *vaddr) +static void dpaa2_eth_free_rx_fd(struct dpaa2_eth_priv *priv, + const struct dpaa2_fd *fd, + void *vaddr) { struct device *dev = priv->net_dev->dev.parent; dma_addr_t addr = dpaa2_fd_get_addr(fd); @@ -100,9 +104,9 @@ free_buf: } /* Build a linear skb based on a single-buffer frame descriptor */ -static struct sk_buff *build_linear_skb(struct dpaa2_eth_channel *ch, - const struct dpaa2_fd *fd, - void *fd_vaddr) +static struct sk_buff *dpaa2_eth_build_linear_skb(struct dpaa2_eth_channel *ch, + const struct dpaa2_fd *fd, + void *fd_vaddr) { struct sk_buff *skb = NULL; u16 fd_offset = dpaa2_fd_get_offset(fd); @@ -121,9 +125,9 @@ static struct sk_buff *build_linear_skb(struct dpaa2_eth_channel *ch, } /* Build a non linear (fragmented) skb based on a S/G table */ -static struct sk_buff *build_frag_skb(struct dpaa2_eth_priv *priv, - struct dpaa2_eth_channel *ch, - struct dpaa2_sg_entry *sgt) +static struct sk_buff *dpaa2_eth_build_frag_skb(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_channel *ch, + struct dpaa2_sg_entry *sgt) { struct sk_buff *skb = NULL; struct device *dev = priv->net_dev->dev.parent; @@ -204,7 +208,8 @@ static struct sk_buff *build_frag_skb(struct dpaa2_eth_priv *priv, /* Free buffers acquired from the buffer pool or which were meant to * be released in the pool */ -static void free_bufs(struct dpaa2_eth_priv *priv, u64 *buf_array, int count) +static void dpaa2_eth_free_bufs(struct dpaa2_eth_priv *priv, u64 *buf_array, + int count) { struct device *dev = priv->net_dev->dev.parent; void *vaddr; @@ -218,9 +223,9 @@ static void free_bufs(struct dpaa2_eth_priv *priv, u64 *buf_array, int count) } } -static void xdp_release_buf(struct dpaa2_eth_priv *priv, - struct dpaa2_eth_channel *ch, - dma_addr_t addr) +static void dpaa2_eth_xdp_release_buf(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_channel *ch, + dma_addr_t addr) { int retries = 0; int err; @@ -238,7 +243,7 @@ static void xdp_release_buf(struct dpaa2_eth_priv *priv, } if (err) { - free_bufs(priv, ch->xdp.drop_bufs, ch->xdp.drop_cnt); + dpaa2_eth_free_bufs(priv, ch->xdp.drop_bufs, ch->xdp.drop_cnt); ch->buf_count -= ch->xdp.drop_cnt; } @@ -274,9 +279,9 @@ static int dpaa2_eth_xdp_flush(struct dpaa2_eth_priv *priv, return total_enqueued; } -static void xdp_tx_flush(struct dpaa2_eth_priv *priv, - struct dpaa2_eth_channel *ch, - struct dpaa2_eth_fq *fq) +static void dpaa2_eth_xdp_tx_flush(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_channel *ch, + struct dpaa2_eth_fq *fq) { struct rtnl_link_stats64 *percpu_stats; struct dpaa2_fd *fds; @@ -295,17 +300,17 @@ static void xdp_tx_flush(struct dpaa2_eth_priv *priv, ch->stats.xdp_tx++; } for (i = enqueued; i < fq->xdp_tx_fds.num; i++) { - xdp_release_buf(priv, ch, dpaa2_fd_get_addr(&fds[i])); + dpaa2_eth_xdp_release_buf(priv, ch, dpaa2_fd_get_addr(&fds[i])); percpu_stats->tx_errors++; ch->stats.xdp_tx_err++; } fq->xdp_tx_fds.num = 0; } -static void xdp_enqueue(struct dpaa2_eth_priv *priv, - struct dpaa2_eth_channel *ch, - struct dpaa2_fd *fd, - void *buf_start, u16 queue_id) +static void dpaa2_eth_xdp_enqueue(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_channel *ch, + struct dpaa2_fd *fd, + void *buf_start, u16 queue_id) { struct dpaa2_faead *faead; struct dpaa2_fd *dest_fd; @@ -333,13 +338,13 @@ static void xdp_enqueue(struct dpaa2_eth_priv *priv, if (fq->xdp_tx_fds.num < DEV_MAP_BULK_SIZE) return; - xdp_tx_flush(priv, ch, fq); + dpaa2_eth_xdp_tx_flush(priv, ch, fq); } -static u32 run_xdp(struct dpaa2_eth_priv *priv, - struct dpaa2_eth_channel *ch, - struct dpaa2_eth_fq *rx_fq, - struct dpaa2_fd *fd, void *vaddr) +static u32 dpaa2_eth_run_xdp(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_channel *ch, + struct dpaa2_eth_fq *rx_fq, + struct dpaa2_fd *fd, void *vaddr) { dma_addr_t addr = dpaa2_fd_get_addr(fd); struct bpf_prog *xdp_prog; @@ -372,7 +377,7 @@ static u32 run_xdp(struct dpaa2_eth_priv *priv, case XDP_PASS: break; case XDP_TX: - xdp_enqueue(priv, ch, fd, vaddr, rx_fq->flowid); + dpaa2_eth_xdp_enqueue(priv, ch, fd, vaddr, rx_fq->flowid); break; default: bpf_warn_invalid_xdp_action(xdp_act); @@ -381,7 +386,7 @@ static u32 run_xdp(struct dpaa2_eth_priv *priv, trace_xdp_exception(priv->net_dev, xdp_prog, xdp_act); fallthrough; case XDP_DROP: - xdp_release_buf(priv, ch, addr); + dpaa2_eth_xdp_release_buf(priv, ch, addr); ch->stats.xdp_drop++; break; case XDP_REDIRECT: @@ -441,7 +446,7 @@ static void dpaa2_eth_rx(struct dpaa2_eth_priv *priv, percpu_extras = this_cpu_ptr(priv->percpu_extras); if (fd_format == dpaa2_fd_single) { - xdp_act = run_xdp(priv, ch, fq, (struct dpaa2_fd *)fd, vaddr); + xdp_act = dpaa2_eth_run_xdp(priv, ch, fq, (struct dpaa2_fd *)fd, vaddr); if (xdp_act != XDP_PASS) { percpu_stats->rx_packets++; percpu_stats->rx_bytes += dpaa2_fd_get_len(fd); @@ -450,13 +455,13 @@ static void dpaa2_eth_rx(struct dpaa2_eth_priv *priv, dma_unmap_page(dev, addr, priv->rx_buf_size, DMA_BIDIRECTIONAL); - skb = build_linear_skb(ch, fd, vaddr); + skb = dpaa2_eth_build_linear_skb(ch, fd, vaddr); } else if (fd_format == dpaa2_fd_sg) { WARN_ON(priv->xdp_prog); dma_unmap_page(dev, addr, priv->rx_buf_size, DMA_BIDIRECTIONAL); - skb = build_frag_skb(priv, ch, buf_data); + skb = dpaa2_eth_build_frag_skb(priv, ch, buf_data); free_pages((unsigned long)vaddr, 0); percpu_extras->rx_sg_frames++; percpu_extras->rx_sg_bytes += dpaa2_fd_get_len(fd); @@ -485,7 +490,7 @@ static void dpaa2_eth_rx(struct dpaa2_eth_priv *priv, /* Check if we need to validate the L4 csum */ if (likely(dpaa2_fd_get_frc(fd) & DPAA2_FD_FRC_FASV)) { status = le32_to_cpu(fas->status); - validate_rx_csum(priv, status, skb); + dpaa2_eth_validate_rx_csum(priv, status, skb); } skb->protocol = eth_type_trans(skb, priv->net_dev); @@ -499,19 +504,71 @@ static void dpaa2_eth_rx(struct dpaa2_eth_priv *priv, return; err_build_skb: - free_rx_fd(priv, fd, vaddr); + dpaa2_eth_free_rx_fd(priv, fd, vaddr); err_frame_format: percpu_stats->rx_dropped++; } +/* Processing of Rx frames received on the error FQ + * We check and print the error bits and then free the frame + */ +static void dpaa2_eth_rx_err(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_channel *ch, + const struct dpaa2_fd *fd, + struct dpaa2_eth_fq *fq __always_unused) +{ + struct device *dev = priv->net_dev->dev.parent; + dma_addr_t addr = dpaa2_fd_get_addr(fd); + u8 fd_format = dpaa2_fd_get_format(fd); + struct rtnl_link_stats64 *percpu_stats; + struct dpaa2_eth_trap_item *trap_item; + struct dpaa2_fapr *fapr; + struct sk_buff *skb; + void *buf_data; + void *vaddr; + + vaddr = dpaa2_iova_to_virt(priv->iommu_domain, addr); + dma_sync_single_for_cpu(dev, addr, priv->rx_buf_size, + DMA_BIDIRECTIONAL); + + buf_data = vaddr + dpaa2_fd_get_offset(fd); + + if (fd_format == dpaa2_fd_single) { + dma_unmap_page(dev, addr, priv->rx_buf_size, + DMA_BIDIRECTIONAL); + skb = dpaa2_eth_build_linear_skb(ch, fd, vaddr); + } else if (fd_format == dpaa2_fd_sg) { + dma_unmap_page(dev, addr, priv->rx_buf_size, + DMA_BIDIRECTIONAL); + skb = dpaa2_eth_build_frag_skb(priv, ch, buf_data); + free_pages((unsigned long)vaddr, 0); + } else { + /* We don't support any other format */ + dpaa2_eth_free_rx_fd(priv, fd, vaddr); + goto err_frame_format; + } + + fapr = dpaa2_get_fapr(vaddr, false); + trap_item = dpaa2_eth_dl_get_trap(priv, fapr); + if (trap_item) + devlink_trap_report(priv->devlink, skb, trap_item->trap_ctx, + &priv->devlink_port, NULL); + consume_skb(skb); + +err_frame_format: + percpu_stats = this_cpu_ptr(priv->percpu_stats); + percpu_stats->rx_errors++; + ch->buf_count--; +} + /* Consume all frames pull-dequeued into the store. This is the simplest way to * make sure we don't accidentally issue another volatile dequeue which would * overwrite (leak) frames already in the store. * * Observance of NAPI budget is not our concern, leaving that to the caller. */ -static int consume_frames(struct dpaa2_eth_channel *ch, - struct dpaa2_eth_fq **src) +static int dpaa2_eth_consume_frames(struct dpaa2_eth_channel *ch, + struct dpaa2_eth_fq **src) { struct dpaa2_eth_priv *priv = ch->priv; struct dpaa2_eth_fq *fq = NULL; @@ -559,11 +616,57 @@ static int consume_frames(struct dpaa2_eth_channel *ch, return cleaned; } +static int dpaa2_eth_ptp_parse(struct sk_buff *skb, + u8 *msgtype, u8 *twostep, u8 *udp, + u16 *correction_offset, + u16 *origintimestamp_offset) +{ + unsigned int ptp_class; + struct ptp_header *hdr; + unsigned int type; + u8 *base; + + ptp_class = ptp_classify_raw(skb); + if (ptp_class == PTP_CLASS_NONE) + return -EINVAL; + + hdr = ptp_parse_header(skb, ptp_class); + if (!hdr) + return -EINVAL; + + *msgtype = ptp_get_msgtype(hdr, ptp_class); + *twostep = hdr->flag_field[0] & 0x2; + + type = ptp_class & PTP_CLASS_PMASK; + if (type == PTP_CLASS_IPV4 || + type == PTP_CLASS_IPV6) + *udp = 1; + else + *udp = 0; + + base = skb_mac_header(skb); + *correction_offset = (u8 *)&hdr->correction - base; + *origintimestamp_offset = (u8 *)hdr + sizeof(struct ptp_header) - base; + + return 0; +} + /* Configure the egress frame annotation for timestamp update */ -static void enable_tx_tstamp(struct dpaa2_fd *fd, void *buf_start) +static void dpaa2_eth_enable_tx_tstamp(struct dpaa2_eth_priv *priv, + struct dpaa2_fd *fd, + void *buf_start, + struct sk_buff *skb) { + struct ptp_tstamp origin_timestamp; + struct dpni_single_step_cfg cfg; + u8 msgtype, twostep, udp; struct dpaa2_faead *faead; + struct dpaa2_fas *fas; + struct timespec64 ts; + u16 offset1, offset2; u32 ctrl, frc; + __le64 *ns; + u8 *data; /* Mark the egress frame annotation area as valid */ frc = dpaa2_fd_get_frc(fd); @@ -579,12 +682,52 @@ static void enable_tx_tstamp(struct dpaa2_fd *fd, void *buf_start) ctrl = DPAA2_FAEAD_A2V | DPAA2_FAEAD_UPDV | DPAA2_FAEAD_UPD; faead = dpaa2_get_faead(buf_start, true); faead->ctrl = cpu_to_le32(ctrl); + + if (skb->cb[0] == TX_TSTAMP_ONESTEP_SYNC) { + if (dpaa2_eth_ptp_parse(skb, &msgtype, &twostep, &udp, + &offset1, &offset2) || + msgtype != 0 || twostep) { + WARN_ONCE(1, "Bad packet for one-step timestamping\n"); + return; + } + + /* Mark the frame annotation status as valid */ + frc = dpaa2_fd_get_frc(fd); + dpaa2_fd_set_frc(fd, frc | DPAA2_FD_FRC_FASV); + + /* Mark the PTP flag for one step timestamping */ + fas = dpaa2_get_fas(buf_start, true); + fas->status = cpu_to_le32(DPAA2_FAS_PTP); + + dpaa2_ptp->caps.gettime64(&dpaa2_ptp->caps, &ts); + ns = dpaa2_get_ts(buf_start, true); + *ns = cpu_to_le64(timespec64_to_ns(&ts) / + DPAA2_PTP_CLK_PERIOD_NS); + + /* Update current time to PTP message originTimestamp field */ + ns_to_ptp_tstamp(&origin_timestamp, le64_to_cpup(ns)); + data = skb_mac_header(skb); + *(__be16 *)(data + offset2) = htons(origin_timestamp.sec_msb); + *(__be32 *)(data + offset2 + 2) = + htonl(origin_timestamp.sec_lsb); + *(__be32 *)(data + offset2 + 6) = htonl(origin_timestamp.nsec); + + cfg.en = 1; + cfg.ch_update = udp; + cfg.offset = offset1; + cfg.peer_delay = 0; + + if (dpni_set_single_step_cfg(priv->mc_io, 0, priv->mc_token, + &cfg)) + WARN_ONCE(1, "Failed to set single step register"); + } } /* Create a frame descriptor based on a fragmented skb */ -static int build_sg_fd(struct dpaa2_eth_priv *priv, - struct sk_buff *skb, - struct dpaa2_fd *fd) +static int dpaa2_eth_build_sg_fd(struct dpaa2_eth_priv *priv, + struct sk_buff *skb, + struct dpaa2_fd *fd, + void **swa_addr) { struct device *dev = priv->net_dev->dev.parent; void *sgt_buf = NULL; @@ -606,7 +749,7 @@ static int build_sg_fd(struct dpaa2_eth_priv *priv, if (unlikely(PAGE_SIZE / sizeof(struct scatterlist) < nr_frags + 1)) return -EINVAL; - scl = kcalloc(nr_frags + 1, sizeof(struct scatterlist), GFP_ATOMIC); + scl = kmalloc_array(nr_frags + 1, sizeof(struct scatterlist), GFP_ATOMIC); if (unlikely(!scl)) return -ENOMEM; @@ -653,6 +796,7 @@ static int build_sg_fd(struct dpaa2_eth_priv *priv, * skb backpointer in the software annotation area. We'll need * all of them on Tx Conf. */ + *swa_addr = (void *)sgt_buf; swa = (struct dpaa2_eth_swa *)sgt_buf; swa->type = DPAA2_ETH_SWA_SG; swa->sg.skb = skb; @@ -672,9 +816,6 @@ static int build_sg_fd(struct dpaa2_eth_priv *priv, dpaa2_fd_set_len(fd, skb->len); dpaa2_fd_set_ctrl(fd, FD_CTRL_PTA); - if (priv->tx_tstamp && skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) - enable_tx_tstamp(fd, sgt_buf); - return 0; dma_map_single_failed: @@ -692,9 +833,10 @@ dma_map_sg_failed: * enough for the HW requirements, thus instead of realloc-ing the skb we * create a SG frame descriptor with only one entry. */ -static int build_sg_fd_single_buf(struct dpaa2_eth_priv *priv, - struct sk_buff *skb, - struct dpaa2_fd *fd) +static int dpaa2_eth_build_sg_fd_single_buf(struct dpaa2_eth_priv *priv, + struct sk_buff *skb, + struct dpaa2_fd *fd, + void **swa_addr) { struct device *dev = priv->net_dev->dev.parent; struct dpaa2_eth_sgt_cache *sgt_cache; @@ -732,6 +874,7 @@ static int build_sg_fd_single_buf(struct dpaa2_eth_priv *priv, dpaa2_sg_set_final(sgt, true); /* Store the skb backpointer in the SGT buffer */ + *swa_addr = (void *)sgt_buf; swa = (struct dpaa2_eth_swa *)sgt_buf; swa->type = DPAA2_ETH_SWA_SINGLE; swa->single.skb = skb; @@ -750,9 +893,6 @@ static int build_sg_fd_single_buf(struct dpaa2_eth_priv *priv, dpaa2_fd_set_len(fd, skb->len); dpaa2_fd_set_ctrl(fd, FD_CTRL_PTA); - if (priv->tx_tstamp && skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) - enable_tx_tstamp(fd, sgt_buf); - return 0; sgt_map_failed: @@ -767,16 +907,17 @@ data_map_failed: } /* Create a frame descriptor based on a linear skb */ -static int build_single_fd(struct dpaa2_eth_priv *priv, - struct sk_buff *skb, - struct dpaa2_fd *fd) +static int dpaa2_eth_build_single_fd(struct dpaa2_eth_priv *priv, + struct sk_buff *skb, + struct dpaa2_fd *fd, + void **swa_addr) { struct device *dev = priv->net_dev->dev.parent; u8 *buffer_start, *aligned_start; struct dpaa2_eth_swa *swa; dma_addr_t addr; - buffer_start = skb->data - dpaa2_eth_needed_headroom(priv, skb); + buffer_start = skb->data - dpaa2_eth_needed_headroom(skb); /* If there's enough room to align the FD address, do it. * It will help hardware optimize accesses. @@ -790,6 +931,7 @@ static int build_single_fd(struct dpaa2_eth_priv *priv, * (in the private data area) such that we can release it * on Tx confirm */ + *swa_addr = (void *)buffer_start; swa = (struct dpaa2_eth_swa *)buffer_start; swa->type = DPAA2_ETH_SWA_SINGLE; swa->single.skb = skb; @@ -806,9 +948,6 @@ static int build_single_fd(struct dpaa2_eth_priv *priv, dpaa2_fd_set_format(fd, dpaa2_fd_single); dpaa2_fd_set_ctrl(fd, FD_CTRL_PTA); - if (priv->tx_tstamp && skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) - enable_tx_tstamp(fd, buffer_start); - return 0; } @@ -819,9 +958,9 @@ static int build_single_fd(struct dpaa2_eth_priv *priv, * This can be called either from dpaa2_eth_tx_conf() or on the error path of * dpaa2_eth_tx(). */ -static void free_tx_fd(const struct dpaa2_eth_priv *priv, - struct dpaa2_eth_fq *fq, - const struct dpaa2_fd *fd, bool in_napi) +static void dpaa2_eth_free_tx_fd(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_fq *fq, + const struct dpaa2_fd *fd, bool in_napi) { struct device *dev = priv->net_dev->dev.parent; dma_addr_t fd_addr, sg_addr; @@ -892,7 +1031,7 @@ static void free_tx_fd(const struct dpaa2_eth_priv *priv, } /* Get the timestamp value */ - if (priv->tx_tstamp && skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) { + if (skb->cb[0] == TX_TSTAMP) { struct skb_shared_hwtstamps shhwtstamps; __le64 *ts = dpaa2_get_ts(buffer_start, true); u64 ns; @@ -902,6 +1041,8 @@ static void free_tx_fd(const struct dpaa2_eth_priv *priv, ns = DPAA2_PTP_CLK_PERIOD_NS * le64_to_cpup(ts); shhwtstamps.hwtstamp = ns_to_ktime(ns); skb_tstamp_tx(skb, &shhwtstamps); + } else if (skb->cb[0] == TX_TSTAMP_ONESTEP_SYNC) { + mutex_unlock(&priv->onestep_tstamp_lock); } /* Free SGT buffer allocated on tx */ @@ -921,7 +1062,8 @@ static void free_tx_fd(const struct dpaa2_eth_priv *priv, napi_consume_skb(skb, in_napi); } -static netdev_tx_t dpaa2_eth_tx(struct sk_buff *skb, struct net_device *net_dev) +static netdev_tx_t __dpaa2_eth_tx(struct sk_buff *skb, + struct net_device *net_dev) { struct dpaa2_eth_priv *priv = netdev_priv(net_dev); struct dpaa2_fd fd; @@ -934,11 +1076,12 @@ static netdev_tx_t dpaa2_eth_tx(struct sk_buff *skb, struct net_device *net_dev) u32 fd_len; u8 prio = 0; int err, i; + void *swa; percpu_stats = this_cpu_ptr(priv->percpu_stats); percpu_extras = this_cpu_ptr(priv->percpu_extras); - needed_headroom = dpaa2_eth_needed_headroom(priv, skb); + needed_headroom = dpaa2_eth_needed_headroom(skb); /* We'll be holding a back-reference to the skb until Tx Confirmation; * we don't want that overwritten by a concurrent Tx with a cloned skb. @@ -954,17 +1097,17 @@ static netdev_tx_t dpaa2_eth_tx(struct sk_buff *skb, struct net_device *net_dev) memset(&fd, 0, sizeof(fd)); if (skb_is_nonlinear(skb)) { - err = build_sg_fd(priv, skb, &fd); + err = dpaa2_eth_build_sg_fd(priv, skb, &fd, &swa); percpu_extras->tx_sg_frames++; percpu_extras->tx_sg_bytes += skb->len; } else if (skb_headroom(skb) < needed_headroom) { - err = build_sg_fd_single_buf(priv, skb, &fd); + err = dpaa2_eth_build_sg_fd_single_buf(priv, skb, &fd, &swa); percpu_extras->tx_sg_frames++; percpu_extras->tx_sg_bytes += skb->len; percpu_extras->tx_converted_sg_frames++; percpu_extras->tx_converted_sg_bytes += skb->len; } else { - err = build_single_fd(priv, skb, &fd); + err = dpaa2_eth_build_single_fd(priv, skb, &fd, &swa); } if (unlikely(err)) { @@ -972,6 +1115,9 @@ static netdev_tx_t dpaa2_eth_tx(struct sk_buff *skb, struct net_device *net_dev) goto err_build_fd; } + if (skb->cb[0]) + dpaa2_eth_enable_tx_tstamp(priv, &fd, swa, skb); + /* Tracing point */ trace_dpaa2_tx_fd(net_dev, &fd); @@ -1010,7 +1156,7 @@ static netdev_tx_t dpaa2_eth_tx(struct sk_buff *skb, struct net_device *net_dev) if (unlikely(err < 0)) { percpu_stats->tx_errors++; /* Clean up everything, including freeing the skb */ - free_tx_fd(priv, fq, &fd, false); + dpaa2_eth_free_tx_fd(priv, fq, &fd, false); netdev_tx_completed_queue(nq, 1, fd_len); } else { percpu_stats->tx_packets++; @@ -1025,6 +1171,63 @@ err_build_fd: return NETDEV_TX_OK; } +static void dpaa2_eth_tx_onestep_tstamp(struct work_struct *work) +{ + struct dpaa2_eth_priv *priv = container_of(work, struct dpaa2_eth_priv, + tx_onestep_tstamp); + struct sk_buff *skb; + + while (true) { + skb = skb_dequeue(&priv->tx_skbs); + if (!skb) + return; + + /* Lock just before TX one-step timestamping packet, + * and release the lock in dpaa2_eth_free_tx_fd when + * confirm the packet has been sent on hardware, or + * when clean up during transmit failure. + */ + mutex_lock(&priv->onestep_tstamp_lock); + __dpaa2_eth_tx(skb, priv->net_dev); + } +} + +static netdev_tx_t dpaa2_eth_tx(struct sk_buff *skb, struct net_device *net_dev) +{ + struct dpaa2_eth_priv *priv = netdev_priv(net_dev); + u8 msgtype, twostep, udp; + u16 offset1, offset2; + + /* Utilize skb->cb[0] for timestamping request per skb */ + skb->cb[0] = 0; + + if ((skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) && dpaa2_ptp) { + if (priv->tx_tstamp_type == HWTSTAMP_TX_ON) + skb->cb[0] = TX_TSTAMP; + else if (priv->tx_tstamp_type == HWTSTAMP_TX_ONESTEP_SYNC) + skb->cb[0] = TX_TSTAMP_ONESTEP_SYNC; + } + + /* TX for one-step timestamping PTP Sync packet */ + if (skb->cb[0] == TX_TSTAMP_ONESTEP_SYNC) { + if (!dpaa2_eth_ptp_parse(skb, &msgtype, &twostep, &udp, + &offset1, &offset2)) + if (msgtype == 0 && twostep == 0) { + skb_queue_tail(&priv->tx_skbs, skb); + queue_work(priv->dpaa2_ptp_wq, + &priv->tx_onestep_tstamp); + return NETDEV_TX_OK; + } + /* Use two-step timestamping if not one-step timestamping + * PTP Sync packet + */ + skb->cb[0] = TX_TSTAMP; + } + + /* TX for other packets */ + return __dpaa2_eth_tx(skb, net_dev); +} + /* Tx confirmation frame processing routine */ static void dpaa2_eth_tx_conf(struct dpaa2_eth_priv *priv, struct dpaa2_eth_channel *ch __always_unused, @@ -1045,7 +1248,7 @@ static void dpaa2_eth_tx_conf(struct dpaa2_eth_priv *priv, /* Check frame errors in the FD field */ fd_errors = dpaa2_fd_get_ctrl(fd) & DPAA2_FD_TX_ERR_MASK; - free_tx_fd(priv, fq, fd, true); + dpaa2_eth_free_tx_fd(priv, fq, fd, true); if (likely(!fd_errors)) return; @@ -1059,7 +1262,7 @@ static void dpaa2_eth_tx_conf(struct dpaa2_eth_priv *priv, percpu_stats->tx_errors++; } -static int set_rx_csum(struct dpaa2_eth_priv *priv, bool enable) +static int dpaa2_eth_set_rx_csum(struct dpaa2_eth_priv *priv, bool enable) { int err; @@ -1082,7 +1285,7 @@ static int set_rx_csum(struct dpaa2_eth_priv *priv, bool enable) return 0; } -static int set_tx_csum(struct dpaa2_eth_priv *priv, bool enable) +static int dpaa2_eth_set_tx_csum(struct dpaa2_eth_priv *priv, bool enable) { int err; @@ -1106,8 +1309,8 @@ static int set_tx_csum(struct dpaa2_eth_priv *priv, bool enable) /* Perform a single release command to add buffers * to the specified buffer pool */ -static int add_bufs(struct dpaa2_eth_priv *priv, - struct dpaa2_eth_channel *ch, u16 bpid) +static int dpaa2_eth_add_bufs(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_channel *ch, u16 bpid) { struct device *dev = priv->net_dev->dev.parent; u64 buf_array[DPAA2_ETH_BUFS_PER_CMD]; @@ -1155,7 +1358,7 @@ release_bufs: * not much else we can do about it */ if (err) { - free_bufs(priv, buf_array, i); + dpaa2_eth_free_bufs(priv, buf_array, i); return 0; } @@ -1173,7 +1376,7 @@ err_alloc: return 0; } -static int seed_pool(struct dpaa2_eth_priv *priv, u16 bpid) +static int dpaa2_eth_seed_pool(struct dpaa2_eth_priv *priv, u16 bpid) { int i, j; int new_count; @@ -1181,7 +1384,7 @@ static int seed_pool(struct dpaa2_eth_priv *priv, u16 bpid) for (j = 0; j < priv->num_channels; j++) { for (i = 0; i < DPAA2_ETH_NUM_BUFS; i += DPAA2_ETH_BUFS_PER_CMD) { - new_count = add_bufs(priv, priv->channel[j], bpid); + new_count = dpaa2_eth_add_bufs(priv, priv->channel[j], bpid); priv->channel[j]->buf_count += new_count; if (new_count < DPAA2_ETH_BUFS_PER_CMD) { @@ -1193,11 +1396,11 @@ static int seed_pool(struct dpaa2_eth_priv *priv, u16 bpid) return 0; } -/** +/* * Drain the specified number of buffers from the DPNI's private buffer pool. * @count must not exceeed DPAA2_ETH_BUFS_PER_CMD */ -static void drain_bufs(struct dpaa2_eth_priv *priv, int count) +static void dpaa2_eth_drain_bufs(struct dpaa2_eth_priv *priv, int count) { u64 buf_array[DPAA2_ETH_BUFS_PER_CMD]; int retries = 0; @@ -1213,17 +1416,17 @@ static void drain_bufs(struct dpaa2_eth_priv *priv, int count) netdev_err(priv->net_dev, "dpaa2_io_service_acquire() failed\n"); return; } - free_bufs(priv, buf_array, ret); + dpaa2_eth_free_bufs(priv, buf_array, ret); retries = 0; } while (ret); } -static void drain_pool(struct dpaa2_eth_priv *priv) +static void dpaa2_eth_drain_pool(struct dpaa2_eth_priv *priv) { int i; - drain_bufs(priv, DPAA2_ETH_BUFS_PER_CMD); - drain_bufs(priv, 1); + dpaa2_eth_drain_bufs(priv, DPAA2_ETH_BUFS_PER_CMD); + dpaa2_eth_drain_bufs(priv, 1); for (i = 0; i < priv->num_channels; i++) priv->channel[i]->buf_count = 0; @@ -1232,9 +1435,9 @@ static void drain_pool(struct dpaa2_eth_priv *priv) /* Function is called from softirq context only, so we don't need to guard * the access to percpu count */ -static int refill_pool(struct dpaa2_eth_priv *priv, - struct dpaa2_eth_channel *ch, - u16 bpid) +static int dpaa2_eth_refill_pool(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_channel *ch, + u16 bpid) { int new_count; @@ -1242,7 +1445,7 @@ static int refill_pool(struct dpaa2_eth_priv *priv, return 0; do { - new_count = add_bufs(priv, ch, bpid); + new_count = dpaa2_eth_add_bufs(priv, ch, bpid); if (unlikely(!new_count)) { /* Out of memory; abort for now, we'll try later on */ break; @@ -1272,7 +1475,7 @@ static void dpaa2_eth_sgt_cache_drain(struct dpaa2_eth_priv *priv) } } -static int pull_channel(struct dpaa2_eth_channel *ch) +static int dpaa2_eth_pull_channel(struct dpaa2_eth_channel *ch) { int err; int dequeues = -1; @@ -1319,14 +1522,14 @@ static int dpaa2_eth_poll(struct napi_struct *napi, int budget) ch->rx_list = &rx_list; do { - err = pull_channel(ch); + err = dpaa2_eth_pull_channel(ch); if (unlikely(err)) break; /* Refill pool if appropriate */ - refill_pool(priv, ch, priv->bpid); + dpaa2_eth_refill_pool(priv, ch, priv->bpid); - store_cleaned = consume_frames(ch, &fq); + store_cleaned = dpaa2_eth_consume_frames(ch, &fq); if (store_cleaned <= 0) break; if (fq->type == DPAA2_RX_FQ) { @@ -1375,12 +1578,12 @@ out: if (ch->xdp.res & XDP_REDIRECT) xdp_do_flush_map(); else if (rx_cleaned && ch->xdp.res & XDP_TX) - xdp_tx_flush(priv, ch, &priv->fq[flowid]); + dpaa2_eth_xdp_tx_flush(priv, ch, &priv->fq[flowid]); return work_done; } -static void enable_ch_napi(struct dpaa2_eth_priv *priv) +static void dpaa2_eth_enable_ch_napi(struct dpaa2_eth_priv *priv) { struct dpaa2_eth_channel *ch; int i; @@ -1391,7 +1594,7 @@ static void enable_ch_napi(struct dpaa2_eth_priv *priv) } } -static void disable_ch_napi(struct dpaa2_eth_priv *priv) +static void dpaa2_eth_disable_ch_napi(struct dpaa2_eth_priv *priv) { struct dpaa2_eth_channel *ch; int i; @@ -1465,7 +1668,7 @@ set_cgtd: priv->rx_cgtd_enabled = td.enable; } -static int link_state_update(struct dpaa2_eth_priv *priv) +static int dpaa2_eth_link_state_update(struct dpaa2_eth_priv *priv) { struct dpni_link_state state = {0}; bool tx_pause; @@ -1517,7 +1720,7 @@ static int dpaa2_eth_open(struct net_device *net_dev) struct dpaa2_eth_priv *priv = netdev_priv(net_dev); int err; - err = seed_pool(priv, priv->bpid); + err = dpaa2_eth_seed_pool(priv, priv->bpid); if (err) { /* Not much to do; the buffer pool, though not filled up, * may still contain some buffers which would enable us @@ -1541,7 +1744,7 @@ static int dpaa2_eth_open(struct net_device *net_dev) */ netif_carrier_off(net_dev); } - enable_ch_napi(priv); + dpaa2_eth_enable_ch_napi(priv); err = dpni_enable(priv->mc_io, 0, priv->mc_token); if (err < 0) { @@ -1549,30 +1752,19 @@ static int dpaa2_eth_open(struct net_device *net_dev) goto enable_err; } - if (!priv->mac) { - /* If the DPMAC object has already processed the link up - * interrupt, we have to learn the link state ourselves. - */ - err = link_state_update(priv); - if (err < 0) { - netdev_err(net_dev, "Can't update link state\n"); - goto link_state_err; - } - } else { + if (priv->mac) phylink_start(priv->mac->phylink); - } return 0; -link_state_err: enable_err: - disable_ch_napi(priv); - drain_pool(priv); + dpaa2_eth_disable_ch_napi(priv); + dpaa2_eth_drain_pool(priv); return err; } /* Total number of in-flight frames on ingress queues */ -static u32 ingress_fq_count(struct dpaa2_eth_priv *priv) +static u32 dpaa2_eth_ingress_fq_count(struct dpaa2_eth_priv *priv) { struct dpaa2_eth_fq *fq; u32 fcnt = 0, bcnt = 0, total = 0; @@ -1591,13 +1783,13 @@ static u32 ingress_fq_count(struct dpaa2_eth_priv *priv) return total; } -static void wait_for_ingress_fq_empty(struct dpaa2_eth_priv *priv) +static void dpaa2_eth_wait_for_ingress_fq_empty(struct dpaa2_eth_priv *priv) { int retries = 10; u32 pending; do { - pending = ingress_fq_count(priv); + pending = dpaa2_eth_ingress_fq_count(priv); if (pending) msleep(100); } while (pending && --retries); @@ -1605,7 +1797,7 @@ static void wait_for_ingress_fq_empty(struct dpaa2_eth_priv *priv) #define DPNI_TX_PENDING_VER_MAJOR 7 #define DPNI_TX_PENDING_VER_MINOR 13 -static void wait_for_egress_fq_empty(struct dpaa2_eth_priv *priv) +static void dpaa2_eth_wait_for_egress_fq_empty(struct dpaa2_eth_priv *priv) { union dpni_statistics stats; int retries = 10; @@ -1651,7 +1843,7 @@ static int dpaa2_eth_stop(struct net_device *net_dev) * on WRIOP. After it finishes, wait until all remaining frames on Rx * and Tx conf queues are consumed on NAPI poll. */ - wait_for_egress_fq_empty(priv); + dpaa2_eth_wait_for_egress_fq_empty(priv); do { dpni_disable(priv->mc_io, 0, priv->mc_token); @@ -1667,11 +1859,11 @@ static int dpaa2_eth_stop(struct net_device *net_dev) */ } - wait_for_ingress_fq_empty(priv); - disable_ch_napi(priv); + dpaa2_eth_wait_for_ingress_fq_empty(priv); + dpaa2_eth_disable_ch_napi(priv); /* Empty the buffer pool */ - drain_pool(priv); + dpaa2_eth_drain_pool(priv); /* Empty the Scatter-Gather Buffer cache */ dpaa2_eth_sgt_cache_drain(priv); @@ -1725,8 +1917,8 @@ static void dpaa2_eth_get_stats(struct net_device *net_dev, /* Copy mac unicast addresses from @net_dev to @priv. * Its sole purpose is to make dpaa2_eth_set_rx_mode() more readable. */ -static void add_uc_hw_addr(const struct net_device *net_dev, - struct dpaa2_eth_priv *priv) +static void dpaa2_eth_add_uc_hw_addr(const struct net_device *net_dev, + struct dpaa2_eth_priv *priv) { struct netdev_hw_addr *ha; int err; @@ -1744,8 +1936,8 @@ static void add_uc_hw_addr(const struct net_device *net_dev, /* Copy mac multicast addresses from @net_dev to @priv * Its sole purpose is to make dpaa2_eth_set_rx_mode() more readable. */ -static void add_mc_hw_addr(const struct net_device *net_dev, - struct dpaa2_eth_priv *priv) +static void dpaa2_eth_add_mc_hw_addr(const struct net_device *net_dev, + struct dpaa2_eth_priv *priv) { struct netdev_hw_addr *ha; int err; @@ -1810,7 +2002,7 @@ static void dpaa2_eth_set_rx_mode(struct net_device *net_dev) err = dpni_clear_mac_filters(mc_io, 0, mc_token, 1, 0); if (err) netdev_warn(net_dev, "Can't clear uc filters\n"); - add_uc_hw_addr(net_dev, priv); + dpaa2_eth_add_uc_hw_addr(net_dev, priv); /* Finally, clear uc promisc and set mc promisc as requested. */ err = dpni_set_unicast_promisc(mc_io, 0, mc_token, 0); @@ -1833,8 +2025,8 @@ static void dpaa2_eth_set_rx_mode(struct net_device *net_dev) err = dpni_clear_mac_filters(mc_io, 0, mc_token, 1, 1); if (err) netdev_warn(net_dev, "Can't clear mac filters\n"); - add_mc_hw_addr(net_dev, priv); - add_uc_hw_addr(net_dev, priv); + dpaa2_eth_add_mc_hw_addr(net_dev, priv); + dpaa2_eth_add_uc_hw_addr(net_dev, priv); /* Now we can clear both ucast and mcast promisc, without risking * to drop legitimate frames anymore. @@ -1868,14 +2060,14 @@ static int dpaa2_eth_set_features(struct net_device *net_dev, if (changed & NETIF_F_RXCSUM) { enable = !!(features & NETIF_F_RXCSUM); - err = set_rx_csum(priv, enable); + err = dpaa2_eth_set_rx_csum(priv, enable); if (err) return err; } if (changed & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)) { enable = !!(features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)); - err = set_tx_csum(priv, enable); + err = dpaa2_eth_set_tx_csum(priv, enable); if (err) return err; } @@ -1888,15 +2080,17 @@ static int dpaa2_eth_ts_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) struct dpaa2_eth_priv *priv = netdev_priv(dev); struct hwtstamp_config config; + if (!dpaa2_ptp) + return -EINVAL; + if (copy_from_user(&config, rq->ifr_data, sizeof(config))) return -EFAULT; switch (config.tx_type) { case HWTSTAMP_TX_OFF: - priv->tx_tstamp = false; - break; case HWTSTAMP_TX_ON: - priv->tx_tstamp = true; + case HWTSTAMP_TX_ONESTEP_SYNC: + priv->tx_tstamp_type = config.tx_type; break; default: return -ERANGE; @@ -1944,7 +2138,7 @@ static bool xdp_mtu_valid(struct dpaa2_eth_priv *priv, int mtu) return true; } -static int set_rx_mfl(struct dpaa2_eth_priv *priv, int mtu, bool has_xdp) +static int dpaa2_eth_set_rx_mfl(struct dpaa2_eth_priv *priv, int mtu, bool has_xdp) { int mfl, err; @@ -1978,7 +2172,7 @@ static int dpaa2_eth_change_mtu(struct net_device *dev, int new_mtu) if (!xdp_mtu_valid(priv, new_mtu)) return -EINVAL; - err = set_rx_mfl(priv, new_mtu, true); + err = dpaa2_eth_set_rx_mfl(priv, new_mtu, true); if (err) return err; @@ -1987,7 +2181,7 @@ out: return 0; } -static int update_rx_buffer_headroom(struct dpaa2_eth_priv *priv, bool has_xdp) +static int dpaa2_eth_update_rx_buffer_headroom(struct dpaa2_eth_priv *priv, bool has_xdp) { struct dpni_buffer_layout buf_layout = {0}; int err; @@ -2013,7 +2207,7 @@ static int update_rx_buffer_headroom(struct dpaa2_eth_priv *priv, bool has_xdp) return 0; } -static int setup_xdp(struct net_device *dev, struct bpf_prog *prog) +static int dpaa2_eth_setup_xdp(struct net_device *dev, struct bpf_prog *prog) { struct dpaa2_eth_priv *priv = netdev_priv(dev); struct dpaa2_eth_channel *ch; @@ -2039,10 +2233,10 @@ static int setup_xdp(struct net_device *dev, struct bpf_prog *prog) * so we are sure no old format buffers will be used from now on. */ if (need_update) { - err = set_rx_mfl(priv, dev->mtu, !!prog); + err = dpaa2_eth_set_rx_mfl(priv, dev->mtu, !!prog); if (err) goto out_err; - err = update_rx_buffer_headroom(priv, !!prog); + err = dpaa2_eth_update_rx_buffer_headroom(priv, !!prog); if (err) goto out_err; } @@ -2079,7 +2273,7 @@ static int dpaa2_eth_xdp(struct net_device *dev, struct netdev_bpf *xdp) { switch (xdp->command) { case XDP_SETUP_PROG: - return setup_xdp(dev, xdp->prog); + return dpaa2_eth_setup_xdp(dev, xdp->prog); default: return -EINVAL; } @@ -2091,7 +2285,6 @@ static int dpaa2_eth_xdp_create_fd(struct net_device *net_dev, struct xdp_frame *xdpf, struct dpaa2_fd *fd) { - struct dpaa2_eth_priv *priv = netdev_priv(net_dev); struct device *dev = net_dev->dev.parent; unsigned int needed_headroom; struct dpaa2_eth_swa *swa; @@ -2101,7 +2294,7 @@ static int dpaa2_eth_xdp_create_fd(struct net_device *net_dev, /* We require a minimum headroom to be able to transmit the frame. * Otherwise return an error and let the original net_device handle it */ - needed_headroom = dpaa2_eth_needed_headroom(priv, NULL); + needed_headroom = dpaa2_eth_needed_headroom(NULL); if (xdpf->headroom < needed_headroom) return -EINVAL; @@ -2316,7 +2509,7 @@ static const struct net_device_ops dpaa2_eth_ops = { .ndo_setup_tc = dpaa2_eth_setup_tc, }; -static void cdan_cb(struct dpaa2_io_notification_ctx *ctx) +static void dpaa2_eth_cdan_cb(struct dpaa2_io_notification_ctx *ctx) { struct dpaa2_eth_channel *ch; @@ -2329,7 +2522,7 @@ static void cdan_cb(struct dpaa2_io_notification_ctx *ctx) } /* Allocate and configure a DPCON object */ -static struct fsl_mc_device *setup_dpcon(struct dpaa2_eth_priv *priv) +static struct fsl_mc_device *dpaa2_eth_setup_dpcon(struct dpaa2_eth_priv *priv) { struct fsl_mc_device *dpcon; struct device *dev = priv->net_dev->dev.parent; @@ -2373,16 +2566,15 @@ free: return ERR_PTR(err); } -static void free_dpcon(struct dpaa2_eth_priv *priv, - struct fsl_mc_device *dpcon) +static void dpaa2_eth_free_dpcon(struct dpaa2_eth_priv *priv, + struct fsl_mc_device *dpcon) { dpcon_disable(priv->mc_io, 0, dpcon->mc_handle); dpcon_close(priv->mc_io, 0, dpcon->mc_handle); fsl_mc_object_free(dpcon); } -static struct dpaa2_eth_channel * -alloc_channel(struct dpaa2_eth_priv *priv) +static struct dpaa2_eth_channel *dpaa2_eth_alloc_channel(struct dpaa2_eth_priv *priv) { struct dpaa2_eth_channel *channel; struct dpcon_attr attr; @@ -2393,7 +2585,7 @@ alloc_channel(struct dpaa2_eth_priv *priv) if (!channel) return NULL; - channel->dpcon = setup_dpcon(priv); + channel->dpcon = dpaa2_eth_setup_dpcon(priv); if (IS_ERR(channel->dpcon)) { err = PTR_ERR(channel->dpcon); goto err_setup; @@ -2413,23 +2605,23 @@ alloc_channel(struct dpaa2_eth_priv *priv) return channel; err_get_attr: - free_dpcon(priv, channel->dpcon); + dpaa2_eth_free_dpcon(priv, channel->dpcon); err_setup: kfree(channel); return ERR_PTR(err); } -static void free_channel(struct dpaa2_eth_priv *priv, - struct dpaa2_eth_channel *channel) +static void dpaa2_eth_free_channel(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_channel *channel) { - free_dpcon(priv, channel->dpcon); + dpaa2_eth_free_dpcon(priv, channel->dpcon); kfree(channel); } /* DPIO setup: allocate and configure QBMan channels, setup core affinity * and register data availability notifications */ -static int setup_dpio(struct dpaa2_eth_priv *priv) +static int dpaa2_eth_setup_dpio(struct dpaa2_eth_priv *priv) { struct dpaa2_io_notification_ctx *nctx; struct dpaa2_eth_channel *channel; @@ -2449,7 +2641,7 @@ static int setup_dpio(struct dpaa2_eth_priv *priv) cpumask_clear(&priv->dpio_cpumask); for_each_online_cpu(i) { /* Try to allocate a channel */ - channel = alloc_channel(priv); + channel = dpaa2_eth_alloc_channel(priv); if (IS_ERR_OR_NULL(channel)) { err = PTR_ERR_OR_ZERO(channel); if (err != -EPROBE_DEFER) @@ -2462,7 +2654,7 @@ static int setup_dpio(struct dpaa2_eth_priv *priv) nctx = &channel->nctx; nctx->is_cdan = 1; - nctx->cb = cdan_cb; + nctx->cb = dpaa2_eth_cdan_cb; nctx->id = channel->ch_id; nctx->desired_cpu = i; @@ -2510,14 +2702,14 @@ static int setup_dpio(struct dpaa2_eth_priv *priv) err_set_cdan: dpaa2_io_service_deregister(channel->dpio, nctx, dev); err_service_reg: - free_channel(priv, channel); + dpaa2_eth_free_channel(priv, channel); err_alloc_ch: if (err == -EPROBE_DEFER) { for (i = 0; i < priv->num_channels; i++) { channel = priv->channel[i]; nctx = &channel->nctx; dpaa2_io_service_deregister(channel->dpio, nctx, dev); - free_channel(priv, channel); + dpaa2_eth_free_channel(priv, channel); } priv->num_channels = 0; return err; @@ -2534,7 +2726,7 @@ err_alloc_ch: return 0; } -static void free_dpio(struct dpaa2_eth_priv *priv) +static void dpaa2_eth_free_dpio(struct dpaa2_eth_priv *priv) { struct device *dev = priv->net_dev->dev.parent; struct dpaa2_eth_channel *ch; @@ -2544,12 +2736,12 @@ static void free_dpio(struct dpaa2_eth_priv *priv) for (i = 0; i < priv->num_channels; i++) { ch = priv->channel[i]; dpaa2_io_service_deregister(ch->dpio, &ch->nctx, dev); - free_channel(priv, ch); + dpaa2_eth_free_channel(priv, ch); } } -static struct dpaa2_eth_channel *get_affine_channel(struct dpaa2_eth_priv *priv, - int cpu) +static struct dpaa2_eth_channel *dpaa2_eth_get_affine_channel(struct dpaa2_eth_priv *priv, + int cpu) { struct device *dev = priv->net_dev->dev.parent; int i; @@ -2566,7 +2758,7 @@ static struct dpaa2_eth_channel *get_affine_channel(struct dpaa2_eth_priv *priv, return priv->channel[0]; } -static void set_fq_affinity(struct dpaa2_eth_priv *priv) +static void dpaa2_eth_set_fq_affinity(struct dpaa2_eth_priv *priv) { struct device *dev = priv->net_dev->dev.parent; struct dpaa2_eth_fq *fq; @@ -2583,6 +2775,7 @@ static void set_fq_affinity(struct dpaa2_eth_priv *priv) fq = &priv->fq[i]; switch (fq->type) { case DPAA2_RX_FQ: + case DPAA2_RX_ERR_FQ: fq->target_cpu = rx_cpu; rx_cpu = cpumask_next(rx_cpu, &priv->dpio_cpumask); if (rx_cpu >= nr_cpu_ids) @@ -2597,13 +2790,13 @@ static void set_fq_affinity(struct dpaa2_eth_priv *priv) default: dev_err(dev, "Unknown FQ type: %d\n", fq->type); } - fq->channel = get_affine_channel(priv, fq->target_cpu); + fq->channel = dpaa2_eth_get_affine_channel(priv, fq->target_cpu); } update_xps(priv); } -static void setup_fqs(struct dpaa2_eth_priv *priv) +static void dpaa2_eth_setup_fqs(struct dpaa2_eth_priv *priv) { int i, j; @@ -2626,12 +2819,16 @@ static void setup_fqs(struct dpaa2_eth_priv *priv) } } + /* We have exactly one Rx error queue per DPNI */ + priv->fq[priv->num_fqs].type = DPAA2_RX_ERR_FQ; + priv->fq[priv->num_fqs++].consume = dpaa2_eth_rx_err; + /* For each FQ, decide on which core to process incoming frames */ - set_fq_affinity(priv); + dpaa2_eth_set_fq_affinity(priv); } /* Allocate and configure one buffer pool for each interface */ -static int setup_dpbp(struct dpaa2_eth_priv *priv) +static int dpaa2_eth_setup_dpbp(struct dpaa2_eth_priv *priv) { int err; struct fsl_mc_device *dpbp_dev; @@ -2690,15 +2887,15 @@ err_open: return err; } -static void free_dpbp(struct dpaa2_eth_priv *priv) +static void dpaa2_eth_free_dpbp(struct dpaa2_eth_priv *priv) { - drain_pool(priv); + dpaa2_eth_drain_pool(priv); dpbp_disable(priv->mc_io, 0, priv->dpbp_dev->mc_handle); dpbp_close(priv->mc_io, 0, priv->dpbp_dev->mc_handle); fsl_mc_object_free(priv->dpbp_dev); } -static int set_buffer_layout(struct dpaa2_eth_priv *priv) +static int dpaa2_eth_set_buffer_layout(struct dpaa2_eth_priv *priv) { struct device *dev = priv->net_dev->dev.parent; struct dpni_buffer_layout buf_layout = {0}; @@ -2723,8 +2920,10 @@ static int set_buffer_layout(struct dpaa2_eth_priv *priv) /* tx buffer */ buf_layout.private_data_size = DPAA2_ETH_SWA_SIZE; buf_layout.pass_timestamp = true; + buf_layout.pass_frame_status = true; buf_layout.options = DPNI_BUF_LAYOUT_OPT_PRIVATE_DATA_SIZE | - DPNI_BUF_LAYOUT_OPT_TIMESTAMP; + DPNI_BUF_LAYOUT_OPT_TIMESTAMP | + DPNI_BUF_LAYOUT_OPT_FRAME_STATUS; err = dpni_set_buffer_layout(priv->mc_io, 0, priv->mc_token, DPNI_QUEUE_TX, &buf_layout); if (err) { @@ -2733,7 +2932,8 @@ static int set_buffer_layout(struct dpaa2_eth_priv *priv) } /* tx-confirm buffer */ - buf_layout.options = DPNI_BUF_LAYOUT_OPT_TIMESTAMP; + buf_layout.options = DPNI_BUF_LAYOUT_OPT_TIMESTAMP | + DPNI_BUF_LAYOUT_OPT_FRAME_STATUS; err = dpni_set_buffer_layout(priv->mc_io, 0, priv->mc_token, DPNI_QUEUE_TX_CONFIRM, &buf_layout); if (err) { @@ -2815,7 +3015,7 @@ static inline int dpaa2_eth_enqueue_fq_multiple(struct dpaa2_eth_priv *priv, return 0; } -static void set_enqueue_mode(struct dpaa2_eth_priv *priv) +static void dpaa2_eth_set_enqueue_mode(struct dpaa2_eth_priv *priv) { if (dpaa2_eth_cmp_dpni_ver(priv, DPNI_ENQUEUE_FQID_VER_MAJOR, DPNI_ENQUEUE_FQID_VER_MINOR) < 0) @@ -2824,7 +3024,7 @@ static void set_enqueue_mode(struct dpaa2_eth_priv *priv) priv->enqueue = dpaa2_eth_enqueue_fq_multiple; } -static int set_pause(struct dpaa2_eth_priv *priv) +static int dpaa2_eth_set_pause(struct dpaa2_eth_priv *priv) { struct device *dev = priv->net_dev->dev.parent; struct dpni_link_cfg link_cfg = {0}; @@ -2851,7 +3051,7 @@ static int set_pause(struct dpaa2_eth_priv *priv) return 0; } -static void update_tx_fqids(struct dpaa2_eth_priv *priv) +static void dpaa2_eth_update_tx_fqids(struct dpaa2_eth_priv *priv) { struct dpni_queue_id qid = {0}; struct dpaa2_eth_fq *fq; @@ -2893,7 +3093,7 @@ out_err: } /* Configure ingress classification based on VLAN PCP */ -static int set_vlan_qos(struct dpaa2_eth_priv *priv) +static int dpaa2_eth_set_vlan_qos(struct dpaa2_eth_priv *priv) { struct device *dev = priv->net_dev->dev.parent; struct dpkg_profile_cfg kg_cfg = {0}; @@ -3005,7 +3205,7 @@ out_free_tbl: } /* Configure the DPNI object this interface is associated with */ -static int setup_dpni(struct fsl_mc_device *ls_dev) +static int dpaa2_eth_setup_dpni(struct fsl_mc_device *ls_dev) { struct device *dev = &ls_dev->dev; struct dpaa2_eth_priv *priv; @@ -3053,20 +3253,20 @@ static int setup_dpni(struct fsl_mc_device *ls_dev) goto close; } - err = set_buffer_layout(priv); + err = dpaa2_eth_set_buffer_layout(priv); if (err) goto close; - set_enqueue_mode(priv); + dpaa2_eth_set_enqueue_mode(priv); /* Enable pause frame support */ if (dpaa2_eth_has_pause_support(priv)) { - err = set_pause(priv); + err = dpaa2_eth_set_pause(priv); if (err) goto close; } - err = set_vlan_qos(priv); + err = dpaa2_eth_set_vlan_qos(priv); if (err && err != -EOPNOTSUPP) goto close; @@ -3086,7 +3286,7 @@ close: return err; } -static void free_dpni(struct dpaa2_eth_priv *priv) +static void dpaa2_eth_free_dpni(struct dpaa2_eth_priv *priv) { int err; @@ -3098,8 +3298,8 @@ static void free_dpni(struct dpaa2_eth_priv *priv) dpni_close(priv->mc_io, 0, priv->mc_token); } -static int setup_rx_flow(struct dpaa2_eth_priv *priv, - struct dpaa2_eth_fq *fq) +static int dpaa2_eth_setup_rx_flow(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_fq *fq) { struct device *dev = priv->net_dev->dev.parent; struct dpni_queue queue; @@ -3150,8 +3350,8 @@ static int setup_rx_flow(struct dpaa2_eth_priv *priv, return 0; } -static int setup_tx_flow(struct dpaa2_eth_priv *priv, - struct dpaa2_eth_fq *fq) +static int dpaa2_eth_setup_tx_flow(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_fq *fq) { struct device *dev = priv->net_dev->dev.parent; struct dpni_queue queue; @@ -3198,6 +3398,38 @@ static int setup_tx_flow(struct dpaa2_eth_priv *priv, return 0; } +static int setup_rx_err_flow(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_fq *fq) +{ + struct device *dev = priv->net_dev->dev.parent; + struct dpni_queue q = { { 0 } }; + struct dpni_queue_id qid; + u8 q_opt = DPNI_QUEUE_OPT_USER_CTX | DPNI_QUEUE_OPT_DEST; + int err; + + err = dpni_get_queue(priv->mc_io, 0, priv->mc_token, + DPNI_QUEUE_RX_ERR, 0, 0, &q, &qid); + if (err) { + dev_err(dev, "dpni_get_queue() failed (%d)\n", err); + return err; + } + + fq->fqid = qid.fqid; + + q.destination.id = fq->channel->dpcon_id; + q.destination.type = DPNI_DEST_DPCON; + q.destination.priority = 1; + q.user_context = (u64)(uintptr_t)fq; + err = dpni_set_queue(priv->mc_io, 0, priv->mc_token, + DPNI_QUEUE_RX_ERR, 0, 0, q_opt, &q); + if (err) { + dev_err(dev, "dpni_set_queue() failed (%d)\n", err); + return err; + } + + return 0; +} + /* Supported header fields for Rx hash distribution key */ static const struct dpaa2_eth_dist_fields dist_fields[] = { { @@ -3266,7 +3498,7 @@ static const struct dpaa2_eth_dist_fields dist_fields[] = { }; /* Configure the Rx hash key using the legacy API */ -static int config_legacy_hash_key(struct dpaa2_eth_priv *priv, dma_addr_t key) +static int dpaa2_eth_config_legacy_hash_key(struct dpaa2_eth_priv *priv, dma_addr_t key) { struct device *dev = priv->net_dev->dev.parent; struct dpni_rx_tc_dist_cfg dist_cfg; @@ -3291,7 +3523,7 @@ static int config_legacy_hash_key(struct dpaa2_eth_priv *priv, dma_addr_t key) } /* Configure the Rx hash key using the new API */ -static int config_hash_key(struct dpaa2_eth_priv *priv, dma_addr_t key) +static int dpaa2_eth_config_hash_key(struct dpaa2_eth_priv *priv, dma_addr_t key) { struct device *dev = priv->net_dev->dev.parent; struct dpni_rx_dist_cfg dist_cfg; @@ -3311,13 +3543,19 @@ static int config_hash_key(struct dpaa2_eth_priv *priv, dma_addr_t key) dev_err(dev, "dpni_set_rx_hash_dist failed\n"); break; } + + /* If the flow steering / hashing key is shared between all + * traffic classes, install it just once + */ + if (priv->dpni_attrs.options & DPNI_OPT_SHARED_FS) + break; } return err; } /* Configure the Rx flow classification key */ -static int config_cls_key(struct dpaa2_eth_priv *priv, dma_addr_t key) +static int dpaa2_eth_config_cls_key(struct dpaa2_eth_priv *priv, dma_addr_t key) { struct device *dev = priv->net_dev->dev.parent; struct dpni_rx_dist_cfg dist_cfg; @@ -3337,6 +3575,12 @@ static int config_cls_key(struct dpaa2_eth_priv *priv, dma_addr_t key) dev_err(dev, "dpni_set_rx_fs_dist failed\n"); break; } + + /* If the flow steering / hashing key is shared between all + * traffic classes, install it just once + */ + if (priv->dpni_attrs.options & DPNI_OPT_SHARED_FS) + break; } return err; @@ -3452,11 +3696,11 @@ static int dpaa2_eth_set_dist_key(struct net_device *net_dev, if (type == DPAA2_ETH_RX_DIST_HASH) { if (dpaa2_eth_has_legacy_dist(priv)) - err = config_legacy_hash_key(priv, key_iova); + err = dpaa2_eth_config_legacy_hash_key(priv, key_iova); else - err = config_hash_key(priv, key_iova); + err = dpaa2_eth_config_hash_key(priv, key_iova); } else { - err = config_cls_key(priv, key_iova); + err = dpaa2_eth_config_cls_key(priv, key_iova); } dma_unmap_single(dev, key_iova, DPAA2_CLASSIFIER_DMA_SIZE, @@ -3531,7 +3775,7 @@ out: /* Bind the DPNI to its needed objects and resources: buffer pool, DPIOs, * frame queues and channels */ -static int bind_dpni(struct dpaa2_eth_priv *priv) +static int dpaa2_eth_bind_dpni(struct dpaa2_eth_priv *priv) { struct net_device *net_dev = priv->net_dev; struct device *dev = net_dev->dev.parent; @@ -3579,10 +3823,13 @@ static int bind_dpni(struct dpaa2_eth_priv *priv) for (i = 0; i < priv->num_fqs; i++) { switch (priv->fq[i].type) { case DPAA2_RX_FQ: - err = setup_rx_flow(priv, &priv->fq[i]); + err = dpaa2_eth_setup_rx_flow(priv, &priv->fq[i]); break; case DPAA2_TX_CONF_FQ: - err = setup_tx_flow(priv, &priv->fq[i]); + err = dpaa2_eth_setup_tx_flow(priv, &priv->fq[i]); + break; + case DPAA2_RX_ERR_FQ: + err = setup_rx_err_flow(priv, &priv->fq[i]); break; default: dev_err(dev, "Invalid FQ type %d\n", priv->fq[i].type); @@ -3603,7 +3850,7 @@ static int bind_dpni(struct dpaa2_eth_priv *priv) } /* Allocate rings for storing incoming frame descriptors */ -static int alloc_rings(struct dpaa2_eth_priv *priv) +static int dpaa2_eth_alloc_rings(struct dpaa2_eth_priv *priv) { struct net_device *net_dev = priv->net_dev; struct device *dev = net_dev->dev.parent; @@ -3630,7 +3877,7 @@ err_ring: return -ENOMEM; } -static void free_rings(struct dpaa2_eth_priv *priv) +static void dpaa2_eth_free_rings(struct dpaa2_eth_priv *priv) { int i; @@ -3638,7 +3885,7 @@ static void free_rings(struct dpaa2_eth_priv *priv) dpaa2_io_store_destroy(priv->channel[i]->store); } -static int set_mac_addr(struct dpaa2_eth_priv *priv) +static int dpaa2_eth_set_mac_addr(struct dpaa2_eth_priv *priv) { struct net_device *net_dev = priv->net_dev; struct device *dev = net_dev->dev.parent; @@ -3703,7 +3950,7 @@ static int set_mac_addr(struct dpaa2_eth_priv *priv) return 0; } -static int netdev_init(struct net_device *net_dev) +static int dpaa2_eth_netdev_init(struct net_device *net_dev) { struct device *dev = net_dev->dev.parent; struct dpaa2_eth_priv *priv = netdev_priv(net_dev); @@ -3716,7 +3963,7 @@ static int netdev_init(struct net_device *net_dev) net_dev->netdev_ops = &dpaa2_eth_ops; net_dev->ethtool_ops = &dpaa2_ethtool_ops; - err = set_mac_addr(priv); + err = dpaa2_eth_set_mac_addr(priv); if (err) return err; @@ -3771,13 +4018,13 @@ static int netdev_init(struct net_device *net_dev) return 0; } -static int poll_link_state(void *arg) +static int dpaa2_eth_poll_link_state(void *arg) { struct dpaa2_eth_priv *priv = (struct dpaa2_eth_priv *)arg; int err; while (!kthread_should_stop()) { - err = link_state_update(priv); + err = dpaa2_eth_link_state_update(priv); if (unlikely(err)) return err; @@ -3847,11 +4094,11 @@ static irqreturn_t dpni_irq0_handler_thread(int irq_num, void *arg) } if (status & DPNI_IRQ_EVENT_LINK_CHANGED) - link_state_update(netdev_priv(net_dev)); + dpaa2_eth_link_state_update(netdev_priv(net_dev)); if (status & DPNI_IRQ_EVENT_ENDPOINT_CHANGED) { - set_mac_addr(netdev_priv(net_dev)); - update_tx_fqids(priv); + dpaa2_eth_set_mac_addr(netdev_priv(net_dev)); + dpaa2_eth_update_tx_fqids(priv); rtnl_lock(); if (priv->mac) @@ -3864,7 +4111,7 @@ static irqreturn_t dpni_irq0_handler_thread(int irq_num, void *arg) return IRQ_HANDLED; } -static int setup_irqs(struct fsl_mc_device *ls_dev) +static int dpaa2_eth_setup_irqs(struct fsl_mc_device *ls_dev) { int err = 0; struct fsl_mc_device_irq *irq; @@ -3910,7 +4157,7 @@ free_mc_irq: return err; } -static void add_ch_napi(struct dpaa2_eth_priv *priv) +static void dpaa2_eth_add_ch_napi(struct dpaa2_eth_priv *priv) { int i; struct dpaa2_eth_channel *ch; @@ -3923,7 +4170,7 @@ static void add_ch_napi(struct dpaa2_eth_priv *priv) } } -static void del_ch_napi(struct dpaa2_eth_priv *priv) +static void dpaa2_eth_del_ch_napi(struct dpaa2_eth_priv *priv) { int i; struct dpaa2_eth_channel *ch; @@ -3958,6 +4205,19 @@ static int dpaa2_eth_probe(struct fsl_mc_device *dpni_dev) priv->iommu_domain = iommu_get_domain_for_dev(dev); + priv->tx_tstamp_type = HWTSTAMP_TX_OFF; + priv->rx_tstamp = false; + + priv->dpaa2_ptp_wq = alloc_workqueue("dpaa2_ptp_wq", 0, 0); + if (!priv->dpaa2_ptp_wq) { + err = -ENOMEM; + goto err_wq_alloc; + } + + INIT_WORK(&priv->tx_onestep_tstamp, dpaa2_eth_tx_onestep_tstamp); + + skb_queue_head_init(&priv->tx_skbs); + /* Obtain a MC portal */ err = fsl_mc_portal_allocate(dpni_dev, FSL_MC_IO_ATOMIC_CONTEXT_PORTAL, &priv->mc_io); @@ -3970,26 +4230,26 @@ static int dpaa2_eth_probe(struct fsl_mc_device *dpni_dev) } /* MC objects initialization and configuration */ - err = setup_dpni(dpni_dev); + err = dpaa2_eth_setup_dpni(dpni_dev); if (err) goto err_dpni_setup; - err = setup_dpio(priv); + err = dpaa2_eth_setup_dpio(priv); if (err) goto err_dpio_setup; - setup_fqs(priv); + dpaa2_eth_setup_fqs(priv); - err = setup_dpbp(priv); + err = dpaa2_eth_setup_dpbp(priv); if (err) goto err_dpbp_setup; - err = bind_dpni(priv); + err = dpaa2_eth_bind_dpni(priv); if (err) goto err_bind; /* Add a NAPI context for each channel */ - add_ch_napi(priv); + dpaa2_eth_add_ch_napi(priv); /* Percpu statistics */ priv->percpu_stats = alloc_percpu(*priv->percpu_stats); @@ -4012,21 +4272,21 @@ static int dpaa2_eth_probe(struct fsl_mc_device *dpni_dev) goto err_alloc_sgt_cache; } - err = netdev_init(net_dev); + err = dpaa2_eth_netdev_init(net_dev); if (err) goto err_netdev_init; /* Configure checksum offload based on current interface flags */ - err = set_rx_csum(priv, !!(net_dev->features & NETIF_F_RXCSUM)); + err = dpaa2_eth_set_rx_csum(priv, !!(net_dev->features & NETIF_F_RXCSUM)); if (err) goto err_csum; - err = set_tx_csum(priv, !!(net_dev->features & - (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM))); + err = dpaa2_eth_set_tx_csum(priv, + !!(net_dev->features & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM))); if (err) goto err_csum; - err = alloc_rings(priv); + err = dpaa2_eth_alloc_rings(priv); if (err) goto err_alloc_rings; @@ -4039,10 +4299,10 @@ static int dpaa2_eth_probe(struct fsl_mc_device *dpni_dev) } #endif - err = setup_irqs(dpni_dev); + err = dpaa2_eth_setup_irqs(dpni_dev); if (err) { netdev_warn(net_dev, "Failed to set link interrupt, fall back to polling\n"); - priv->poll_thread = kthread_run(poll_link_state, priv, + priv->poll_thread = kthread_run(dpaa2_eth_poll_link_state, priv, "%s_poll_link", net_dev->name); if (IS_ERR(priv->poll_thread)) { dev_err(dev, "Error starting polling thread\n"); @@ -4055,6 +4315,18 @@ static int dpaa2_eth_probe(struct fsl_mc_device *dpni_dev) if (err) goto err_connect_mac; + err = dpaa2_eth_dl_register(priv); + if (err) + goto err_dl_register; + + err = dpaa2_eth_dl_traps_register(priv); + if (err) + goto err_dl_trap_register; + + err = dpaa2_eth_dl_port_add(priv); + if (err) + goto err_dl_port_add; + err = register_netdev(net_dev); if (err < 0) { dev_err(dev, "register_netdev() failed\n"); @@ -4069,6 +4341,12 @@ static int dpaa2_eth_probe(struct fsl_mc_device *dpni_dev) return 0; err_netdev_reg: + dpaa2_eth_dl_port_del(priv); +err_dl_port_add: + dpaa2_eth_dl_traps_unregister(priv); +err_dl_trap_register: + dpaa2_eth_dl_unregister(priv); +err_dl_register: dpaa2_eth_disconnect_mac(priv); err_connect_mac: if (priv->do_link_poll) @@ -4076,7 +4354,7 @@ err_connect_mac: else fsl_mc_free_irqs(dpni_dev); err_poll_thread: - free_rings(priv); + dpaa2_eth_free_rings(priv); err_alloc_rings: err_csum: err_netdev_init: @@ -4086,16 +4364,18 @@ err_alloc_sgt_cache: err_alloc_percpu_extras: free_percpu(priv->percpu_stats); err_alloc_percpu_stats: - del_ch_napi(priv); + dpaa2_eth_del_ch_napi(priv); err_bind: - free_dpbp(priv); + dpaa2_eth_free_dpbp(priv); err_dpbp_setup: - free_dpio(priv); + dpaa2_eth_free_dpio(priv); err_dpio_setup: - free_dpni(priv); + dpaa2_eth_free_dpni(priv); err_dpni_setup: fsl_mc_portal_free(priv->mc_io); err_portal_alloc: + destroy_workqueue(priv->dpaa2_ptp_wq); +err_wq_alloc: dev_set_drvdata(dev, NULL); free_netdev(net_dev); @@ -4121,20 +4401,24 @@ static int dpaa2_eth_remove(struct fsl_mc_device *ls_dev) unregister_netdev(net_dev); + dpaa2_eth_dl_port_del(priv); + dpaa2_eth_dl_traps_unregister(priv); + dpaa2_eth_dl_unregister(priv); + if (priv->do_link_poll) kthread_stop(priv->poll_thread); else fsl_mc_free_irqs(ls_dev); - free_rings(priv); + dpaa2_eth_free_rings(priv); free_percpu(priv->sgt_cache); free_percpu(priv->percpu_stats); free_percpu(priv->percpu_extras); - del_ch_napi(priv); - free_dpbp(priv); - free_dpio(priv); - free_dpni(priv); + dpaa2_eth_del_ch_napi(priv); + dpaa2_eth_free_dpbp(priv); + dpaa2_eth_free_dpio(priv); + dpaa2_eth_free_dpni(priv); fsl_mc_portal_free(priv->mc_io); diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.h b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.h index 7f3c41dc98f2..d236b8695c39 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.h +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.h @@ -10,6 +10,8 @@ #include <linux/netdevice.h> #include <linux/if_vlan.h> #include <linux/fsl/mc.h> +#include <linux/net_tstamp.h> +#include <net/devlink.h> #include <soc/fsl/dpaa2-io.h> #include <soc/fsl/dpaa2-fd.h> @@ -180,6 +182,49 @@ struct dpaa2_fas { */ #define DPAA2_TS_OFFSET 0x8 +/* Frame annotation parse results */ +struct dpaa2_fapr { + /* 64-bit word 1 */ + __le32 faf_lo; + __le16 faf_ext; + __le16 nxt_hdr; + /* 64-bit word 2 */ + __le64 faf_hi; + /* 64-bit word 3 */ + u8 last_ethertype_offset; + u8 vlan_tci_offset_n; + u8 vlan_tci_offset_1; + u8 llc_snap_offset; + u8 eth_offset; + u8 ip1_pid_offset; + u8 shim_offset_2; + u8 shim_offset_1; + /* 64-bit word 4 */ + u8 l5_offset; + u8 l4_offset; + u8 gre_offset; + u8 l3_offset_n; + u8 l3_offset_1; + u8 mpls_offset_n; + u8 mpls_offset_1; + u8 pppoe_offset; + /* 64-bit word 5 */ + __le16 running_sum; + __le16 gross_running_sum; + u8 ipv6_frag_offset; + u8 nxt_hdr_offset; + u8 routing_hdr_offset_2; + u8 routing_hdr_offset_1; + /* 64-bit word 6 */ + u8 reserved[5]; /* Soft-parsing context */ + u8 ip_proto_offset_n; + u8 nxt_hdr_frag_offset; + u8 parse_error_code; +}; + +#define DPAA2_FAPR_OFFSET 0x10 +#define DPAA2_FAPR_SIZE sizeof((struct dpaa2_fapr)) + /* Frame annotation egress action descriptor */ #define DPAA2_FAEAD_OFFSET 0x58 @@ -194,6 +239,24 @@ struct dpaa2_faead { #define DPAA2_FAEAD_EBDDV 0x00002000 #define DPAA2_FAEAD_UPD 0x00000010 +struct ptp_tstamp { + u16 sec_msb; + u32 sec_lsb; + u32 nsec; +}; + +static inline void ns_to_ptp_tstamp(struct ptp_tstamp *tstamp, u64 ns) +{ + u64 sec, nsec; + + sec = ns; + nsec = do_div(sec, 1000000000); + + tstamp->sec_lsb = sec & 0xFFFFFFFF; + tstamp->sec_msb = (sec >> 32) & 0xFFFF; + tstamp->nsec = nsec; +} + /* Accessors for the hardware annotation fields that we use */ static inline void *dpaa2_get_hwa(void *buf_addr, bool swa) { @@ -210,6 +273,11 @@ static inline __le64 *dpaa2_get_ts(void *buf_addr, bool swa) return dpaa2_get_hwa(buf_addr, swa) + DPAA2_TS_OFFSET; } +static inline struct dpaa2_fapr *dpaa2_get_fapr(void *buf_addr, bool swa) +{ + return dpaa2_get_hwa(buf_addr, swa) + DPAA2_FAPR_OFFSET; +} + static inline struct dpaa2_faead *dpaa2_get_faead(void *buf_addr, bool swa) { return dpaa2_get_hwa(buf_addr, swa) + DPAA2_FAEAD_OFFSET; @@ -324,8 +392,10 @@ struct dpaa2_eth_ch_stats { #define DPAA2_ETH_MAX_RX_QUEUES \ (DPAA2_ETH_MAX_RX_QUEUES_PER_TC * DPAA2_ETH_MAX_TCS) #define DPAA2_ETH_MAX_TX_QUEUES 16 +#define DPAA2_ETH_MAX_RX_ERR_QUEUES 1 #define DPAA2_ETH_MAX_QUEUES (DPAA2_ETH_MAX_RX_QUEUES + \ - DPAA2_ETH_MAX_TX_QUEUES) + DPAA2_ETH_MAX_TX_QUEUES + \ + DPAA2_ETH_MAX_RX_ERR_QUEUES) #define DPAA2_ETH_MAX_NETDEV_QUEUES \ (DPAA2_ETH_MAX_TX_QUEUES * DPAA2_ETH_MAX_TCS) @@ -334,6 +404,7 @@ struct dpaa2_eth_ch_stats { enum dpaa2_eth_fq_type { DPAA2_RX_FQ = 0, DPAA2_TX_CONF_FQ, + DPAA2_RX_ERR_FQ }; struct dpaa2_eth_priv; @@ -407,6 +478,15 @@ struct dpaa2_eth_sgt_cache { u16 count; }; +struct dpaa2_eth_trap_item { + void *trap_ctx; +}; + +struct dpaa2_eth_trap_data { + struct dpaa2_eth_trap_item *trap_items_arr; + struct dpaa2_eth_priv *priv; +}; + /* Driver private data */ struct dpaa2_eth_priv { struct net_device *net_dev; @@ -433,8 +513,8 @@ struct dpaa2_eth_priv { u16 bpid; struct iommu_domain *iommu_domain; - bool tx_tstamp; /* Tx timestamping enabled */ - bool rx_tstamp; /* Rx timestamping enabled */ + enum hwtstamp_tx_types tx_tstamp_type; /* Tx timestamping type */ + bool rx_tstamp; /* Rx timestamping enabled */ u16 tx_qdid; struct fsl_mc_io *mc_io; @@ -473,8 +553,29 @@ struct dpaa2_eth_priv { #endif struct dpaa2_mac *mac; + struct workqueue_struct *dpaa2_ptp_wq; + struct work_struct tx_onestep_tstamp; + struct sk_buff_head tx_skbs; + /* The one-step timestamping configuration on hardware + * registers could only be done when no one-step + * timestamping frames are in flight. So we use a mutex + * lock here to make sure the lock is released by last + * one-step timestamping packet through TX confirmation + * queue before transmit current packet. + */ + struct mutex onestep_tstamp_lock; + struct devlink *devlink; + struct dpaa2_eth_trap_data *trap_data; + struct devlink_port devlink_port; }; +struct dpaa2_eth_devlink_priv { + struct dpaa2_eth_priv *dpaa2_priv; +}; + +#define TX_TSTAMP 0x1 +#define TX_TSTAMP_ONESTEP_SYNC 0x2 + #define DPAA2_RXH_SUPPORTED (RXH_L2DA | RXH_VLAN | RXH_L3_PROTO \ | RXH_IP_SRC | RXH_IP_DST | RXH_L4_B_0_1 \ | RXH_L4_B_2_3) @@ -491,6 +592,7 @@ struct dpaa2_eth_priv { extern const struct ethtool_ops dpaa2_ethtool_ops; extern int dpaa2_phc_index; +extern struct ptp_qoriq *dpaa2_ptp; static inline int dpaa2_eth_cmp_dpni_ver(struct dpaa2_eth_priv *priv, u16 ver_major, u16 ver_minor) @@ -560,9 +662,7 @@ static inline bool dpaa2_eth_rx_pause_enabled(u64 link_options) return !!(link_options & DPNI_LINK_OPT_PAUSE); } -static inline -unsigned int dpaa2_eth_needed_headroom(struct dpaa2_eth_priv *priv, - struct sk_buff *skb) +static inline unsigned int dpaa2_eth_needed_headroom(struct sk_buff *skb) { unsigned int headroom = DPAA2_ETH_SWA_SIZE; @@ -579,7 +679,7 @@ unsigned int dpaa2_eth_needed_headroom(struct dpaa2_eth_priv *priv, return 0; /* If we have Tx timestamping, need 128B hardware annotation */ - if (priv->tx_tstamp && skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) + if (skb->cb[0]) headroom += DPAA2_ETH_TX_HWA_SIZE; return headroom; @@ -604,4 +704,15 @@ void dpaa2_eth_set_rx_taildrop(struct dpaa2_eth_priv *priv, extern const struct dcbnl_rtnl_ops dpaa2_eth_dcbnl_ops; +int dpaa2_eth_dl_register(struct dpaa2_eth_priv *priv); +void dpaa2_eth_dl_unregister(struct dpaa2_eth_priv *priv); + +int dpaa2_eth_dl_port_add(struct dpaa2_eth_priv *priv); +void dpaa2_eth_dl_port_del(struct dpaa2_eth_priv *priv); + +int dpaa2_eth_dl_traps_register(struct dpaa2_eth_priv *priv); +void dpaa2_eth_dl_traps_unregister(struct dpaa2_eth_priv *priv); + +struct dpaa2_eth_trap_item *dpaa2_eth_dl_get_trap(struct dpaa2_eth_priv *priv, + struct dpaa2_fapr *fapr); #endif /* __DPAA2_H */ diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c index 8356f1fbbee1..f981a523e13a 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) /* Copyright 2014-2016 Freescale Semiconductor Inc. * Copyright 2016 NXP + * Copyright 2020 NXP */ #include <linux/net_tstamp.h> @@ -316,8 +317,8 @@ static void dpaa2_eth_get_ethtool_stats(struct net_device *net_dev, dpaa2_mac_get_ethtool_stats(priv->mac, data + i); } -static int prep_eth_rule(struct ethhdr *eth_value, struct ethhdr *eth_mask, - void *key, void *mask, u64 *fields) +static int dpaa2_eth_prep_eth_rule(struct ethhdr *eth_value, struct ethhdr *eth_mask, + void *key, void *mask, u64 *fields) { int off; @@ -345,9 +346,9 @@ static int prep_eth_rule(struct ethhdr *eth_value, struct ethhdr *eth_mask, return 0; } -static int prep_uip_rule(struct ethtool_usrip4_spec *uip_value, - struct ethtool_usrip4_spec *uip_mask, - void *key, void *mask, u64 *fields) +static int dpaa2_eth_prep_uip_rule(struct ethtool_usrip4_spec *uip_value, + struct ethtool_usrip4_spec *uip_mask, + void *key, void *mask, u64 *fields) { int off; u32 tmp_value, tmp_mask; @@ -400,9 +401,9 @@ static int prep_uip_rule(struct ethtool_usrip4_spec *uip_value, return 0; } -static int prep_l4_rule(struct ethtool_tcpip4_spec *l4_value, - struct ethtool_tcpip4_spec *l4_mask, - void *key, void *mask, u8 l4_proto, u64 *fields) +static int dpaa2_eth_prep_l4_rule(struct ethtool_tcpip4_spec *l4_value, + struct ethtool_tcpip4_spec *l4_mask, + void *key, void *mask, u8 l4_proto, u64 *fields) { int off; @@ -451,9 +452,9 @@ static int prep_l4_rule(struct ethtool_tcpip4_spec *l4_value, return 0; } -static int prep_ext_rule(struct ethtool_flow_ext *ext_value, - struct ethtool_flow_ext *ext_mask, - void *key, void *mask, u64 *fields) +static int dpaa2_eth_prep_ext_rule(struct ethtool_flow_ext *ext_value, + struct ethtool_flow_ext *ext_mask, + void *key, void *mask, u64 *fields) { int off; @@ -470,9 +471,9 @@ static int prep_ext_rule(struct ethtool_flow_ext *ext_value, return 0; } -static int prep_mac_ext_rule(struct ethtool_flow_ext *ext_value, - struct ethtool_flow_ext *ext_mask, - void *key, void *mask, u64 *fields) +static int dpaa2_eth_prep_mac_ext_rule(struct ethtool_flow_ext *ext_value, + struct ethtool_flow_ext *ext_mask, + void *key, void *mask, u64 *fields) { int off; @@ -486,32 +487,32 @@ static int prep_mac_ext_rule(struct ethtool_flow_ext *ext_value, return 0; } -static int prep_cls_rule(struct ethtool_rx_flow_spec *fs, void *key, void *mask, - u64 *fields) +static int dpaa2_eth_prep_cls_rule(struct ethtool_rx_flow_spec *fs, void *key, + void *mask, u64 *fields) { int err; switch (fs->flow_type & 0xFF) { case ETHER_FLOW: - err = prep_eth_rule(&fs->h_u.ether_spec, &fs->m_u.ether_spec, - key, mask, fields); + err = dpaa2_eth_prep_eth_rule(&fs->h_u.ether_spec, &fs->m_u.ether_spec, + key, mask, fields); break; case IP_USER_FLOW: - err = prep_uip_rule(&fs->h_u.usr_ip4_spec, - &fs->m_u.usr_ip4_spec, key, mask, fields); + err = dpaa2_eth_prep_uip_rule(&fs->h_u.usr_ip4_spec, + &fs->m_u.usr_ip4_spec, key, mask, fields); break; case TCP_V4_FLOW: - err = prep_l4_rule(&fs->h_u.tcp_ip4_spec, &fs->m_u.tcp_ip4_spec, - key, mask, IPPROTO_TCP, fields); + err = dpaa2_eth_prep_l4_rule(&fs->h_u.tcp_ip4_spec, &fs->m_u.tcp_ip4_spec, + key, mask, IPPROTO_TCP, fields); break; case UDP_V4_FLOW: - err = prep_l4_rule(&fs->h_u.udp_ip4_spec, &fs->m_u.udp_ip4_spec, - key, mask, IPPROTO_UDP, fields); + err = dpaa2_eth_prep_l4_rule(&fs->h_u.udp_ip4_spec, &fs->m_u.udp_ip4_spec, + key, mask, IPPROTO_UDP, fields); break; case SCTP_V4_FLOW: - err = prep_l4_rule(&fs->h_u.sctp_ip4_spec, - &fs->m_u.sctp_ip4_spec, key, mask, - IPPROTO_SCTP, fields); + err = dpaa2_eth_prep_l4_rule(&fs->h_u.sctp_ip4_spec, + &fs->m_u.sctp_ip4_spec, key, mask, + IPPROTO_SCTP, fields); break; default: return -EOPNOTSUPP; @@ -521,14 +522,14 @@ static int prep_cls_rule(struct ethtool_rx_flow_spec *fs, void *key, void *mask, return err; if (fs->flow_type & FLOW_EXT) { - err = prep_ext_rule(&fs->h_ext, &fs->m_ext, key, mask, fields); + err = dpaa2_eth_prep_ext_rule(&fs->h_ext, &fs->m_ext, key, mask, fields); if (err) return err; } if (fs->flow_type & FLOW_MAC_EXT) { - err = prep_mac_ext_rule(&fs->h_ext, &fs->m_ext, key, mask, - fields); + err = dpaa2_eth_prep_mac_ext_rule(&fs->h_ext, &fs->m_ext, key, + mask, fields); if (err) return err; } @@ -536,9 +537,9 @@ static int prep_cls_rule(struct ethtool_rx_flow_spec *fs, void *key, void *mask, return 0; } -static int do_cls_rule(struct net_device *net_dev, - struct ethtool_rx_flow_spec *fs, - bool add) +static int dpaa2_eth_do_cls_rule(struct net_device *net_dev, + struct ethtool_rx_flow_spec *fs, + bool add) { struct dpaa2_eth_priv *priv = netdev_priv(net_dev); struct device *dev = net_dev->dev.parent; @@ -561,7 +562,7 @@ static int do_cls_rule(struct net_device *net_dev, return -ENOMEM; /* Fill the key and mask memory areas */ - err = prep_cls_rule(fs, key_buf, key_buf + rule_cfg.key_size, &fields); + err = dpaa2_eth_prep_cls_rule(fs, key_buf, key_buf + rule_cfg.key_size, &fields); if (err) goto free_mem; @@ -617,7 +618,7 @@ static int do_cls_rule(struct net_device *net_dev, err = dpni_remove_fs_entry(priv->mc_io, 0, priv->mc_token, i, &rule_cfg); - if (err) + if (err || priv->dpni_attrs.options & DPNI_OPT_SHARED_FS) break; } @@ -629,7 +630,7 @@ free_mem: return err; } -static int num_rules(struct dpaa2_eth_priv *priv) +static int dpaa2_eth_num_cls_rules(struct dpaa2_eth_priv *priv) { int i, rules = 0; @@ -640,9 +641,9 @@ static int num_rules(struct dpaa2_eth_priv *priv) return rules; } -static int update_cls_rule(struct net_device *net_dev, - struct ethtool_rx_flow_spec *new_fs, - unsigned int location) +static int dpaa2_eth_update_cls_rule(struct net_device *net_dev, + struct ethtool_rx_flow_spec *new_fs, + unsigned int location) { struct dpaa2_eth_priv *priv = netdev_priv(net_dev); struct dpaa2_eth_cls_rule *rule; @@ -658,13 +659,14 @@ static int update_cls_rule(struct net_device *net_dev, /* If a rule is present at the specified location, delete it. */ if (rule->in_use) { - err = do_cls_rule(net_dev, &rule->fs, false); + err = dpaa2_eth_do_cls_rule(net_dev, &rule->fs, false); if (err) return err; rule->in_use = 0; - if (!dpaa2_eth_fs_mask_enabled(priv) && !num_rules(priv)) + if (!dpaa2_eth_fs_mask_enabled(priv) && + !dpaa2_eth_num_cls_rules(priv)) priv->rx_cls_fields = 0; } @@ -672,7 +674,7 @@ static int update_cls_rule(struct net_device *net_dev, if (!new_fs) return err; - err = do_cls_rule(net_dev, new_fs, true); + err = dpaa2_eth_do_cls_rule(net_dev, new_fs, true); if (err) return err; @@ -702,7 +704,7 @@ static int dpaa2_eth_get_rxnfc(struct net_device *net_dev, break; case ETHTOOL_GRXCLSRLCNT: rxnfc->rule_cnt = 0; - rxnfc->rule_cnt = num_rules(priv); + rxnfc->rule_cnt = dpaa2_eth_num_cls_rules(priv); rxnfc->data = max_rules; break; case ETHTOOL_GRXCLSRULE: @@ -744,10 +746,10 @@ static int dpaa2_eth_set_rxnfc(struct net_device *net_dev, err = dpaa2_eth_set_hash(net_dev, rxnfc->data); break; case ETHTOOL_SRXCLSRLINS: - err = update_cls_rule(net_dev, &rxnfc->fs, rxnfc->fs.location); + err = dpaa2_eth_update_cls_rule(net_dev, &rxnfc->fs, rxnfc->fs.location); break; case ETHTOOL_SRXCLSRLDEL: - err = update_cls_rule(net_dev, NULL, rxnfc->fs.location); + err = dpaa2_eth_update_cls_rule(net_dev, NULL, rxnfc->fs.location); break; default: err = -EOPNOTSUPP; @@ -762,6 +764,9 @@ EXPORT_SYMBOL(dpaa2_phc_index); static int dpaa2_eth_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info) { + if (!dpaa2_ptp) + return ethtool_op_get_ts_info(dev, info); + info->so_timestamping = SOF_TIMESTAMPING_TX_HARDWARE | SOF_TIMESTAMPING_RX_HARDWARE | SOF_TIMESTAMPING_RAW_HARDWARE; @@ -769,7 +774,8 @@ static int dpaa2_eth_get_ts_info(struct net_device *dev, info->phc_index = dpaa2_phc_index; info->tx_types = (1 << HWTSTAMP_TX_OFF) | - (1 << HWTSTAMP_TX_ON); + (1 << HWTSTAMP_TX_ON) | + (1 << HWTSTAMP_TX_ONESTEP_SYNC); info->rx_filters = (1 << HWTSTAMP_FILTER_NONE) | (1 << HWTSTAMP_FILTER_ALL); diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c index 3ee236c5fc37..90cd243070d7 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.c @@ -15,6 +15,18 @@ static int phy_mode(enum dpmac_eth_if eth_if, phy_interface_t *if_mode) case DPMAC_ETH_IF_RGMII: *if_mode = PHY_INTERFACE_MODE_RGMII; break; + case DPMAC_ETH_IF_USXGMII: + *if_mode = PHY_INTERFACE_MODE_USXGMII; + break; + case DPMAC_ETH_IF_QSGMII: + *if_mode = PHY_INTERFACE_MODE_QSGMII; + break; + case DPMAC_ETH_IF_SGMII: + *if_mode = PHY_INTERFACE_MODE_SGMII; + break; + case DPMAC_ETH_IF_XFI: + *if_mode = PHY_INTERFACE_MODE_10GBASER; + break; default: return -EINVAL; } @@ -67,6 +79,10 @@ static bool dpaa2_mac_phy_mode_mismatch(struct dpaa2_mac *mac, phy_interface_t interface) { switch (interface) { + case PHY_INTERFACE_MODE_10GBASER: + case PHY_INTERFACE_MODE_USXGMII: + case PHY_INTERFACE_MODE_QSGMII: + case PHY_INTERFACE_MODE_SGMII: case PHY_INTERFACE_MODE_RGMII: case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_RXID: @@ -95,6 +111,17 @@ static void dpaa2_mac_validate(struct phylink_config *config, phylink_set(mask, Asym_Pause); switch (state->interface) { + case PHY_INTERFACE_MODE_NA: + case PHY_INTERFACE_MODE_10GBASER: + case PHY_INTERFACE_MODE_USXGMII: + phylink_set(mask, 10000baseT_Full); + if (state->interface == PHY_INTERFACE_MODE_10GBASER) + break; + phylink_set(mask, 5000baseT_Full); + phylink_set(mask, 2500baseT_Full); + fallthrough; + case PHY_INTERFACE_MODE_SGMII: + case PHY_INTERFACE_MODE_QSGMII: case PHY_INTERFACE_MODE_RGMII: case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII_RXID: @@ -227,6 +254,51 @@ out: return fixed; } +static int dpaa2_pcs_create(struct dpaa2_mac *mac, + struct device_node *dpmac_node, int id) +{ + struct mdio_device *mdiodev; + struct device_node *node; + + node = of_parse_phandle(dpmac_node, "pcs-handle", 0); + if (!node) { + /* do not error out on old DTS files */ + netdev_warn(mac->net_dev, "pcs-handle node not found\n"); + return 0; + } + + if (!of_device_is_available(node)) { + netdev_err(mac->net_dev, "pcs-handle node not available\n"); + return -ENODEV; + } + + mdiodev = of_mdio_find_device(node); + of_node_put(node); + if (!mdiodev) + return -EPROBE_DEFER; + + mac->pcs = lynx_pcs_create(mdiodev); + if (!mac->pcs) { + netdev_err(mac->net_dev, "lynx_pcs_create() failed\n"); + put_device(&mdiodev->dev); + return -ENOMEM; + } + + return 0; +} + +static void dpaa2_pcs_destroy(struct dpaa2_mac *mac) +{ + struct lynx_pcs *pcs = mac->pcs; + + if (pcs) { + struct device *dev = &pcs->mdio->dev; + lynx_pcs_destroy(pcs); + put_device(dev); + mac->pcs = NULL; + } +} + int dpaa2_mac_connect(struct dpaa2_mac *mac) { struct fsl_mc_device *dpmac_dev = mac->mc_dev; @@ -278,6 +350,13 @@ int dpaa2_mac_connect(struct dpaa2_mac *mac) goto err_put_node; } + if (attr.link_type == DPMAC_LINK_TYPE_PHY && + attr.eth_if != DPMAC_ETH_IF_RGMII) { + err = dpaa2_pcs_create(mac, dpmac_node, attr.id); + if (err) + goto err_put_node; + } + mac->phylink_config.dev = &net_dev->dev; mac->phylink_config.type = PHYLINK_NETDEV; @@ -286,10 +365,13 @@ int dpaa2_mac_connect(struct dpaa2_mac *mac) &dpaa2_mac_phylink_ops); if (IS_ERR(phylink)) { err = PTR_ERR(phylink); - goto err_put_node; + goto err_pcs_destroy; } mac->phylink = phylink; + if (mac->pcs) + phylink_set_pcs(mac->phylink, &mac->pcs->pcs); + err = phylink_of_phy_connect(mac->phylink, dpmac_node, 0); if (err) { netdev_err(net_dev, "phylink_of_phy_connect() = %d\n", err); @@ -302,6 +384,8 @@ int dpaa2_mac_connect(struct dpaa2_mac *mac) err_phylink_destroy: phylink_destroy(mac->phylink); +err_pcs_destroy: + dpaa2_pcs_destroy(mac); err_put_node: of_node_put(dpmac_node); err_close_dpmac: @@ -316,6 +400,8 @@ void dpaa2_mac_disconnect(struct dpaa2_mac *mac) phylink_disconnect_phy(mac->phylink); phylink_destroy(mac->phylink); + dpaa2_pcs_destroy(mac); + dpmac_close(mac->mc_io, 0, mac->mc_dev->mc_handle); } diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.h b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.h index 2130d9c7d40e..955a52856210 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.h +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-mac.h @@ -7,6 +7,7 @@ #include <linux/of_mdio.h> #include <linux/of_net.h> #include <linux/phylink.h> +#include <linux/pcs-lynx.h> #include "dpmac.h" #include "dpmac-cmd.h" @@ -21,6 +22,7 @@ struct dpaa2_mac { struct phylink *phylink; phy_interface_t if_mode; enum dpmac_link_type if_link_type; + struct lynx_pcs *pcs; }; bool dpaa2_mac_is_type_fixed(struct fsl_mc_device *dpmac_dev, diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.c index cc1b7f85e433..32b5faa87bb8 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.c @@ -2,6 +2,7 @@ /* * Copyright 2013-2016 Freescale Semiconductor Inc. * Copyright 2016-2018 NXP + * Copyright 2020 NXP */ #include <linux/module.h> @@ -9,7 +10,6 @@ #include <linux/of_address.h> #include <linux/msi.h> #include <linux/fsl/mc.h> -#include <linux/fsl/ptp_qoriq.h> #include "dpaa2-ptp.h" @@ -201,6 +201,7 @@ static int dpaa2_ptp_probe(struct fsl_mc_device *mc_dev) goto err_free_threaded_irq; dpaa2_phc_index = ptp_qoriq->phc_index; + dpaa2_ptp = ptp_qoriq; dev_set_drvdata(dev, ptp_qoriq); return 0; diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.h b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.h index df2458a5e9ef..e1023538b4c3 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.h +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp.h @@ -1,14 +1,18 @@ /* SPDX-License-Identifier: GPL-2.0 */ /* * Copyright 2018 NXP + * Copyright 2020 NXP */ #ifndef __RTC_H #define __RTC_H +#include <linux/fsl/ptp_qoriq.h> + #include "dprtc.h" #include "dprtc-cmd.h" extern int dpaa2_phc_index; +extern struct ptp_qoriq *dpaa2_ptp; #endif diff --git a/drivers/net/ethernet/freescale/dpaa2/dpni-cmd.h b/drivers/net/ethernet/freescale/dpaa2/dpni-cmd.h index 3c06f5fb5759..90453dc7baef 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpni-cmd.h +++ b/drivers/net/ethernet/freescale/dpaa2/dpni-cmd.h @@ -1,6 +1,7 @@ /* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */ /* Copyright 2013-2016 Freescale Semiconductor Inc. * Copyright 2016 NXP + * Copyright 2020 NXP */ #ifndef _FSL_DPNI_CMD_H #define _FSL_DPNI_CMD_H @@ -92,6 +93,9 @@ #define DPNI_CMDID_SET_RX_HASH_DIST DPNI_CMD(0x274) #define DPNI_CMDID_GET_LINK_CFG DPNI_CMD(0x278) +#define DPNI_CMDID_SET_SINGLE_STEP_CFG DPNI_CMD(0x279) +#define DPNI_CMDID_GET_SINGLE_STEP_CFG DPNI_CMD(0x27a) + /* Macros for accessing command fields smaller than 1byte */ #define DPNI_MASK(field) \ GENMASK(DPNI_##field##_SHIFT + DPNI_##field##_SIZE - 1, \ @@ -641,4 +645,21 @@ struct dpni_cmd_set_tx_shaping { u8 coupled; }; +#define DPNI_PTP_ENABLE_SHIFT 0 +#define DPNI_PTP_ENABLE_SIZE 1 +#define DPNI_PTP_CH_UPDATE_SHIFT 1 +#define DPNI_PTP_CH_UPDATE_SIZE 1 + +struct dpni_cmd_single_step_cfg { + __le16 flags; + __le16 offset; + __le32 peer_delay; +}; + +struct dpni_rsp_single_step_cfg { + __le16 flags; + __le16 offset; + __le32 peer_delay; +}; + #endif /* _FSL_DPNI_CMD_H */ diff --git a/drivers/net/ethernet/freescale/dpaa2/dpni.c b/drivers/net/ethernet/freescale/dpaa2/dpni.c index 68ed4c41b282..6ea7db66a632 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpni.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpni.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) /* Copyright 2013-2016 Freescale Semiconductor Inc. * Copyright 2016 NXP + * Copyright 2020 NXP */ #include <linux/kernel.h> #include <linux/errno.h> @@ -1999,3 +2000,81 @@ int dpni_set_tx_shaping(struct fsl_mc_io *mc_io, /* send command to mc*/ return mc_send_command(mc_io, &cmd); } + +/** + * dpni_get_single_step_cfg() - return current configuration for + * single step PTP + * @mc_io: Pointer to MC portal's I/O object + * @cmd_flags: Command flags; one or more of 'MC_CMD_FLAG_' + * @token: Token of DPNI object + * @ptp_cfg: ptp single step configuration + * + * Return: '0' on Success; Error code otherwise. + * + */ +int dpni_get_single_step_cfg(struct fsl_mc_io *mc_io, + u32 cmd_flags, + u16 token, + struct dpni_single_step_cfg *ptp_cfg) +{ + struct dpni_rsp_single_step_cfg *rsp_params; + struct fsl_mc_command cmd = { 0 }; + int err; + + /* prepare command */ + cmd.header = mc_encode_cmd_header(DPNI_CMDID_GET_SINGLE_STEP_CFG, + cmd_flags, token); + /* send command to mc*/ + err = mc_send_command(mc_io, &cmd); + if (err) + return err; + + /* read command response */ + rsp_params = (struct dpni_rsp_single_step_cfg *)cmd.params; + ptp_cfg->offset = le16_to_cpu(rsp_params->offset); + ptp_cfg->en = dpni_get_field(le16_to_cpu(rsp_params->flags), + PTP_ENABLE) ? 1 : 0; + ptp_cfg->ch_update = dpni_get_field(le16_to_cpu(rsp_params->flags), + PTP_CH_UPDATE) ? 1 : 0; + ptp_cfg->peer_delay = le32_to_cpu(rsp_params->peer_delay); + + return err; +} + +/** + * dpni_set_single_step_cfg() - enable/disable and configure single step PTP + * @mc_io: Pointer to MC portal's I/O object + * @cmd_flags: Command flags; one or more of 'MC_CMD_FLAG_' + * @token: Token of DPNI object + * @ptp_cfg: ptp single step configuration + * + * Return: '0' on Success; Error code otherwise. + * + * The function has effect only when dpni object is connected to a dpmac + * object. If the dpni is not connected to a dpmac the configuration will + * be stored inside and applied when connection is made. + */ +int dpni_set_single_step_cfg(struct fsl_mc_io *mc_io, + u32 cmd_flags, + u16 token, + struct dpni_single_step_cfg *ptp_cfg) +{ + struct dpni_cmd_single_step_cfg *cmd_params; + struct fsl_mc_command cmd = { 0 }; + u16 flags; + + /* prepare command */ + cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_SINGLE_STEP_CFG, + cmd_flags, token); + cmd_params = (struct dpni_cmd_single_step_cfg *)cmd.params; + cmd_params->offset = cpu_to_le16(ptp_cfg->offset); + cmd_params->peer_delay = cpu_to_le32(ptp_cfg->peer_delay); + + flags = le16_to_cpu(cmd_params->flags); + dpni_set_field(flags, PTP_ENABLE, !!ptp_cfg->en); + dpni_set_field(flags, PTP_CH_UPDATE, !!ptp_cfg->ch_update); + cmd_params->flags = cpu_to_le16(flags); + + /* send command to mc*/ + return mc_send_command(mc_io, &cmd); +} diff --git a/drivers/net/ethernet/freescale/dpaa2/dpni.h b/drivers/net/ethernet/freescale/dpaa2/dpni.h index 39387991a1f9..e7b9e195b534 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpni.h +++ b/drivers/net/ethernet/freescale/dpaa2/dpni.h @@ -1,6 +1,7 @@ /* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */ /* Copyright 2013-2016 Freescale Semiconductor Inc. * Copyright 2016 NXP + * Copyright 2020 NXP */ #ifndef __FSL_DPNI_H #define __FSL_DPNI_H @@ -74,6 +75,10 @@ struct fsl_mc_io; * Disables the flow steering table. */ #define DPNI_OPT_NO_FS 0x000020 +/** + * Flow steering table is shared between all traffic classes + */ +#define DPNI_OPT_SHARED_FS 0x001000 int dpni_open(struct fsl_mc_io *mc_io, u32 cmd_flags, @@ -1079,4 +1084,34 @@ int dpni_set_tx_shaping(struct fsl_mc_io *mc_io, const struct dpni_tx_shaping_cfg *tx_er_shaper, int coupled); +/** + * struct dpni_single_step_cfg - configure single step PTP (IEEE 1588) + * @en: enable single step PTP. When enabled the PTPv1 functionality + * will not work. If the field is zero, offset and ch_update + * parameters will be ignored + * @offset: start offset from the beginning of the frame where + * timestamp field is found. The offset must respect all MAC + * headers, VLAN tags and other protocol headers + * @ch_update: when set UDP checksum will be updated inside packet + * @peer_delay: For peer-to-peer transparent clocks add this value to the + * correction field in addition to the transient time update. + * The value expresses nanoseconds. + */ +struct dpni_single_step_cfg { + u8 en; + u8 ch_update; + u16 offset; + u32 peer_delay; +}; + +int dpni_set_single_step_cfg(struct fsl_mc_io *mc_io, + u32 cmd_flags, + u16 token, + struct dpni_single_step_cfg *ptp_cfg); + +int dpni_get_single_step_cfg(struct fsl_mc_io *mc_io, + u32 cmd_flags, + u16 token, + struct dpni_single_step_cfg *ptp_cfg); + #endif /* __FSL_DPNI_H */ diff --git a/drivers/net/ethernet/freescale/enetc/Kconfig b/drivers/net/ethernet/freescale/enetc/Kconfig index 37b804f8bd76..0fa18b00c49b 100644 --- a/drivers/net/ethernet/freescale/enetc/Kconfig +++ b/drivers/net/ethernet/freescale/enetc/Kconfig @@ -3,7 +3,8 @@ config FSL_ENETC tristate "ENETC PF driver" depends on PCI && PCI_MSI select FSL_ENETC_MDIO - select PHYLIB + select PHYLINK + select PCS_LYNX select DIMLIB help This driver supports NXP ENETC gigabit ethernet controller PCIe @@ -15,7 +16,7 @@ config FSL_ENETC config FSL_ENETC_VF tristate "ENETC VF driver" depends on PCI && PCI_MSI - select PHYLIB + select PHYLINK select DIMLIB help This driver supports NXP ENETC gigabit ethernet controller PCIe diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c index f78ca7b343d2..52be6e315752 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc.c +++ b/drivers/net/ethernet/freescale/enetc/enetc.c @@ -4,7 +4,6 @@ #include "enetc.h" #include <linux/tcp.h> #include <linux/udp.h> -#include <linux/of_mdio.h> #include <linux/vmalloc.h> /* ENETC overhead: optional extension BD + 1 BD gap */ @@ -1392,38 +1391,24 @@ static void enetc_clear_interrupts(struct enetc_ndev_priv *priv) enetc_rxbdr_wr(&priv->si->hw, i, ENETC_RBIER, 0); } -static void adjust_link(struct net_device *ndev) +static int enetc_phylink_connect(struct net_device *ndev) { struct enetc_ndev_priv *priv = netdev_priv(ndev); - struct phy_device *phydev = ndev->phydev; - - if (priv->active_offloads & ENETC_F_QBV) - enetc_sched_speed_set(ndev); - - phy_print_status(phydev); -} - -static int enetc_phy_connect(struct net_device *ndev) -{ - struct enetc_ndev_priv *priv = netdev_priv(ndev); - struct phy_device *phydev; struct ethtool_eee edata; + int err; - if (!priv->phy_node) + if (!priv->phylink) return 0; /* phy-less mode */ - phydev = of_phy_connect(ndev, priv->phy_node, &adjust_link, - 0, priv->if_mode); - if (!phydev) { + err = phylink_of_phy_connect(priv->phylink, priv->dev->of_node, 0); + if (err) { dev_err(&ndev->dev, "could not attach to PHY\n"); - return -ENODEV; + return err; } - phy_attached_info(phydev); - /* disable EEE autoneg, until ENETC driver supports it */ memset(&edata, 0, sizeof(struct ethtool_eee)); - phy_ethtool_set_eee(phydev, &edata); + phylink_ethtool_set_eee(priv->phylink, &edata); return 0; } @@ -1443,8 +1428,8 @@ void enetc_start(struct net_device *ndev) enable_irq(irq); } - if (ndev->phydev) - phy_start(ndev->phydev); + if (priv->phylink) + phylink_start(priv->phylink); else netif_carrier_on(ndev); @@ -1460,7 +1445,7 @@ int enetc_open(struct net_device *ndev) if (err) return err; - err = enetc_phy_connect(ndev); + err = enetc_phylink_connect(ndev); if (err) goto err_phy_connect; @@ -1490,8 +1475,8 @@ err_set_queues: err_alloc_rx: enetc_free_tx_resources(priv); err_alloc_tx: - if (ndev->phydev) - phy_disconnect(ndev->phydev); + if (priv->phylink) + phylink_disconnect_phy(priv->phylink); err_phy_connect: enetc_free_irqs(priv); @@ -1514,8 +1499,8 @@ void enetc_stop(struct net_device *ndev) napi_disable(&priv->int_vector[i]->napi); } - if (ndev->phydev) - phy_stop(ndev->phydev); + if (priv->phylink) + phylink_stop(priv->phylink); else netif_carrier_off(ndev); @@ -1529,8 +1514,8 @@ int enetc_close(struct net_device *ndev) enetc_stop(ndev); enetc_clear_bdrs(priv); - if (ndev->phydev) - phy_disconnect(ndev->phydev); + if (priv->phylink) + phylink_disconnect_phy(priv->phylink); enetc_free_rxtx_rings(priv); enetc_free_rx_resources(priv); enetc_free_tx_resources(priv); @@ -1780,6 +1765,7 @@ static int enetc_hwtstamp_get(struct net_device *ndev, struct ifreq *ifr) int enetc_ioctl(struct net_device *ndev, struct ifreq *rq, int cmd) { + struct enetc_ndev_priv *priv = netdev_priv(ndev); #ifdef CONFIG_FSL_ENETC_PTP_CLOCK if (cmd == SIOCSHWTSTAMP) return enetc_hwtstamp_set(ndev, rq); @@ -1787,9 +1773,10 @@ int enetc_ioctl(struct net_device *ndev, struct ifreq *rq, int cmd) return enetc_hwtstamp_get(ndev, rq); #endif - if (!ndev->phydev) + if (!priv->phylink) return -EOPNOTSUPP; - return phy_mii_ioctl(ndev->phydev, rq, cmd); + + return phylink_mii_ioctl(priv->phylink, rq, cmd); } int enetc_alloc_msix(struct enetc_ndev_priv *priv) diff --git a/drivers/net/ethernet/freescale/enetc/enetc.h b/drivers/net/ethernet/freescale/enetc/enetc.h index d309803cfeb6..dd0fb0c066d7 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc.h +++ b/drivers/net/ethernet/freescale/enetc/enetc.h @@ -9,7 +9,7 @@ #include <linux/skbuff.h> #include <linux/ethtool.h> #include <linux/if_vlan.h> -#include <linux/phy.h> +#include <linux/phylink.h> #include <linux/dim.h> #include "enetc_hw.h" @@ -264,8 +264,7 @@ struct enetc_ndev_priv { struct psfp_cap psfp_cap; - struct device_node *phy_node; - phy_interface_t if_mode; + struct phylink *phylink; int ic_mode; u32 tx_ictt; }; @@ -323,7 +322,7 @@ int enetc_send_cmd(struct enetc_si *si, struct enetc_cbd *cbd); #ifdef CONFIG_FSL_ENETC_QOS int enetc_setup_tc_taprio(struct net_device *ndev, void *type_data); -void enetc_sched_speed_set(struct net_device *ndev); +void enetc_sched_speed_set(struct enetc_ndev_priv *priv, int speed); int enetc_setup_tc_cbs(struct net_device *ndev, void *type_data); int enetc_setup_tc_txtime(struct net_device *ndev, void *type_data); int enetc_setup_tc_block_cb(enum tc_setup_type type, void *type_data, @@ -388,7 +387,7 @@ static inline int enetc_psfp_disable(struct enetc_ndev_priv *priv) #else #define enetc_setup_tc_taprio(ndev, type_data) -EOPNOTSUPP -#define enetc_sched_speed_set(ndev) (void)0 +#define enetc_sched_speed_set(priv, speed) (void)0 #define enetc_setup_tc_cbs(ndev, type_data) -EOPNOTSUPP #define enetc_setup_tc_txtime(ndev, type_data) -EOPNOTSUPP #define enetc_setup_tc_psfp(ndev, type_data) -EOPNOTSUPP diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c index 1dab83fbca77..8ed1ebd5a183 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c @@ -686,6 +686,28 @@ static int enetc_set_wol(struct net_device *dev, return ret; } +static int enetc_get_link_ksettings(struct net_device *dev, + struct ethtool_link_ksettings *cmd) +{ + struct enetc_ndev_priv *priv = netdev_priv(dev); + + if (!priv->phylink) + return -EOPNOTSUPP; + + return phylink_ethtool_ksettings_get(priv->phylink, cmd); +} + +static int enetc_set_link_ksettings(struct net_device *dev, + const struct ethtool_link_ksettings *cmd) +{ + struct enetc_ndev_priv *priv = netdev_priv(dev); + + if (!priv->phylink) + return -EOPNOTSUPP; + + return phylink_ethtool_ksettings_set(priv->phylink, cmd); +} + static const struct ethtool_ops enetc_pf_ethtool_ops = { .supported_coalesce_params = ETHTOOL_COALESCE_USECS | ETHTOOL_COALESCE_MAX_FRAMES | @@ -704,8 +726,8 @@ static const struct ethtool_ops enetc_pf_ethtool_ops = { .get_ringparam = enetc_get_ringparam, .get_coalesce = enetc_get_coalesce, .set_coalesce = enetc_set_coalesce, - .get_link_ksettings = phy_ethtool_get_link_ksettings, - .set_link_ksettings = phy_ethtool_set_link_ksettings, + .get_link_ksettings = enetc_get_link_ksettings, + .set_link_ksettings = enetc_set_link_ksettings, .get_link = ethtool_op_get_link, .get_ts_info = enetc_get_ts_info, .get_wol = enetc_get_wol, diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf.c b/drivers/net/ethernet/freescale/enetc/enetc_pf.c index 177334f0adb1..419306342ac5 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_pf.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_pf.c @@ -482,8 +482,7 @@ static void enetc_port_si_configure(struct enetc_si *si) enetc_port_wr(hw, ENETC_PSIVLANFMR, ENETC_PSIVLANFMR_VS); } -static void enetc_configure_port_mac(struct enetc_hw *hw, - phy_interface_t phy_mode) +static void enetc_configure_port_mac(struct enetc_hw *hw) { enetc_port_wr(hw, ENETC_PM0_MAXFRM, ENETC_SET_MAXFRM(ENETC_RX_MAXFRM_SIZE)); @@ -492,12 +491,14 @@ static void enetc_configure_port_mac(struct enetc_hw *hw, enetc_port_wr(hw, ENETC_PTXMBAR, 2 * ENETC_MAC_MAXFRM_SIZE); enetc_port_wr(hw, ENETC_PM0_CMD_CFG, ENETC_PM0_CMD_PHY_TX_EN | - ENETC_PM0_CMD_TXP | ENETC_PM0_PROMISC | - ENETC_PM0_TX_EN | ENETC_PM0_RX_EN); + ENETC_PM0_CMD_TXP | ENETC_PM0_PROMISC); enetc_port_wr(hw, ENETC_PM1_CMD_CFG, ENETC_PM0_CMD_PHY_TX_EN | - ENETC_PM0_CMD_TXP | ENETC_PM0_PROMISC | - ENETC_PM0_TX_EN | ENETC_PM0_RX_EN); + ENETC_PM0_CMD_TXP | ENETC_PM0_PROMISC); +} + +static void enetc_mac_config(struct enetc_hw *hw, phy_interface_t phy_mode) +{ /* set auto-speed for RGMII */ if (enetc_port_rd(hw, ENETC_PM0_IF_MODE) & ENETC_PMO_IFM_RG || phy_interface_mode_is_rgmii(phy_mode)) @@ -507,6 +508,17 @@ static void enetc_configure_port_mac(struct enetc_hw *hw, enetc_port_wr(hw, ENETC_PM0_IF_MODE, ENETC_PM0_IFM_XGMII); } +static void enetc_mac_enable(struct enetc_hw *hw, bool en) +{ + u32 val = enetc_port_rd(hw, ENETC_PM0_CMD_CFG); + + val &= ~(ENETC_PM0_TX_EN | ENETC_PM0_RX_EN); + val |= en ? (ENETC_PM0_TX_EN | ENETC_PM0_RX_EN) : 0; + + enetc_port_wr(hw, ENETC_PM0_CMD_CFG, val); + enetc_port_wr(hw, ENETC_PM1_CMD_CFG, val); +} + static void enetc_configure_port_pmac(struct enetc_hw *hw) { u32 temp; @@ -527,7 +539,7 @@ static void enetc_configure_port(struct enetc_pf *pf) enetc_configure_port_pmac(hw); - enetc_configure_port_mac(hw, pf->if_mode); + enetc_configure_port_mac(hw); enetc_port_si_configure(pf->si); @@ -733,11 +745,10 @@ static void enetc_pf_netdev_setup(struct enetc_si *si, struct net_device *ndev, enetc_get_primary_mac_addr(&si->hw, ndev->dev_addr); } -static int enetc_mdio_probe(struct enetc_pf *pf) +static int enetc_mdio_probe(struct enetc_pf *pf, struct device_node *np) { struct device *dev = &pf->si->pdev->dev; struct enetc_mdio_priv *mdio_priv; - struct device_node *np; struct mii_bus *bus; int err; @@ -754,20 +765,12 @@ static int enetc_mdio_probe(struct enetc_pf *pf) mdio_priv->mdio_base = ENETC_EMDIO_BASE; snprintf(bus->id, MII_BUS_ID_SIZE, "%s", dev_name(dev)); - np = of_get_child_by_name(dev->of_node, "mdio"); - if (!np) { - dev_err(dev, "MDIO node missing\n"); - return -EINVAL; - } - err = of_mdiobus_register(bus, np); if (err) { - of_node_put(np); dev_err(dev, "cannot register MDIO bus\n"); return err; } - of_node_put(np); pf->mdio = bus; return 0; @@ -779,69 +782,12 @@ static void enetc_mdio_remove(struct enetc_pf *pf) mdiobus_unregister(pf->mdio); } -static int enetc_of_get_phy(struct enetc_pf *pf) -{ - struct device *dev = &pf->si->pdev->dev; - struct device_node *np = dev->of_node; - struct device_node *mdio_np; - int err; - - pf->phy_node = of_parse_phandle(np, "phy-handle", 0); - if (!pf->phy_node) { - if (!of_phy_is_fixed_link(np)) { - dev_err(dev, "PHY not specified\n"); - return -ENODEV; - } - - err = of_phy_register_fixed_link(np); - if (err < 0) { - dev_err(dev, "fixed link registration failed\n"); - return err; - } - - pf->phy_node = of_node_get(np); - } - - mdio_np = of_get_child_by_name(np, "mdio"); - if (mdio_np) { - of_node_put(mdio_np); - err = enetc_mdio_probe(pf); - if (err) { - of_node_put(pf->phy_node); - return err; - } - } - - err = of_get_phy_mode(np, &pf->if_mode); - if (err) { - dev_err(dev, "missing phy type\n"); - of_node_put(pf->phy_node); - if (of_phy_is_fixed_link(np)) - of_phy_deregister_fixed_link(np); - else - enetc_mdio_remove(pf); - - return -EINVAL; - } - - return 0; -} - -static void enetc_of_put_phy(struct enetc_pf *pf) -{ - struct device_node *np = pf->si->pdev->dev.of_node; - - if (np && of_phy_is_fixed_link(np)) - of_phy_deregister_fixed_link(np); - if (pf->phy_node) - of_node_put(pf->phy_node); -} - -static int enetc_imdio_init(struct enetc_pf *pf, bool is_c45) +static int enetc_imdio_create(struct enetc_pf *pf) { struct device *dev = &pf->si->pdev->dev; struct enetc_mdio_priv *mdio_priv; - struct phy_device *pcs; + struct lynx_pcs *pcs_lynx; + struct mdio_device *pcs; struct mii_bus *bus; int err; @@ -865,15 +811,23 @@ static int enetc_imdio_init(struct enetc_pf *pf, bool is_c45) goto free_mdio_bus; } - pcs = get_phy_device(bus, 0, is_c45); + pcs = mdio_device_create(bus, 0); if (IS_ERR(pcs)) { err = PTR_ERR(pcs); - dev_err(dev, "cannot get internal PCS PHY (%d)\n", err); + dev_err(dev, "cannot create pcs (%d)\n", err); + goto unregister_mdiobus; + } + + pcs_lynx = lynx_pcs_create(pcs); + if (!pcs_lynx) { + mdio_device_free(pcs); + err = -ENOMEM; + dev_err(dev, "cannot create lynx pcs (%d)\n", err); goto unregister_mdiobus; } pf->imdio = bus; - pf->pcs = pcs; + pf->pcs = pcs_lynx; return 0; @@ -886,91 +840,168 @@ free_mdio_bus: static void enetc_imdio_remove(struct enetc_pf *pf) { - if (pf->pcs) - put_device(&pf->pcs->mdio.dev); + if (pf->pcs) { + mdio_device_free(pf->pcs->mdio); + lynx_pcs_destroy(pf->pcs); + } if (pf->imdio) { mdiobus_unregister(pf->imdio); mdiobus_free(pf->imdio); } } -static void enetc_configure_sgmii(struct phy_device *pcs) +static bool enetc_port_has_pcs(struct enetc_pf *pf) +{ + return (pf->if_mode == PHY_INTERFACE_MODE_SGMII || + pf->if_mode == PHY_INTERFACE_MODE_2500BASEX || + pf->if_mode == PHY_INTERFACE_MODE_USXGMII); +} + +static int enetc_mdiobus_create(struct enetc_pf *pf) +{ + struct device *dev = &pf->si->pdev->dev; + struct device_node *mdio_np; + int err; + + mdio_np = of_get_child_by_name(dev->of_node, "mdio"); + if (mdio_np) { + err = enetc_mdio_probe(pf, mdio_np); + + of_node_put(mdio_np); + if (err) + return err; + } + + if (enetc_port_has_pcs(pf)) { + err = enetc_imdio_create(pf); + if (err) { + enetc_mdio_remove(pf); + return err; + } + } + + return 0; +} + +static void enetc_mdiobus_destroy(struct enetc_pf *pf) { - /* SGMII spec requires tx_config_Reg[15:0] to be exactly 0x4001 - * for the MAC PCS in order to acknowledge the AN. - */ - phy_write(pcs, MII_ADVERTISE, ADVERTISE_SGMII | ADVERTISE_LPACK); + enetc_mdio_remove(pf); + enetc_imdio_remove(pf); +} - phy_write(pcs, ENETC_PCS_IF_MODE, - ENETC_PCS_IF_MODE_SGMII_EN | - ENETC_PCS_IF_MODE_USE_SGMII_AN); +static void enetc_pl_mac_validate(struct phylink_config *config, + unsigned long *supported, + struct phylink_link_state *state) +{ + __ETHTOOL_DECLARE_LINK_MODE_MASK(mask) = { 0, }; + + if (state->interface != PHY_INTERFACE_MODE_NA && + state->interface != PHY_INTERFACE_MODE_INTERNAL && + state->interface != PHY_INTERFACE_MODE_SGMII && + state->interface != PHY_INTERFACE_MODE_2500BASEX && + state->interface != PHY_INTERFACE_MODE_USXGMII && + !phy_interface_mode_is_rgmii(state->interface)) { + bitmap_zero(supported, __ETHTOOL_LINK_MODE_MASK_NBITS); + return; + } - /* Adjust link timer for SGMII */ - phy_write(pcs, ENETC_PCS_LINK_TIMER1, ENETC_PCS_LINK_TIMER1_VAL); - phy_write(pcs, ENETC_PCS_LINK_TIMER2, ENETC_PCS_LINK_TIMER2_VAL); + phylink_set_port_modes(mask); + phylink_set(mask, Autoneg); + phylink_set(mask, Pause); + phylink_set(mask, Asym_Pause); + phylink_set(mask, 10baseT_Half); + phylink_set(mask, 10baseT_Full); + phylink_set(mask, 100baseT_Half); + phylink_set(mask, 100baseT_Full); + phylink_set(mask, 100baseT_Half); + phylink_set(mask, 1000baseT_Half); + phylink_set(mask, 1000baseT_Full); + + if (state->interface == PHY_INTERFACE_MODE_INTERNAL || + state->interface == PHY_INTERFACE_MODE_2500BASEX || + state->interface == PHY_INTERFACE_MODE_USXGMII) { + phylink_set(mask, 2500baseT_Full); + phylink_set(mask, 2500baseX_Full); + } - phy_write(pcs, MII_BMCR, BMCR_ANRESTART | BMCR_ANENABLE); + bitmap_and(supported, supported, mask, + __ETHTOOL_LINK_MODE_MASK_NBITS); + bitmap_and(state->advertising, state->advertising, mask, + __ETHTOOL_LINK_MODE_MASK_NBITS); } -static void enetc_configure_2500basex(struct phy_device *pcs) +static void enetc_pl_mac_config(struct phylink_config *config, + unsigned int mode, + const struct phylink_link_state *state) { - phy_write(pcs, ENETC_PCS_IF_MODE, - ENETC_PCS_IF_MODE_SGMII_EN | - ENETC_PCS_IF_MODE_SGMII_SPEED(ENETC_PCS_SPEED_2500)); + struct enetc_pf *pf = phylink_to_enetc_pf(config); + struct enetc_ndev_priv *priv; + + enetc_mac_config(&pf->si->hw, state->interface); - phy_write(pcs, MII_BMCR, BMCR_SPEED1000 | BMCR_FULLDPLX | BMCR_RESET); + priv = netdev_priv(pf->si->ndev); + if (pf->pcs) + phylink_set_pcs(priv->phylink, &pf->pcs->pcs); } -static void enetc_configure_usxgmii(struct phy_device *pcs) +static void enetc_pl_mac_link_up(struct phylink_config *config, + struct phy_device *phy, unsigned int mode, + phy_interface_t interface, int speed, + int duplex, bool tx_pause, bool rx_pause) { - /* Configure device ability for the USXGMII Replicator */ - phy_write_mmd(pcs, MDIO_MMD_VEND2, MII_ADVERTISE, - ADVERTISE_SGMII | ADVERTISE_LPACK | - MDIO_USXGMII_FULL_DUPLEX); - - /* Restart PCS AN */ - phy_write_mmd(pcs, MDIO_MMD_VEND2, MII_BMCR, - BMCR_RESET | BMCR_ANENABLE | BMCR_ANRESTART); + struct enetc_pf *pf = phylink_to_enetc_pf(config); + struct enetc_ndev_priv *priv; + + priv = netdev_priv(pf->si->ndev); + if (priv->active_offloads & ENETC_F_QBV) + enetc_sched_speed_set(priv, speed); + + enetc_mac_enable(&pf->si->hw, true); } -static int enetc_configure_serdes(struct enetc_ndev_priv *priv) +static void enetc_pl_mac_link_down(struct phylink_config *config, + unsigned int mode, + phy_interface_t interface) +{ + struct enetc_pf *pf = phylink_to_enetc_pf(config); + + enetc_mac_enable(&pf->si->hw, false); +} + +static const struct phylink_mac_ops enetc_mac_phylink_ops = { + .validate = enetc_pl_mac_validate, + .mac_config = enetc_pl_mac_config, + .mac_link_up = enetc_pl_mac_link_up, + .mac_link_down = enetc_pl_mac_link_down, +}; + +static int enetc_phylink_create(struct enetc_ndev_priv *priv) { - bool is_c45 = priv->if_mode == PHY_INTERFACE_MODE_USXGMII; struct enetc_pf *pf = enetc_si_priv(priv->si); + struct device *dev = &pf->si->pdev->dev; + struct phylink *phylink; int err; - if (priv->if_mode != PHY_INTERFACE_MODE_SGMII && - priv->if_mode != PHY_INTERFACE_MODE_2500BASEX && - priv->if_mode != PHY_INTERFACE_MODE_USXGMII) - return 0; + pf->phylink_config.dev = &priv->ndev->dev; + pf->phylink_config.type = PHYLINK_NETDEV; - err = enetc_imdio_init(pf, is_c45); - if (err) + phylink = phylink_create(&pf->phylink_config, + of_fwnode_handle(dev->of_node), + pf->if_mode, &enetc_mac_phylink_ops); + if (IS_ERR(phylink)) { + err = PTR_ERR(phylink); return err; - - switch (priv->if_mode) { - case PHY_INTERFACE_MODE_SGMII: - enetc_configure_sgmii(pf->pcs); - break; - case PHY_INTERFACE_MODE_2500BASEX: - enetc_configure_2500basex(pf->pcs); - break; - case PHY_INTERFACE_MODE_USXGMII: - enetc_configure_usxgmii(pf->pcs); - break; - default: - dev_err(&pf->si->pdev->dev, "Unsupported link mode %s\n", - phy_modes(priv->if_mode)); } + priv->phylink = phylink; + return 0; } -static void enetc_teardown_serdes(struct enetc_ndev_priv *priv) +static void enetc_phylink_destroy(struct enetc_ndev_priv *priv) { - struct enetc_pf *pf = enetc_si_priv(priv->si); - - enetc_imdio_remove(pf); + if (priv->phylink) + phylink_destroy(priv->phylink); } static int enetc_pf_probe(struct pci_dev *pdev, @@ -1004,10 +1035,6 @@ static int enetc_pf_probe(struct pci_dev *pdev, pf->si = si; pf->total_vfs = pci_sriov_get_totalvfs(pdev); - err = enetc_of_get_phy(pf); - if (err) - dev_warn(&pdev->dev, "Fallback to PHY-less operation\n"); - enetc_configure_port(pf); enetc_get_si_caps(si); @@ -1022,8 +1049,6 @@ static int enetc_pf_probe(struct pci_dev *pdev, enetc_pf_netdev_setup(si, ndev, &enetc_ndev_ops); priv = netdev_priv(ndev); - priv->phy_node = pf->phy_node; - priv->if_mode = pf->if_mode; enetc_init_si_rings_params(priv); @@ -1039,20 +1064,27 @@ static int enetc_pf_probe(struct pci_dev *pdev, goto err_alloc_msix; } - err = enetc_configure_serdes(priv); - if (err) - dev_warn(&pdev->dev, "Attempted SerDes config but failed\n"); + if (!of_get_phy_mode(pdev->dev.of_node, &pf->if_mode)) { + err = enetc_mdiobus_create(pf); + if (err) + goto err_mdiobus_create; + + err = enetc_phylink_create(priv); + if (err) + goto err_phylink_create; + } err = register_netdev(ndev); if (err) goto err_reg_netdev; - netif_carrier_off(ndev); - return 0; err_reg_netdev: - enetc_teardown_serdes(priv); + enetc_phylink_destroy(priv); +err_phylink_create: + enetc_mdiobus_destroy(pf); +err_mdiobus_create: enetc_free_msix(priv); err_alloc_msix: enetc_free_si_resources(priv); @@ -1060,8 +1092,6 @@ err_alloc_si_res: si->ndev = NULL; free_netdev(ndev); err_alloc_netdev: - enetc_mdio_remove(pf); - enetc_of_put_phy(pf); err_map_pf_space: enetc_pci_remove(pdev); @@ -1074,16 +1104,15 @@ static void enetc_pf_remove(struct pci_dev *pdev) struct enetc_pf *pf = enetc_si_priv(si); struct enetc_ndev_priv *priv; + priv = netdev_priv(si->ndev); + enetc_phylink_destroy(priv); + enetc_mdiobus_destroy(pf); + if (pf->num_vfs) enetc_sriov_configure(pdev, 0); - priv = netdev_priv(si->ndev); unregister_netdev(si->ndev); - enetc_teardown_serdes(priv); - enetc_mdio_remove(pf); - enetc_of_put_phy(pf); - enetc_free_msix(priv); enetc_free_si_resources(priv); diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf.h b/drivers/net/ethernet/freescale/enetc/enetc_pf.h index 0d0ee91282a5..263946c51e37 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_pf.h +++ b/drivers/net/ethernet/freescale/enetc/enetc_pf.h @@ -2,6 +2,7 @@ /* Copyright 2017-2019 NXP */ #include "enetc.h" +#include <linux/pcs-lynx.h> #define ENETC_PF_NUM_RINGS 8 @@ -45,12 +46,15 @@ struct enetc_pf { struct mii_bus *mdio; /* saved for cleanup */ struct mii_bus *imdio; - struct phy_device *pcs; + struct lynx_pcs *pcs; - struct device_node *phy_node; phy_interface_t if_mode; + struct phylink_config phylink_config; }; +#define phylink_to_enetc_pf(config) \ + container_of((config), struct enetc_pf, phylink_config) + int enetc_msg_psi_init(struct enetc_pf *pf); void enetc_msg_psi_free(struct enetc_pf *pf); void enetc_msg_handle_rxmsg(struct enetc_pf *pf, int mbox_id, u16 *status); diff --git a/drivers/net/ethernet/freescale/enetc/enetc_qos.c b/drivers/net/ethernet/freescale/enetc/enetc_qos.c index 1c4a535890da..c81be32bcedf 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_qos.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_qos.c @@ -15,17 +15,14 @@ static u16 enetc_get_max_gcl_len(struct enetc_hw *hw) & ENETC_QBV_MAX_GCL_LEN_MASK; } -void enetc_sched_speed_set(struct net_device *ndev) +void enetc_sched_speed_set(struct enetc_ndev_priv *priv, int speed) { - struct enetc_ndev_priv *priv = netdev_priv(ndev); - struct phy_device *phydev = ndev->phydev; u32 old_speed = priv->speed; - u32 speed, pspeed; + u32 pspeed; - if (phydev->speed == old_speed) + if (speed == old_speed) return; - speed = phydev->speed; switch (speed) { case SPEED_1000: pspeed = ENETC_PMR_PSPEED_1000M; diff --git a/drivers/net/ethernet/freescale/enetc/enetc_vf.c b/drivers/net/ethernet/freescale/enetc/enetc_vf.c index f14576212a0e..7b5c82c7e4e5 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_vf.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_vf.c @@ -78,16 +78,11 @@ static int enetc_vf_set_mac_addr(struct net_device *ndev, void *addr) { struct enetc_ndev_priv *priv = netdev_priv(ndev); struct sockaddr *saddr = addr; - int err; if (!is_valid_ether_addr(saddr->sa_data)) return -EADDRNOTAVAIL; - err = enetc_msg_vsi_set_primary_mac_addr(priv, saddr); - if (err) - return err; - - return 0; + return enetc_msg_vsi_set_primary_mac_addr(priv, saddr); } static int enetc_vf_set_features(struct net_device *ndev, diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index fb37816a74db..8f7eca1e7716 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -1912,6 +1912,27 @@ out: return ret; } +static void fec_enet_phy_reset_after_clk_enable(struct net_device *ndev) +{ + struct fec_enet_private *fep = netdev_priv(ndev); + struct phy_device *phy_dev = ndev->phydev; + + if (phy_dev) { + phy_reset_after_clk_enable(phy_dev); + } else if (fep->phy_node) { + /* + * If the PHY still is not bound to the MAC, but there is + * OF PHY node and a matching PHY device instance already, + * use the OF PHY node to obtain the PHY device instance, + * and then use that PHY device instance when triggering + * the PHY reset. + */ + phy_dev = of_phy_find_device(fep->phy_node); + phy_reset_after_clk_enable(phy_dev); + put_device(&phy_dev->mdio.dev); + } +} + static int fec_enet_clk_enable(struct net_device *ndev, bool enable) { struct fec_enet_private *fep = netdev_priv(ndev); @@ -1938,7 +1959,7 @@ static int fec_enet_clk_enable(struct net_device *ndev, bool enable) if (ret) goto failed_clk_ref; - phy_reset_after_clk_enable(ndev->phydev); + fec_enet_phy_reset_after_clk_enable(ndev); } else { clk_disable_unprepare(fep->clk_enet_out); if (fep->clk_ptp) { @@ -1960,8 +1981,7 @@ failed_clk_ref: mutex_unlock(&fep->ptp_clk_mutex); } failed_clk_ptp: - if (fep->clk_enet_out) - clk_disable_unprepare(fep->clk_enet_out); + clk_disable_unprepare(fep->clk_enet_out); return ret; } @@ -2984,16 +3004,16 @@ fec_enet_open(struct net_device *ndev) /* Init MAC prior to mii bus probe */ fec_restart(ndev); - /* Probe and connect to PHY when open the interface */ - ret = fec_enet_mii_probe(ndev); - if (ret) - goto err_enet_mii_probe; - /* Call phy_reset_after_clk_enable() again if it failed during * phy_reset_after_clk_enable() before because the PHY wasn't probed. */ if (reset_again) - phy_reset_after_clk_enable(ndev->phydev); + fec_enet_phy_reset_after_clk_enable(ndev); + + /* Probe and connect to PHY when open the interface */ + ret = fec_enet_mii_probe(ndev); + if (ret) + goto err_enet_mii_probe; if (fep->quirks & FEC_QUIRK_ERR006687) imx6q_cpuidle_fec_irqs_used(); diff --git a/drivers/net/ethernet/freescale/fec_mpc52xx.c b/drivers/net/ethernet/freescale/fec_mpc52xx.c index 7a3f066e611d..b3bad429e03b 100644 --- a/drivers/net/ethernet/freescale/fec_mpc52xx.c +++ b/drivers/net/ethernet/freescale/fec_mpc52xx.c @@ -74,7 +74,7 @@ struct mpc52xx_fec_priv { static irqreturn_t mpc52xx_fec_interrupt(int, void *); static irqreturn_t mpc52xx_fec_rx_interrupt(int, void *); static irqreturn_t mpc52xx_fec_tx_interrupt(int, void *); -static void mpc52xx_fec_stop(struct net_device *dev); +static void mpc52xx_fec_stop(struct net_device *dev, bool may_sleep); static void mpc52xx_fec_start(struct net_device *dev); static void mpc52xx_fec_reset(struct net_device *dev); @@ -283,7 +283,7 @@ static int mpc52xx_fec_close(struct net_device *dev) netif_stop_queue(dev); - mpc52xx_fec_stop(dev); + mpc52xx_fec_stop(dev, true); mpc52xx_fec_free_rx_buffers(dev, priv->rx_dmatsk); @@ -693,7 +693,7 @@ static void mpc52xx_fec_start(struct net_device *dev) * * stop all activity on fec and empty dma buffers */ -static void mpc52xx_fec_stop(struct net_device *dev) +static void mpc52xx_fec_stop(struct net_device *dev, bool may_sleep) { struct mpc52xx_fec_priv *priv = netdev_priv(dev); struct mpc52xx_fec __iomem *fec = priv->fec; @@ -706,7 +706,7 @@ static void mpc52xx_fec_stop(struct net_device *dev) bcom_disable(priv->rx_dmatsk); /* Wait for tx queue to drain, but only if we're in process context */ - if (!in_interrupt()) { + if (may_sleep) { timeout = jiffies + msecs_to_jiffies(2000); while (time_before(jiffies, timeout) && !bcom_queue_empty(priv->tx_dmatsk)) @@ -738,7 +738,7 @@ static void mpc52xx_fec_reset(struct net_device *dev) struct mpc52xx_fec_priv *priv = netdev_priv(dev); struct mpc52xx_fec __iomem *fec = priv->fec; - mpc52xx_fec_stop(dev); + mpc52xx_fec_stop(dev, false); out_be32(&fec->rfifo_status, in_be32(&fec->rfifo_status)); out_be32(&fec->reset_cntrl, FEC_RESET_CNTRL_RESET_FIFO); diff --git a/drivers/net/ethernet/freescale/fec_ptp.c b/drivers/net/ethernet/freescale/fec_ptp.c index a0c1f4410306..2e344aada4c6 100644 --- a/drivers/net/ethernet/freescale/fec_ptp.c +++ b/drivers/net/ethernet/freescale/fec_ptp.c @@ -512,7 +512,7 @@ int fec_ptp_get(struct net_device *ndev, struct ifreq *ifr) -EFAULT : 0; } -/** +/* * fec_time_keep - call timecounter_read every second to avoid timer overrun * because ENET just support 32bit counter, will timeout in 4s */ @@ -520,13 +520,12 @@ static void fec_time_keep(struct work_struct *work) { struct delayed_work *dwork = to_delayed_work(work); struct fec_enet_private *fep = container_of(dwork, struct fec_enet_private, time_keep); - u64 ns; unsigned long flags; mutex_lock(&fep->ptp_clk_mutex); if (fep->ptp_clk_on) { spin_lock_irqsave(&fep->tmreg_lock, flags); - ns = timecounter_read(&fep->tc); + timecounter_read(&fep->tc); spin_unlock_irqrestore(&fep->tmreg_lock, flags); } mutex_unlock(&fep->ptp_clk_mutex); @@ -567,7 +566,8 @@ static irqreturn_t fec_pps_interrupt(int irq, void *dev_id) /** * fec_ptp_init - * @ndev: The FEC network adapter + * @pdev: The FEC network adapter + * @irq_idx: the interrupt index * * This function performs the required steps for enabling ptp * support. If ptp support has already been loaded it simply calls the diff --git a/drivers/net/ethernet/freescale/fman/fman.c b/drivers/net/ethernet/freescale/fman/fman.c index ef67e8599b39..ce0a121580f6 100644 --- a/drivers/net/ethernet/freescale/fman/fman.c +++ b/drivers/net/ethernet/freescale/fman/fman.c @@ -2063,11 +2063,11 @@ static int fman_set_exception(struct fman *fman, /** * fman_register_intr * @fman: A Pointer to FMan device - * @mod: Calling module + * @module: Calling module * @mod_id: Module id (if more than 1 exists, '0' if not) * @intr_type: Interrupt type (error/normal) selection. - * @f_isr: The interrupt service routine. - * @h_src_arg: Argument to be passed to f_isr. + * @isr_cb: The interrupt service routine. + * @src_arg: Argument to be passed to isr_cb. * * Used to register an event handler to be processed by FMan * @@ -2091,7 +2091,7 @@ EXPORT_SYMBOL(fman_register_intr); /** * fman_unregister_intr * @fman: A Pointer to FMan device - * @mod: Calling module + * @module: Calling module * @mod_id: Module id (if more than 1 exists, '0' if not) * @intr_type: Interrupt type (error/normal) selection. * @@ -2342,8 +2342,8 @@ EXPORT_SYMBOL(fman_get_bmi_max_fifo_size); /** * fman_get_revision - * @fman - Pointer to the FMan module - * @rev_info - A structure of revision information parameters. + * @fman: - Pointer to the FMan module + * @rev_info: - A structure of revision information parameters. * * Returns the FM revision * @@ -2508,7 +2508,7 @@ EXPORT_SYMBOL(fman_get_rx_extra_headroom); /** * fman_bind - * @dev: FMan OF device pointer + * @fm_dev: FMan OF device pointer * * Bind to a specific FMan device. * diff --git a/drivers/net/ethernet/freescale/fman/fman_muram.c b/drivers/net/ethernet/freescale/fman/fman_muram.c index 5ec94d243da0..7ad317e622bc 100644 --- a/drivers/net/ethernet/freescale/fman/fman_muram.c +++ b/drivers/net/ethernet/freescale/fman/fman_muram.c @@ -144,9 +144,9 @@ unsigned long fman_muram_alloc(struct muram_info *muram, size_t size) /** * fman_muram_free_mem - * muram: FM-MURAM module pointer. - * offset: offset of the memory region to be freed. - * size: size of the memory to be freed. + * @muram: FM-MURAM module pointer. + * @offset: offset of the memory region to be freed. + * @size: size of the memory to be freed. * * Free an allocated memory from FM-MURAM partition. */ diff --git a/drivers/net/ethernet/freescale/fman/fman_port.c b/drivers/net/ethernet/freescale/fman/fman_port.c index 624b2eb6f01d..d9baac0dbc7d 100644 --- a/drivers/net/ethernet/freescale/fman/fman_port.c +++ b/drivers/net/ethernet/freescale/fman/fman_port.c @@ -1410,9 +1410,11 @@ err_port_cfg: } EXPORT_SYMBOL(fman_port_config); -/** +/* * fman_port_use_kg_hash - * port: A pointer to a FM Port module. + * @port: A pointer to a FM Port module. + * @enable: enable or disable + * * Sets the HW KeyGen or the BMI as HW Parser next engine, enabling * or bypassing the KeyGen hashing of Rx traffic */ @@ -1430,7 +1432,8 @@ EXPORT_SYMBOL(fman_port_use_kg_hash); /** * fman_port_init - * port: A pointer to a FM Port module. + * @port: A pointer to a FM Port module. + * * Initializes the FM PORT module by defining the software structure and * configuring the hardware registers. * @@ -1524,8 +1527,8 @@ EXPORT_SYMBOL(fman_port_init); /** * fman_port_cfg_buf_prefix_content - * @port A pointer to a FM Port module. - * @buffer_prefix_content A structure of parameters describing + * @port: A pointer to a FM Port module. + * @buffer_prefix_content: A structure of parameters describing * the structure of the buffer. * Out parameter: * Start margin - offset of data from @@ -1570,7 +1573,7 @@ EXPORT_SYMBOL(fman_port_cfg_buf_prefix_content); /** * fman_port_disable - * port: A pointer to a FM Port module. + * @port: A pointer to a FM Port module. * * Gracefully disable an FM port. The port will not start new tasks after all * tasks associated with the port are terminated. @@ -1651,7 +1654,7 @@ EXPORT_SYMBOL(fman_port_disable); /** * fman_port_enable - * port: A pointer to a FM Port module. + * @port: A pointer to a FM Port module. * * A runtime routine provided to allow disable/enable of port. * @@ -1697,7 +1700,7 @@ EXPORT_SYMBOL(fman_port_enable); /** * fman_port_bind - * dev: FMan Port OF device pointer + * @dev: FMan Port OF device pointer * * Bind to a specific FMan Port. * @@ -1713,7 +1716,7 @@ EXPORT_SYMBOL(fman_port_bind); /** * fman_port_get_qman_channel_id - * port: Pointer to the FMan port devuce + * @port: Pointer to the FMan port devuce * * Get the QMan channel ID for the specific port * @@ -1727,7 +1730,7 @@ EXPORT_SYMBOL(fman_port_get_qman_channel_id); /** * fman_port_get_device - * port: Pointer to the FMan port device + * @port: Pointer to the FMan port device * * Get the 'struct device' associated to the specified FMan port device * diff --git a/drivers/net/ethernet/freescale/fman/mac.c b/drivers/net/ethernet/freescale/fman/mac.c index 43427c5b9396..901749a7a318 100644 --- a/drivers/net/ethernet/freescale/fman/mac.c +++ b/drivers/net/ethernet/freescale/fman/mac.c @@ -359,8 +359,8 @@ EXPORT_SYMBOL(fman_set_mac_active_pause); /** * fman_get_pause_cfg * @mac_dev: A pointer to the MAC device - * @rx: Return value for RX setting - * @tx: Return value for TX setting + * @rx_pause: Return value for RX setting + * @tx_pause: Return value for TX setting * * Determine the MAC RX/TX PAUSE frames settings based on PHY * autonegotiation or values set by eththool. diff --git a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c index bf846b42bc74..78e008b81374 100644 --- a/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c +++ b/drivers/net/ethernet/freescale/fs_enet/fs_enet-main.c @@ -562,10 +562,13 @@ fs_enet_start_xmit(struct sk_buff *skb, struct net_device *dev) BD_ENET_TX_TC); CBDS_SC(bdp, BD_ENET_TX_READY); - if ((CBDR_SC(bdp) & BD_ENET_TX_WRAP) == 0) - bdp++, curidx++; - else - bdp = fep->tx_bd_base, curidx = 0; + if ((CBDR_SC(bdp) & BD_ENET_TX_WRAP) == 0) { + bdp++; + curidx++; + } else { + bdp = fep->tx_bd_base; + curidx = 0; + } len = skb_frag_size(frag); CBDW_BUFADDR(bdp, skb_frag_dma_map(fep->dev, frag, 0, len, diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h index ebc37e256922..f5c80229ea96 100644 --- a/drivers/net/ethernet/google/gve/gve.h +++ b/drivers/net/ethernet/google/gve/gve.h @@ -27,6 +27,17 @@ /* 1 for management, 1 for rx, 1 for tx */ #define GVE_MIN_MSIX 3 +/* Numbers of gve tx/rx stats in stats report. */ +#define GVE_TX_STATS_REPORT_NUM 5 +#define GVE_RX_STATS_REPORT_NUM 2 + +/* Interval to schedule a stats report update, 20000ms. */ +#define GVE_STATS_REPORT_TIMER_PERIOD 20000 + +/* Numbers of NIC tx/rx stats in stats report. */ +#define NIC_TX_STATS_REPORT_NUM 0 +#define NIC_RX_STATS_REPORT_NUM 4 + /* Each slot in the desc ring has a 1:1 mapping to a slot in the data ring */ struct gve_rx_desc_queue { struct gve_rx_desc *desc_ring; /* the descriptor ring */ @@ -71,6 +82,11 @@ struct gve_rx_ring { u32 cnt; /* free-running total number of completed packets */ u32 fill_cnt; /* free-running total number of descs and buffs posted */ u32 mask; /* masks the cnt and fill_cnt to the size of the ring */ + u64 rx_copybreak_pkt; /* free-running count of copybreak packets */ + u64 rx_copied_pkt; /* free-running total number of copied packets */ + u64 rx_skb_alloc_fail; /* free-running count of skb alloc fails */ + u64 rx_buf_alloc_fail; /* free-running count of buffer alloc fails */ + u64 rx_desc_err_dropped_pkt; /* free-running count of packets dropped by descriptor error */ u32 q_num; /* queue index */ u32 ntfy_id; /* notification block index */ struct gve_queue_resources *q_resources; /* head and tail pointer idx */ @@ -202,24 +218,63 @@ struct gve_priv { dma_addr_t adminq_bus_addr; u32 adminq_mask; /* masks prod_cnt to adminq size */ u32 adminq_prod_cnt; /* free-running count of AQ cmds executed */ - + u32 adminq_cmd_fail; /* free-running count of AQ cmds failed */ + u32 adminq_timeouts; /* free-running count of AQ cmds timeouts */ + /* free-running count of per AQ cmd executed */ + u32 adminq_describe_device_cnt; + u32 adminq_cfg_device_resources_cnt; + u32 adminq_register_page_list_cnt; + u32 adminq_unregister_page_list_cnt; + u32 adminq_create_tx_queue_cnt; + u32 adminq_create_rx_queue_cnt; + u32 adminq_destroy_tx_queue_cnt; + u32 adminq_destroy_rx_queue_cnt; + u32 adminq_dcfg_device_resources_cnt; + u32 adminq_set_driver_parameter_cnt; + u32 adminq_report_stats_cnt; + u32 adminq_report_link_speed_cnt; + + /* Global stats */ + u32 interface_up_cnt; /* count of times interface turned up since last reset */ + u32 interface_down_cnt; /* count of times interface turned down since last reset */ + u32 reset_cnt; /* count of reset */ + u32 page_alloc_fail; /* count of page alloc fails */ + u32 dma_mapping_error; /* count of dma mapping errors */ + u32 stats_report_trigger_cnt; /* count of device-requested stats-reports since last reset */ struct workqueue_struct *gve_wq; struct work_struct service_task; + struct work_struct stats_report_task; unsigned long service_task_flags; unsigned long state_flags; + + struct gve_stats_report *stats_report; + u64 stats_report_len; + dma_addr_t stats_report_bus; /* dma address for the stats report */ + unsigned long ethtool_flags; + + unsigned long stats_report_timer_period; + struct timer_list stats_report_timer; + + /* Gvnic device link speed from hypervisor. */ + u64 link_speed; }; -enum gve_service_task_flags { - GVE_PRIV_FLAGS_DO_RESET = BIT(1), - GVE_PRIV_FLAGS_RESET_IN_PROGRESS = BIT(2), - GVE_PRIV_FLAGS_PROBE_IN_PROGRESS = BIT(3), +enum gve_service_task_flags_bit { + GVE_PRIV_FLAGS_DO_RESET = 1, + GVE_PRIV_FLAGS_RESET_IN_PROGRESS = 2, + GVE_PRIV_FLAGS_PROBE_IN_PROGRESS = 3, + GVE_PRIV_FLAGS_DO_REPORT_STATS = 4, }; -enum gve_state_flags { - GVE_PRIV_FLAGS_ADMIN_QUEUE_OK = BIT(1), - GVE_PRIV_FLAGS_DEVICE_RESOURCES_OK = BIT(2), - GVE_PRIV_FLAGS_DEVICE_RINGS_OK = BIT(3), - GVE_PRIV_FLAGS_NAPI_ENABLED = BIT(4), +enum gve_state_flags_bit { + GVE_PRIV_FLAGS_ADMIN_QUEUE_OK = 1, + GVE_PRIV_FLAGS_DEVICE_RESOURCES_OK = 2, + GVE_PRIV_FLAGS_DEVICE_RINGS_OK = 3, + GVE_PRIV_FLAGS_NAPI_ENABLED = 4, +}; + +enum gve_ethtool_flags_bit { + GVE_PRIV_FLAGS_REPORT_STATS = 0, }; static inline bool gve_get_do_reset(struct gve_priv *priv) @@ -269,6 +324,22 @@ static inline void gve_clear_probe_in_progress(struct gve_priv *priv) clear_bit(GVE_PRIV_FLAGS_PROBE_IN_PROGRESS, &priv->service_task_flags); } +static inline bool gve_get_do_report_stats(struct gve_priv *priv) +{ + return test_bit(GVE_PRIV_FLAGS_DO_REPORT_STATS, + &priv->service_task_flags); +} + +static inline void gve_set_do_report_stats(struct gve_priv *priv) +{ + set_bit(GVE_PRIV_FLAGS_DO_REPORT_STATS, &priv->service_task_flags); +} + +static inline void gve_clear_do_report_stats(struct gve_priv *priv) +{ + clear_bit(GVE_PRIV_FLAGS_DO_REPORT_STATS, &priv->service_task_flags); +} + static inline bool gve_get_admin_queue_ok(struct gve_priv *priv) { return test_bit(GVE_PRIV_FLAGS_ADMIN_QUEUE_OK, &priv->state_flags); @@ -329,6 +400,16 @@ static inline void gve_clear_napi_enabled(struct gve_priv *priv) clear_bit(GVE_PRIV_FLAGS_NAPI_ENABLED, &priv->state_flags); } +static inline bool gve_get_report_stats(struct gve_priv *priv) +{ + return test_bit(GVE_PRIV_FLAGS_REPORT_STATS, &priv->ethtool_flags); +} + +static inline void gve_clear_report_stats(struct gve_priv *priv) +{ + clear_bit(GVE_PRIV_FLAGS_REPORT_STATS, &priv->ethtool_flags); +} + /* Returns the address of the ntfy_blocks irq doorbell */ static inline __be32 __iomem *gve_irq_doorbell(struct gve_priv *priv, @@ -426,7 +507,8 @@ static inline bool gve_can_recycle_pages(struct net_device *dev) } /* buffers */ -int gve_alloc_page(struct device *dev, struct page **page, dma_addr_t *dma, +int gve_alloc_page(struct gve_priv *priv, struct device *dev, + struct page **page, dma_addr_t *dma, enum dma_data_direction); void gve_free_page(struct device *dev, struct page *page, dma_addr_t dma, enum dma_data_direction); @@ -450,6 +532,8 @@ int gve_reset(struct gve_priv *priv, bool attempt_teardown); int gve_adjust_queues(struct gve_priv *priv, struct gve_queue_config new_rx_config, struct gve_queue_config new_tx_config); +/* report stats handling */ +void gve_handle_report_stats(struct gve_priv *priv); /* exported by ethtool.c */ extern const struct ethtool_ops gve_ethtool_ops; /* needed by ethtool */ diff --git a/drivers/net/ethernet/google/gve/gve_adminq.c b/drivers/net/ethernet/google/gve/gve_adminq.c index c3ba7baf0107..24ae6a28a806 100644 --- a/drivers/net/ethernet/google/gve/gve_adminq.c +++ b/drivers/net/ethernet/google/gve/gve_adminq.c @@ -23,6 +23,20 @@ int gve_adminq_alloc(struct device *dev, struct gve_priv *priv) priv->adminq_mask = (PAGE_SIZE / sizeof(union gve_adminq_command)) - 1; priv->adminq_prod_cnt = 0; + priv->adminq_cmd_fail = 0; + priv->adminq_timeouts = 0; + priv->adminq_describe_device_cnt = 0; + priv->adminq_cfg_device_resources_cnt = 0; + priv->adminq_register_page_list_cnt = 0; + priv->adminq_unregister_page_list_cnt = 0; + priv->adminq_create_tx_queue_cnt = 0; + priv->adminq_create_rx_queue_cnt = 0; + priv->adminq_destroy_tx_queue_cnt = 0; + priv->adminq_destroy_rx_queue_cnt = 0; + priv->adminq_dcfg_device_resources_cnt = 0; + priv->adminq_set_driver_parameter_cnt = 0; + priv->adminq_report_stats_cnt = 0; + priv->adminq_report_link_speed_cnt = 0; /* Setup Admin queue with the device */ iowrite32be(priv->adminq_bus_addr / PAGE_SIZE, @@ -81,17 +95,18 @@ static bool gve_adminq_wait_for_cmd(struct gve_priv *priv, u32 prod_cnt) return false; } -static int gve_adminq_parse_err(struct device *dev, u32 status) +static int gve_adminq_parse_err(struct gve_priv *priv, u32 status) { if (status != GVE_ADMINQ_COMMAND_PASSED && - status != GVE_ADMINQ_COMMAND_UNSET) - dev_err(dev, "AQ command failed with status %d\n", status); - + status != GVE_ADMINQ_COMMAND_UNSET) { + dev_err(&priv->pdev->dev, "AQ command failed with status %d\n", status); + priv->adminq_cmd_fail++; + } switch (status) { case GVE_ADMINQ_COMMAND_PASSED: return 0; case GVE_ADMINQ_COMMAND_UNSET: - dev_err(dev, "parse_aq_err: err and status both unset, this should not be possible.\n"); + dev_err(&priv->pdev->dev, "parse_aq_err: err and status both unset, this should not be possible.\n"); return -EINVAL; case GVE_ADMINQ_COMMAND_ERROR_ABORTED: case GVE_ADMINQ_COMMAND_ERROR_CANCELLED: @@ -116,36 +131,145 @@ static int gve_adminq_parse_err(struct device *dev, u32 status) case GVE_ADMINQ_COMMAND_ERROR_UNIMPLEMENTED: return -ENOTSUPP; default: - dev_err(dev, "parse_aq_err: unknown status code %d\n", status); + dev_err(&priv->pdev->dev, "parse_aq_err: unknown status code %d\n", status); return -EINVAL; } } +/* Flushes all AQ commands currently queued and waits for them to complete. + * If there are failures, it will return the first error. + */ +static int gve_adminq_kick_and_wait(struct gve_priv *priv) +{ + u32 tail, head; + int i; + + tail = ioread32be(&priv->reg_bar0->adminq_event_counter); + head = priv->adminq_prod_cnt; + + gve_adminq_kick_cmd(priv, head); + if (!gve_adminq_wait_for_cmd(priv, head)) { + dev_err(&priv->pdev->dev, "AQ commands timed out, need to reset AQ\n"); + priv->adminq_timeouts++; + return -ENOTRECOVERABLE; + } + + for (i = tail; i < head; i++) { + union gve_adminq_command *cmd; + u32 status, err; + + cmd = &priv->adminq[i & priv->adminq_mask]; + status = be32_to_cpu(READ_ONCE(cmd->status)); + err = gve_adminq_parse_err(priv, status); + if (err) + // Return the first error if we failed. + return err; + } + + return 0; +} + /* This function is not threadsafe - the caller is responsible for any * necessary locks. */ -int gve_adminq_execute_cmd(struct gve_priv *priv, - union gve_adminq_command *cmd_orig) +static int gve_adminq_issue_cmd(struct gve_priv *priv, + union gve_adminq_command *cmd_orig) { union gve_adminq_command *cmd; - u32 status = 0; - u32 prod_cnt; + u32 opcode; + u32 tail; + + tail = ioread32be(&priv->reg_bar0->adminq_event_counter); + + // Check if next command will overflow the buffer. + if (((priv->adminq_prod_cnt + 1) & priv->adminq_mask) == tail) { + int err; + + // Flush existing commands to make room. + err = gve_adminq_kick_and_wait(priv); + if (err) + return err; + + // Retry. + tail = ioread32be(&priv->reg_bar0->adminq_event_counter); + if (((priv->adminq_prod_cnt + 1) & priv->adminq_mask) == tail) { + // This should never happen. We just flushed the + // command queue so there should be enough space. + return -ENOMEM; + } + } cmd = &priv->adminq[priv->adminq_prod_cnt & priv->adminq_mask]; priv->adminq_prod_cnt++; - prod_cnt = priv->adminq_prod_cnt; memcpy(cmd, cmd_orig, sizeof(*cmd_orig)); - - gve_adminq_kick_cmd(priv, prod_cnt); - if (!gve_adminq_wait_for_cmd(priv, prod_cnt)) { - dev_err(&priv->pdev->dev, "AQ command timed out, need to reset AQ\n"); - return -ENOTRECOVERABLE; + opcode = be32_to_cpu(READ_ONCE(cmd->opcode)); + + switch (opcode) { + case GVE_ADMINQ_DESCRIBE_DEVICE: + priv->adminq_describe_device_cnt++; + break; + case GVE_ADMINQ_CONFIGURE_DEVICE_RESOURCES: + priv->adminq_cfg_device_resources_cnt++; + break; + case GVE_ADMINQ_REGISTER_PAGE_LIST: + priv->adminq_register_page_list_cnt++; + break; + case GVE_ADMINQ_UNREGISTER_PAGE_LIST: + priv->adminq_unregister_page_list_cnt++; + break; + case GVE_ADMINQ_CREATE_TX_QUEUE: + priv->adminq_create_tx_queue_cnt++; + break; + case GVE_ADMINQ_CREATE_RX_QUEUE: + priv->adminq_create_rx_queue_cnt++; + break; + case GVE_ADMINQ_DESTROY_TX_QUEUE: + priv->adminq_destroy_tx_queue_cnt++; + break; + case GVE_ADMINQ_DESTROY_RX_QUEUE: + priv->adminq_destroy_rx_queue_cnt++; + break; + case GVE_ADMINQ_DECONFIGURE_DEVICE_RESOURCES: + priv->adminq_dcfg_device_resources_cnt++; + break; + case GVE_ADMINQ_SET_DRIVER_PARAMETER: + priv->adminq_set_driver_parameter_cnt++; + break; + case GVE_ADMINQ_REPORT_STATS: + priv->adminq_report_stats_cnt++; + break; + case GVE_ADMINQ_REPORT_LINK_SPEED: + priv->adminq_report_link_speed_cnt++; + break; + default: + dev_err(&priv->pdev->dev, "unknown AQ command opcode %d\n", opcode); } - memcpy(cmd_orig, cmd, sizeof(*cmd)); - status = be32_to_cpu(READ_ONCE(cmd->status)); - return gve_adminq_parse_err(&priv->pdev->dev, status); + return 0; +} + +/* This function is not threadsafe - the caller is responsible for any + * necessary locks. + * The caller is also responsible for making sure there are no commands + * waiting to be executed. + */ +static int gve_adminq_execute_cmd(struct gve_priv *priv, union gve_adminq_command *cmd_orig) +{ + u32 tail, head; + int err; + + tail = ioread32be(&priv->reg_bar0->adminq_event_counter); + head = priv->adminq_prod_cnt; + if (tail != head) + // This is not a valid path + return -EINVAL; + + err = gve_adminq_issue_cmd(priv, cmd_orig); + if (err) + return err; + + return gve_adminq_kick_and_wait(priv); } /* The device specifies that the management vector can either be the first irq @@ -190,29 +314,50 @@ int gve_adminq_deconfigure_device_resources(struct gve_priv *priv) return gve_adminq_execute_cmd(priv, &cmd); } -int gve_adminq_create_tx_queue(struct gve_priv *priv, u32 queue_index) +static int gve_adminq_create_tx_queue(struct gve_priv *priv, u32 queue_index) { struct gve_tx_ring *tx = &priv->tx[queue_index]; union gve_adminq_command cmd; + int err; memset(&cmd, 0, sizeof(cmd)); cmd.opcode = cpu_to_be32(GVE_ADMINQ_CREATE_TX_QUEUE); cmd.create_tx_queue = (struct gve_adminq_create_tx_queue) { .queue_id = cpu_to_be32(queue_index), .reserved = 0, - .queue_resources_addr = cpu_to_be64(tx->q_resources_bus), + .queue_resources_addr = + cpu_to_be64(tx->q_resources_bus), .tx_ring_addr = cpu_to_be64(tx->bus), .queue_page_list_id = cpu_to_be32(tx->tx_fifo.qpl->id), .ntfy_id = cpu_to_be32(tx->ntfy_id), }; - return gve_adminq_execute_cmd(priv, &cmd); + err = gve_adminq_issue_cmd(priv, &cmd); + if (err) + return err; + + return 0; +} + +int gve_adminq_create_tx_queues(struct gve_priv *priv, u32 num_queues) +{ + int err; + int i; + + for (i = 0; i < num_queues; i++) { + err = gve_adminq_create_tx_queue(priv, i); + if (err) + return err; + } + + return gve_adminq_kick_and_wait(priv); } -int gve_adminq_create_rx_queue(struct gve_priv *priv, u32 queue_index) +static int gve_adminq_create_rx_queue(struct gve_priv *priv, u32 queue_index) { struct gve_rx_ring *rx = &priv->rx[queue_index]; union gve_adminq_command cmd; + int err; memset(&cmd, 0, sizeof(cmd)); cmd.opcode = cpu_to_be32(GVE_ADMINQ_CREATE_RX_QUEUE); @@ -227,12 +372,31 @@ int gve_adminq_create_rx_queue(struct gve_priv *priv, u32 queue_index) .queue_page_list_id = cpu_to_be32(rx->data.qpl->id), }; - return gve_adminq_execute_cmd(priv, &cmd); + err = gve_adminq_issue_cmd(priv, &cmd); + if (err) + return err; + + return 0; } -int gve_adminq_destroy_tx_queue(struct gve_priv *priv, u32 queue_index) +int gve_adminq_create_rx_queues(struct gve_priv *priv, u32 num_queues) +{ + int err; + int i; + + for (i = 0; i < num_queues; i++) { + err = gve_adminq_create_rx_queue(priv, i); + if (err) + return err; + } + + return gve_adminq_kick_and_wait(priv); +} + +static int gve_adminq_destroy_tx_queue(struct gve_priv *priv, u32 queue_index) { union gve_adminq_command cmd; + int err; memset(&cmd, 0, sizeof(cmd)); cmd.opcode = cpu_to_be32(GVE_ADMINQ_DESTROY_TX_QUEUE); @@ -240,12 +404,31 @@ int gve_adminq_destroy_tx_queue(struct gve_priv *priv, u32 queue_index) .queue_id = cpu_to_be32(queue_index), }; - return gve_adminq_execute_cmd(priv, &cmd); + err = gve_adminq_issue_cmd(priv, &cmd); + if (err) + return err; + + return 0; +} + +int gve_adminq_destroy_tx_queues(struct gve_priv *priv, u32 num_queues) +{ + int err; + int i; + + for (i = 0; i < num_queues; i++) { + err = gve_adminq_destroy_tx_queue(priv, i); + if (err) + return err; + } + + return gve_adminq_kick_and_wait(priv); } -int gve_adminq_destroy_rx_queue(struct gve_priv *priv, u32 queue_index) +static int gve_adminq_destroy_rx_queue(struct gve_priv *priv, u32 queue_index) { union gve_adminq_command cmd; + int err; memset(&cmd, 0, sizeof(cmd)); cmd.opcode = cpu_to_be32(GVE_ADMINQ_DESTROY_RX_QUEUE); @@ -253,7 +436,25 @@ int gve_adminq_destroy_rx_queue(struct gve_priv *priv, u32 queue_index) .queue_id = cpu_to_be32(queue_index), }; - return gve_adminq_execute_cmd(priv, &cmd); + err = gve_adminq_issue_cmd(priv, &cmd); + if (err) + return err; + + return 0; +} + +int gve_adminq_destroy_rx_queues(struct gve_priv *priv, u32 num_queues) +{ + int err; + int i; + + for (i = 0; i < num_queues; i++) { + err = gve_adminq_destroy_rx_queue(priv, i); + if (err) + return err; + } + + return gve_adminq_kick_and_wait(priv); } int gve_adminq_describe_device(struct gve_priv *priv) @@ -283,8 +484,7 @@ int gve_adminq_describe_device(struct gve_priv *priv) priv->tx_desc_cnt = be16_to_cpu(descriptor->tx_queue_entries); if (priv->tx_desc_cnt * sizeof(priv->tx->desc[0]) < PAGE_SIZE) { - netif_err(priv, drv, priv->dev, "Tx desc count %d too low\n", - priv->tx_desc_cnt); + dev_err(&priv->pdev->dev, "Tx desc count %d too low\n", priv->tx_desc_cnt); err = -EINVAL; goto free_device_descriptor; } @@ -293,8 +493,7 @@ int gve_adminq_describe_device(struct gve_priv *priv) < PAGE_SIZE || priv->rx_desc_cnt * sizeof(priv->rx->data.data_ring[0]) < PAGE_SIZE) { - netif_err(priv, drv, priv->dev, "Rx desc count %d too low\n", - priv->rx_desc_cnt); + dev_err(&priv->pdev->dev, "Rx desc count %d too low\n", priv->rx_desc_cnt); err = -EINVAL; goto free_device_descriptor; } @@ -302,8 +501,7 @@ int gve_adminq_describe_device(struct gve_priv *priv) be64_to_cpu(descriptor->max_registered_pages); mtu = be16_to_cpu(descriptor->mtu); if (mtu < ETH_MIN_MTU) { - netif_err(priv, drv, priv->dev, "MTU %d below minimum MTU\n", - mtu); + dev_err(&priv->pdev->dev, "MTU %d below minimum MTU\n", mtu); err = -EINVAL; goto free_device_descriptor; } @@ -311,12 +509,12 @@ int gve_adminq_describe_device(struct gve_priv *priv) priv->num_event_counters = be16_to_cpu(descriptor->counters); ether_addr_copy(priv->dev->dev_addr, descriptor->mac); mac = descriptor->mac; - netif_info(priv, drv, priv->dev, "MAC addr: %pM\n", mac); + dev_info(&priv->pdev->dev, "MAC addr: %pM\n", mac); priv->tx_pages_per_qpl = be16_to_cpu(descriptor->tx_pages_per_qpl); priv->rx_pages_per_qpl = be16_to_cpu(descriptor->rx_pages_per_qpl); if (priv->rx_pages_per_qpl < priv->rx_desc_cnt) { - netif_err(priv, drv, priv->dev, "rx_pages_per_qpl cannot be smaller than rx_desc_cnt, setting rx_desc_cnt down to %d.\n", - priv->rx_pages_per_qpl); + dev_err(&priv->pdev->dev, "rx_pages_per_qpl cannot be smaller than rx_desc_cnt, setting rx_desc_cnt down to %d.\n", + priv->rx_pages_per_qpl); priv->rx_desc_cnt = priv->rx_pages_per_qpl; } priv->default_num_queues = be16_to_cpu(descriptor->default_num_queues); @@ -385,3 +583,46 @@ int gve_adminq_set_mtu(struct gve_priv *priv, u64 mtu) return gve_adminq_execute_cmd(priv, &cmd); } + +int gve_adminq_report_stats(struct gve_priv *priv, u64 stats_report_len, + dma_addr_t stats_report_addr, u64 interval) +{ + union gve_adminq_command cmd; + + memset(&cmd, 0, sizeof(cmd)); + cmd.opcode = cpu_to_be32(GVE_ADMINQ_REPORT_STATS); + cmd.report_stats = (struct gve_adminq_report_stats) { + .stats_report_len = cpu_to_be64(stats_report_len), + .stats_report_addr = cpu_to_be64(stats_report_addr), + .interval = cpu_to_be64(interval), + }; + + return gve_adminq_execute_cmd(priv, &cmd); +} + +int gve_adminq_report_link_speed(struct gve_priv *priv) +{ + union gve_adminq_command gvnic_cmd; + dma_addr_t link_speed_region_bus; + __be64 *link_speed_region; + int err; + + link_speed_region = + dma_alloc_coherent(&priv->pdev->dev, sizeof(*link_speed_region), + &link_speed_region_bus, GFP_KERNEL); + + if (!link_speed_region) + return -ENOMEM; + + memset(&gvnic_cmd, 0, sizeof(gvnic_cmd)); + gvnic_cmd.opcode = cpu_to_be32(GVE_ADMINQ_REPORT_LINK_SPEED); + gvnic_cmd.report_link_speed.link_speed_address = + cpu_to_be64(link_speed_region_bus); + + err = gve_adminq_execute_cmd(priv, &gvnic_cmd); + + priv->link_speed = be64_to_cpu(*link_speed_region); + dma_free_coherent(&priv->pdev->dev, sizeof(*link_speed_region), link_speed_region, + link_speed_region_bus); + return err; +} diff --git a/drivers/net/ethernet/google/gve/gve_adminq.h b/drivers/net/ethernet/google/gve/gve_adminq.h index 4dfa06edc0f8..281de8326bc5 100644 --- a/drivers/net/ethernet/google/gve/gve_adminq.h +++ b/drivers/net/ethernet/google/gve/gve_adminq.h @@ -21,6 +21,8 @@ enum gve_adminq_opcodes { GVE_ADMINQ_DESTROY_RX_QUEUE = 0x8, GVE_ADMINQ_DECONFIGURE_DEVICE_RESOURCES = 0x9, GVE_ADMINQ_SET_DRIVER_PARAMETER = 0xB, + GVE_ADMINQ_REPORT_STATS = 0xC, + GVE_ADMINQ_REPORT_LINK_SPEED = 0xD }; /* Admin queue status codes */ @@ -172,6 +174,51 @@ struct gve_adminq_set_driver_parameter { static_assert(sizeof(struct gve_adminq_set_driver_parameter) == 16); +struct gve_adminq_report_stats { + __be64 stats_report_len; + __be64 stats_report_addr; + __be64 interval; +}; + +static_assert(sizeof(struct gve_adminq_report_stats) == 24); + +struct gve_adminq_report_link_speed { + __be64 link_speed_address; +}; + +static_assert(sizeof(struct gve_adminq_report_link_speed) == 8); + +struct stats { + __be32 stat_name; + __be32 queue_id; + __be64 value; +}; + +static_assert(sizeof(struct stats) == 16); + +struct gve_stats_report { + __be64 written_count; + struct stats stats[0]; +}; + +static_assert(sizeof(struct gve_stats_report) == 8); + +enum gve_stat_names { + // stats from gve + TX_WAKE_CNT = 1, + TX_STOP_CNT = 2, + TX_FRAMES_SENT = 3, + TX_BYTES_SENT = 4, + TX_LAST_COMPLETION_PROCESSED = 5, + RX_NEXT_EXPECTED_SEQUENCE = 6, + RX_BUFFERS_POSTED = 7, + // stats from NIC + RX_QUEUE_DROP_CNT = 65, + RX_NO_BUFFERS_POSTED = 66, + RX_DROPS_PACKET_OVER_MRU = 67, + RX_DROPS_INVALID_CHECKSUM = 68, +}; + union gve_adminq_command { struct { __be32 opcode; @@ -187,6 +234,8 @@ union gve_adminq_command { struct gve_adminq_register_page_list reg_page_list; struct gve_adminq_unregister_page_list unreg_page_list; struct gve_adminq_set_driver_parameter set_driver_param; + struct gve_adminq_report_stats report_stats; + struct gve_adminq_report_link_speed report_link_speed; }; }; u8 reserved[64]; @@ -197,8 +246,6 @@ static_assert(sizeof(union gve_adminq_command) == 64); int gve_adminq_alloc(struct device *dev, struct gve_priv *priv); void gve_adminq_free(struct device *dev, struct gve_priv *priv); void gve_adminq_release(struct gve_priv *priv); -int gve_adminq_execute_cmd(struct gve_priv *priv, - union gve_adminq_command *cmd_orig); int gve_adminq_describe_device(struct gve_priv *priv); int gve_adminq_configure_device_resources(struct gve_priv *priv, dma_addr_t counter_array_bus_addr, @@ -206,12 +253,15 @@ int gve_adminq_configure_device_resources(struct gve_priv *priv, dma_addr_t db_array_bus_addr, u32 num_ntfy_blks); int gve_adminq_deconfigure_device_resources(struct gve_priv *priv); -int gve_adminq_create_tx_queue(struct gve_priv *priv, u32 queue_id); -int gve_adminq_destroy_tx_queue(struct gve_priv *priv, u32 queue_id); -int gve_adminq_create_rx_queue(struct gve_priv *priv, u32 queue_id); -int gve_adminq_destroy_rx_queue(struct gve_priv *priv, u32 queue_id); +int gve_adminq_create_tx_queues(struct gve_priv *priv, u32 num_queues); +int gve_adminq_destroy_tx_queues(struct gve_priv *priv, u32 queue_id); +int gve_adminq_create_rx_queues(struct gve_priv *priv, u32 num_queues); +int gve_adminq_destroy_rx_queues(struct gve_priv *priv, u32 queue_id); int gve_adminq_register_page_list(struct gve_priv *priv, struct gve_queue_page_list *qpl); int gve_adminq_unregister_page_list(struct gve_priv *priv, u32 page_list_id); int gve_adminq_set_mtu(struct gve_priv *priv, u64 mtu); +int gve_adminq_report_stats(struct gve_priv *priv, u64 stats_report_len, + dma_addr_t stats_report_addr, u64 interval); +int gve_adminq_report_link_speed(struct gve_priv *priv); #endif /* _GVE_ADMINQ_H */ diff --git a/drivers/net/ethernet/google/gve/gve_ethtool.c b/drivers/net/ethernet/google/gve/gve_ethtool.c index d8fa816f4473..7b44769bd87c 100644 --- a/drivers/net/ethernet/google/gve/gve_ethtool.c +++ b/drivers/net/ethernet/google/gve/gve_ethtool.c @@ -6,6 +6,7 @@ #include <linux/rtnetlink.h> #include "gve.h" +#include "gve_adminq.h" static void gve_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *info) @@ -34,41 +35,84 @@ static u32 gve_get_msglevel(struct net_device *netdev) static const char gve_gstrings_main_stats[][ETH_GSTRING_LEN] = { "rx_packets", "tx_packets", "rx_bytes", "tx_bytes", "rx_dropped", "tx_dropped", "tx_timeouts", + "rx_skb_alloc_fail", "rx_buf_alloc_fail", "rx_desc_err_dropped_pkt", + "interface_up_cnt", "interface_down_cnt", "reset_cnt", + "page_alloc_fail", "dma_mapping_error", "stats_report_trigger_cnt", +}; + +static const char gve_gstrings_rx_stats[][ETH_GSTRING_LEN] = { + "rx_posted_desc[%u]", "rx_completed_desc[%u]", "rx_bytes[%u]", + "rx_dropped_pkt[%u]", "rx_copybreak_pkt[%u]", "rx_copied_pkt[%u]", + "rx_queue_drop_cnt[%u]", "rx_no_buffers_posted[%u]", + "rx_drops_packet_over_mru[%u]", "rx_drops_invalid_checksum[%u]", +}; + +static const char gve_gstrings_tx_stats[][ETH_GSTRING_LEN] = { + "tx_posted_desc[%u]", "tx_completed_desc[%u]", "tx_bytes[%u]", + "tx_wake[%u]", "tx_stop[%u]", "tx_event_counter[%u]", +}; + +static const char gve_gstrings_adminq_stats[][ETH_GSTRING_LEN] = { + "adminq_prod_cnt", "adminq_cmd_fail", "adminq_timeouts", + "adminq_describe_device_cnt", "adminq_cfg_device_resources_cnt", + "adminq_register_page_list_cnt", "adminq_unregister_page_list_cnt", + "adminq_create_tx_queue_cnt", "adminq_create_rx_queue_cnt", + "adminq_destroy_tx_queue_cnt", "adminq_destroy_rx_queue_cnt", + "adminq_dcfg_device_resources_cnt", "adminq_set_driver_parameter_cnt", + "adminq_report_stats_cnt", "adminq_report_link_speed_cnt" +}; + +static const char gve_gstrings_priv_flags[][ETH_GSTRING_LEN] = { + "report-stats", }; #define GVE_MAIN_STATS_LEN ARRAY_SIZE(gve_gstrings_main_stats) -#define NUM_GVE_TX_CNTS 5 -#define NUM_GVE_RX_CNTS 2 +#define GVE_ADMINQ_STATS_LEN ARRAY_SIZE(gve_gstrings_adminq_stats) +#define NUM_GVE_TX_CNTS ARRAY_SIZE(gve_gstrings_tx_stats) +#define NUM_GVE_RX_CNTS ARRAY_SIZE(gve_gstrings_rx_stats) +#define GVE_PRIV_FLAGS_STR_LEN ARRAY_SIZE(gve_gstrings_priv_flags) static void gve_get_strings(struct net_device *netdev, u32 stringset, u8 *data) { struct gve_priv *priv = netdev_priv(netdev); char *s = (char *)data; - int i; + int i, j; - if (stringset != ETH_SS_STATS) - return; + switch (stringset) { + case ETH_SS_STATS: + memcpy(s, *gve_gstrings_main_stats, + sizeof(gve_gstrings_main_stats)); + s += sizeof(gve_gstrings_main_stats); + + for (i = 0; i < priv->rx_cfg.num_queues; i++) { + for (j = 0; j < NUM_GVE_RX_CNTS; j++) { + snprintf(s, ETH_GSTRING_LEN, + gve_gstrings_rx_stats[j], i); + s += ETH_GSTRING_LEN; + } + } - memcpy(s, *gve_gstrings_main_stats, - sizeof(gve_gstrings_main_stats)); - s += sizeof(gve_gstrings_main_stats); - for (i = 0; i < priv->rx_cfg.num_queues; i++) { - snprintf(s, ETH_GSTRING_LEN, "rx_desc_cnt[%u]", i); - s += ETH_GSTRING_LEN; - snprintf(s, ETH_GSTRING_LEN, "rx_desc_fill_cnt[%u]", i); - s += ETH_GSTRING_LEN; - } - for (i = 0; i < priv->tx_cfg.num_queues; i++) { - snprintf(s, ETH_GSTRING_LEN, "tx_req[%u]", i); - s += ETH_GSTRING_LEN; - snprintf(s, ETH_GSTRING_LEN, "tx_done[%u]", i); - s += ETH_GSTRING_LEN; - snprintf(s, ETH_GSTRING_LEN, "tx_wake[%u]", i); - s += ETH_GSTRING_LEN; - snprintf(s, ETH_GSTRING_LEN, "tx_stop[%u]", i); - s += ETH_GSTRING_LEN; - snprintf(s, ETH_GSTRING_LEN, "tx_event_counter[%u]", i); - s += ETH_GSTRING_LEN; + for (i = 0; i < priv->tx_cfg.num_queues; i++) { + for (j = 0; j < NUM_GVE_TX_CNTS; j++) { + snprintf(s, ETH_GSTRING_LEN, + gve_gstrings_tx_stats[j], i); + s += ETH_GSTRING_LEN; + } + } + + memcpy(s, *gve_gstrings_adminq_stats, + sizeof(gve_gstrings_adminq_stats)); + s += sizeof(gve_gstrings_adminq_stats); + break; + + case ETH_SS_PRIV_FLAGS: + memcpy(s, *gve_gstrings_priv_flags, + sizeof(gve_gstrings_priv_flags)); + s += sizeof(gve_gstrings_priv_flags); + break; + + default: + break; } } @@ -78,9 +122,11 @@ static int gve_get_sset_count(struct net_device *netdev, int sset) switch (sset) { case ETH_SS_STATS: - return GVE_MAIN_STATS_LEN + + return GVE_MAIN_STATS_LEN + GVE_ADMINQ_STATS_LEN + (priv->rx_cfg.num_queues * NUM_GVE_RX_CNTS) + (priv->tx_cfg.num_queues * NUM_GVE_TX_CNTS); + case ETH_SS_PRIV_FLAGS: + return GVE_PRIV_FLAGS_STR_LEN; default: return -EOPNOTSUPP; } @@ -90,24 +136,56 @@ static void gve_get_ethtool_stats(struct net_device *netdev, struct ethtool_stats *stats, u64 *data) { - struct gve_priv *priv = netdev_priv(netdev); - u64 rx_pkts, rx_bytes, tx_pkts, tx_bytes; + u64 tmp_rx_pkts, tmp_rx_bytes, tmp_rx_skb_alloc_fail, tmp_rx_buf_alloc_fail, + tmp_rx_desc_err_dropped_pkt, tmp_tx_pkts, tmp_tx_bytes; + u64 rx_buf_alloc_fail, rx_desc_err_dropped_pkt, rx_pkts, + rx_skb_alloc_fail, rx_bytes, tx_pkts, tx_bytes; + int stats_idx, base_stats_idx, max_stats_idx; + struct stats *report_stats; + int *rx_qid_to_stats_idx; + int *tx_qid_to_stats_idx; + struct gve_priv *priv; + bool skip_nic_stats; unsigned int start; int ring; - int i; + int i, j; ASSERT_RTNL(); - for (rx_pkts = 0, rx_bytes = 0, ring = 0; + priv = netdev_priv(netdev); + report_stats = priv->stats_report->stats; + rx_qid_to_stats_idx = kmalloc_array(priv->rx_cfg.num_queues, + sizeof(int), GFP_KERNEL); + if (!rx_qid_to_stats_idx) + return; + tx_qid_to_stats_idx = kmalloc_array(priv->tx_cfg.num_queues, + sizeof(int), GFP_KERNEL); + if (!tx_qid_to_stats_idx) { + kfree(rx_qid_to_stats_idx); + return; + } + for (rx_pkts = 0, rx_bytes = 0, rx_skb_alloc_fail = 0, + rx_buf_alloc_fail = 0, rx_desc_err_dropped_pkt = 0, ring = 0; ring < priv->rx_cfg.num_queues; ring++) { if (priv->rx) { do { + struct gve_rx_ring *rx = &priv->rx[ring]; + start = u64_stats_fetch_begin(&priv->rx[ring].statss); - rx_pkts += priv->rx[ring].rpackets; - rx_bytes += priv->rx[ring].rbytes; + tmp_rx_pkts = rx->rpackets; + tmp_rx_bytes = rx->rbytes; + tmp_rx_skb_alloc_fail = rx->rx_skb_alloc_fail; + tmp_rx_buf_alloc_fail = rx->rx_buf_alloc_fail; + tmp_rx_desc_err_dropped_pkt = + rx->rx_desc_err_dropped_pkt; } while (u64_stats_fetch_retry(&priv->rx[ring].statss, start)); + rx_pkts += tmp_rx_pkts; + rx_bytes += tmp_rx_bytes; + rx_skb_alloc_fail += tmp_rx_skb_alloc_fail; + rx_buf_alloc_fail += tmp_rx_buf_alloc_fail; + rx_desc_err_dropped_pkt += tmp_rx_desc_err_dropped_pkt; } } for (tx_pkts = 0, tx_bytes = 0, ring = 0; @@ -116,10 +194,12 @@ gve_get_ethtool_stats(struct net_device *netdev, do { start = u64_stats_fetch_begin(&priv->tx[ring].statss); - tx_pkts += priv->tx[ring].pkt_done; - tx_bytes += priv->tx[ring].bytes_done; + tmp_tx_pkts = priv->tx[ring].pkt_done; + tmp_tx_bytes = priv->tx[ring].bytes_done; } while (u64_stats_fetch_retry(&priv->tx[ring].statss, start)); + tx_pkts += tmp_tx_pkts; + tx_bytes += tmp_tx_bytes; } } @@ -128,22 +208,102 @@ gve_get_ethtool_stats(struct net_device *netdev, data[i++] = tx_pkts; data[i++] = rx_bytes; data[i++] = tx_bytes; - /* Skip rx_dropped and tx_dropped */ - i += 2; + /* total rx dropped packets */ + data[i++] = rx_skb_alloc_fail + rx_buf_alloc_fail + + rx_desc_err_dropped_pkt; + /* Skip tx_dropped */ + i++; + data[i++] = priv->tx_timeo_cnt; + data[i++] = rx_skb_alloc_fail; + data[i++] = rx_buf_alloc_fail; + data[i++] = rx_desc_err_dropped_pkt; + data[i++] = priv->interface_up_cnt; + data[i++] = priv->interface_down_cnt; + data[i++] = priv->reset_cnt; + data[i++] = priv->page_alloc_fail; + data[i++] = priv->dma_mapping_error; + data[i++] = priv->stats_report_trigger_cnt; i = GVE_MAIN_STATS_LEN; + /* For rx cross-reporting stats, start from nic rx stats in report */ + base_stats_idx = GVE_TX_STATS_REPORT_NUM * priv->tx_cfg.num_queues + + GVE_RX_STATS_REPORT_NUM * priv->rx_cfg.num_queues; + max_stats_idx = NIC_RX_STATS_REPORT_NUM * priv->rx_cfg.num_queues + + base_stats_idx; + /* Preprocess the stats report for rx, map queue id to start index */ + skip_nic_stats = false; + for (stats_idx = base_stats_idx; stats_idx < max_stats_idx; + stats_idx += NIC_RX_STATS_REPORT_NUM) { + u32 stat_name = be32_to_cpu(report_stats[stats_idx].stat_name); + u32 queue_id = be32_to_cpu(report_stats[stats_idx].queue_id); + + if (stat_name == 0) { + /* no stats written by NIC yet */ + skip_nic_stats = true; + break; + } + rx_qid_to_stats_idx[queue_id] = stats_idx; + } /* walk RX rings */ if (priv->rx) { for (ring = 0; ring < priv->rx_cfg.num_queues; ring++) { struct gve_rx_ring *rx = &priv->rx[ring]; - data[i++] = rx->cnt; data[i++] = rx->fill_cnt; + data[i++] = rx->cnt; + do { + start = + u64_stats_fetch_begin(&priv->rx[ring].statss); + tmp_rx_bytes = rx->rbytes; + tmp_rx_skb_alloc_fail = rx->rx_skb_alloc_fail; + tmp_rx_buf_alloc_fail = rx->rx_buf_alloc_fail; + tmp_rx_desc_err_dropped_pkt = + rx->rx_desc_err_dropped_pkt; + } while (u64_stats_fetch_retry(&priv->rx[ring].statss, + start)); + data[i++] = tmp_rx_bytes; + /* rx dropped packets */ + data[i++] = tmp_rx_skb_alloc_fail + + tmp_rx_buf_alloc_fail + + tmp_rx_desc_err_dropped_pkt; + data[i++] = rx->rx_copybreak_pkt; + data[i++] = rx->rx_copied_pkt; + /* stats from NIC */ + if (skip_nic_stats) { + /* skip NIC rx stats */ + i += NIC_RX_STATS_REPORT_NUM; + continue; + } + for (j = 0; j < NIC_RX_STATS_REPORT_NUM; j++) { + u64 value = + be64_to_cpu(report_stats[rx_qid_to_stats_idx[ring] + j].value); + + data[i++] = value; + } } } else { i += priv->rx_cfg.num_queues * NUM_GVE_RX_CNTS; } + + /* For tx cross-reporting stats, start from nic tx stats in report */ + base_stats_idx = max_stats_idx; + max_stats_idx = NIC_TX_STATS_REPORT_NUM * priv->tx_cfg.num_queues + + max_stats_idx; + /* Preprocess the stats report for tx, map queue id to start index */ + skip_nic_stats = false; + for (stats_idx = base_stats_idx; stats_idx < max_stats_idx; + stats_idx += NIC_TX_STATS_REPORT_NUM) { + u32 stat_name = be32_to_cpu(report_stats[stats_idx].stat_name); + u32 queue_id = be32_to_cpu(report_stats[stats_idx].queue_id); + + if (stat_name == 0) { + /* no stats written by NIC yet */ + skip_nic_stats = true; + break; + } + tx_qid_to_stats_idx[queue_id] = stats_idx; + } /* walk TX rings */ if (priv->tx) { for (ring = 0; ring < priv->tx_cfg.num_queues; ring++) { @@ -151,14 +311,51 @@ gve_get_ethtool_stats(struct net_device *netdev, data[i++] = tx->req; data[i++] = tx->done; + do { + start = + u64_stats_fetch_begin(&priv->tx[ring].statss); + tmp_tx_bytes = tx->bytes_done; + } while (u64_stats_fetch_retry(&priv->tx[ring].statss, + start)); + data[i++] = tmp_tx_bytes; data[i++] = tx->wake_queue; data[i++] = tx->stop_queue; data[i++] = be32_to_cpu(gve_tx_load_event_counter(priv, tx)); + /* stats from NIC */ + if (skip_nic_stats) { + /* skip NIC tx stats */ + i += NIC_TX_STATS_REPORT_NUM; + continue; + } + for (j = 0; j < NIC_TX_STATS_REPORT_NUM; j++) { + u64 value = + be64_to_cpu(report_stats[tx_qid_to_stats_idx[ring] + j].value); + data[i++] = value; + } } } else { i += priv->tx_cfg.num_queues * NUM_GVE_TX_CNTS; } + + kfree(rx_qid_to_stats_idx); + kfree(tx_qid_to_stats_idx); + /* AQ Stats */ + data[i++] = priv->adminq_prod_cnt; + data[i++] = priv->adminq_cmd_fail; + data[i++] = priv->adminq_timeouts; + data[i++] = priv->adminq_describe_device_cnt; + data[i++] = priv->adminq_cfg_device_resources_cnt; + data[i++] = priv->adminq_register_page_list_cnt; + data[i++] = priv->adminq_unregister_page_list_cnt; + data[i++] = priv->adminq_create_tx_queue_cnt; + data[i++] = priv->adminq_create_rx_queue_cnt; + data[i++] = priv->adminq_destroy_tx_queue_cnt; + data[i++] = priv->adminq_destroy_rx_queue_cnt; + data[i++] = priv->adminq_dcfg_device_resources_cnt; + data[i++] = priv->adminq_set_driver_parameter_cnt; + data[i++] = priv->adminq_report_stats_cnt; + data[i++] = priv->adminq_report_link_speed_cnt; } static void gve_get_channels(struct net_device *netdev, @@ -230,6 +427,95 @@ static int gve_user_reset(struct net_device *netdev, u32 *flags) return -EOPNOTSUPP; } +static int gve_get_tunable(struct net_device *netdev, + const struct ethtool_tunable *etuna, void *value) +{ + struct gve_priv *priv = netdev_priv(netdev); + + switch (etuna->id) { + case ETHTOOL_RX_COPYBREAK: + *(u32 *)value = priv->rx_copybreak; + return 0; + default: + return -EOPNOTSUPP; + } +} + +static int gve_set_tunable(struct net_device *netdev, + const struct ethtool_tunable *etuna, + const void *value) +{ + struct gve_priv *priv = netdev_priv(netdev); + u32 len; + + switch (etuna->id) { + case ETHTOOL_RX_COPYBREAK: + len = *(u32 *)value; + if (len > PAGE_SIZE / 2) + return -EINVAL; + priv->rx_copybreak = len; + return 0; + default: + return -EOPNOTSUPP; + } +} + +static u32 gve_get_priv_flags(struct net_device *netdev) +{ + struct gve_priv *priv = netdev_priv(netdev); + u32 ret_flags = 0; + + /* Only 1 flag exists currently: report-stats (BIT(O)), so set that flag. */ + if (priv->ethtool_flags & BIT(0)) + ret_flags |= BIT(0); + return ret_flags; +} + +static int gve_set_priv_flags(struct net_device *netdev, u32 flags) +{ + struct gve_priv *priv = netdev_priv(netdev); + u64 ori_flags, new_flags; + + ori_flags = READ_ONCE(priv->ethtool_flags); + new_flags = ori_flags; + + /* Only one priv flag exists: report-stats (BIT(0))*/ + if (flags & BIT(0)) + new_flags |= BIT(0); + else + new_flags &= ~(BIT(0)); + priv->ethtool_flags = new_flags; + /* start report-stats timer when user turns report stats on. */ + if (flags & BIT(0)) { + mod_timer(&priv->stats_report_timer, + round_jiffies(jiffies + + msecs_to_jiffies(priv->stats_report_timer_period))); + } + /* Zero off gve stats when report-stats turned off and */ + /* delete report stats timer. */ + if (!(flags & BIT(0)) && (ori_flags & BIT(0))) { + int tx_stats_num = GVE_TX_STATS_REPORT_NUM * + priv->tx_cfg.num_queues; + int rx_stats_num = GVE_RX_STATS_REPORT_NUM * + priv->rx_cfg.num_queues; + + memset(priv->stats_report->stats, 0, (tx_stats_num + rx_stats_num) * + sizeof(struct stats)); + del_timer_sync(&priv->stats_report_timer); + } + return 0; +} + +static int gve_get_link_ksettings(struct net_device *netdev, + struct ethtool_link_ksettings *cmd) +{ + struct gve_priv *priv = netdev_priv(netdev); + int err = gve_adminq_report_link_speed(priv); + + cmd->base.speed = priv->link_speed; + return err; +} + const struct ethtool_ops gve_ethtool_ops = { .get_drvinfo = gve_get_drvinfo, .get_strings = gve_get_strings, @@ -242,4 +528,9 @@ const struct ethtool_ops gve_ethtool_ops = { .get_link = ethtool_op_get_link, .get_ringparam = gve_get_ringparam, .reset = gve_user_reset, + .get_tunable = gve_get_tunable, + .set_tunable = gve_set_tunable, + .get_priv_flags = gve_get_priv_flags, + .set_priv_flags = gve_set_priv_flags, + .get_link_ksettings = gve_get_link_ksettings }; diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c index e032563ceefd..48a433154ce0 100644 --- a/drivers/net/ethernet/google/gve/gve_main.c +++ b/drivers/net/ethernet/google/gve/gve_main.c @@ -78,6 +78,66 @@ static void gve_free_counter_array(struct gve_priv *priv) priv->counter_array = NULL; } +/* NIC requests to report stats */ +static void gve_stats_report_task(struct work_struct *work) +{ + struct gve_priv *priv = container_of(work, struct gve_priv, + stats_report_task); + if (gve_get_do_report_stats(priv)) { + gve_handle_report_stats(priv); + gve_clear_do_report_stats(priv); + } +} + +static void gve_stats_report_schedule(struct gve_priv *priv) +{ + if (!gve_get_probe_in_progress(priv) && + !gve_get_reset_in_progress(priv)) { + gve_set_do_report_stats(priv); + queue_work(priv->gve_wq, &priv->stats_report_task); + } +} + +static void gve_stats_report_timer(struct timer_list *t) +{ + struct gve_priv *priv = from_timer(priv, t, stats_report_timer); + + mod_timer(&priv->stats_report_timer, + round_jiffies(jiffies + + msecs_to_jiffies(priv->stats_report_timer_period))); + gve_stats_report_schedule(priv); +} + +static int gve_alloc_stats_report(struct gve_priv *priv) +{ + int tx_stats_num, rx_stats_num; + + tx_stats_num = (GVE_TX_STATS_REPORT_NUM + NIC_TX_STATS_REPORT_NUM) * + priv->tx_cfg.num_queues; + rx_stats_num = (GVE_RX_STATS_REPORT_NUM + NIC_RX_STATS_REPORT_NUM) * + priv->rx_cfg.num_queues; + priv->stats_report_len = sizeof(struct gve_stats_report) + + (tx_stats_num + rx_stats_num) * + sizeof(struct stats); + priv->stats_report = + dma_alloc_coherent(&priv->pdev->dev, priv->stats_report_len, + &priv->stats_report_bus, GFP_KERNEL); + if (!priv->stats_report) + return -ENOMEM; + /* Set up timer for the report-stats task */ + timer_setup(&priv->stats_report_timer, gve_stats_report_timer, 0); + priv->stats_report_timer_period = GVE_STATS_REPORT_TIMER_PERIOD; + return 0; +} + +static void gve_free_stats_report(struct gve_priv *priv) +{ + del_timer_sync(&priv->stats_report_timer); + dma_free_coherent(&priv->pdev->dev, priv->stats_report_len, + priv->stats_report, priv->stats_report_bus); + priv->stats_report = NULL; +} + static irqreturn_t gve_mgmnt_intr(int irq, void *arg) { struct gve_priv *priv = arg; @@ -270,6 +330,9 @@ static int gve_setup_device_resources(struct gve_priv *priv) err = gve_alloc_notify_blocks(priv); if (err) goto abort_with_counter; + err = gve_alloc_stats_report(priv); + if (err) + goto abort_with_ntfy_blocks; err = gve_adminq_configure_device_resources(priv, priv->counter_array_bus, priv->num_event_counters, @@ -279,10 +342,18 @@ static int gve_setup_device_resources(struct gve_priv *priv) dev_err(&priv->pdev->dev, "could not setup device_resources: err=%d\n", err); err = -ENXIO; - goto abort_with_ntfy_blocks; + goto abort_with_stats_report; } + err = gve_adminq_report_stats(priv, priv->stats_report_len, + priv->stats_report_bus, + GVE_STATS_REPORT_TIMER_PERIOD); + if (err) + dev_err(&priv->pdev->dev, + "Failed to report stats: err=%d\n", err); gve_set_device_resources_ok(priv); return 0; +abort_with_stats_report: + gve_free_stats_report(priv); abort_with_ntfy_blocks: gve_free_notify_blocks(priv); abort_with_counter: @@ -298,6 +369,13 @@ static void gve_teardown_device_resources(struct gve_priv *priv) /* Tell device its resources are being freed */ if (gve_get_device_resources_ok(priv)) { + /* detach the stats report */ + err = gve_adminq_report_stats(priv, 0, 0x0, GVE_STATS_REPORT_TIMER_PERIOD); + if (err) { + dev_err(&priv->pdev->dev, + "Failed to detach stats report: err=%d\n", err); + gve_trigger_reset(priv); + } err = gve_adminq_deconfigure_device_resources(priv); if (err) { dev_err(&priv->pdev->dev, @@ -308,6 +386,7 @@ static void gve_teardown_device_resources(struct gve_priv *priv) } gve_free_counter_array(priv); gve_free_notify_blocks(priv); + gve_free_stats_report(priv); gve_clear_device_resources_ok(priv); } @@ -371,36 +450,37 @@ static int gve_create_rings(struct gve_priv *priv) int err; int i; - for (i = 0; i < priv->tx_cfg.num_queues; i++) { - err = gve_adminq_create_tx_queue(priv, i); - if (err) { - netif_err(priv, drv, priv->dev, "failed to create tx queue %d\n", - i); - /* This failure will trigger a reset - no need to clean - * up - */ - return err; - } - netif_dbg(priv, drv, priv->dev, "created tx queue %d\n", i); + err = gve_adminq_create_tx_queues(priv, priv->tx_cfg.num_queues); + if (err) { + netif_err(priv, drv, priv->dev, "failed to create %d tx queues\n", + priv->tx_cfg.num_queues); + /* This failure will trigger a reset - no need to clean + * up + */ + return err; } - for (i = 0; i < priv->rx_cfg.num_queues; i++) { - err = gve_adminq_create_rx_queue(priv, i); - if (err) { - netif_err(priv, drv, priv->dev, "failed to create rx queue %d\n", - i); - /* This failure will trigger a reset - no need to clean - * up - */ - return err; - } - /* Rx data ring has been prefilled with packet buffers at - * queue allocation time. - * Write the doorbell to provide descriptor slots and packet - * buffers to the NIC. + netif_dbg(priv, drv, priv->dev, "created %d tx queues\n", + priv->tx_cfg.num_queues); + + err = gve_adminq_create_rx_queues(priv, priv->rx_cfg.num_queues); + if (err) { + netif_err(priv, drv, priv->dev, "failed to create %d rx queues\n", + priv->rx_cfg.num_queues); + /* This failure will trigger a reset - no need to clean + * up */ - gve_rx_write_doorbell(priv, &priv->rx[i]); - netif_dbg(priv, drv, priv->dev, "created rx queue %d\n", i); + return err; } + netif_dbg(priv, drv, priv->dev, "created %d rx queues\n", + priv->rx_cfg.num_queues); + + /* Rx data ring has been prefilled with packet buffers at queue + * allocation time. + * Write the doorbell to provide descriptor slots and packet buffers + * to the NIC. + */ + for (i = 0; i < priv->rx_cfg.num_queues; i++) + gve_rx_write_doorbell(priv, &priv->rx[i]); return 0; } @@ -458,34 +538,23 @@ free_tx: static int gve_destroy_rings(struct gve_priv *priv) { int err; - int i; - for (i = 0; i < priv->tx_cfg.num_queues; i++) { - err = gve_adminq_destroy_tx_queue(priv, i); - if (err) { - netif_err(priv, drv, priv->dev, - "failed to destroy tx queue %d\n", - i); - /* This failure will trigger a reset - no need to clean - * up - */ - return err; - } - netif_dbg(priv, drv, priv->dev, "destroyed tx queue %d\n", i); + err = gve_adminq_destroy_tx_queues(priv, priv->tx_cfg.num_queues); + if (err) { + netif_err(priv, drv, priv->dev, + "failed to destroy tx queues\n"); + /* This failure will trigger a reset - no need to clean up */ + return err; } - for (i = 0; i < priv->rx_cfg.num_queues; i++) { - err = gve_adminq_destroy_rx_queue(priv, i); - if (err) { - netif_err(priv, drv, priv->dev, - "failed to destroy rx queue %d\n", - i); - /* This failure will trigger a reset - no need to clean - * up - */ - return err; - } - netif_dbg(priv, drv, priv->dev, "destroyed rx queue %d\n", i); + netif_dbg(priv, drv, priv->dev, "destroyed tx queues\n"); + err = gve_adminq_destroy_rx_queues(priv, priv->rx_cfg.num_queues); + if (err) { + netif_err(priv, drv, priv->dev, + "failed to destroy rx queues\n"); + /* This failure will trigger a reset - no need to clean up */ + return err; } + netif_dbg(priv, drv, priv->dev, "destroyed rx queues\n"); return 0; } @@ -514,14 +583,18 @@ static void gve_free_rings(struct gve_priv *priv) } } -int gve_alloc_page(struct device *dev, struct page **page, dma_addr_t *dma, +int gve_alloc_page(struct gve_priv *priv, struct device *dev, + struct page **page, dma_addr_t *dma, enum dma_data_direction dir) { *page = alloc_page(GFP_KERNEL); - if (!*page) + if (!*page) { + priv->page_alloc_fail++; return -ENOMEM; + } *dma = dma_map_page(dev, *page, 0, PAGE_SIZE, dir); if (dma_mapping_error(dev, *dma)) { + priv->dma_mapping_error++; put_page(*page); return -ENOMEM; } @@ -556,7 +629,7 @@ static int gve_alloc_queue_page_list(struct gve_priv *priv, u32 id, return -ENOMEM; for (i = 0; i < pages; i++) { - err = gve_alloc_page(&priv->pdev->dev, &qpl->pages[i], + err = gve_alloc_page(priv, &priv->pdev->dev, &qpl->pages[i], &qpl->page_buses[i], gve_qpl_dma_dir(priv, id)); /* caller handles clean up */ @@ -695,8 +768,14 @@ static int gve_open(struct net_device *dev) goto reset; gve_set_device_rings_ok(priv); + if (gve_get_report_stats(priv)) + mod_timer(&priv->stats_report_timer, + round_jiffies(jiffies + + msecs_to_jiffies(priv->stats_report_timer_period))); + gve_turnup(priv); - netif_carrier_on(dev); + queue_work(priv->gve_wq, &priv->service_task); + priv->interface_up_cnt++; return 0; free_rings: @@ -735,9 +814,11 @@ static int gve_close(struct net_device *dev) goto err; gve_clear_device_rings_ok(priv); } + del_timer_sync(&priv->stats_report_timer); gve_free_rings(priv); gve_free_qpls(priv); + priv->interface_down_cnt++; return 0; err: @@ -817,6 +898,7 @@ static void gve_turndown(struct gve_priv *priv) netif_tx_disable(priv->dev); gve_clear_napi_enabled(priv); + gve_clear_report_stats(priv); } static void gve_turnup(struct gve_priv *priv) @@ -867,6 +949,10 @@ static void gve_handle_status(struct gve_priv *priv, u32 status) dev_info(&priv->pdev->dev, "Device requested reset.\n"); gve_set_do_reset(priv); } + if (GVE_DEVICE_STATUS_REPORT_STATS_MASK & status) { + priv->stats_report_trigger_cnt++; + gve_set_do_report_stats(priv); + } } static void gve_handle_reset(struct gve_priv *priv) @@ -885,16 +971,95 @@ static void gve_handle_reset(struct gve_priv *priv) } } -/* Handle NIC status register changes and reset requests */ +void gve_handle_report_stats(struct gve_priv *priv) +{ + int idx, stats_idx = 0, tx_bytes; + unsigned int start = 0; + struct stats *stats = priv->stats_report->stats; + + if (!gve_get_report_stats(priv)) + return; + + be64_add_cpu(&priv->stats_report->written_count, 1); + /* tx stats */ + if (priv->tx) { + for (idx = 0; idx < priv->tx_cfg.num_queues; idx++) { + do { + start = u64_stats_fetch_begin(&priv->tx[idx].statss); + tx_bytes = priv->tx[idx].bytes_done; + } while (u64_stats_fetch_retry(&priv->tx[idx].statss, start)); + stats[stats_idx++] = (struct stats) { + .stat_name = cpu_to_be32(TX_WAKE_CNT), + .value = cpu_to_be64(priv->tx[idx].wake_queue), + .queue_id = cpu_to_be32(idx), + }; + stats[stats_idx++] = (struct stats) { + .stat_name = cpu_to_be32(TX_STOP_CNT), + .value = cpu_to_be64(priv->tx[idx].stop_queue), + .queue_id = cpu_to_be32(idx), + }; + stats[stats_idx++] = (struct stats) { + .stat_name = cpu_to_be32(TX_FRAMES_SENT), + .value = cpu_to_be64(priv->tx[idx].req), + .queue_id = cpu_to_be32(idx), + }; + stats[stats_idx++] = (struct stats) { + .stat_name = cpu_to_be32(TX_BYTES_SENT), + .value = cpu_to_be64(tx_bytes), + .queue_id = cpu_to_be32(idx), + }; + stats[stats_idx++] = (struct stats) { + .stat_name = cpu_to_be32(TX_LAST_COMPLETION_PROCESSED), + .value = cpu_to_be64(priv->tx[idx].done), + .queue_id = cpu_to_be32(idx), + }; + } + } + /* rx stats */ + if (priv->rx) { + for (idx = 0; idx < priv->rx_cfg.num_queues; idx++) { + stats[stats_idx++] = (struct stats) { + .stat_name = cpu_to_be32(RX_NEXT_EXPECTED_SEQUENCE), + .value = cpu_to_be64(priv->rx[idx].desc.seqno), + .queue_id = cpu_to_be32(idx), + }; + stats[stats_idx++] = (struct stats) { + .stat_name = cpu_to_be32(RX_BUFFERS_POSTED), + .value = cpu_to_be64(priv->rx[0].fill_cnt), + .queue_id = cpu_to_be32(idx), + }; + } + } +} + +static void gve_handle_link_status(struct gve_priv *priv, bool link_status) +{ + if (!gve_get_napi_enabled(priv)) + return; + + if (link_status == netif_carrier_ok(priv->dev)) + return; + + if (link_status) { + netdev_info(priv->dev, "Device link is up.\n"); + netif_carrier_on(priv->dev); + } else { + netdev_info(priv->dev, "Device link is down.\n"); + netif_carrier_off(priv->dev); + } +} + +/* Handle NIC status register changes, reset requests and report stats */ static void gve_service_task(struct work_struct *work) { struct gve_priv *priv = container_of(work, struct gve_priv, service_task); + u32 status = ioread32be(&priv->reg_bar0->device_status); - gve_handle_status(priv, - ioread32be(&priv->reg_bar0->device_status)); + gve_handle_status(priv, status); gve_handle_reset(priv); + gve_handle_link_status(priv, GVE_DEVICE_STATUS_LINK_STATUS_MASK & status); } static int gve_init_priv(struct gve_priv *priv, bool skip_describe_device) @@ -924,7 +1089,7 @@ static int gve_init_priv(struct gve_priv *priv, bool skip_describe_device) priv->dev->max_mtu = PAGE_SIZE; err = gve_adminq_set_mtu(priv, priv->dev->mtu); if (err) { - netif_err(priv, drv, priv->dev, "Could not set mtu"); + dev_err(&priv->pdev->dev, "Could not set mtu"); goto err; } } @@ -964,10 +1129,10 @@ static int gve_init_priv(struct gve_priv *priv, bool skip_describe_device) priv->rx_cfg.num_queues); } - netif_info(priv, drv, priv->dev, "TX queues %d, RX queues %d\n", - priv->tx_cfg.num_queues, priv->rx_cfg.num_queues); - netif_info(priv, drv, priv->dev, "Max TX queues %d, Max RX queues %d\n", - priv->tx_cfg.max_queues, priv->rx_cfg.max_queues); + dev_info(&priv->pdev->dev, "TX queues %d, RX queues %d\n", + priv->tx_cfg.num_queues, priv->rx_cfg.num_queues); + dev_info(&priv->pdev->dev, "Max TX queues %d, Max RX queues %d\n", + priv->tx_cfg.max_queues, priv->rx_cfg.max_queues); setup_device: err = gve_setup_device_resources(priv); @@ -1047,6 +1212,10 @@ int gve_reset(struct gve_priv *priv, bool attempt_teardown) /* Set it all back up */ err = gve_reset_recovery(priv, was_up); gve_clear_reset_in_progress(priv); + priv->reset_cnt++; + priv->interface_up_cnt = 0; + priv->interface_down_cnt = 0; + priv->stats_report_trigger_cnt = 0; return err; } @@ -1149,6 +1318,7 @@ static int gve_probe(struct pci_dev *pdev, const struct pci_device_id *ent) priv->db_bar2 = db_bar; priv->service_task_flags = 0x0; priv->state_flags = 0x0; + priv->ethtool_flags = 0x0; gve_set_probe_in_progress(priv); priv->gve_wq = alloc_ordered_workqueue("gve", 0); @@ -1158,6 +1328,7 @@ static int gve_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto abort_with_netdev; } INIT_WORK(&priv->service_task, gve_service_task); + INIT_WORK(&priv->stats_report_task, gve_stats_report_task); priv->tx_cfg.max_queues = max_tx_queues; priv->rx_cfg.max_queues = max_rx_queues; diff --git a/drivers/net/ethernet/google/gve/gve_register.h b/drivers/net/ethernet/google/gve/gve_register.h index 84ab8893aadd..fb655463c357 100644 --- a/drivers/net/ethernet/google/gve/gve_register.h +++ b/drivers/net/ethernet/google/gve/gve_register.h @@ -23,5 +23,6 @@ struct gve_registers { enum gve_device_status_flags { GVE_DEVICE_STATUS_RESET_MASK = BIT(1), GVE_DEVICE_STATUS_LINK_STATUS_MASK = BIT(2), + GVE_DEVICE_STATUS_REPORT_STATS_MASK = BIT(3), }; #endif /* _GVE_REGISTER_H_ */ diff --git a/drivers/net/ethernet/google/gve/gve_rx.c b/drivers/net/ethernet/google/gve/gve_rx.c index 9f52e72ff641..008fa897a3e6 100644 --- a/drivers/net/ethernet/google/gve/gve_rx.c +++ b/drivers/net/ethernet/google/gve/gve_rx.c @@ -225,7 +225,8 @@ static enum pkt_hash_types gve_rss_type(__be16 pkt_flags) return PKT_HASH_TYPE_L2; } -static struct sk_buff *gve_rx_copy(struct net_device *dev, +static struct sk_buff *gve_rx_copy(struct gve_rx_ring *rx, + struct net_device *dev, struct napi_struct *napi, struct gve_rx_slot_page_info *page_info, u16 len) @@ -242,6 +243,11 @@ static struct sk_buff *gve_rx_copy(struct net_device *dev, skb_copy_to_linear_data(skb, va, len); skb->protocol = eth_type_trans(skb, dev); + + u64_stats_update_begin(&rx->statss); + rx->rx_copied_pkt++; + u64_stats_update_end(&rx->statss); + return skb; } @@ -284,8 +290,12 @@ static bool gve_rx(struct gve_rx_ring *rx, struct gve_rx_desc *rx_desc, u16 len; /* drop this packet */ - if (unlikely(rx_desc->flags_seq & GVE_RXF_ERR)) + if (unlikely(rx_desc->flags_seq & GVE_RXF_ERR)) { + u64_stats_update_begin(&rx->statss); + rx->rx_desc_err_dropped_pkt++; + u64_stats_update_end(&rx->statss); return true; + } len = be16_to_cpu(rx_desc->len) - GVE_RX_PAD; page_info = &rx->data.page_info[idx]; @@ -300,11 +310,14 @@ static bool gve_rx(struct gve_rx_ring *rx, struct gve_rx_desc *rx_desc, if (PAGE_SIZE == 4096) { if (len <= priv->rx_copybreak) { /* Just copy small packets */ - skb = gve_rx_copy(dev, napi, page_info, len); + skb = gve_rx_copy(rx, dev, napi, page_info, len); + u64_stats_update_begin(&rx->statss); + rx->rx_copybreak_pkt++; + u64_stats_update_end(&rx->statss); goto have_skb; } if (unlikely(!gve_can_recycle_pages(dev))) { - skb = gve_rx_copy(dev, napi, page_info, len); + skb = gve_rx_copy(rx, dev, napi, page_info, len); goto have_skb; } pagecount = page_count(page_info->page); @@ -314,8 +327,12 @@ static bool gve_rx(struct gve_rx_ring *rx, struct gve_rx_desc *rx_desc, * stack. */ skb = gve_rx_add_frags(dev, napi, page_info, len); - if (!skb) + if (!skb) { + u64_stats_update_begin(&rx->statss); + rx->rx_skb_alloc_fail++; + u64_stats_update_end(&rx->statss); return true; + } /* Make sure the kernel stack can't release the page */ get_page(page_info->page); /* "flip" to other packet buffer on this page */ @@ -324,21 +341,25 @@ static bool gve_rx(struct gve_rx_ring *rx, struct gve_rx_desc *rx_desc, /* We have previously passed the other half of this * page up the stack, but it has not yet been freed. */ - skb = gve_rx_copy(dev, napi, page_info, len); + skb = gve_rx_copy(rx, dev, napi, page_info, len); } else { WARN(pagecount < 1, "Pagecount should never be < 1"); return false; } } else { - skb = gve_rx_copy(dev, napi, page_info, len); + skb = gve_rx_copy(rx, dev, napi, page_info, len); } have_skb: /* We didn't manage to allocate an skb but we haven't had any * reset worthy failures. */ - if (!skb) + if (!skb) { + u64_stats_update_begin(&rx->statss); + rx->rx_skb_alloc_fail++; + u64_stats_update_end(&rx->statss); return true; + } if (likely(feat & NETIF_F_RXCSUM)) { /* NIC passes up the partial sum */ diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.c b/drivers/net/ethernet/hisilicon/hns/hnae.c index 08339278c722..00fafc0f8512 100644 --- a/drivers/net/ethernet/hisilicon/hns/hnae.c +++ b/drivers/net/ethernet/hisilicon/hns/hnae.c @@ -270,7 +270,7 @@ static void hnae_fini_queue(struct hnae_queue *q) hnae_fini_ring(&q->rx_ring); } -/** +/* * ae_chain - define ae chain head */ static RAW_NOTIFIER_HEAD(ae_chain); @@ -438,7 +438,7 @@ EXPORT_SYMBOL(hnae_ae_register); /** * hnae_ae_unregister - unregisters a HNAE AE engine - * @cdev: the device to unregister + * @hdev: the device to unregister */ void hnae_ae_unregister(struct hnae_ae_dev *hdev) { diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c index b43dec0560a8..b98244f75ab9 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c @@ -13,8 +13,6 @@ #include "hns_dsaf_ppe.h" #include "hns_dsaf_rcb.h" -#define AE_NAME_PORT_ID_IDX 6 - static struct hns_mac_cb *hns_get_mac_cb(struct hnae_handle *handle) { struct hnae_vf_cb *vf_cb = hns_ae_get_vf_cb(handle); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c index 9a907947ba19..4a448138b4ec 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c @@ -374,11 +374,12 @@ static void hns_mac_param_get(struct mac_params *param, } /** - *hns_mac_queue_config_bc_en - set broadcast rx&tx enable - *@mac_cb: mac device - *@queue: queue number - *@en:enable - *retuen 0 - success , negative --fail + * hns_mac_queue_config_bc_en - set broadcast rx&tx enable + * @mac_cb: mac device + * @port_num: queue number + * @vlan_id: vlan id` + * @enable: enable + * return 0 - success , negative --fail */ static int hns_mac_port_config_bc_en(struct hns_mac_cb *mac_cb, u32 port_num, u16 vlan_id, bool enable) @@ -408,11 +409,11 @@ static int hns_mac_port_config_bc_en(struct hns_mac_cb *mac_cb, } /** - *hns_mac_vm_config_bc_en - set broadcast rx&tx enable - *@mac_cb: mac device - *@vmid: vm id - *@en:enable - *retuen 0 - success , negative --fail + * hns_mac_vm_config_bc_en - set broadcast rx&tx enable + * @mac_cb: mac device + * @vmid: vm id + * @enable: enable + * return 0 - success , negative --fail */ int hns_mac_vm_config_bc_en(struct hns_mac_cb *mac_cb, u32 vmid, bool enable) { @@ -542,8 +543,8 @@ void hns_mac_stop(struct hns_mac_cb *mac_cb) /** * hns_mac_get_autoneg - get auto autonegotiation * @mac_cb: mac control block - * @enable: enable or not - * retuen 0 - success , negative --fail + * @auto_neg: output pointer to autoneg result + * return 0 - success , negative --fail */ void hns_mac_get_autoneg(struct hns_mac_cb *mac_cb, u32 *auto_neg) { @@ -560,7 +561,7 @@ void hns_mac_get_autoneg(struct hns_mac_cb *mac_cb, u32 *auto_neg) * @mac_cb: mac control block * @rx_en: rx enable status * @tx_en: tx enable status - * retuen 0 - success , negative --fail + * return 0 - success , negative --fail */ void hns_mac_get_pauseparam(struct hns_mac_cb *mac_cb, u32 *rx_en, u32 *tx_en) { @@ -578,7 +579,7 @@ void hns_mac_get_pauseparam(struct hns_mac_cb *mac_cb, u32 *rx_en, u32 *tx_en) * hns_mac_set_autoneg - set auto autonegotiation * @mac_cb: mac control block * @enable: enable or not - * retuen 0 - success , negative --fail + * return 0 - success , negative --fail */ int hns_mac_set_autoneg(struct hns_mac_cb *mac_cb, u8 enable) { @@ -623,7 +624,7 @@ int hns_mac_set_pauseparam(struct hns_mac_cb *mac_cb, u32 rx_en, u32 tx_en) /** * hns_mac_init_ex - mac init * @mac_cb: mac control block - * retuen 0 - success , negative --fail + * return 0 - success , negative --fail */ static int hns_mac_init_ex(struct hns_mac_cb *mac_cb) { @@ -800,7 +801,6 @@ static const struct { /** *hns_mac_get_info - get mac information from device node *@mac_cb: mac device - *@np:device node * return: 0 --success, negative --fail */ static int hns_mac_get_info(struct hns_mac_cb *mac_cb) @@ -951,7 +951,7 @@ static int hns_mac_get_info(struct hns_mac_cb *mac_cb) /** * hns_mac_get_mode - get mac mode * @phy_if: phy interface - * retuen 0 - gmac, 1 - xgmac , negative --fail + * return 0 - gmac, 1 - xgmac , negative --fail */ static int hns_mac_get_mode(phy_interface_t phy_if) { diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c index acfa86e5296f..87d3db4666df 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c @@ -207,7 +207,7 @@ static int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) /** * hns_dsaf_sbm_link_sram_init_en - config dsaf_sbm_init_en - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_sbm_link_sram_init_en(struct dsaf_device *dsaf_dev) { @@ -216,8 +216,8 @@ static void hns_dsaf_sbm_link_sram_init_en(struct dsaf_device *dsaf_dev) /** * hns_dsaf_reg_cnt_clr_ce - config hns_dsaf_reg_cnt_clr_ce - * @dsaf_id: dsa fabric id - * @hns_dsaf_reg_cnt_clr_ce: config value + * @dsaf_dev: dsa fabric id + * @reg_cnt_clr_ce: config value */ static void hns_dsaf_reg_cnt_clr_ce(struct dsaf_device *dsaf_dev, u32 reg_cnt_clr_ce) @@ -228,8 +228,8 @@ hns_dsaf_reg_cnt_clr_ce(struct dsaf_device *dsaf_dev, u32 reg_cnt_clr_ce) /** * hns_ppe_qid_cfg - config ppe qid - * @dsaf_id: dsa fabric id - * @pppe_qid_cfg: value array + * @dsaf_dev: dsa fabric id + * @qid_cfg: value array */ static void hns_dsaf_ppe_qid_cfg(struct dsaf_device *dsaf_dev, u32 qid_cfg) @@ -285,8 +285,8 @@ static void hns_dsaf_inner_qid_cfg(struct dsaf_device *dsaf_dev) /** * hns_dsaf_sw_port_type_cfg - cfg sw type - * @dsaf_id: dsa fabric id - * @psw_port_type: array + * @dsaf_dev: dsa fabric id + * @port_type: array */ static void hns_dsaf_sw_port_type_cfg(struct dsaf_device *dsaf_dev, enum dsaf_sw_port_type port_type) @@ -303,8 +303,8 @@ static void hns_dsaf_sw_port_type_cfg(struct dsaf_device *dsaf_dev, /** * hns_dsaf_stp_port_type_cfg - cfg stp type - * @dsaf_id: dsa fabric id - * @pstp_port_type: array + * @dsaf_dev: dsa fabric id + * @port_type: array */ static void hns_dsaf_stp_port_type_cfg(struct dsaf_device *dsaf_dev, enum dsaf_stp_port_type port_type) @@ -323,7 +323,7 @@ static void hns_dsaf_stp_port_type_cfg(struct dsaf_device *dsaf_dev, (AE_IS_VER1((dev)->dsaf_ver) ? DSAF_SBM_NUM : DSAFV2_SBM_NUM) /** * hns_dsaf_sbm_cfg - config sbm - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_sbm_cfg(struct dsaf_device *dsaf_dev) { @@ -342,7 +342,7 @@ static void hns_dsaf_sbm_cfg(struct dsaf_device *dsaf_dev) /** * hns_dsaf_sbm_cfg_mib_en - config sbm - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static int hns_dsaf_sbm_cfg_mib_en(struct dsaf_device *dsaf_dev) { @@ -387,7 +387,7 @@ static int hns_dsaf_sbm_cfg_mib_en(struct dsaf_device *dsaf_dev) /** * hns_dsaf_sbm_bp_wl_cfg - config sbm - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_sbm_bp_wl_cfg(struct dsaf_device *dsaf_dev) { @@ -556,7 +556,7 @@ static void hns_dsafv2_sbm_bp_wl_cfg(struct dsaf_device *dsaf_dev) /** * hns_dsaf_voq_bp_all_thrd_cfg - voq - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_voq_bp_all_thrd_cfg(struct dsaf_device *dsaf_dev) { @@ -599,7 +599,7 @@ static void hns_dsaf_tbl_tcam_match_cfg( /** * hns_dsaf_tbl_tcam_data_cfg - tbl - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id * @ptbl_tcam_data: addr */ static void hns_dsaf_tbl_tcam_data_cfg( @@ -614,8 +614,8 @@ static void hns_dsaf_tbl_tcam_data_cfg( /** * dsaf_tbl_tcam_mcast_cfg - tbl - * @dsaf_id: dsa fabric id - * @ptbl_tcam_mcast: addr + * @dsaf_dev: dsa fabric id + * @mcast: addr */ static void hns_dsaf_tbl_tcam_mcast_cfg( struct dsaf_device *dsaf_dev, @@ -648,8 +648,8 @@ static void hns_dsaf_tbl_tcam_mcast_cfg( /** * hns_dsaf_tbl_tcam_ucast_cfg - tbl - * @dsaf_id: dsa fabric id - * @ptbl_tcam_ucast: addr + * @dsaf_dev: dsa fabric id + * @tbl_tcam_ucast: addr */ static void hns_dsaf_tbl_tcam_ucast_cfg( struct dsaf_device *dsaf_dev, @@ -674,8 +674,8 @@ static void hns_dsaf_tbl_tcam_ucast_cfg( /** * hns_dsaf_tbl_line_cfg - tbl - * @dsaf_id: dsa fabric id - * @ptbl_lin: addr + * @dsaf_dev: dsa fabric id + * @tbl_lin: addr */ static void hns_dsaf_tbl_line_cfg(struct dsaf_device *dsaf_dev, struct dsaf_tbl_line_cfg *tbl_lin) @@ -695,7 +695,7 @@ static void hns_dsaf_tbl_line_cfg(struct dsaf_device *dsaf_dev, /** * hns_dsaf_tbl_tcam_mcast_pul - tbl - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_tbl_tcam_mcast_pul(struct dsaf_device *dsaf_dev) { @@ -710,7 +710,7 @@ static void hns_dsaf_tbl_tcam_mcast_pul(struct dsaf_device *dsaf_dev) /** * hns_dsaf_tbl_line_pul - tbl - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_tbl_line_pul(struct dsaf_device *dsaf_dev) { @@ -725,7 +725,7 @@ static void hns_dsaf_tbl_line_pul(struct dsaf_device *dsaf_dev) /** * hns_dsaf_tbl_tcam_data_mcast_pul - tbl - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_tbl_tcam_data_mcast_pul( struct dsaf_device *dsaf_dev) @@ -743,7 +743,7 @@ static void hns_dsaf_tbl_tcam_data_mcast_pul( /** * hns_dsaf_tbl_tcam_data_ucast_pul - tbl - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_tbl_tcam_data_ucast_pul( struct dsaf_device *dsaf_dev) @@ -768,8 +768,7 @@ void hns_dsaf_set_promisc_mode(struct dsaf_device *dsaf_dev, u32 en) /** * hns_dsaf_tbl_stat_en - tbl - * @dsaf_id: dsa fabric id - * @ptbl_stat_en: addr + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_tbl_stat_en(struct dsaf_device *dsaf_dev) { @@ -785,7 +784,7 @@ static void hns_dsaf_tbl_stat_en(struct dsaf_device *dsaf_dev) /** * hns_dsaf_rocee_bp_en - rocee back press enable - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_rocee_bp_en(struct dsaf_device *dsaf_dev) { @@ -852,9 +851,9 @@ static void hns_dsaf_int_tbl_src_clr(struct dsaf_device *dsaf_dev, /** * hns_dsaf_single_line_tbl_cfg - INT - * @dsaf_id: dsa fabric id - * @address: - * @ptbl_line: + * @dsaf_dev: dsa fabric id + * @address: the address + * @ptbl_line: the line */ static void hns_dsaf_single_line_tbl_cfg( struct dsaf_device *dsaf_dev, @@ -876,9 +875,10 @@ static void hns_dsaf_single_line_tbl_cfg( /** * hns_dsaf_tcam_uc_cfg - INT - * @dsaf_id: dsa fabric id - * @address, - * @ptbl_tcam_data, + * @dsaf_dev: dsa fabric id + * @address: the address + * @ptbl_tcam_data: the data + * @ptbl_tcam_ucast: unicast */ static void hns_dsaf_tcam_uc_cfg( struct dsaf_device *dsaf_dev, u32 address, @@ -904,7 +904,8 @@ static void hns_dsaf_tcam_uc_cfg( * @dsaf_dev: dsa fabric device struct pointer * @address: tcam index * @ptbl_tcam_data: tcam data struct pointer - * @ptbl_tcam_mcast: tcam mask struct pointer, it must be null for HNSv1 + * @ptbl_tcam_mask: tcam mask struct pointer, it must be null for HNSv1 + * @ptbl_tcam_mcast: tcam data struct pointer */ static void hns_dsaf_tcam_mc_cfg( struct dsaf_device *dsaf_dev, u32 address, @@ -933,8 +934,10 @@ static void hns_dsaf_tcam_mc_cfg( /** * hns_dsaf_tcam_uc_cfg_vague - INT * @dsaf_dev: dsa fabric device struct pointer - * @address, - * @ptbl_tcam_data, + * @address: the address + * @tcam_data: the data + * @tcam_mask: the mask + * @tcam_uc: the unicast data */ static void hns_dsaf_tcam_uc_cfg_vague(struct dsaf_device *dsaf_dev, u32 address, @@ -960,10 +963,10 @@ static void hns_dsaf_tcam_uc_cfg_vague(struct dsaf_device *dsaf_dev, /** * hns_dsaf_tcam_mc_cfg_vague - INT * @dsaf_dev: dsa fabric device struct pointer - * @address, - * @ptbl_tcam_data, - * @ptbl_tcam_mask - * @ptbl_tcam_mcast + * @address: the address + * @tcam_data: the data + * @tcam_mask: the mask + * @tcam_mc: the multicast data */ static void hns_dsaf_tcam_mc_cfg_vague(struct dsaf_device *dsaf_dev, u32 address, @@ -988,8 +991,8 @@ static void hns_dsaf_tcam_mc_cfg_vague(struct dsaf_device *dsaf_dev, /** * hns_dsaf_tcam_mc_invld - INT - * @dsaf_id: dsa fabric id - * @address + * @dsaf_dev: dsa fabric id + * @address: the address */ static void hns_dsaf_tcam_mc_invld(struct dsaf_device *dsaf_dev, u32 address) { @@ -1024,10 +1027,10 @@ hns_dsaf_tcam_addr_get(struct dsaf_drv_tbl_tcam_key *mac_key, u8 *addr) /** * hns_dsaf_tcam_uc_get - INT - * @dsaf_id: dsa fabric id - * @address - * @ptbl_tcam_data - * @ptbl_tcam_ucast + * @dsaf_dev: dsa fabric id + * @address: the address + * @ptbl_tcam_data: the data + * @ptbl_tcam_ucast: unicast */ static void hns_dsaf_tcam_uc_get( struct dsaf_device *dsaf_dev, u32 address, @@ -1077,10 +1080,10 @@ static void hns_dsaf_tcam_uc_get( /** * hns_dsaf_tcam_mc_get - INT - * @dsaf_id: dsa fabric id - * @address - * @ptbl_tcam_data - * @ptbl_tcam_ucast + * @dsaf_dev: dsa fabric id + * @address: the address + * @ptbl_tcam_data: the data + * @ptbl_tcam_mcast: tcam multicast data */ static void hns_dsaf_tcam_mc_get( struct dsaf_device *dsaf_dev, u32 address, @@ -1127,7 +1130,7 @@ static void hns_dsaf_tcam_mc_get( /** * hns_dsaf_tbl_line_init - INT - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_tbl_line_init(struct dsaf_device *dsaf_dev) { @@ -1141,7 +1144,7 @@ static void hns_dsaf_tbl_line_init(struct dsaf_device *dsaf_dev) /** * hns_dsaf_tbl_tcam_init - INT - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_tbl_tcam_init(struct dsaf_device *dsaf_dev) { @@ -1156,7 +1159,9 @@ static void hns_dsaf_tbl_tcam_init(struct dsaf_device *dsaf_dev) /** * hns_dsaf_pfc_en_cfg - dsaf pfc pause cfg - * @mac_cb: mac contrl block + * @dsaf_dev: dsa fabric id + * @mac_id: mac contrl block + * @tc_en: traffic class */ static void hns_dsaf_pfc_en_cfg(struct dsaf_device *dsaf_dev, int mac_id, int tc_en) @@ -1209,8 +1214,7 @@ void hns_dsaf_get_rx_mac_pause_en(struct dsaf_device *dsaf_dev, int mac_id, /** * hns_dsaf_tbl_tcam_init - INT - * @dsaf_id: dsa fabric id - * @dsaf_mode + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_comm_init(struct dsaf_device *dsaf_dev) { @@ -1263,7 +1267,7 @@ static void hns_dsaf_comm_init(struct dsaf_device *dsaf_dev) /** * hns_dsaf_inode_init - INT - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_inode_init(struct dsaf_device *dsaf_dev) { @@ -1315,7 +1319,7 @@ static void hns_dsaf_inode_init(struct dsaf_device *dsaf_dev) /** * hns_dsaf_sbm_init - INT - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static int hns_dsaf_sbm_init(struct dsaf_device *dsaf_dev) { @@ -1369,7 +1373,7 @@ static int hns_dsaf_sbm_init(struct dsaf_device *dsaf_dev) /** * hns_dsaf_tbl_init - INT - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_tbl_init(struct dsaf_device *dsaf_dev) { @@ -1381,7 +1385,7 @@ static void hns_dsaf_tbl_init(struct dsaf_device *dsaf_dev) /** * hns_dsaf_voq_init - INT - * @dsaf_id: dsa fabric id + * @dsaf_dev: dsa fabric id */ static void hns_dsaf_voq_init(struct dsaf_device *dsaf_dev) { @@ -1435,7 +1439,7 @@ static void hns_dsaf_remove_hw(struct dsaf_device *dsaf_dev) /** * hns_dsaf_init - init dsa fabric * @dsaf_dev: dsa fabric device struct pointer - * retuen 0 - success , negative --fail + * return 0 - success , negative --fail */ static int hns_dsaf_init(struct dsaf_device *dsaf_dev) { @@ -2099,7 +2103,7 @@ static struct dsaf_device *hns_dsaf_alloc_dev(struct device *dev, /** * hns_dsaf_free_dev - free dev mem - * @dev: struct device pointer + * @dsaf_dev: struct device pointer */ static void hns_dsaf_free_dev(struct dsaf_device *dsaf_dev) { @@ -2108,9 +2112,9 @@ static void hns_dsaf_free_dev(struct dsaf_device *dsaf_dev) /** * dsaf_pfc_unit_cnt - set pfc unit count - * @dsaf_id: dsa fabric id - * @pport_rate: value array - * @pdsaf_pfc_unit_cnt: value array + * @dsaf_dev: dsa fabric id + * @mac_id: id in use + * @rate: value array */ static void hns_dsaf_pfc_unit_cnt(struct dsaf_device *dsaf_dev, int mac_id, enum dsaf_port_rate_mode rate) @@ -2139,8 +2143,9 @@ static void hns_dsaf_pfc_unit_cnt(struct dsaf_device *dsaf_dev, int mac_id, /** * dsaf_port_work_rate_cfg - fifo - * @dsaf_id: dsa fabric id - * @xge_ge_work_mode + * @dsaf_dev: dsa fabric id + * @mac_id: mac contrl block + * @rate_mode: value array */ static void hns_dsaf_port_work_rate_cfg(struct dsaf_device *dsaf_dev, int mac_id, @@ -2253,7 +2258,8 @@ void hns_dsaf_update_stats(struct dsaf_device *dsaf_dev, u32 node_num) /** *hns_dsaf_get_regs - dump dsaf regs - *@dsaf_dev: dsaf device + *@ddev: dsaf device + *@port: port *@data:data for value of regs */ void hns_dsaf_get_regs(struct dsaf_device *ddev, u32 port, void *data) @@ -2690,6 +2696,7 @@ void hns_dsaf_get_stats(struct dsaf_device *ddev, u64 *data, int port) /** *hns_dsaf_get_sset_count - get dsaf string set count + *@dsaf_dev: dsaf device *@stringset: type of values in data *return dsaf string name count */ @@ -2711,6 +2718,7 @@ int hns_dsaf_get_sset_count(struct dsaf_device *dsaf_dev, int stringset) *@stringset:srting set index *@data:strings name value *@port:port index + *@dsaf_dev: dsaf device */ void hns_dsaf_get_strings(int stringset, u8 *data, int port, struct dsaf_device *dsaf_dev) @@ -2943,7 +2951,7 @@ int hns_dsaf_wait_pkt_clean(struct dsaf_device *dsaf_dev, int port) /** * dsaf_probe - probo dsaf dev * @pdev: dasf platform device - * retuen 0 - success , negative --fail + * return 0 - success , negative --fail */ static int hns_dsaf_probe(struct platform_device *pdev) { @@ -3038,8 +3046,8 @@ module_platform_driver(g_dsaf_driver); /** * hns_dsaf_roce_reset - reset dsaf and roce * @dsaf_fwnode: Pointer to framework node for the dasf - * @enable: false - request reset , true - drop reset - * retuen 0 - success , negative -fail + * @dereset: false - request reset , true - drop reset + * return 0 - success , negative -fail */ int hns_dsaf_roce_reset(struct fwnode_handle *dsaf_fwnode, bool dereset) { diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c index a769273b36f7..a9aca8c24e90 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c @@ -330,11 +330,12 @@ static void hns_dsaf_xge_srst_by_port_acpi(struct dsaf_device *dsaf_dev, * hns_dsaf_srst_chns - reset dsaf channels * @dsaf_dev: dsaf device struct pointer * @msk: xbar channels mask value: + * @dereset: false - request reset , true - drop reset + * * bit0-5 for xge0-5 * bit6-11 for ppe0-5 * bit12-17 for roce0-5 * bit18-19 for com/dfx - * @dereset: false - request reset , true - drop reset */ static void hns_dsaf_srst_chns(struct dsaf_device *dsaf_dev, u32 msk, bool dereset) @@ -353,11 +354,12 @@ hns_dsaf_srst_chns(struct dsaf_device *dsaf_dev, u32 msk, bool dereset) * hns_dsaf_srst_chns - reset dsaf channels * @dsaf_dev: dsaf device struct pointer * @msk: xbar channels mask value: + * @dereset: false - request reset , true - drop reset + * * bit0-5 for xge0-5 * bit6-11 for ppe0-5 * bit12-17 for roce0-5 * bit18-19 for com/dfx - * @dereset: false - request reset , true - drop reset */ static void hns_dsaf_srst_chns_acpi(struct dsaf_device *dsaf_dev, u32 msk, bool dereset) @@ -612,7 +614,8 @@ static int hns_mac_get_sfp_prsnt_acpi(struct hns_mac_cb *mac_cb, int *sfp_prsnt) /** * hns_mac_config_sds_loopback - set loop back for serdes * @mac_cb: mac control block - * retuen 0 == success + * @en: enable or disable + * return 0 == success */ static int hns_mac_config_sds_loopback(struct hns_mac_cb *mac_cb, bool en) { diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.c index 2b34b553acf3..d0f8b1fff333 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.c @@ -66,8 +66,8 @@ hns_ppe_common_get_ioaddr(struct ppe_common_cb *ppe_common) /** * hns_ppe_common_get_cfg - get ppe common config * @dsaf_dev: dasf device - * comm_index: common index - * retuen 0 - success , negative --fail + * @comm_index: common index + * return 0 - success , negative --fail */ static int hns_ppe_common_get_cfg(struct dsaf_device *dsaf_dev, int comm_index) { @@ -143,7 +143,7 @@ static void hns_ppe_set_vlan_strip(struct hns_ppe_cb *ppe_cb, int en) /** * hns_ppe_checksum_hw - set ppe checksum caculate - * @ppe_device: ppe device + * @ppe_cb: ppe device * @value: value */ static void hns_ppe_checksum_hw(struct hns_ppe_cb *ppe_cb, u32 value) @@ -179,7 +179,7 @@ static void hns_ppe_set_qid(struct ppe_common_cb *ppe_common, u32 qid) /** * hns_ppe_set_port_mode - set port mode - * @ppe_device: ppe device + * @ppe_cb: ppe device * @mode: port mode */ static void hns_ppe_set_port_mode(struct hns_ppe_cb *ppe_cb, @@ -344,7 +344,7 @@ static void hns_ppe_init_hw(struct hns_ppe_cb *ppe_cb) /** * ppe_uninit_hw - uninit ppe - * @ppe_device: ppe device + * @ppe_cb: ppe device */ static void hns_ppe_uninit_hw(struct hns_ppe_cb *ppe_cb) { @@ -384,7 +384,8 @@ void hns_ppe_uninit(struct dsaf_device *dsaf_dev) /** * hns_ppe_reset - reinit ppe/rcb hw * @dsaf_dev: dasf device - * retuen void + * @ppe_common_index: the index + * return void */ void hns_ppe_reset_common(struct dsaf_device *dsaf_dev, u8 ppe_common_index) { @@ -455,7 +456,7 @@ int hns_ppe_get_regs_count(void) /** * ppe_get_strings - get ppe srting - * @ppe_device: ppe device + * @ppe_cb: ppe device * @stringset: string set type * @data: output string */ @@ -513,7 +514,7 @@ void hns_ppe_get_stats(struct hns_ppe_cb *ppe_cb, u64 *data) /** * hns_ppe_init - init ppe device * @dsaf_dev: dasf device - * retuen 0 - success , negative --fail + * return 0 - success , negative --fail */ int hns_ppe_init(struct dsaf_device *dsaf_dev) { diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c index 5453597ec629..b6c8910cf7ba 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c @@ -34,7 +34,7 @@ /** *hns_rcb_wait_fbd_clean - clean fbd *@qs: ring struct pointer array - *@qnum: num of array + *@q_num: num of array *@flag: tx or rx flag */ void hns_rcb_wait_fbd_clean(struct hnae_queue **qs, int q_num, u32 flag) @@ -191,7 +191,8 @@ void hns_rcbv2_int_clr_hw(struct hnae_queue *q, u32 flag) /** *hns_rcb_ring_enable_hw - enable ring - *@ring: rcb ring + *@q: rcb ring + *@val: value to write */ void hns_rcb_ring_enable_hw(struct hnae_queue *q, u32 val) { @@ -844,7 +845,7 @@ void hns_rcb_update_stats(struct hnae_queue *queue) /** *hns_rcb_get_stats - get rcb statistic - *@ring: rcb ring + *@queue: rcb ring *@data:statistic value */ void hns_rcb_get_stats(struct hnae_queue *queue, u64 *data) diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.c index 0a3dbab2dfc9..7e3609ce112a 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.c @@ -130,7 +130,7 @@ static void hns_xgmac_lf_rf_control_init(struct mac_driver *mac_drv) /** *hns_xgmac_enable - enable xgmac port - *@drv: mac driver + *@mac_drv: mac driver *@mode: mode of mac port */ static void hns_xgmac_enable(void *mac_drv, enum mac_commom_mode mode) @@ -242,7 +242,8 @@ static void hns_xgmac_config_pad_and_crc(void *mac_drv, u8 newval) /** *hns_xgmac_pausefrm_cfg - set pause param about xgmac *@mac_drv: mac driver - *@newval:enable of pad and crc + *@rx_en: enable receive + *@tx_en: enable transmit */ static void hns_xgmac_pausefrm_cfg(void *mac_drv, u32 rx_en, u32 tx_en) { @@ -490,7 +491,6 @@ static void hns_xgmac_get_link_status(void *mac_drv, u32 *link_stat) /** *hns_xgmac_get_regs - dump xgmac regs *@mac_drv: mac driver - *@cmd:ethtool cmd *@data:data for value of regs */ static void hns_xgmac_get_regs(void *mac_drv, void *data) diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c index 22522f8a5299..858cb293152a 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c @@ -11,6 +11,7 @@ #include <linux/io.h> #include <linux/ip.h> #include <linux/ipv6.h> +#include <linux/irq.h> #include <linux/module.h> #include <linux/phy.h> #include <linux/platform_device.h> @@ -557,10 +558,7 @@ static int hns_nic_poll_rx_skb(struct hns_nic_ring_data *ring_data, va = (unsigned char *)desc_cb->buf + desc_cb->page_offset; /* prefetch first cache line of first page */ - prefetch(va); -#if L1_CACHE_BYTES < 128 - prefetch(va + L1_CACHE_BYTES); -#endif + net_prefetch(va); skb = *out_skb = napi_alloc_skb(&ring_data->napi, HNS_RX_HEAD_SIZE); @@ -754,6 +752,8 @@ static void hns_update_rx_rate(struct hnae_ring *ring) /** * smooth_alg - smoothing algrithm for adjusting coalesce parameter + * @new_param: new value + * @old_param: old value **/ static u32 smooth_alg(u32 new_param, u32 old_param) { @@ -1293,6 +1293,7 @@ static int hns_nic_init_irq(struct hns_nic_priv *priv) rd->ring->ring_name[RCB_RING_NAME_LEN - 1] = '\0'; + irq_set_status_flags(rd->ring->irq, IRQ_NOAUTOEN); ret = request_irq(rd->ring->irq, hns_irq_handle, 0, rd->ring->ring_name, rd); if (ret) { @@ -1300,7 +1301,6 @@ static int hns_nic_init_irq(struct hns_nic_priv *priv) rd->ring->irq); goto out_free_irq; } - disable_irq(rd->ring->irq); cpu = hns_nic_init_affinity_mask(h->q_num, i, rd->ring, &rd->mask); @@ -1831,9 +1831,8 @@ static int hns_nic_uc_unsync(struct net_device *netdev, } /** - * nic_set_multicast_list - set mutl mac address - * @netdev: net device - * @p: mac address + * hns_set_multicast_list - set mutl mac address + * @ndev: net device * * return void */ diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c index 14e60c9e491d..7165da0ee9aa 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c @@ -462,7 +462,7 @@ static int __lb_clean_rings(struct hns_nic_priv *priv, } /** - * nic_run_loopback_test - run loopback test + * __lb_run_test - run loopback test * @ndev: net device * @loop_mode: loopback mode */ @@ -971,7 +971,7 @@ static void hns_get_strings(struct net_device *netdev, u32 stringset, u8 *data) } /** - * nic_get_sset_count - get string set count witch returned by nic_get_strings. + * hns_get_sset_count - get string set count returned by nic_get_strings * @netdev: net device * @stringset: string set index, 0: self test string; 1: statistics string. * @@ -1027,7 +1027,7 @@ static int hns_phy_led_set(struct net_device *netdev, int value) } /** - * nic_set_phys_id - set phy identify LED. + * hns_set_phys_id - set phy identify LED. * @netdev: net device * @state: LED state. * @@ -1125,7 +1125,7 @@ static void hns_get_regs(struct net_device *net_dev, struct ethtool_regs *cmd, } /** - * nic_get_regs_len - get total register len. + * hns_get_regs_len - get total register len. * @net_dev: net device * * Return total register len. diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h index 088550db2de7..912c51e327d6 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h +++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h @@ -34,6 +34,13 @@ #define HNAE3_MIN_VECTOR_NUM 2 /* first one for misc, another for IO */ +/* Device version */ +#define HNAE3_DEVICE_VERSION_V1 0x00020 +#define HNAE3_DEVICE_VERSION_V2 0x00021 +#define HNAE3_DEVICE_VERSION_V3 0x00030 + +#define HNAE3_PCI_REVISION_BIT_SIZE 8 + /* Device IDs */ #define HNAE3_DEV_ID_GE 0xA220 #define HNAE3_DEV_ID_25GE 0xA221 @@ -42,8 +49,9 @@ #define HNAE3_DEV_ID_50GE_RDMA 0xA224 #define HNAE3_DEV_ID_50GE_RDMA_MACSEC 0xA225 #define HNAE3_DEV_ID_100G_RDMA_MACSEC 0xA226 -#define HNAE3_DEV_ID_100G_VF 0xA22E -#define HNAE3_DEV_ID_100G_RDMA_DCB_PFC_VF 0xA22F +#define HNAE3_DEV_ID_200G_RDMA 0xA228 +#define HNAE3_DEV_ID_VF 0xA22E +#define HNAE3_DEV_ID_RDMA_DCB_PFC_VF 0xA22F #define HNAE3_CLASS_NAME_SIZE 16 @@ -53,8 +61,6 @@ #define HNAE3_KNIC_CLIENT_INITED_B 0x3 #define HNAE3_UNIC_CLIENT_INITED_B 0x4 #define HNAE3_ROCE_CLIENT_INITED_B 0x5 -#define HNAE3_DEV_SUPPORT_FD_B 0x6 -#define HNAE3_DEV_SUPPORT_GRO_B 0x7 #define HNAE3_DEV_SUPPORT_ROCE_DCB_BITS (BIT(HNAE3_DEV_SUPPORT_DCB_B) |\ BIT(HNAE3_DEV_SUPPORT_ROCE_B)) @@ -65,11 +71,67 @@ #define hnae3_dev_dcb_supported(hdev) \ hnae3_get_bit((hdev)->ae_dev->flag, HNAE3_DEV_SUPPORT_DCB_B) +enum HNAE3_DEV_CAP_BITS { + HNAE3_DEV_SUPPORT_FD_B, + HNAE3_DEV_SUPPORT_GRO_B, + HNAE3_DEV_SUPPORT_FEC_B, + HNAE3_DEV_SUPPORT_UDP_GSO_B, + HNAE3_DEV_SUPPORT_QB_B, + HNAE3_DEV_SUPPORT_FD_FORWARD_TC_B, + HNAE3_DEV_SUPPORT_PTP_B, + HNAE3_DEV_SUPPORT_INT_QL_B, + HNAE3_DEV_SUPPORT_SIMPLE_BD_B, + HNAE3_DEV_SUPPORT_TX_PUSH_B, + HNAE3_DEV_SUPPORT_PHY_IMP_B, + HNAE3_DEV_SUPPORT_TQP_TXRX_INDEP_B, + HNAE3_DEV_SUPPORT_HW_PAD_B, + HNAE3_DEV_SUPPORT_STASH_B, +}; + #define hnae3_dev_fd_supported(hdev) \ - hnae3_get_bit((hdev)->ae_dev->flag, HNAE3_DEV_SUPPORT_FD_B) + test_bit(HNAE3_DEV_SUPPORT_FD_B, (hdev)->ae_dev->caps) #define hnae3_dev_gro_supported(hdev) \ - hnae3_get_bit((hdev)->ae_dev->flag, HNAE3_DEV_SUPPORT_GRO_B) + test_bit(HNAE3_DEV_SUPPORT_GRO_B, (hdev)->ae_dev->caps) + +#define hnae3_dev_fec_supported(hdev) \ + test_bit(HNAE3_DEV_SUPPORT_FEC_B, (hdev)->ae_dev->caps) + +#define hnae3_dev_udp_gso_supported(hdev) \ + test_bit(HNAE3_DEV_SUPPORT_UDP_GSO_B, (hdev)->ae_dev->caps) + +#define hnae3_dev_qb_supported(hdev) \ + test_bit(HNAE3_DEV_SUPPORT_QB_B, (hdev)->ae_dev->caps) + +#define hnae3_dev_fd_forward_tc_supported(hdev) \ + test_bit(HNAE3_DEV_SUPPORT_FD_FORWARD_TC_B, (hdev)->ae_dev->caps) + +#define hnae3_dev_ptp_supported(hdev) \ + test_bit(HNAE3_DEV_SUPPORT_PTP_B, (hdev)->ae_dev->caps) + +#define hnae3_dev_int_ql_supported(hdev) \ + test_bit(HNAE3_DEV_SUPPORT_INT_QL_B, (hdev)->ae_dev->caps) + +#define hnae3_dev_simple_bd_supported(hdev) \ + test_bit(HNAE3_DEV_SUPPORT_SIMPLE_BD_B, (hdev)->ae_dev->caps) + +#define hnae3_dev_tx_push_supported(hdev) \ + test_bit(HNAE3_DEV_SUPPORT_TX_PUSH_B, (hdev)->ae_dev->caps) + +#define hnae3_dev_phy_imp_supported(hdev) \ + test_bit(HNAE3_DEV_SUPPORT_PHY_IMP_B, (hdev)->ae_dev->caps) + +#define hnae3_dev_tqp_txrx_indep_supported(hdev) \ + test_bit(HNAE3_DEV_SUPPORT_TQP_TXRX_INDEP_B, (hdev)->ae_dev->caps) + +#define hnae3_dev_hw_pad_supported(hdev) \ + test_bit(HNAE3_DEV_SUPPORT_HW_PAD_B, (hdev)->ae_dev->caps) + +#define hnae3_dev_stash_supported(hdev) \ + test_bit(HNAE3_DEV_SUPPORT_STASH_B, (hdev)->ae_dev->caps) + +#define hnae3_ae_dev_tqp_txrx_indep_supported(ae_dev) \ + test_bit(HNAE3_DEV_SUPPORT_TQP_TXRX_INDEP_B, (ae_dev)->caps) #define ring_ptr_move_fw(ring, p) \ ((ring)->p = ((ring)->p + 1) % (ring)->desc_num) @@ -152,6 +214,7 @@ enum hnae3_hw_error_type { HNAE3_PPU_POISON_ERROR, HNAE3_CMDQ_ECC_ERROR, HNAE3_IMP_RD_POISON_ERROR, + HNAE3_ROCEE_AXI_RESP_ERROR, }; enum hnae3_reset_type { @@ -207,6 +270,17 @@ struct hnae3_ring_chain_node { #define HNAE3_IS_TX_RING(node) \ (((node)->flag & (1 << HNAE3_RING_TYPE_B)) == HNAE3_RING_TYPE_TX) +/* device specification info from firmware */ +struct hnae3_dev_specs { + u32 mac_entry_num; /* number of mac-vlan table entry */ + u32 mng_entry_num; /* number of manager table entry */ + u32 max_tm_rate; + u16 rss_ind_tbl_size; + u16 rss_key_size; + u16 int_ql_max; /* max value of interrupt coalesce based on INT_QL */ + u8 max_non_tso_bd_num; /* max BD number of one non-TSO packet */ +}; + struct hnae3_client_ops { int (*init_instance)(struct hnae3_handle *handle); void (*uninit_instance)(struct hnae3_handle *handle, bool reset); @@ -227,12 +301,16 @@ struct hnae3_client { struct list_head node; }; +#define HNAE3_DEV_CAPS_MAX_NUM 96 struct hnae3_ae_dev { struct pci_dev *pdev; const struct hnae3_ae_ops *ops; struct list_head node; u32 flag; unsigned long hw_err_reset_req; + struct hnae3_dev_specs dev_specs; + u32 dev_version; + unsigned long caps[BITS_TO_LONGS(HNAE3_DEV_CAPS_MAX_NUM)]; void *priv; }; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c index fe7fb565da19..dc9a85745e62 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c @@ -15,11 +15,12 @@ static struct dentry *hns3_dbgfs_root; static int hns3_dbg_queue_info(struct hnae3_handle *h, const char *cmd_buf) { + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(h->pdev); struct hns3_nic_priv *priv = h->priv; struct hns3_enet_ring *ring; u32 base_add_l, base_add_h; u32 queue_num, queue_max; - u32 value, i = 0; + u32 value, i; int cnt; if (!priv->ring) { @@ -118,8 +119,25 @@ static int hns3_dbg_queue_info(struct hnae3_handle *h, value = readl_relaxed(ring->tqp->io_base + HNS3_RING_TX_RING_PKTNUM_RECORD_REG); - dev_info(&h->pdev->dev, "TX(%u) RING PKTNUM: %u\n\n", i, - value); + dev_info(&h->pdev->dev, "TX(%u) RING PKTNUM: %u\n", i, value); + + value = readl_relaxed(ring->tqp->io_base + HNS3_RING_EN_REG); + dev_info(&h->pdev->dev, "TX/RX(%u) RING EN: %s\n", i, + value ? "enable" : "disable"); + + if (hnae3_ae_dev_tqp_txrx_indep_supported(ae_dev)) { + value = readl_relaxed(ring->tqp->io_base + + HNS3_RING_TX_EN_REG); + dev_info(&h->pdev->dev, "TX(%u) RING EN: %s\n", i, + value ? "enable" : "disable"); + + value = readl_relaxed(ring->tqp->io_base + + HNS3_RING_RX_EN_REG); + dev_info(&h->pdev->dev, "RX(%u) RING EN: %s\n", i, + value ? "enable" : "disable"); + } + + dev_info(&h->pdev->dev, "\n"); } return 0; @@ -244,6 +262,8 @@ static void hns3_dbg_help(struct hnae3_handle *h) dev_info(&h->pdev->dev, "queue info <number>\n"); dev_info(&h->pdev->dev, "queue map\n"); dev_info(&h->pdev->dev, "bd info <q_num> <bd index>\n"); + dev_info(&h->pdev->dev, "dev capability\n"); + dev_info(&h->pdev->dev, "dev spec\n"); if (!hns3_is_phys_func(h->pdev)) return; @@ -264,6 +284,7 @@ static void hns3_dbg_help(struct hnae3_handle *h) dev_info(&h->pdev->dev, "dump qs shaper [qs id]\n"); dev_info(&h->pdev->dev, "dump uc mac list <func id>\n"); dev_info(&h->pdev->dev, "dump mc mac list <func id>\n"); + dev_info(&h->pdev->dev, "dump intr\n"); memset(printf_buf, 0, HNS3_DBG_BUF_LEN); strncat(printf_buf, "dump reg [[bios common] [ssu <port_id>]", @@ -284,6 +305,52 @@ static void hns3_dbg_help(struct hnae3_handle *h) dev_info(&h->pdev->dev, "%s", printf_buf); } +static void hns3_dbg_dev_caps(struct hnae3_handle *h) +{ + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(h->pdev); + unsigned long *caps; + + caps = ae_dev->caps; + + dev_info(&h->pdev->dev, "support FD: %s\n", + test_bit(HNAE3_DEV_SUPPORT_FD_B, caps) ? "yes" : "no"); + dev_info(&h->pdev->dev, "support GRO: %s\n", + test_bit(HNAE3_DEV_SUPPORT_GRO_B, caps) ? "yes" : "no"); + dev_info(&h->pdev->dev, "support FEC: %s\n", + test_bit(HNAE3_DEV_SUPPORT_FEC_B, caps) ? "yes" : "no"); + dev_info(&h->pdev->dev, "support UDP GSO: %s\n", + test_bit(HNAE3_DEV_SUPPORT_UDP_GSO_B, caps) ? "yes" : "no"); + dev_info(&h->pdev->dev, "support PTP: %s\n", + test_bit(HNAE3_DEV_SUPPORT_PTP_B, caps) ? "yes" : "no"); + dev_info(&h->pdev->dev, "support INT QL: %s\n", + test_bit(HNAE3_DEV_SUPPORT_INT_QL_B, caps) ? "yes" : "no"); +} + +static void hns3_dbg_dev_specs(struct hnae3_handle *h) +{ + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(h->pdev); + struct hnae3_dev_specs *dev_specs = &ae_dev->dev_specs; + struct hnae3_knic_private_info *kinfo = &h->kinfo; + struct hns3_nic_priv *priv = h->priv; + + dev_info(priv->dev, "MAC entry num: %u\n", dev_specs->mac_entry_num); + dev_info(priv->dev, "MNG entry num: %u\n", dev_specs->mng_entry_num); + dev_info(priv->dev, "MAX non tso bd num: %u\n", + dev_specs->max_non_tso_bd_num); + dev_info(priv->dev, "RSS ind tbl size: %u\n", + dev_specs->rss_ind_tbl_size); + dev_info(priv->dev, "RSS key size: %u\n", dev_specs->rss_key_size); + dev_info(priv->dev, "RSS size: %u\n", kinfo->rss_size); + dev_info(priv->dev, "Allocated RSS size: %u\n", kinfo->req_rss_size); + dev_info(priv->dev, "Task queue pairs numbers: %u\n", kinfo->num_tqps); + + dev_info(priv->dev, "RX buffer length: %u\n", kinfo->rx_buf_len); + dev_info(priv->dev, "Desc num per TX queue: %u\n", kinfo->num_tx_desc); + dev_info(priv->dev, "Desc num per RX queue: %u\n", kinfo->num_rx_desc); + dev_info(priv->dev, "Total number of enabled TCs: %u\n", kinfo->num_tc); + dev_info(priv->dev, "MAX INT QL: %u\n", dev_specs->int_ql_max); +} + static ssize_t hns3_dbg_cmd_read(struct file *filp, char __user *buffer, size_t count, loff_t *ppos) { @@ -359,6 +426,10 @@ static ssize_t hns3_dbg_cmd_write(struct file *filp, const char __user *buffer, ret = hns3_dbg_queue_map(handle); else if (strncmp(cmd_buf, "bd info", 7) == 0) ret = hns3_dbg_bd_info(handle, cmd_buf); + else if (strncmp(cmd_buf, "dev capability", 14) == 0) + hns3_dbg_dev_caps(handle); + else if (strncmp(cmd_buf, "dev spec", 8) == 0) + hns3_dbg_dev_specs(handle); else if (handle->ae_algo->ops->dbg_run_cmd) ret = handle->ae_algo->ops->dbg_run_cmd(handle, cmd_buf); else diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index a4f1d515e5e0..a362516a3185 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -81,8 +81,10 @@ static const struct pci_device_id hns3_pci_tbl[] = { HNAE3_DEV_SUPPORT_ROCE_DCB_BITS}, {PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_100G_RDMA_MACSEC), HNAE3_DEV_SUPPORT_ROCE_DCB_BITS}, - {PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_100G_VF), 0}, - {PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_100G_RDMA_DCB_PFC_VF), + {PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_200G_RDMA), + HNAE3_DEV_SUPPORT_ROCE_DCB_BITS}, + {PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_VF), 0}, + {PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_RDMA_DCB_PFC_VF), HNAE3_DEV_SUPPORT_ROCE_DCB_BITS}, /* required last entry */ {0, } @@ -623,27 +625,15 @@ void hns3_request_update_promisc_mode(struct hnae3_handle *handle) ops->request_update_promisc_mode(handle); } -int hns3_update_promisc_mode(struct net_device *netdev, u8 promisc_flags) -{ - struct hns3_nic_priv *priv = netdev_priv(netdev); - struct hnae3_handle *h = priv->ae_handle; - - if (h->ae_algo->ops->set_promisc_mode) { - return h->ae_algo->ops->set_promisc_mode(h, - promisc_flags & HNAE3_UPE, - promisc_flags & HNAE3_MPE); - } - - return 0; -} - void hns3_enable_vlan_filter(struct net_device *netdev, bool enable) { struct hns3_nic_priv *priv = netdev_priv(netdev); struct hnae3_handle *h = priv->ae_handle; + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(h->pdev); bool last_state; - if (h->pdev->revision >= 0x21 && h->ae_algo->ops->enable_vlan_filter) { + if (ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2 && + h->ae_algo->ops->enable_vlan_filter) { last_state = h->netdev_flags & HNAE3_VLAN_FLTR ? true : false; if (enable != last_state) { netdev_info(netdev, @@ -706,12 +696,19 @@ static int hns3_set_tso(struct sk_buff *skb, u32 *paylen, /* normal or tunnel packet */ l4_offset = l4.hdr - skb->data; - hdr_len = (l4.tcp->doff << 2) + l4_offset; /* remove payload length from inner pseudo checksum when tso */ l4_paylen = skb->len - l4_offset; - csum_replace_by_diff(&l4.tcp->check, - (__force __wsum)htonl(l4_paylen)); + + if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) { + hdr_len = sizeof(*l4.udp) + l4_offset; + csum_replace_by_diff(&l4.udp->check, + (__force __wsum)htonl(l4_paylen)); + } else { + hdr_len = (l4.tcp->doff << 2) + l4_offset; + csum_replace_by_diff(&l4.tcp->check, + (__force __wsum)htonl(l4_paylen)); + } /* find the txbd field values */ *paylen = skb->len - hdr_len; @@ -1194,21 +1191,23 @@ static unsigned int hns3_skb_bd_num(struct sk_buff *skb, unsigned int *bd_size, return bd_num; } -static unsigned int hns3_tx_bd_num(struct sk_buff *skb, unsigned int *bd_size) +static unsigned int hns3_tx_bd_num(struct sk_buff *skb, unsigned int *bd_size, + u8 max_non_tso_bd_num) { struct sk_buff *frag_skb; unsigned int bd_num = 0; /* If the total len is within the max bd limit */ if (likely(skb->len <= HNS3_MAX_BD_SIZE && !skb_has_frag_list(skb) && - skb_shinfo(skb)->nr_frags < HNS3_MAX_NON_TSO_BD_NUM)) + skb_shinfo(skb)->nr_frags < max_non_tso_bd_num)) return skb_shinfo(skb)->nr_frags + 1U; /* The below case will always be linearized, return * HNS3_MAX_BD_NUM_TSO + 1U to make sure it is linearized. */ if (unlikely(skb->len > HNS3_MAX_TSO_SIZE || - (!skb_is_gso(skb) && skb->len > HNS3_MAX_NON_TSO_SIZE))) + (!skb_is_gso(skb) && skb->len > + HNS3_MAX_NON_TSO_SIZE(max_non_tso_bd_num)))) return HNS3_MAX_TSO_BD_NUM + 1U; bd_num = hns3_skb_bd_num(skb, bd_size, bd_num); @@ -1233,31 +1232,34 @@ static unsigned int hns3_gso_hdr_len(struct sk_buff *skb) return skb_inner_transport_offset(skb) + inner_tcp_hdrlen(skb); } -/* HW need every continuous 8 buffer data to be larger than MSS, - * we simplify it by ensuring skb_headlen + the first continuous - * 7 frags to to be larger than gso header len + mss, and the remaining - * continuous 7 frags to be larger than MSS except the last 7 frags. +/* HW need every continuous max_non_tso_bd_num buffer data to be larger + * than MSS, we simplify it by ensuring skb_headlen + the first continuous + * max_non_tso_bd_num - 1 frags to be larger than gso header len + mss, + * and the remaining continuous max_non_tso_bd_num - 1 frags to be larger + * than MSS except the last max_non_tso_bd_num - 1 frags. */ static bool hns3_skb_need_linearized(struct sk_buff *skb, unsigned int *bd_size, - unsigned int bd_num) + unsigned int bd_num, u8 max_non_tso_bd_num) { unsigned int tot_len = 0; int i; - for (i = 0; i < HNS3_MAX_NON_TSO_BD_NUM - 1U; i++) + for (i = 0; i < max_non_tso_bd_num - 1U; i++) tot_len += bd_size[i]; - /* ensure the first 8 frags is greater than mss + header */ - if (tot_len + bd_size[HNS3_MAX_NON_TSO_BD_NUM - 1U] < + /* ensure the first max_non_tso_bd_num frags is greater than + * mss + header + */ + if (tot_len + bd_size[max_non_tso_bd_num - 1U] < skb_shinfo(skb)->gso_size + hns3_gso_hdr_len(skb)) return true; - /* ensure every continuous 7 buffer is greater than mss - * except the last one. + /* ensure every continuous max_non_tso_bd_num - 1 buffer is greater + * than mss except the last one. */ - for (i = 0; i < bd_num - HNS3_MAX_NON_TSO_BD_NUM; i++) { + for (i = 0; i < bd_num - max_non_tso_bd_num; i++) { tot_len -= bd_size[i]; - tot_len += bd_size[i + HNS3_MAX_NON_TSO_BD_NUM - 1U]; + tot_len += bd_size[i + max_non_tso_bd_num - 1U]; if (tot_len < skb_shinfo(skb)->gso_size) return true; @@ -1268,7 +1270,7 @@ static bool hns3_skb_need_linearized(struct sk_buff *skb, unsigned int *bd_size, void hns3_shinfo_pack(struct skb_shared_info *shinfo, __u32 *size) { - int i = 0; + int i; for (i = 0; i < MAX_SKB_FRAGS; i++) size[i] = skb_frag_size(&shinfo->frags[i]); @@ -1279,14 +1281,16 @@ static int hns3_nic_maybe_stop_tx(struct hns3_enet_ring *ring, struct sk_buff *skb) { struct hns3_nic_priv *priv = netdev_priv(netdev); + u8 max_non_tso_bd_num = priv->max_non_tso_bd_num; unsigned int bd_size[HNS3_MAX_TSO_BD_NUM + 1U]; unsigned int bd_num; - bd_num = hns3_tx_bd_num(skb, bd_size); - if (unlikely(bd_num > HNS3_MAX_NON_TSO_BD_NUM)) { + bd_num = hns3_tx_bd_num(skb, bd_size, max_non_tso_bd_num); + if (unlikely(bd_num > max_non_tso_bd_num)) { if (bd_num <= HNS3_MAX_TSO_BD_NUM && skb_is_gso(skb) && - !hns3_skb_need_linearized(skb, bd_size, bd_num)) { - trace_hns3_over_8bd(skb); + !hns3_skb_need_linearized(skb, bd_size, bd_num, + max_non_tso_bd_num)) { + trace_hns3_over_max_bd(skb); goto out; } @@ -1296,8 +1300,8 @@ static int hns3_nic_maybe_stop_tx(struct hns3_enet_ring *ring, bd_num = hns3_tx_bd_count(skb->len); if ((skb_is_gso(skb) && bd_num > HNS3_MAX_TSO_BD_NUM) || (!skb_is_gso(skb) && - bd_num > HNS3_MAX_NON_TSO_BD_NUM)) { - trace_hns3_over_8bd(skb); + bd_num > max_non_tso_bd_num)) { + trace_hns3_over_max_bd(skb); return -ENOMEM; } @@ -1397,6 +1401,27 @@ static int hns3_fill_skb_to_desc(struct hns3_enet_ring *ring, return bd_num; } +static void hns3_tx_doorbell(struct hns3_enet_ring *ring, int num, + bool doorbell) +{ + ring->pending_buf += num; + + if (!doorbell) { + u64_stats_update_begin(&ring->syncp); + ring->stats.tx_more++; + u64_stats_update_end(&ring->syncp); + return; + } + + if (!ring->pending_buf) + return; + + writel(ring->pending_buf, + ring->tqp->io_base + HNS3_RING_TX_RING_TAIL_REG); + ring->pending_buf = 0; + WRITE_ONCE(ring->last_to_use, ring->next_to_use); +} + netdev_tx_t hns3_nic_net_xmit(struct sk_buff *skb, struct net_device *netdev) { struct hns3_nic_priv *priv = netdev_priv(netdev); @@ -1405,11 +1430,14 @@ netdev_tx_t hns3_nic_net_xmit(struct sk_buff *skb, struct net_device *netdev) int pre_ntu, next_to_use_head; struct sk_buff *frag_skb; int bd_num = 0; + bool doorbell; int ret; /* Hardware can only handle short frames above 32 bytes */ - if (skb_put_padto(skb, HNS3_MIN_TX_LEN)) + if (skb_put_padto(skb, HNS3_MIN_TX_LEN)) { + hns3_tx_doorbell(ring, 0, !netdev_xmit_more()); return NETDEV_TX_OK; + } /* Prefetch the data used later */ prefetch(skb->data); @@ -1420,6 +1448,7 @@ netdev_tx_t hns3_nic_net_xmit(struct sk_buff *skb, struct net_device *netdev) u64_stats_update_begin(&ring->syncp); ring->stats.tx_busy++; u64_stats_update_end(&ring->syncp); + hns3_tx_doorbell(ring, 0, true); return NETDEV_TX_BUSY; } else if (ret == -ENOMEM) { u64_stats_update_begin(&ring->syncp); @@ -1460,11 +1489,9 @@ netdev_tx_t hns3_nic_net_xmit(struct sk_buff *skb, struct net_device *netdev) /* Complete translate all packets */ dev_queue = netdev_get_tx_queue(netdev, ring->queue_index); - netdev_tx_sent_queue(dev_queue, skb->len); - - wmb(); /* Commit all data before submit */ - - hnae3_queue_xmit(ring->tqp, bd_num); + doorbell = __netdev_tx_sent_queue(dev_queue, skb->len, + netdev_xmit_more()); + hns3_tx_doorbell(ring, bd_num, doorbell); return NETDEV_TX_OK; @@ -1473,6 +1500,7 @@ fill_err: out_err_tx_ok: dev_kfree_skb_any(skb); + hns3_tx_doorbell(ring, 0, !netdev_xmit_more()); return NETDEV_TX_OK; } @@ -1853,13 +1881,13 @@ static bool hns3_get_tx_timeo_queue_info(struct net_device *ndev) tx_ring->next_to_clean, napi->state); netdev_info(ndev, - "tx_pkts: %llu, tx_bytes: %llu, io_err_cnt: %llu, sw_err_cnt: %llu\n", + "tx_pkts: %llu, tx_bytes: %llu, sw_err_cnt: %llu, tx_pending: %d\n", tx_ring->stats.tx_pkts, tx_ring->stats.tx_bytes, - tx_ring->stats.io_err_cnt, tx_ring->stats.sw_err_cnt); + tx_ring->stats.sw_err_cnt, tx_ring->pending_buf); netdev_info(ndev, - "seg_pkt_cnt: %llu, tx_err_cnt: %llu, restart_queue: %llu, tx_busy: %llu\n", - tx_ring->stats.seg_pkt_cnt, tx_ring->stats.tx_err_cnt, + "seg_pkt_cnt: %llu, tx_more: %llu, restart_queue: %llu, tx_busy: %llu\n", + tx_ring->stats.seg_pkt_cnt, tx_ring->stats.tx_more, tx_ring->stats.restart_queue, tx_ring->stats.tx_busy); /* When mac received many pause frames continuous, it's unable to send @@ -2034,9 +2062,10 @@ bool hns3_is_phys_func(struct pci_dev *pdev) case HNAE3_DEV_ID_50GE_RDMA: case HNAE3_DEV_ID_50GE_RDMA_MACSEC: case HNAE3_DEV_ID_100G_RDMA_MACSEC: + case HNAE3_DEV_ID_200G_RDMA: return true; - case HNAE3_DEV_ID_100G_VF: - case HNAE3_DEV_ID_100G_RDMA_DCB_PFC_VF: + case HNAE3_DEV_ID_VF: + case HNAE3_DEV_ID_RDMA_DCB_PFC_VF: return false; default: dev_warn(&pdev->dev, "un-recognized pci device-id %u", @@ -2061,15 +2090,6 @@ static void hns3_disable_sriov(struct pci_dev *pdev) pci_disable_sriov(pdev); } -static void hns3_get_dev_capability(struct pci_dev *pdev, - struct hnae3_ae_dev *ae_dev) -{ - if (pdev->revision >= 0x21) { - hnae3_set_bit(ae_dev->flag, HNAE3_DEV_SUPPORT_FD_B, 1); - hnae3_set_bit(ae_dev->flag, HNAE3_DEV_SUPPORT_GRO_B, 1); - } -} - /* hns3_probe - Device initialization routine * @pdev: PCI device information struct * @ent: entry in hns3_pci_tbl @@ -2091,7 +2111,6 @@ static int hns3_probe(struct pci_dev *pdev, const struct pci_device_id *ent) ae_dev->pdev = pdev; ae_dev->flag = ent->driver_data; - hns3_get_dev_capability(pdev, ae_dev); pci_set_drvdata(pdev, ae_dev); ret = hnae3_register_ae_dev(ae_dev); @@ -2252,6 +2271,7 @@ static void hns3_set_default_feature(struct net_device *netdev) { struct hnae3_handle *h = hns3_get_handle(netdev); struct pci_dev *pdev = h->pdev; + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev); netdev->priv_flags |= IFF_UNICAST_FLT; @@ -2289,7 +2309,7 @@ static void hns3_set_default_feature(struct net_device *netdev) NETIF_F_GSO_UDP_TUNNEL_CSUM | NETIF_F_SCTP_CRC | NETIF_F_FRAGLIST; - if (pdev->revision >= 0x21) { + if (ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) { netdev->hw_features |= NETIF_F_GRO_HW; netdev->features |= NETIF_F_GRO_HW; @@ -2298,6 +2318,13 @@ static void hns3_set_default_feature(struct net_device *netdev) netdev->features |= NETIF_F_NTUPLE; } } + + if (test_bit(HNAE3_DEV_SUPPORT_UDP_GSO_B, ae_dev->caps)) { + netdev->hw_features |= NETIF_F_GSO_UDP_L4; + netdev->features |= NETIF_F_GSO_UDP_L4; + netdev->vlan_features |= NETIF_F_GSO_UDP_L4; + netdev->hw_enc_features |= NETIF_F_GSO_UDP_L4; + } } static int hns3_alloc_buffer(struct hns3_enet_ring *ring, @@ -2316,17 +2343,19 @@ static int hns3_alloc_buffer(struct hns3_enet_ring *ring, cb->buf = page_address(p); cb->length = hns3_page_size(ring); cb->type = DESC_TYPE_PAGE; + page_ref_add(p, USHRT_MAX - 1); + cb->pagecnt_bias = USHRT_MAX; return 0; } static void hns3_free_buffer(struct hns3_enet_ring *ring, - struct hns3_desc_cb *cb) + struct hns3_desc_cb *cb, int budget) { if (cb->type == DESC_TYPE_SKB) - dev_kfree_skb_any((struct sk_buff *)cb->priv); - else if (!HNAE3_IS_TX_RING(ring)) - put_page((struct page *)cb->priv); + napi_consume_skb(cb->priv, budget); + else if (!HNAE3_IS_TX_RING(ring) && cb->pagecnt_bias) + __page_frag_cache_drain(cb->priv, cb->pagecnt_bias); memset(cb, 0, sizeof(*cb)); } @@ -2358,7 +2387,8 @@ static void hns3_buffer_detach(struct hns3_enet_ring *ring, int i) ring->desc[i].addr = 0; } -static void hns3_free_buffer_detach(struct hns3_enet_ring *ring, int i) +static void hns3_free_buffer_detach(struct hns3_enet_ring *ring, int i, + int budget) { struct hns3_desc_cb *cb = &ring->desc_cb[i]; @@ -2366,7 +2396,7 @@ static void hns3_free_buffer_detach(struct hns3_enet_ring *ring, int i) return; hns3_buffer_detach(ring, i); - hns3_free_buffer(ring, cb); + hns3_free_buffer(ring, cb, budget); } static void hns3_free_buffers(struct hns3_enet_ring *ring) @@ -2374,7 +2404,7 @@ static void hns3_free_buffers(struct hns3_enet_ring *ring) int i; for (i = 0; i < ring->desc_num; i++) - hns3_free_buffer_detach(ring, i); + hns3_free_buffer_detach(ring, i, 0); } /* free desc along with its attached buffer */ @@ -2419,7 +2449,7 @@ static int hns3_alloc_and_map_buffer(struct hns3_enet_ring *ring, return 0; out_with_buf: - hns3_free_buffer(ring, cb); + hns3_free_buffer(ring, cb, 0); out: return ret; } @@ -2451,7 +2481,7 @@ static int hns3_alloc_ring_buffers(struct hns3_enet_ring *ring) out_buffer_fail: for (j = i - 1; j >= 0; j--) - hns3_free_buffer_detach(ring, j); + hns3_free_buffer_detach(ring, j, 0); return ret; } @@ -2478,71 +2508,62 @@ static void hns3_reuse_buffer(struct hns3_enet_ring *ring, int i) DMA_FROM_DEVICE); } -static void hns3_nic_reclaim_desc(struct hns3_enet_ring *ring, int head, - int *bytes, int *pkts) +static bool hns3_nic_reclaim_desc(struct hns3_enet_ring *ring, + int *bytes, int *pkts, int budget) { + /* pair with ring->last_to_use update in hns3_tx_doorbell(), + * smp_store_release() is not used in hns3_tx_doorbell() because + * the doorbell operation already have the needed barrier operation. + */ + int ltu = smp_load_acquire(&ring->last_to_use); int ntc = ring->next_to_clean; struct hns3_desc_cb *desc_cb; + bool reclaimed = false; + struct hns3_desc *desc; + + while (ltu != ntc) { + desc = &ring->desc[ntc]; + + if (le16_to_cpu(desc->tx.bdtp_fe_sc_vld_ra_ri) & + BIT(HNS3_TXD_VLD_B)) + break; - while (head != ntc) { desc_cb = &ring->desc_cb[ntc]; (*pkts) += (desc_cb->type == DESC_TYPE_SKB); (*bytes) += desc_cb->length; /* desc_cb will be cleaned, after hnae3_free_buffer_detach */ - hns3_free_buffer_detach(ring, ntc); + hns3_free_buffer_detach(ring, ntc, budget); if (++ntc == ring->desc_num) ntc = 0; /* Issue prefetch for next Tx descriptor */ prefetch(&ring->desc_cb[ntc]); + reclaimed = true; } + if (unlikely(!reclaimed)) + return false; + /* This smp_store_release() pairs with smp_load_acquire() in * ring_space called by hns3_nic_net_xmit. */ smp_store_release(&ring->next_to_clean, ntc); + return true; } -static int is_valid_clean_head(struct hns3_enet_ring *ring, int h) -{ - int u = ring->next_to_use; - int c = ring->next_to_clean; - - if (unlikely(h > ring->desc_num)) - return 0; - - return u > c ? (h > c && h <= u) : (h > c || h <= u); -} - -void hns3_clean_tx_ring(struct hns3_enet_ring *ring) +void hns3_clean_tx_ring(struct hns3_enet_ring *ring, int budget) { struct net_device *netdev = ring_to_netdev(ring); struct hns3_nic_priv *priv = netdev_priv(netdev); struct netdev_queue *dev_queue; int bytes, pkts; - int head; - - head = readl_relaxed(ring->tqp->io_base + HNS3_RING_TX_RING_HEAD_REG); - - if (is_ring_empty(ring) || head == ring->next_to_clean) - return; /* no data to poll */ - - rmb(); /* Make sure head is ready before touch any data */ - - if (unlikely(!is_valid_clean_head(ring, head))) { - hns3_rl_err(netdev, "wrong head (%d, %d-%d)\n", head, - ring->next_to_use, ring->next_to_clean); - - u64_stats_update_begin(&ring->syncp); - ring->stats.io_err_cnt++; - u64_stats_update_end(&ring->syncp); - return; - } bytes = 0; pkts = 0; - hns3_nic_reclaim_desc(ring, head, &bytes, &pkts); + + if (unlikely(!hns3_nic_reclaim_desc(ring, &bytes, &pkts, budget))) + return; ring->tqp_vector->tx_group.total_bytes += bytes; ring->tqp_vector->tx_group.total_packets += pkts; @@ -2614,8 +2635,7 @@ static void hns3_nic_alloc_rx_buffers(struct hns3_enet_ring *ring, ring_ptr_move_fw(ring, next_to_use); } - wmb(); /* Make all data has been write before submit */ - writel_relaxed(i, ring->tqp->io_base + HNS3_RING_RX_RING_HEAD_REG); + writel(i, ring->tqp->io_base + HNS3_RING_RX_RING_HEAD_REG); } static bool hns3_page_is_reusable(struct page *page) @@ -2624,6 +2644,11 @@ static bool hns3_page_is_reusable(struct page *page) !page_is_pfmemalloc(page); } +static bool hns3_can_reuse_page(struct hns3_desc_cb *cb) +{ + return (page_count(cb->priv) - cb->pagecnt_bias) == 1; +} + static void hns3_nic_reuse_page(struct sk_buff *skb, int i, struct hns3_enet_ring *ring, int pull_len, struct hns3_desc_cb *desc_cb) @@ -2632,6 +2657,7 @@ static void hns3_nic_reuse_page(struct sk_buff *skb, int i, int size = le16_to_cpu(desc->rx.size); u32 truesize = hns3_buf_size(ring); + desc_cb->pagecnt_bias--; skb_add_rx_frag(skb, i, desc_cb->priv, desc_cb->page_offset + pull_len, size - pull_len, truesize); @@ -2639,20 +2665,27 @@ static void hns3_nic_reuse_page(struct sk_buff *skb, int i, * when page_offset rollback to zero, flag default unreuse */ if (unlikely(!hns3_page_is_reusable(desc_cb->priv)) || - (!desc_cb->page_offset && page_count(desc_cb->priv) > 1)) + (!desc_cb->page_offset && !hns3_can_reuse_page(desc_cb))) { + __page_frag_cache_drain(desc_cb->priv, desc_cb->pagecnt_bias); return; + } /* Move offset up to the next cache line */ desc_cb->page_offset += truesize; if (desc_cb->page_offset + truesize <= hns3_page_size(ring)) { desc_cb->reuse_flag = 1; - /* Bump ref count on page before it is given */ - get_page(desc_cb->priv); - } else if (page_count(desc_cb->priv) == 1) { + } else if (hns3_can_reuse_page(desc_cb)) { desc_cb->reuse_flag = 1; desc_cb->page_offset = 0; - get_page(desc_cb->priv); + } else if (desc_cb->pagecnt_bias) { + __page_frag_cache_drain(desc_cb->priv, desc_cb->pagecnt_bias); + return; + } + + if (unlikely(!desc_cb->pagecnt_bias)) { + page_ref_add(desc_cb->priv, USHRT_MAX); + desc_cb->pagecnt_bias = USHRT_MAX; } } @@ -2782,8 +2815,9 @@ static bool hns3_parse_vlan_tag(struct hns3_enet_ring *ring, { struct hnae3_handle *handle = ring->tqp->handle; struct pci_dev *pdev = ring->tqp->handle->pdev; + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev); - if (pdev->revision == 0x20) { + if (unlikely(ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2)) { *vlan_tag = le16_to_cpu(desc->rx.ot_vlan_tag); if (!(*vlan_tag & VLAN_VID_MASK)) *vlan_tag = le16_to_cpu(desc->rx.vlan_tag); @@ -2828,6 +2862,16 @@ static bool hns3_parse_vlan_tag(struct hns3_enet_ring *ring, } } +static void hns3_rx_ring_move_fw(struct hns3_enet_ring *ring) +{ + ring->desc[ring->next_to_clean].rx.bd_base_info &= + cpu_to_le32(~BIT(HNS3_RXD_VLD_B)); + ring->next_to_clean += 1; + + if (unlikely(ring->next_to_clean == ring->desc_num)) + ring->next_to_clean = 0; +} + static int hns3_alloc_skb(struct hns3_enet_ring *ring, unsigned int length, unsigned char *va) { @@ -2860,9 +2904,10 @@ static int hns3_alloc_skb(struct hns3_enet_ring *ring, unsigned int length, if (likely(hns3_page_is_reusable(desc_cb->priv))) desc_cb->reuse_flag = 1; else /* This page cannot be reused so discard it */ - put_page(desc_cb->priv); + __page_frag_cache_drain(desc_cb->priv, + desc_cb->pagecnt_bias); - ring_ptr_move_fw(ring, next_to_clean); + hns3_rx_ring_move_fw(ring); return 0; } u64_stats_update_begin(&ring->syncp); @@ -2873,7 +2918,7 @@ static int hns3_alloc_skb(struct hns3_enet_ring *ring, unsigned int length, __skb_put(skb, ring->pull_len); hns3_nic_reuse_page(skb, ring->frag_num++, ring, ring->pull_len, desc_cb); - ring_ptr_move_fw(ring, next_to_clean); + hns3_rx_ring_move_fw(ring); return 0; } @@ -2928,7 +2973,7 @@ static int hns3_add_frag(struct hns3_enet_ring *ring) hns3_nic_reuse_page(skb, ring->frag_num++, ring, 0, desc_cb); trace_hns3_rx_desc(ring); - ring_ptr_move_fw(ring, next_to_clean); + hns3_rx_ring_move_fw(ring); ring->pending_buf++; } while (!(bd_base_info & BIT(HNS3_RXD_FE_B))); @@ -3070,35 +3115,32 @@ static int hns3_handle_rx_bd(struct hns3_enet_ring *ring) prefetch(desc); - length = le16_to_cpu(desc->rx.size); - bd_base_info = le32_to_cpu(desc->rx.bd_base_info); + if (!skb) { + bd_base_info = le32_to_cpu(desc->rx.bd_base_info); - /* Check valid BD */ - if (unlikely(!(bd_base_info & BIT(HNS3_RXD_VLD_B)))) - return -ENXIO; + /* Check valid BD */ + if (unlikely(!(bd_base_info & BIT(HNS3_RXD_VLD_B)))) + return -ENXIO; + + dma_rmb(); + length = le16_to_cpu(desc->rx.size); - if (!skb) { ring->va = desc_cb->buf + desc_cb->page_offset; dma_sync_single_for_cpu(ring_to_dev(ring), desc_cb->dma + desc_cb->page_offset, hns3_buf_size(ring), DMA_FROM_DEVICE); - } - /* Prefetch first cache line of first page - * Idea is to cache few bytes of the header of the packet. Our L1 Cache - * line size is 64B so need to prefetch twice to make it 128B. But in - * actual we can have greater size of caches with 128B Level 1 cache - * lines. In such a case, single fetch would suffice to cache in the - * relevant part of the header. - */ - prefetch(ring->va); -#if L1_CACHE_BYTES < 128 - prefetch(ring->va + L1_CACHE_BYTES); -#endif + /* Prefetch first cache line of first page. + * Idea is to cache few bytes of the header of the packet. + * Our L1 Cache line size is 64B so need to prefetch twice to make + * it 128B. But in actual we can have greater size of caches with + * 128B Level 1 cache lines. In such a case, single fetch would + * suffice to cache in the relevant part of the header. + */ + net_prefetch(ring->va); - if (!skb) { ret = hns3_alloc_skb(ring, length, ring->va); skb = ring->skb; @@ -3138,19 +3180,11 @@ int hns3_clean_rx_ring(struct hns3_enet_ring *ring, int budget, #define RCB_NOF_ALLOC_RX_BUFF_ONCE 16 int unused_count = hns3_desc_unused(ring); int recv_pkts = 0; - int recv_bds = 0; - int err, num; + int err; - num = readl_relaxed(ring->tqp->io_base + HNS3_RING_RX_RING_FBDNUM_REG); - num -= unused_count; unused_count -= ring->pending_buf; - if (num <= 0) - goto out; - - rmb(); /* Make sure num taken effect before the other data is touched */ - - while (recv_pkts < budget && recv_bds < num) { + while (recv_pkts < budget) { /* Reuse or realloc buffers */ if (unused_count >= RCB_NOF_ALLOC_RX_BUFF_ONCE) { hns3_nic_alloc_rx_buffers(ring, unused_count); @@ -3168,7 +3202,6 @@ int hns3_clean_rx_ring(struct hns3_enet_ring *ring, int budget, recv_pkts++; } - recv_bds += ring->pending_buf; unused_count += ring->pending_buf; ring->skb = NULL; ring->pending_buf = 0; @@ -3337,7 +3370,7 @@ static int hns3_nic_common_poll(struct napi_struct *napi, int budget) * budget and be more aggressive about cleaning up the Tx descriptors. */ hns3_for_each_ring(ring, tqp_vector->tx_group) - hns3_clean_tx_ring(ring); + hns3_clean_tx_ring(ring, budget); /* make sure rx ring budget not smaller than 1 */ if (tqp_vector->num_tqps > 1) @@ -3496,7 +3529,7 @@ static int hns3_nic_init_vector_data(struct hns3_nic_priv *priv) struct hnae3_ring_chain_node vector_ring_chain; struct hnae3_handle *h = priv->ae_handle; struct hns3_enet_tqp_vector *tqp_vector; - int ret = 0; + int ret; int i; hns3_nic_set_cpumask(priv); @@ -3673,12 +3706,10 @@ static void hns3_ring_get_cfg(struct hnae3_queue *q, struct hns3_nic_priv *priv, ring = &priv->ring[q->tqp_index]; desc_num = priv->ae_handle->kinfo.num_tx_desc; ring->queue_index = q->tqp_index; - ring->io_base = (u8 __iomem *)q->io_base + HNS3_TX_REG_OFFSET; } else { ring = &priv->ring[q->tqp_index + queue_num]; desc_num = priv->ae_handle->kinfo.num_rx_desc; ring->queue_index = q->tqp_index; - ring->io_base = q->io_base; } hnae3_set_bit(ring->flag, HNAE3_RING_TYPE_B, ring_type); @@ -3692,6 +3723,7 @@ static void hns3_ring_get_cfg(struct hnae3_queue *q, struct hns3_nic_priv *priv, ring->desc_num = desc_num; ring->next_to_use = 0; ring->next_to_clean = 0; + ring->last_to_use = 0; } static void hns3_queue_to_ring(struct hnae3_queue *tqp, @@ -3771,6 +3803,7 @@ void hns3_fini_ring(struct hns3_enet_ring *ring) ring->desc_cb = NULL; ring->next_to_clean = 0; ring->next_to_use = 0; + ring->last_to_use = 0; ring->pending_buf = 0; if (ring->skb) { dev_kfree_skb_any(ring->skb); @@ -3979,6 +4012,7 @@ static void hns3_info_show(struct hns3_nic_priv *priv) static int hns3_client_init(struct hnae3_handle *handle) { struct pci_dev *pdev = handle->pdev; + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev); u16 alloc_tqps, max_rss_size; struct hns3_nic_priv *priv; struct net_device *netdev; @@ -3995,6 +4029,7 @@ static int hns3_client_init(struct hnae3_handle *handle) priv->netdev = netdev; priv->ae_handle = handle; priv->tx_timeout_count = 0; + priv->max_non_tso_bd_num = ae_dev->dev_specs.max_non_tso_bd_num; set_bit(HNS3_NIC_STATE_DOWN, &priv->state); handle->msg_enable = netif_msg_init(debug, DEFAULT_MSG_LEVEL); @@ -4181,9 +4216,11 @@ static void hns3_clear_tx_ring(struct hns3_enet_ring *ring) { while (ring->next_to_clean != ring->next_to_use) { ring->desc[ring->next_to_clean].tx.bdtp_fe_sc_vld_ra_ri = 0; - hns3_free_buffer_detach(ring, ring->next_to_clean); + hns3_free_buffer_detach(ring, ring->next_to_clean, 0); ring_ptr_move_fw(ring, next_to_clean); } + + ring->pending_buf = 0; } static int hns3_clear_rx_ring(struct hns3_enet_ring *ring) @@ -4286,6 +4323,7 @@ int hns3_nic_reset_all_ring(struct hnae3_handle *h) hns3_clear_tx_ring(&priv->ring[i]); priv->ring[i].next_to_clean = 0; priv->ring[i].next_to_use = 0; + priv->ring[i].last_to_use = 0; rx_ring = &priv->ring[i + h->kinfo.num_tqps]; hns3_init_ring_hw(rx_ring); @@ -4582,6 +4620,8 @@ static const struct hns3_hw_error_info hns3_hw_err[] = { .msg = "IMP CMDQ error" }, { .type = HNAE3_IMP_RD_POISON_ERROR, .msg = "IMP RD poison" }, + { .type = HNAE3_ROCEE_AXI_RESP_ERROR, + .msg = "ROCEE AXI RESP error" }, }; static void hns3_process_hw_error(struct hnae3_handle *handle, diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h index 9922c5fd7f94..1c81dea0da1e 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h @@ -42,12 +42,9 @@ enum hns3_nic_state { #define HNS3_RING_TX_RING_PKTNUM_RECORD_REG 0x0006C #define HNS3_RING_TX_RING_EBD_OFFSET_REG 0x00070 #define HNS3_RING_TX_RING_BD_ERR_REG 0x00074 -#define HNS3_RING_PREFETCH_EN_REG 0x0007C -#define HNS3_RING_CFG_VF_NUM_REG 0x00080 -#define HNS3_RING_ASID_REG 0x0008C #define HNS3_RING_EN_REG 0x00090 - -#define HNS3_TX_REG_OFFSET 0x40 +#define HNS3_RING_RX_EN_REG 0x00098 +#define HNS3_RING_TX_EN_REG 0x000D4 #define HNS3_RX_HEAD_SIZE 256 @@ -172,13 +169,12 @@ enum hns3_nic_state { #define HNS3_VECTOR_INITED 1 #define HNS3_MAX_BD_SIZE 65535 -#define HNS3_MAX_NON_TSO_BD_NUM 8U #define HNS3_MAX_TSO_BD_NUM 63U #define HNS3_MAX_TSO_SIZE \ (HNS3_MAX_BD_SIZE * HNS3_MAX_TSO_BD_NUM) -#define HNS3_MAX_NON_TSO_SIZE \ - (HNS3_MAX_BD_SIZE * HNS3_MAX_NON_TSO_BD_NUM) +#define HNS3_MAX_NON_TSO_SIZE(max_non_tso_bd_num) \ + (HNS3_MAX_BD_SIZE * (max_non_tso_bd_num)) #define HNS3_VECTOR_GL0_OFFSET 0x100 #define HNS3_VECTOR_GL1_OFFSET 0x200 @@ -292,6 +288,7 @@ struct hns3_desc_cb { /* desc type, used by the ring user to mark the type of the priv data */ u16 type; + u16 pagecnt_bias; }; enum hns3_pkt_l3type { @@ -348,14 +345,13 @@ enum hns3_pkt_ol4type { }; struct ring_stats { - u64 io_err_cnt; u64 sw_err_cnt; u64 seg_pkt_cnt; union { struct { u64 tx_pkts; u64 tx_bytes; - u64 tx_err_cnt; + u64 tx_more; u64 restart_queue; u64 tx_busy; u64 tx_copy; @@ -380,7 +376,6 @@ struct ring_stats { }; struct hns3_enet_ring { - u8 __iomem *io_base; /* base io address for the ring */ struct hns3_desc *desc; /* dma map address space */ struct hns3_desc_cb *desc_cb; struct hns3_enet_ring *next; @@ -402,8 +397,10 @@ struct hns3_enet_ring { * next_to_use */ int next_to_clean; - - u32 pull_len; /* head length for current packet */ + union { + int last_to_use; /* last idx used by xmit */ + u32 pull_len; /* memcpy len for current rx packet */ + }; u32 frag_num; void *va; /* first buffer address for current packet */ @@ -479,6 +476,7 @@ struct hns3_nic_priv { struct hns3_enet_ring *ring; struct hns3_enet_tqp_vector *tqp_vector; u16 vector_num; + u8 max_non_tso_bd_num; u64 tx_timeout_count; @@ -518,11 +516,6 @@ static inline int ring_space(struct hns3_enet_ring *ring) (begin - end)) - 1; } -static inline int is_ring_empty(struct hns3_enet_ring *ring) -{ - return ring->next_to_use == ring->next_to_clean; -} - static inline u32 hns3_read_reg(void __iomem *base, u32 reg) { return readl(base + reg); @@ -548,9 +541,6 @@ static inline bool hns3_nic_resetting(struct net_device *netdev) #define hns3_write_dev(a, reg, value) \ hns3_write_reg((a)->io_base, (reg), (value)) -#define hnae3_queue_xmit(tqp, buf_num) writel_relaxed(buf_num, \ - (tqp)->io_base + HNS3_RING_TX_RING_TAIL_REG) - #define ring_to_dev(ring) ((ring)->dev) #define ring_to_netdev(ring) ((ring)->tqp_vector->napi.dev) @@ -588,7 +578,7 @@ void hns3_ethtool_set_ops(struct net_device *netdev); int hns3_set_channels(struct net_device *netdev, struct ethtool_channels *ch); -void hns3_clean_tx_ring(struct hns3_enet_ring *ring); +void hns3_clean_tx_ring(struct hns3_enet_ring *ring, int budget); int hns3_init_all_ring(struct hns3_nic_priv *priv); int hns3_uninit_all_ring(struct hns3_nic_priv *priv); int hns3_nic_reset_all_ring(struct hnae3_handle *h); @@ -607,7 +597,6 @@ void hns3_set_vector_coalesce_rl(struct hns3_enet_tqp_vector *tqp_vector, u32 rl_value); void hns3_enable_vlan_filter(struct net_device *netdev, bool enable); -int hns3_update_promisc_mode(struct net_device *netdev, u8 promisc_flags); void hns3_request_update_promisc_mode(struct hnae3_handle *handle); #ifdef CONFIG_HNS3_DCB diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c index 2622e04e8eed..6b07b2771172 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c @@ -27,12 +27,11 @@ struct hns3_sfp_type { static const struct hns3_stats hns3_txq_stats[] = { /* Tx per-queue statistics */ - HNS3_TQP_STAT("io_err_cnt", io_err_cnt), HNS3_TQP_STAT("dropped", sw_err_cnt), HNS3_TQP_STAT("seg_pkt_cnt", seg_pkt_cnt), HNS3_TQP_STAT("packets", tx_pkts), HNS3_TQP_STAT("bytes", tx_bytes), - HNS3_TQP_STAT("errors", tx_err_cnt), + HNS3_TQP_STAT("more", tx_more), HNS3_TQP_STAT("wake", restart_queue), HNS3_TQP_STAT("busy", tx_busy), HNS3_TQP_STAT("copy", tx_copy), @@ -46,7 +45,6 @@ static const struct hns3_stats hns3_txq_stats[] = { static const struct hns3_stats hns3_rxq_stats[] = { /* Rx per-queue statistics */ - HNS3_TQP_STAT("io_err_cnt", io_err_cnt), HNS3_TQP_STAT("dropped", sw_err_cnt), HNS3_TQP_STAT("seg_pkt_cnt", seg_pkt_cnt), HNS3_TQP_STAT("packets", rx_pkts), @@ -79,6 +77,7 @@ static const struct hns3_stats hns3_rxq_stats[] = { static int hns3_lp_setup(struct net_device *ndev, enum hnae3_loop loop, bool en) { struct hnae3_handle *h = hns3_get_handle(ndev); + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(h->pdev); bool vlan_filter_enable; int ret; @@ -98,7 +97,7 @@ static int hns3_lp_setup(struct net_device *ndev, enum hnae3_loop loop, bool en) break; } - if (ret || h->pdev->revision >= 0x21) + if (ret || ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) return ret; if (en) { @@ -149,6 +148,7 @@ static void hns3_lp_setup_skb(struct sk_buff *skb) struct net_device *ndev = skb->dev; struct hnae3_handle *handle; + struct hnae3_ae_dev *ae_dev; unsigned char *packet; struct ethhdr *ethh; unsigned int i; @@ -165,7 +165,8 @@ static void hns3_lp_setup_skb(struct sk_buff *skb) * the purpose of mac or serdes selftest. */ handle = hns3_get_handle(ndev); - if (handle->pdev->revision == 0x20) + ae_dev = pci_get_drvdata(handle->pdev); + if (ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2) ethh->h_dest[5] += HNS3_NIC_LB_DST_MAC_ADDR; eth_zero_addr(ethh->h_source); ethh->h_proto = htons(ETH_P_ARP); @@ -232,7 +233,7 @@ static void hns3_lb_clear_tx_ring(struct hns3_nic_priv *priv, u32 start_ringid, for (i = start_ringid; i <= end_ringid; i++) { struct hns3_enet_ring *ring = &priv->ring[i]; - hns3_clean_tx_ring(ring); + hns3_clean_tx_ring(ring, 0); } } @@ -310,9 +311,6 @@ static void hns3_self_test(struct net_device *ndev, struct hnae3_handle *h = priv->ae_handle; int st_param[HNS3_SELF_TEST_TYPE_NUM][2]; bool if_running = netif_running(ndev); -#if IS_ENABLED(CONFIG_VLAN_8021Q) - bool dis_vlan_filter; -#endif int test_index = 0; u32 i; @@ -349,9 +347,7 @@ static void hns3_self_test(struct net_device *ndev, #if IS_ENABLED(CONFIG_VLAN_8021Q) /* Disable the vlan filter for selftest does not support it */ - dis_vlan_filter = (ndev->features & NETIF_F_HW_VLAN_CTAG_FILTER) && - h->ae_algo->ops->enable_vlan_filter; - if (dis_vlan_filter) + if (h->ae_algo->ops->enable_vlan_filter) h->ae_algo->ops->enable_vlan_filter(h, false); #endif @@ -388,7 +384,7 @@ static void hns3_self_test(struct net_device *ndev, h->ae_algo->ops->halt_autoneg(h, false); #if IS_ENABLED(CONFIG_VLAN_8021Q) - if (dis_vlan_filter) + if (h->ae_algo->ops->enable_vlan_filter) h->ae_algo->ops->enable_vlan_filter(h, true); #endif @@ -763,6 +759,7 @@ static int hns3_set_link_ksettings(struct net_device *netdev, const struct ethtool_link_ksettings *cmd) { struct hnae3_handle *handle = hns3_get_handle(netdev); + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(handle->pdev); const struct hnae3_ae_ops *ops = handle->ae_algo->ops; int ret; @@ -784,7 +781,7 @@ static int hns3_set_link_ksettings(struct net_device *netdev, return phy_ethtool_ksettings_set(netdev->phydev, cmd); } - if (handle->pdev->revision == 0x20) + if (ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2) return -EOPNOTSUPP; ret = hns3_check_ksettings_param(netdev, cmd); @@ -848,11 +845,12 @@ static int hns3_set_rss(struct net_device *netdev, const u32 *indir, const u8 *key, const u8 hfunc) { struct hnae3_handle *h = hns3_get_handle(netdev); + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(h->pdev); if (!h->ae_algo->ops->set_rss) return -EOPNOTSUPP; - if ((h->pdev->revision == 0x20 && + if ((ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2 && hfunc != ETH_RSS_HASH_TOP) || (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP && hfunc != ETH_RSS_HASH_XOR)) { netdev_err(netdev, "hash func not supported\n"); @@ -1073,9 +1071,6 @@ static int hns3_nway_reset(struct net_device *netdev) if (phy) return genphy_restart_aneg(phy); - if (handle->pdev->revision == 0x20) - return -EOPNOTSUPP; - return ops->restart_autoneg(handle); } @@ -1363,11 +1358,12 @@ static int hns3_get_fecparam(struct net_device *netdev, struct ethtool_fecparam *fec) { struct hnae3_handle *handle = hns3_get_handle(netdev); + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(handle->pdev); const struct hnae3_ae_ops *ops = handle->ae_algo->ops; u8 fec_ability; u8 fec_mode; - if (handle->pdev->revision == 0x20) + if (!test_bit(HNAE3_DEV_SUPPORT_FEC_B, ae_dev->caps)) return -EOPNOTSUPP; if (!ops->get_fec) @@ -1385,10 +1381,11 @@ static int hns3_set_fecparam(struct net_device *netdev, struct ethtool_fecparam *fec) { struct hnae3_handle *handle = hns3_get_handle(netdev); + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(handle->pdev); const struct hnae3_ae_ops *ops = handle->ae_algo->ops; u32 fec_mode; - if (handle->pdev->revision == 0x20) + if (!test_bit(HNAE3_DEV_SUPPORT_FEC_B, ae_dev->caps)) return -EOPNOTSUPP; if (!ops->set_fec) @@ -1406,11 +1403,13 @@ static int hns3_get_module_info(struct net_device *netdev, #define HNS3_SFF_8636_V1_3 0x03 struct hnae3_handle *handle = hns3_get_handle(netdev); + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(handle->pdev); const struct hnae3_ae_ops *ops = handle->ae_algo->ops; struct hns3_sfp_type sfp_type; int ret; - if (handle->pdev->revision == 0x20 || !ops->get_module_eeprom) + if (ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2 || + !ops->get_module_eeprom) return -EOPNOTSUPP; memset(&sfp_type, 0, sizeof(sfp_type)); @@ -1454,9 +1453,11 @@ static int hns3_get_module_eeprom(struct net_device *netdev, struct ethtool_eeprom *ee, u8 *data) { struct hnae3_handle *handle = hns3_get_handle(netdev); + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(handle->pdev); const struct hnae3_ae_ops *ops = handle->ae_algo->ops; - if (handle->pdev->revision == 0x20 || !ops->get_module_eeprom) + if (ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2 || + !ops->get_module_eeprom) return -EOPNOTSUPP; if (!ee->len) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h b/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h index 7bddcca148a5..5153e5d41bbd 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h @@ -53,7 +53,7 @@ DECLARE_EVENT_CLASS(hns3_skb_template, ) ); -DEFINE_EVENT(hns3_skb_template, hns3_over_8bd, +DEFINE_EVENT(hns3_skb_template, hns3_over_max_bd, TP_PROTO(struct sk_buff *skb), TP_ARGS(skb)); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c index 1d6c328bd9fb..e6321dda0f3f 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c @@ -261,7 +261,7 @@ int hclge_cmd_send(struct hclge_hw *hw, struct hclge_desc *desc, int num) bool complete = false; u32 timeout = 0; int handle = 0; - int retval = 0; + int retval; int ntc; spin_lock_bh(&hw->cmq.csq.lock); @@ -330,9 +330,37 @@ int hclge_cmd_send(struct hclge_hw *hw, struct hclge_desc *desc, int num) return retval; } -static enum hclge_cmd_status hclge_cmd_query_firmware_version( - struct hclge_hw *hw, u32 *version) +static void hclge_set_default_capability(struct hclge_dev *hdev) { + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev); + + set_bit(HNAE3_DEV_SUPPORT_FD_B, ae_dev->caps); + set_bit(HNAE3_DEV_SUPPORT_GRO_B, ae_dev->caps); + set_bit(HNAE3_DEV_SUPPORT_FEC_B, ae_dev->caps); +} + +static void hclge_parse_capability(struct hclge_dev *hdev, + struct hclge_query_version_cmd *cmd) +{ + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev); + u32 caps; + + caps = __le32_to_cpu(cmd->caps[0]); + + if (hnae3_get_bit(caps, HCLGE_CAP_UDP_GSO_B)) + set_bit(HNAE3_DEV_SUPPORT_UDP_GSO_B, ae_dev->caps); + if (hnae3_get_bit(caps, HCLGE_CAP_PTP_B)) + set_bit(HNAE3_DEV_SUPPORT_PTP_B, ae_dev->caps); + if (hnae3_get_bit(caps, HCLGE_CAP_INT_QL_B)) + set_bit(HNAE3_DEV_SUPPORT_INT_QL_B, ae_dev->caps); + if (hnae3_get_bit(caps, HCLGE_CAP_TQP_TXRX_INDEP_B)) + set_bit(HNAE3_DEV_SUPPORT_TQP_TXRX_INDEP_B, ae_dev->caps); +} + +static enum hclge_cmd_status +hclge_cmd_query_version_and_capability(struct hclge_dev *hdev) +{ + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev); struct hclge_query_version_cmd *resp; struct hclge_desc desc; int ret; @@ -340,9 +368,20 @@ static enum hclge_cmd_status hclge_cmd_query_firmware_version( hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_QUERY_FW_VER, 1); resp = (struct hclge_query_version_cmd *)desc.data; - ret = hclge_cmd_send(hw, &desc, 1); - if (!ret) - *version = le32_to_cpu(resp->firmware); + ret = hclge_cmd_send(&hdev->hw, &desc, 1); + if (ret) + return ret; + + hdev->fw_version = le32_to_cpu(resp->firmware); + + ae_dev->dev_version = le32_to_cpu(resp->hardware) << + HNAE3_PCI_REVISION_BIT_SIZE; + ae_dev->dev_version |= hdev->pdev->revision; + + if (ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) + hclge_set_default_capability(hdev); + + hclge_parse_capability(hdev, resp); return ret; } @@ -402,7 +441,6 @@ static int hclge_firmware_compat_config(struct hclge_dev *hdev) int hclge_cmd_init(struct hclge_dev *hdev) { - u32 version; int ret; spin_lock_bh(&hdev->hw.cmq.csq.lock); @@ -431,22 +469,23 @@ int hclge_cmd_init(struct hclge_dev *hdev) goto err_cmd_init; } - ret = hclge_cmd_query_firmware_version(&hdev->hw, &version); + /* get version and device capabilities */ + ret = hclge_cmd_query_version_and_capability(hdev); if (ret) { dev_err(&hdev->pdev->dev, - "firmware version query failed %d\n", ret); + "failed to query version and capabilities, ret = %d\n", + ret); goto err_cmd_init; } - hdev->fw_version = version; dev_info(&hdev->pdev->dev, "The firmware version is %lu.%lu.%lu.%lu\n", - hnae3_get_field(version, HNAE3_FW_VERSION_BYTE3_MASK, + hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE3_MASK, HNAE3_FW_VERSION_BYTE3_SHIFT), - hnae3_get_field(version, HNAE3_FW_VERSION_BYTE2_MASK, + hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE2_MASK, HNAE3_FW_VERSION_BYTE2_SHIFT), - hnae3_get_field(version, HNAE3_FW_VERSION_BYTE1_MASK, + hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE1_MASK, HNAE3_FW_VERSION_BYTE1_SHIFT), - hnae3_get_field(version, HNAE3_FW_VERSION_BYTE0_MASK, + hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE0_MASK, HNAE3_FW_VERSION_BYTE0_SHIFT)); /* ask the firmware to enable some features, driver can work without diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h index 463f29151ef0..096e26a2e16b 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h @@ -115,7 +115,8 @@ enum hclge_opcode_type { HCLGE_OPC_DFX_RCB_REG = 0x004D, HCLGE_OPC_DFX_TQP_REG = 0x004E, HCLGE_OPC_DFX_SSU_REG_2 = 0x004F, - HCLGE_OPC_DFX_QUERY_CHIP_CAP = 0x0050, + + HCLGE_OPC_QUERY_DEV_SPECS = 0x0050, /* MAC command */ HCLGE_OPC_CONFIG_MAC_MODE = 0x0301, @@ -362,9 +363,26 @@ struct hclge_rx_priv_buff_cmd { u8 rsv[6]; }; +enum HCLGE_CAP_BITS { + HCLGE_CAP_UDP_GSO_B, + HCLGE_CAP_QB_B, + HCLGE_CAP_FD_FORWARD_TC_B, + HCLGE_CAP_PTP_B, + HCLGE_CAP_INT_QL_B, + HCLGE_CAP_SIMPLE_BD_B, + HCLGE_CAP_TX_PUSH_B, + HCLGE_CAP_PHY_IMP_B, + HCLGE_CAP_TQP_TXRX_INDEP_B, + HCLGE_CAP_HW_PAD_B, + HCLGE_CAP_STASH_B, +}; + +#define HCLGE_QUERY_CAP_LENGTH 3 struct hclge_query_version_cmd { __le32 firmware; - __le32 firmware_rsv[5]; + __le32 hardware; + __le32 rsv; + __le32 caps[HCLGE_QUERY_CAP_LENGTH]; /* capabilities of device */ }; #define HCLGE_RX_PRIV_EN_B 15 @@ -491,6 +509,8 @@ struct hclge_pf_res_cmd { #define HCLGE_CFG_RSS_SIZE_M GENMASK(31, 24) #define HCLGE_CFG_SPEED_ABILITY_S 0 #define HCLGE_CFG_SPEED_ABILITY_M GENMASK(7, 0) +#define HCLGE_CFG_SPEED_ABILITY_EXT_S 10 +#define HCLGE_CFG_SPEED_ABILITY_EXT_M GENMASK(15, 10) #define HCLGE_CFG_UMV_TBL_SPACE_S 16 #define HCLGE_CFG_UMV_TBL_SPACE_M GENMASK(31, 16) @@ -1069,6 +1089,20 @@ struct hclge_sfp_info_bd0_cmd { u8 data[HCLGE_SFP_INFO_BD0_LEN]; }; +#define HCLGE_QUERY_DEV_SPECS_BD_NUM 4 + +struct hclge_dev_specs_0_cmd { + __le32 rsv0; + __le32 mac_entry_num; + __le32 mng_entry_num; + __le16 rss_ind_tbl_size; + __le16 rss_key_size; + __le16 int_ql_max; + u8 max_non_tso_bd_num; + u8 rsv1; + __le32 max_tm_rate; +}; + int hclge_cmd_init(struct hclge_dev *hdev); static inline void hclge_write_reg(void __iomem *base, u32 reg, u32 value) { diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_dcb.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_dcb.c index d6c3952aba04..3606240025a8 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_dcb.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_dcb.c @@ -2,7 +2,9 @@ // Copyright (c) 2016-2017 Hisilicon Limited. #include "hclge_main.h" +#include "hclge_dcb.h" #include "hclge_tm.h" +#include "hclge_dcb.h" #include "hnae3.h" #define BW_PERCENT 100 diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c index 26f6f068b01d..16df050e72cf 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c @@ -8,7 +8,7 @@ #include "hclge_tm.h" #include "hnae3.h" -static struct hclge_dbg_reg_type_info hclge_dbg_reg_info[] = { +static const struct hclge_dbg_reg_type_info hclge_dbg_reg_info[] = { { .reg_type = "bios common", .dfx_msg = &hclge_dbg_bios_common_reg[0], .reg_msg = { .msg_num = ARRAY_SIZE(hclge_dbg_bios_common_reg), @@ -115,14 +115,14 @@ static int hclge_dbg_cmd_send(struct hclge_dev *hdev, } static void hclge_dbg_dump_reg_common(struct hclge_dev *hdev, - struct hclge_dbg_reg_type_info *reg_info, + const struct hclge_dbg_reg_type_info *reg_info, const char *cmd_buf) { #define IDX_OFFSET 1 const char *s = &cmd_buf[strlen(reg_info->reg_type) + IDX_OFFSET]; - struct hclge_dbg_dfx_message *dfx_message = reg_info->dfx_msg; - struct hclge_dbg_reg_common_msg *reg_msg = ®_info->reg_msg; + const struct hclge_dbg_dfx_message *dfx_message = reg_info->dfx_msg; + const struct hclge_dbg_reg_common_msg *reg_msg = ®_info->reg_msg; struct hclge_desc *desc_src; struct hclge_desc *desc; int entries_per_desc; @@ -399,7 +399,7 @@ err_dcb_cmd_send: static void hclge_dbg_dump_reg_cmd(struct hclge_dev *hdev, const char *cmd_buf) { - struct hclge_dbg_reg_type_info *reg_info; + const struct hclge_dbg_reg_type_info *reg_info; bool has_dump = false; int i; @@ -428,17 +428,13 @@ static void hclge_dbg_dump_reg_cmd(struct hclge_dev *hdev, const char *cmd_buf) } } -static void hclge_title_idx_print(struct hclge_dev *hdev, bool flag, int index, - char *title_buf, char *true_buf, - char *false_buf) +static void hclge_print_tc_info(struct hclge_dev *hdev, bool flag, int index) { if (flag) - dev_info(&hdev->pdev->dev, "%s(%d): %s weight: %u\n", - title_buf, index, true_buf, - hdev->tm_info.pg_info[0].tc_dwrr[index]); + dev_info(&hdev->pdev->dev, "tc(%d): no sp mode weight: %u\n", + index, hdev->tm_info.pg_info[0].tc_dwrr[index]); else - dev_info(&hdev->pdev->dev, "%s(%d): %s\n", title_buf, index, - false_buf); + dev_info(&hdev->pdev->dev, "tc(%d): sp mode\n", index); } static void hclge_dbg_dump_tc(struct hclge_dev *hdev) @@ -469,8 +465,7 @@ static void hclge_dbg_dump_tc(struct hclge_dev *hdev) ets_weight->weight_offset); for (i = 0; i < HNAE3_MAX_TC; i++) - hclge_title_idx_print(hdev, ets_weight->tc_weight[i], i, - "tc", "no sp mode", "sp mode"); + hclge_print_tc_info(hdev, ets_weight->tc_weight[i], i); } static void hclge_dbg_dump_tm_pg(struct hclge_dev *hdev) @@ -1170,6 +1165,14 @@ static void hclge_dbg_dump_serv_info(struct hclge_dev *hdev) hdev->serv_processed_cnt); } +static void hclge_dbg_dump_interrupt(struct hclge_dev *hdev) +{ + dev_info(&hdev->pdev->dev, "num_nic_msi: %u\n", hdev->num_nic_msi); + dev_info(&hdev->pdev->dev, "num_roce_msi: %u\n", hdev->num_roce_msi); + dev_info(&hdev->pdev->dev, "num_msi_used: %u\n", hdev->num_msi_used); + dev_info(&hdev->pdev->dev, "num_msi_left: %u\n", hdev->num_msi_left); +} + static void hclge_dbg_get_m7_stats_info(struct hclge_dev *hdev) { struct hclge_desc *desc_src, *desc_tmp; @@ -1494,6 +1497,7 @@ int hclge_dbg_run_cmd(struct hnae3_handle *handle, const char *cmd_buf) #define DUMP_REG "dump reg" #define DUMP_TM_MAP "dump tm map" #define DUMP_LOOPBACK "dump loopback" +#define DUMP_INTERRUPT "dump intr" struct hclge_vport *vport = hclge_get_vport(handle); struct hclge_dev *hdev = vport->back; @@ -1541,6 +1545,9 @@ int hclge_dbg_run_cmd(struct hnae3_handle *handle, const char *cmd_buf) hclge_dbg_dump_mac_list(hdev, &cmd_buf[sizeof("dump mc mac list")], false); + } else if (strncmp(cmd_buf, DUMP_INTERRUPT, + strlen(DUMP_INTERRUPT)) == 0) { + hclge_dbg_dump_interrupt(hdev); } else { dev_info(&hdev->pdev->dev, "unknown command\n"); return -EINVAL; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.h index 38b79321c4c4..a9066e6ff697 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.h @@ -81,13 +81,13 @@ struct hclge_dbg_dfx_message { #define HCLGE_DBG_MAC_REG_TYPE_LEN 32 struct hclge_dbg_reg_type_info { const char *reg_type; - struct hclge_dbg_dfx_message *dfx_msg; + const struct hclge_dbg_dfx_message *dfx_msg; struct hclge_dbg_reg_common_msg reg_msg; }; #pragma pack() -static struct hclge_dbg_dfx_message hclge_dbg_bios_common_reg[] = { +static const struct hclge_dbg_dfx_message hclge_dbg_bios_common_reg[] = { {false, "Reserved"}, {true, "BP_CPU_STATE"}, {true, "DFX_MSIX_INFO_NIC_0"}, @@ -103,7 +103,7 @@ static struct hclge_dbg_dfx_message hclge_dbg_bios_common_reg[] = { {false, "Reserved"}, }; -static struct hclge_dbg_dfx_message hclge_dbg_ssu_reg_0[] = { +static const struct hclge_dbg_dfx_message hclge_dbg_ssu_reg_0[] = { {false, "Reserved"}, {true, "SSU_ETS_PORT_STATUS"}, {true, "SSU_ETS_TCG_STATUS"}, @@ -175,7 +175,7 @@ static struct hclge_dbg_dfx_message hclge_dbg_ssu_reg_0[] = { {false, "Reserved"}, }; -static struct hclge_dbg_dfx_message hclge_dbg_ssu_reg_1[] = { +static const struct hclge_dbg_dfx_message hclge_dbg_ssu_reg_1[] = { {true, "prt_id"}, {true, "PACKET_TC_CURR_BUFFER_CNT_0"}, {true, "PACKET_TC_CURR_BUFFER_CNT_1"}, @@ -282,7 +282,7 @@ static struct hclge_dbg_dfx_message hclge_dbg_ssu_reg_1[] = { {false, "Reserved"}, }; -static struct hclge_dbg_dfx_message hclge_dbg_ssu_reg_2[] = { +static const struct hclge_dbg_dfx_message hclge_dbg_ssu_reg_2[] = { {true, "OQ_INDEX"}, {true, "QUEUE_CNT"}, {false, "Reserved"}, @@ -291,7 +291,7 @@ static struct hclge_dbg_dfx_message hclge_dbg_ssu_reg_2[] = { {false, "Reserved"}, }; -static struct hclge_dbg_dfx_message hclge_dbg_igu_egu_reg[] = { +static const struct hclge_dbg_dfx_message hclge_dbg_igu_egu_reg[] = { {true, "prt_id"}, {true, "IGU_RX_ERR_PKT"}, {true, "IGU_RX_NO_SOF_PKT"}, @@ -356,7 +356,7 @@ static struct hclge_dbg_dfx_message hclge_dbg_igu_egu_reg[] = { {false, "Reserved"}, }; -static struct hclge_dbg_dfx_message hclge_dbg_rpu_reg_0[] = { +static const struct hclge_dbg_dfx_message hclge_dbg_rpu_reg_0[] = { {true, "tc_queue_num"}, {true, "FSM_DFX_ST0"}, {true, "FSM_DFX_ST1"}, @@ -365,7 +365,7 @@ static struct hclge_dbg_dfx_message hclge_dbg_rpu_reg_0[] = { {true, "BUF_WAIT_TIMEOUT_QID"}, }; -static struct hclge_dbg_dfx_message hclge_dbg_rpu_reg_1[] = { +static const struct hclge_dbg_dfx_message hclge_dbg_rpu_reg_1[] = { {false, "Reserved"}, {true, "FIFO_DFX_ST0"}, {true, "FIFO_DFX_ST1"}, @@ -381,7 +381,7 @@ static struct hclge_dbg_dfx_message hclge_dbg_rpu_reg_1[] = { {false, "Reserved"}, }; -static struct hclge_dbg_dfx_message hclge_dbg_ncsi_reg[] = { +static const struct hclge_dbg_dfx_message hclge_dbg_ncsi_reg[] = { {false, "Reserved"}, {true, "NCSI_EGU_TX_FIFO_STS"}, {true, "NCSI_PAUSE_STATUS"}, @@ -453,7 +453,7 @@ static struct hclge_dbg_dfx_message hclge_dbg_ncsi_reg[] = { {true, "NCSI_MAC_RX_PAUSE_FRAMES"}, }; -static struct hclge_dbg_dfx_message hclge_dbg_rtc_reg[] = { +static const struct hclge_dbg_dfx_message hclge_dbg_rtc_reg[] = { {false, "Reserved"}, {true, "LGE_IGU_AFIFO_DFX_0"}, {true, "LGE_IGU_AFIFO_DFX_1"}, @@ -483,7 +483,7 @@ static struct hclge_dbg_dfx_message hclge_dbg_rtc_reg[] = { {false, "Reserved"}, }; -static struct hclge_dbg_dfx_message hclge_dbg_ppp_reg[] = { +static const struct hclge_dbg_dfx_message hclge_dbg_ppp_reg[] = { {false, "Reserved"}, {true, "DROP_FROM_PRT_PKT_CNT"}, {true, "DROP_FROM_HOST_PKT_CNT"}, @@ -639,7 +639,7 @@ static struct hclge_dbg_dfx_message hclge_dbg_ppp_reg[] = { {false, "Reserved"}, }; -static struct hclge_dbg_dfx_message hclge_dbg_rcb_reg[] = { +static const struct hclge_dbg_dfx_message hclge_dbg_rcb_reg[] = { {false, "Reserved"}, {true, "FSM_DFX_ST0"}, {true, "FSM_DFX_ST1"}, @@ -711,7 +711,7 @@ static struct hclge_dbg_dfx_message hclge_dbg_rcb_reg[] = { {false, "Reserved"}, }; -static struct hclge_dbg_dfx_message hclge_dbg_tqp_reg[] = { +static const struct hclge_dbg_dfx_message hclge_dbg_tqp_reg[] = { {true, "q_num"}, {true, "RCB_CFG_RX_RING_TAIL"}, {true, "RCB_CFG_RX_RING_HEAD"}, diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c index 50d5ef71756b..9ee55ee0487d 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c @@ -729,7 +729,7 @@ static int hclge_config_ncsi_hw_err_int(struct hclge_dev *hdev, bool en) struct hclge_desc desc; int ret; - if (hdev->pdev->revision < 0x21) + if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2) return 0; /* configure NCSI error interrupts */ @@ -808,7 +808,7 @@ static int hclge_config_ppp_error_interrupt(struct hclge_dev *hdev, u32 cmd, cpu_to_le32(HCLGE_PPP_MPF_ECC_ERR_INT0_EN_MASK); desc[1].data[1] = cpu_to_le32(HCLGE_PPP_MPF_ECC_ERR_INT1_EN_MASK); - if (hdev->pdev->revision >= 0x21) + if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) desc[1].data[2] = cpu_to_le32(HCLGE_PPP_PF_ERR_INT_EN_MASK); } else if (cmd == HCLGE_PPP_CMD1_INT_CMD) { @@ -1041,7 +1041,7 @@ static int hclge_config_ssu_hw_err_int(struct hclge_dev *hdev, bool en) hclge_cmd_setup_basic_desc(&desc[1], HCLGE_SSU_COMMON_INT_CMD, false); if (en) { - if (hdev->pdev->revision >= 0x21) + if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) desc[0].data[0] = cpu_to_le32(HCLGE_SSU_COMMON_INT_EN); else @@ -1507,6 +1507,8 @@ hclge_log_and_clear_rocee_ras_error(struct hclge_dev *hdev) reset_type = HNAE3_FUNC_RESET; + hclge_report_hw_error(hdev, HNAE3_ROCEE_AXI_RESP_ERROR); + ret = hclge_log_rocee_axi_error(hdev); if (ret) return HNAE3_GLOBAL_RESET; @@ -1548,7 +1550,8 @@ int hclge_config_rocee_ras_interrupt(struct hclge_dev *hdev, bool en) struct hclge_desc desc; int ret; - if (hdev->pdev->revision < 0x21 || !hnae3_dev_roce_supported(hdev)) + if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2 || + !hnae3_dev_roce_supported(hdev)) return 0; hclge_cmd_setup_basic_desc(&desc, HCLGE_CONFIG_ROCEE_RAS_INT_EN, false); @@ -1574,8 +1577,7 @@ static void hclge_handle_rocee_ras_error(struct hnae3_ae_dev *ae_dev) struct hclge_dev *hdev = ae_dev->priv; enum hnae3_reset_type reset_type; - if (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state) || - hdev->pdev->revision < 0x21) + if (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state)) return; reset_type = hclge_log_and_clear_rocee_ras_error(hdev); @@ -1661,7 +1663,7 @@ pci_ers_result_t hclge_handle_hw_ras_error(struct hnae3_ae_dev *ae_dev) } /* Handling Non-fatal Rocee RAS errors */ - if (hdev->pdev->revision >= 0x21 && + if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2 && status & HCLGE_RAS_REG_ROCEE_ERR_MASK) { dev_err(dev, "ROCEE Non-Fatal RAS error identified\n"); hclge_handle_rocee_ras_error(ae_dev); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index d553ed7ee64c..1f026408ad38 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -84,6 +84,7 @@ static const struct pci_device_id ae_algo_pci_tbl[] = { {PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_50GE_RDMA), 0}, {PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_50GE_RDMA_MACSEC), 0}, {PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_100G_RDMA_MACSEC), 0}, + {PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_200G_RDMA), 0}, /* required last entry */ {0, } }; @@ -622,7 +623,7 @@ static u8 *hclge_tqps_get_strings(struct hnae3_handle *handle, u8 *data) { struct hnae3_knic_private_info *kinfo = &handle->kinfo; u8 *buff = data; - int i = 0; + int i; for (i = 0; i < kinfo->num_tqps; i++) { struct hclge_tqp *tqp = container_of(handle->kinfo.tqp[i], @@ -739,7 +740,7 @@ static int hclge_get_sset_count(struct hnae3_handle *handle, int stringset) if (stringset == ETH_SS_TEST) { /* clear loopback bit flags at first */ handle->flags = (handle->flags & (~HCLGE_LOOPBACK_TEST_FLAGS)); - if (hdev->pdev->revision >= 0x21 || + if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2 || hdev->hw.mac.speed == HCLGE_MAC_SPEED_10M || hdev->hw.mac.speed == HCLGE_MAC_SPEED_100M || hdev->hw.mac.speed == HCLGE_MAC_SPEED_1G) { @@ -965,6 +966,9 @@ static int hclge_parse_speed(int speed_cmd, int *speed) case 5: *speed = HCLGE_MAC_SPEED_100G; break; + case 8: + *speed = HCLGE_MAC_SPEED_200G; + break; default: return -EINVAL; } @@ -1004,6 +1008,9 @@ static int hclge_check_port_speed(struct hnae3_handle *handle, u32 speed) case HCLGE_MAC_SPEED_100G: speed_bit = HCLGE_SUPPORT_100G_BIT; break; + case HCLGE_MAC_SPEED_200G: + speed_bit = HCLGE_SUPPORT_200G_BIT; + break; default: return -EINVAL; } @@ -1014,7 +1021,7 @@ static int hclge_check_port_speed(struct hnae3_handle *handle, u32 speed) return -EINVAL; } -static void hclge_convert_setting_sr(struct hclge_mac *mac, u8 speed_ability) +static void hclge_convert_setting_sr(struct hclge_mac *mac, u16 speed_ability) { if (speed_ability & HCLGE_SUPPORT_10G_BIT) linkmode_set_bit(ETHTOOL_LINK_MODE_10000baseSR_Full_BIT, @@ -1031,9 +1038,12 @@ static void hclge_convert_setting_sr(struct hclge_mac *mac, u8 speed_ability) if (speed_ability & HCLGE_SUPPORT_100G_BIT) linkmode_set_bit(ETHTOOL_LINK_MODE_100000baseSR4_Full_BIT, mac->supported); + if (speed_ability & HCLGE_SUPPORT_200G_BIT) + linkmode_set_bit(ETHTOOL_LINK_MODE_200000baseSR4_Full_BIT, + mac->supported); } -static void hclge_convert_setting_lr(struct hclge_mac *mac, u8 speed_ability) +static void hclge_convert_setting_lr(struct hclge_mac *mac, u16 speed_ability) { if (speed_ability & HCLGE_SUPPORT_10G_BIT) linkmode_set_bit(ETHTOOL_LINK_MODE_10000baseLR_Full_BIT, @@ -1050,9 +1060,13 @@ static void hclge_convert_setting_lr(struct hclge_mac *mac, u8 speed_ability) if (speed_ability & HCLGE_SUPPORT_100G_BIT) linkmode_set_bit(ETHTOOL_LINK_MODE_100000baseLR4_ER4_Full_BIT, mac->supported); + if (speed_ability & HCLGE_SUPPORT_200G_BIT) + linkmode_set_bit( + ETHTOOL_LINK_MODE_200000baseLR4_ER4_FR4_Full_BIT, + mac->supported); } -static void hclge_convert_setting_cr(struct hclge_mac *mac, u8 speed_ability) +static void hclge_convert_setting_cr(struct hclge_mac *mac, u16 speed_ability) { if (speed_ability & HCLGE_SUPPORT_10G_BIT) linkmode_set_bit(ETHTOOL_LINK_MODE_10000baseCR_Full_BIT, @@ -1069,9 +1083,12 @@ static void hclge_convert_setting_cr(struct hclge_mac *mac, u8 speed_ability) if (speed_ability & HCLGE_SUPPORT_100G_BIT) linkmode_set_bit(ETHTOOL_LINK_MODE_100000baseCR4_Full_BIT, mac->supported); + if (speed_ability & HCLGE_SUPPORT_200G_BIT) + linkmode_set_bit(ETHTOOL_LINK_MODE_200000baseCR4_Full_BIT, + mac->supported); } -static void hclge_convert_setting_kr(struct hclge_mac *mac, u8 speed_ability) +static void hclge_convert_setting_kr(struct hclge_mac *mac, u16 speed_ability) { if (speed_ability & HCLGE_SUPPORT_1G_BIT) linkmode_set_bit(ETHTOOL_LINK_MODE_1000baseKX_Full_BIT, @@ -1091,6 +1108,9 @@ static void hclge_convert_setting_kr(struct hclge_mac *mac, u8 speed_ability) if (speed_ability & HCLGE_SUPPORT_100G_BIT) linkmode_set_bit(ETHTOOL_LINK_MODE_100000baseKR4_Full_BIT, mac->supported); + if (speed_ability & HCLGE_SUPPORT_200G_BIT) + linkmode_set_bit(ETHTOOL_LINK_MODE_200000baseKR4_Full_BIT, + mac->supported); } static void hclge_convert_setting_fec(struct hclge_mac *mac) @@ -1115,6 +1135,7 @@ static void hclge_convert_setting_fec(struct hclge_mac *mac) BIT(HNAE3_FEC_AUTO); break; case HCLGE_MAC_SPEED_100G: + case HCLGE_MAC_SPEED_200G: linkmode_set_bit(ETHTOOL_LINK_MODE_FEC_RS_BIT, mac->supported); mac->fec_ability = BIT(HNAE3_FEC_RS) | BIT(HNAE3_FEC_AUTO); break; @@ -1125,7 +1146,7 @@ static void hclge_convert_setting_fec(struct hclge_mac *mac) } static void hclge_parse_fiber_link_mode(struct hclge_dev *hdev, - u8 speed_ability) + u16 speed_ability) { struct hclge_mac *mac = &hdev->hw.mac; @@ -1136,7 +1157,7 @@ static void hclge_parse_fiber_link_mode(struct hclge_dev *hdev, hclge_convert_setting_sr(mac, speed_ability); hclge_convert_setting_lr(mac, speed_ability); hclge_convert_setting_cr(mac, speed_ability); - if (hdev->pdev->revision >= 0x21) + if (hnae3_dev_fec_supported(hdev)) hclge_convert_setting_fec(mac); linkmode_set_bit(ETHTOOL_LINK_MODE_FIBRE_BIT, mac->supported); @@ -1145,12 +1166,12 @@ static void hclge_parse_fiber_link_mode(struct hclge_dev *hdev, } static void hclge_parse_backplane_link_mode(struct hclge_dev *hdev, - u8 speed_ability) + u16 speed_ability) { struct hclge_mac *mac = &hdev->hw.mac; hclge_convert_setting_kr(mac, speed_ability); - if (hdev->pdev->revision >= 0x21) + if (hnae3_dev_fec_supported(hdev)) hclge_convert_setting_fec(mac); linkmode_set_bit(ETHTOOL_LINK_MODE_Backplane_BIT, mac->supported); linkmode_set_bit(ETHTOOL_LINK_MODE_Pause_BIT, mac->supported); @@ -1158,7 +1179,7 @@ static void hclge_parse_backplane_link_mode(struct hclge_dev *hdev, } static void hclge_parse_copper_link_mode(struct hclge_dev *hdev, - u8 speed_ability) + u16 speed_ability) { unsigned long *supported = hdev->hw.mac.supported; @@ -1188,7 +1209,7 @@ static void hclge_parse_copper_link_mode(struct hclge_dev *hdev, linkmode_set_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT, supported); } -static void hclge_parse_link_mode(struct hclge_dev *hdev, u8 speed_ability) +static void hclge_parse_link_mode(struct hclge_dev *hdev, u16 speed_ability) { u8 media_type = hdev->hw.mac.media_type; @@ -1200,8 +1221,11 @@ static void hclge_parse_link_mode(struct hclge_dev *hdev, u8 speed_ability) hclge_parse_backplane_link_mode(hdev, speed_ability); } -static u32 hclge_get_max_speed(u8 speed_ability) +static u32 hclge_get_max_speed(u16 speed_ability) { + if (speed_ability & HCLGE_SUPPORT_200G_BIT) + return HCLGE_MAC_SPEED_200G; + if (speed_ability & HCLGE_SUPPORT_100G_BIT) return HCLGE_MAC_SPEED_100G; @@ -1231,8 +1255,11 @@ static u32 hclge_get_max_speed(u8 speed_ability) static void hclge_parse_cfg(struct hclge_cfg *cfg, struct hclge_desc *desc) { +#define SPEED_ABILITY_EXT_SHIFT 8 + struct hclge_cfg_param_cmd *req; u64 mac_addr_tmp_high; + u16 speed_ability_ext; u64 mac_addr_tmp; unsigned int i; @@ -1281,6 +1308,11 @@ static void hclge_parse_cfg(struct hclge_cfg *cfg, struct hclge_desc *desc) cfg->speed_ability = hnae3_get_field(__le32_to_cpu(req->param[1]), HCLGE_CFG_SPEED_ABILITY_M, HCLGE_CFG_SPEED_ABILITY_S); + speed_ability_ext = hnae3_get_field(__le32_to_cpu(req->param[1]), + HCLGE_CFG_SPEED_ABILITY_EXT_M, + HCLGE_CFG_SPEED_ABILITY_EXT_S); + cfg->speed_ability |= speed_ability_ext << SPEED_ABILITY_EXT_SHIFT; + cfg->umv_space = hnae3_get_field(__le32_to_cpu(req->param[1]), HCLGE_CFG_UMV_TBL_SPACE_M, HCLGE_CFG_UMV_TBL_SPACE_S); @@ -1324,6 +1356,78 @@ static int hclge_get_cfg(struct hclge_dev *hdev, struct hclge_cfg *hcfg) return 0; } +static void hclge_set_default_dev_specs(struct hclge_dev *hdev) +{ +#define HCLGE_MAX_NON_TSO_BD_NUM 8U + + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev); + + ae_dev->dev_specs.max_non_tso_bd_num = HCLGE_MAX_NON_TSO_BD_NUM; + ae_dev->dev_specs.rss_ind_tbl_size = HCLGE_RSS_IND_TBL_SIZE; + ae_dev->dev_specs.rss_key_size = HCLGE_RSS_KEY_SIZE; + ae_dev->dev_specs.max_tm_rate = HCLGE_ETHER_MAX_RATE; +} + +static void hclge_parse_dev_specs(struct hclge_dev *hdev, + struct hclge_desc *desc) +{ + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev); + struct hclge_dev_specs_0_cmd *req0; + + req0 = (struct hclge_dev_specs_0_cmd *)desc[0].data; + + ae_dev->dev_specs.max_non_tso_bd_num = req0->max_non_tso_bd_num; + ae_dev->dev_specs.rss_ind_tbl_size = + le16_to_cpu(req0->rss_ind_tbl_size); + ae_dev->dev_specs.rss_key_size = le16_to_cpu(req0->rss_key_size); + ae_dev->dev_specs.max_tm_rate = le32_to_cpu(req0->max_tm_rate); +} + +static void hclge_check_dev_specs(struct hclge_dev *hdev) +{ + struct hnae3_dev_specs *dev_specs = &hdev->ae_dev->dev_specs; + + if (!dev_specs->max_non_tso_bd_num) + dev_specs->max_non_tso_bd_num = HCLGE_MAX_NON_TSO_BD_NUM; + if (!dev_specs->rss_ind_tbl_size) + dev_specs->rss_ind_tbl_size = HCLGE_RSS_IND_TBL_SIZE; + if (!dev_specs->rss_key_size) + dev_specs->rss_key_size = HCLGE_RSS_KEY_SIZE; + if (!dev_specs->max_tm_rate) + dev_specs->max_tm_rate = HCLGE_ETHER_MAX_RATE; +} + +static int hclge_query_dev_specs(struct hclge_dev *hdev) +{ + struct hclge_desc desc[HCLGE_QUERY_DEV_SPECS_BD_NUM]; + int ret; + int i; + + /* set default specifications as devices lower than version V3 do not + * support querying specifications from firmware. + */ + if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V3) { + hclge_set_default_dev_specs(hdev); + return 0; + } + + for (i = 0; i < HCLGE_QUERY_DEV_SPECS_BD_NUM - 1; i++) { + hclge_cmd_setup_basic_desc(&desc[i], HCLGE_OPC_QUERY_DEV_SPECS, + true); + desc[i].flag |= cpu_to_le16(HCLGE_CMD_FLAG_NEXT); + } + hclge_cmd_setup_basic_desc(&desc[i], HCLGE_OPC_QUERY_DEV_SPECS, true); + + ret = hclge_cmd_send(&hdev->hw, desc, HCLGE_QUERY_DEV_SPECS_BD_NUM); + if (ret) + return ret; + + hclge_parse_dev_specs(hdev, desc); + hclge_check_dev_specs(hdev); + + return 0; +} + static int hclge_get_cap(struct hclge_dev *hdev) { int ret; @@ -2422,6 +2526,10 @@ static int hclge_cfg_mac_speed_dup_hw(struct hclge_dev *hdev, int speed, hnae3_set_field(req->speed_dup, HCLGE_CFG_SPEED_M, HCLGE_CFG_SPEED_S, 5); break; + case HCLGE_MAC_SPEED_200G: + hnae3_set_field(req->speed_dup, HCLGE_CFG_SPEED_M, + HCLGE_CFG_SPEED_S, 8); + break; default: dev_err(&hdev->pdev->dev, "invalid speed (%d)\n", speed); return -EINVAL; @@ -2856,7 +2964,7 @@ static int hclge_update_port_info(struct hclge_dev *hdev) if (!hdev->support_sfp_query) return 0; - if (hdev->pdev->revision >= 0x21) + if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) ret = hclge_get_sfp_info(hdev, mac); else ret = hclge_get_sfp_speed(hdev, &speed); @@ -2868,7 +2976,7 @@ static int hclge_update_port_info(struct hclge_dev *hdev) return ret; } - if (hdev->pdev->revision >= 0x21) { + if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) { if (mac->speed_type == QUERY_ACTIVE_SPEED) { hclge_update_port_capability(mac); return 0; @@ -3211,7 +3319,7 @@ static int hclge_notify_roce_client(struct hclge_dev *hdev, enum hnae3_reset_notify_type type) { struct hnae3_client *client = hdev->roce_client; - int ret = 0; + int ret; u16 i; if (!test_bit(HCLGE_STATE_ROCE_REGISTERED, &hdev->state) || !client) @@ -3533,7 +3641,7 @@ static void hclge_clear_reset_cause(struct hclge_dev *hdev) /* For revision 0x20, the reset interrupt source * can only be cleared after hardware reset done */ - if (hdev->pdev->revision == 0x20) + if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2) hclge_write_dev(&hdev->hw, HCLGE_MISC_RESET_STS_REG, clearval); @@ -3944,6 +4052,9 @@ static void hclge_periodic_service_task(struct hclge_dev *hdev) { unsigned long delta = round_jiffies_relative(HZ); + if (test_bit(HCLGE_STATE_RST_FAIL, &hdev->state)) + return; + /* Always handle the link updating to make sure link state is * updated when it is triggered by mbx. */ @@ -4537,7 +4648,7 @@ static void hclge_rss_init_cfg(struct hclge_dev *hdev) int i, rss_algo = HCLGE_RSS_HASH_ALGO_TOEPLITZ; struct hclge_vport *vport = hdev->vport; - if (hdev->pdev->revision >= 0x21) + if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) rss_algo = HCLGE_RSS_HASH_ALGO_SIMPLE; for (i = 0; i < hdev->num_vmdq_vport + 1; i++) { @@ -4737,13 +4848,14 @@ static int hclge_set_promisc_mode(struct hnae3_handle *handle, bool en_uc_pmc, bool en_mc_pmc) { struct hclge_vport *vport = hclge_get_vport(handle); + struct hclge_dev *hdev = vport->back; bool en_bc_pmc = true; - /* For revision 0x20, if broadcast promisc enabled, vlan filter is - * always bypassed. So broadcast promisc should be disabled until - * user enable promisc mode + /* For device whose version below V2, if broadcast promisc enabled, + * vlan filter is always bypassed. So broadcast promisc should be + * disabled until user enable promisc mode */ - if (handle->pdev->revision == 0x20) + if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2) en_bc_pmc = handle->netdev_flags & HNAE3_BPE ? true : false; return hclge_set_vport_promisc_mode(vport, en_uc_pmc, en_mc_pmc, @@ -6758,7 +6870,7 @@ static int hclge_set_loopback(struct hnae3_handle *handle, * the same, the packets are looped back in the SSU. If SSU loopback * is disabled, packets can reach MAC even if SMAC is the same as DMAC. */ - if (hdev->pdev->revision >= 0x21) { + if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) { u8 switch_param = en ? 0 : BIT(HCLGE_SWITCH_ALW_LPBK_B); ret = hclge_config_switch_param(hdev, PF_VPORT_ID, switch_param, @@ -8260,7 +8372,7 @@ static void hclge_enable_vlan_filter(struct hnae3_handle *handle, bool enable) struct hclge_vport *vport = hclge_get_vport(handle); struct hclge_dev *hdev = vport->back; - if (hdev->pdev->revision >= 0x21) { + if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) { hclge_set_vlan_filter_ctrl(hdev, HCLGE_FILTER_TYPE_VF, HCLGE_FILTER_FE_EGRESS, enable, 0); hclge_set_vlan_filter_ctrl(hdev, HCLGE_FILTER_TYPE_PORT, @@ -8620,7 +8732,7 @@ static int hclge_init_vlan_config(struct hclge_dev *hdev) int ret; int i; - if (hdev->pdev->revision >= 0x21) { + if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) { /* for revision 0x21, vf vlan filter is per function */ for (i = 0; i < hdev->num_alloc_vport; i++) { vport = &hdev->vport[i]; @@ -8975,7 +9087,7 @@ static int hclge_set_vf_vlan_filter(struct hnae3_handle *handle, int vfid, u16 state; int ret; - if (hdev->pdev->revision == 0x20) + if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2) return -EOPNOTSUPP; vport = hclge_get_vf_vport(hdev, vfid); @@ -9950,6 +10062,13 @@ static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev) if (ret) goto err_cmd_uninit; + ret = hclge_query_dev_specs(hdev); + if (ret) { + dev_err(&pdev->dev, "failed to query dev specifications, ret = %d.\n", + ret); + goto err_cmd_uninit; + } + ret = hclge_configure(hdev); if (ret) { dev_err(&pdev->dev, "Configure dev error, ret = %d.\n", ret); @@ -10147,7 +10266,7 @@ static int hclge_set_vf_spoofchk(struct hnae3_handle *handle, int vf, u32 new_spoofchk = enable ? 1 : 0; int ret; - if (hdev->pdev->revision == 0x20) + if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2) return -EOPNOTSUPP; vport = hclge_get_vf_vport(hdev, vf); @@ -10180,7 +10299,7 @@ static int hclge_reset_vport_spoofchk(struct hclge_dev *hdev) int ret; int i; - if (hdev->pdev->revision == 0x20) + if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2) return 0; /* resume the vf spoof check state after reset */ @@ -10200,6 +10319,7 @@ static int hclge_set_vf_trust(struct hnae3_handle *handle, int vf, bool enable) { struct hclge_vport *vport = hclge_get_vport(handle); struct hclge_dev *hdev = vport->back; + struct hnae3_ae_dev *ae_dev = hdev->ae_dev; u32 new_trusted = enable ? 1 : 0; bool en_bc_pmc; int ret; @@ -10213,7 +10333,7 @@ static int hclge_set_vf_trust(struct hnae3_handle *handle, int vf, bool enable) /* Disable promisc mode for VF if it is not trusted any more. */ if (!enable && vport->vf_info.promisc_enable) { - en_bc_pmc = hdev->pdev->revision != 0x20; + en_bc_pmc = ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2; ret = hclge_set_vport_promisc_mode(vport, false, false, en_bc_pmc); if (ret) @@ -11090,7 +11210,7 @@ static void hclge_sync_promisc_mode(struct hclge_dev *hdev) { struct hclge_vport *vport = &hdev->vport[0]; struct hnae3_handle *handle = &vport->nic; - u8 tmp_flags = 0; + u8 tmp_flags; int ret; if (vport->last_promisc_flags != vport->overflow_promisc_flags) { diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h index 9bbdd4557c27..64e6afdb61b8 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h @@ -199,6 +199,7 @@ enum HLCGE_PORT_TYPE { #define HCLGE_SUPPORT_40G_BIT BIT(5) #define HCLGE_SUPPORT_100M_BIT BIT(6) #define HCLGE_SUPPORT_10M_BIT BIT(7) +#define HCLGE_SUPPORT_200G_BIT BIT(8) #define HCLGE_SUPPORT_GE \ (HCLGE_SUPPORT_1G_BIT | HCLGE_SUPPORT_100M_BIT | HCLGE_SUPPORT_10M_BIT) @@ -238,7 +239,8 @@ enum HCLGE_MAC_SPEED { HCLGE_MAC_SPEED_25G = 25000, /* 25000 Mbps = 25 Gbps */ HCLGE_MAC_SPEED_40G = 40000, /* 40000 Mbps = 40 Gbps */ HCLGE_MAC_SPEED_50G = 50000, /* 50000 Mbps = 50 Gbps */ - HCLGE_MAC_SPEED_100G = 100000 /* 100000 Mbps = 100 Gbps */ + HCLGE_MAC_SPEED_100G = 100000, /* 100000 Mbps = 100 Gbps */ + HCLGE_MAC_SPEED_200G = 200000 /* 200000 Mbps = 200 Gbps */ }; enum HCLGE_MAC_DUPLEX { @@ -266,7 +268,7 @@ struct hclge_mac { u32 fec_mode; /* active fec mode */ u32 user_fec_mode; u32 fec_ability; - int link; /* store the link status of mac & phy (if phy exit) */ + int link; /* store the link status of mac & phy (if phy exists) */ struct phy_device *phydev; struct mii_bus *mdio_bus; phy_interface_t phy_if; @@ -349,7 +351,7 @@ struct hclge_cfg { u8 mac_addr[ETH_ALEN]; u8 default_speed; u32 numa_node_map; - u8 speed_ability; + u16 speed_ability; u16 umv_space; }; @@ -749,7 +751,6 @@ struct hclge_dev { u16 num_tx_desc; /* desc num of per tx queue */ u16 num_rx_desc; /* desc num of per rx queue */ u8 hw_tc_map; - u8 tc_num_last_time; enum hclge_fc_mode fc_mode_last_time; u8 support_sfp_query; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c index 28db13253a5e..15f69fa86323 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c @@ -23,14 +23,11 @@ enum hclge_shaper_level { #define HCLGE_SHAPER_BS_U_DEF 5 #define HCLGE_SHAPER_BS_S_DEF 20 -#define HCLGE_ETHER_MAX_RATE 100000 - /* hclge_shaper_para_calc: calculate ir parameter for the shaper * @ir: Rate to be config, its unit is Mbps * @shaper_level: the shaper level. eg: port, pg, priority, queueset - * @ir_b: IR_B parameter of IR shaper - * @ir_u: IR_U parameter of IR shaper - * @ir_s: IR_S parameter of IR shaper + * @ir_para: parameters of IR shaper + * @max_tm_rate: max tm rate is available to config * * the formula: * @@ -41,7 +38,8 @@ enum hclge_shaper_level { * @return: 0: calculate sucessful, negative: fail */ static int hclge_shaper_para_calc(u32 ir, u8 shaper_level, - u8 *ir_b, u8 *ir_u, u8 *ir_s) + struct hclge_shaper_ir_para *ir_para, + u32 max_tm_rate) { #define DIVISOR_CLK (1000 * 8) #define DIVISOR_IR_B_126 (126 * DIVISOR_CLK) @@ -59,7 +57,7 @@ static int hclge_shaper_para_calc(u32 ir, u8 shaper_level, /* Calc tick */ if (shaper_level >= HCLGE_SHAPER_LVL_CNT || - ir > HCLGE_ETHER_MAX_RATE) + ir > max_tm_rate) return -EINVAL; tick = tick_array[shaper_level]; @@ -74,9 +72,9 @@ static int hclge_shaper_para_calc(u32 ir, u8 shaper_level, ir_calc = (DIVISOR_IR_B_126 + (tick >> 1) - 1) / tick; if (ir_calc == ir) { - *ir_b = 126; - *ir_u = 0; - *ir_s = 0; + ir_para->ir_b = 126; + ir_para->ir_u = 0; + ir_para->ir_s = 0; return 0; } else if (ir_calc > ir) { @@ -86,8 +84,8 @@ static int hclge_shaper_para_calc(u32 ir, u8 shaper_level, ir_calc = DIVISOR_IR_B_126 / (tick * (1 << ir_s_calc)); } - *ir_b = (ir * tick * (1 << ir_s_calc) + (DIVISOR_CLK >> 1)) / - DIVISOR_CLK; + ir_para->ir_b = (ir * tick * (1 << ir_s_calc) + + (DIVISOR_CLK >> 1)) / DIVISOR_CLK; } else { /* Increasing the numerator to select ir_u value */ u32 numerator; @@ -99,15 +97,16 @@ static int hclge_shaper_para_calc(u32 ir, u8 shaper_level, } if (ir_calc == ir) { - *ir_b = 126; + ir_para->ir_b = 126; } else { u32 denominator = DIVISOR_CLK * (1 << --ir_u_calc); - *ir_b = (ir * tick + (denominator >> 1)) / denominator; + ir_para->ir_b = (ir * tick + (denominator >> 1)) / + denominator; } } - *ir_u = ir_u_calc; - *ir_s = ir_s_calc; + ir_para->ir_u = ir_u_calc; + ir_para->ir_s = ir_s_calc; return 0; } @@ -400,21 +399,22 @@ static int hclge_tm_pg_shapping_cfg(struct hclge_dev *hdev, static int hclge_tm_port_shaper_cfg(struct hclge_dev *hdev) { struct hclge_port_shapping_cmd *shap_cfg_cmd; + struct hclge_shaper_ir_para ir_para; struct hclge_desc desc; - u8 ir_u, ir_b, ir_s; u32 shapping_para; int ret; - ret = hclge_shaper_para_calc(hdev->hw.mac.speed, - HCLGE_SHAPER_LVL_PORT, - &ir_b, &ir_u, &ir_s); + ret = hclge_shaper_para_calc(hdev->hw.mac.speed, HCLGE_SHAPER_LVL_PORT, + &ir_para, + hdev->ae_dev->dev_specs.max_tm_rate); if (ret) return ret; hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TM_PORT_SHAPPING, false); shap_cfg_cmd = (struct hclge_port_shapping_cmd *)desc.data; - shapping_para = hclge_tm_get_shapping_para(ir_b, ir_u, ir_s, + shapping_para = hclge_tm_get_shapping_para(ir_para.ir_b, ir_para.ir_u, + ir_para.ir_s, HCLGE_SHAPER_BS_U_DEF, HCLGE_SHAPER_BS_S_DEF); @@ -515,21 +515,23 @@ int hclge_tm_qs_shaper_cfg(struct hclge_vport *vport, int max_tx_rate) { struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo; struct hclge_qs_shapping_cmd *shap_cfg_cmd; + struct hclge_shaper_ir_para ir_para; struct hclge_dev *hdev = vport->back; struct hclge_desc desc; - u8 ir_b, ir_u, ir_s; u32 shaper_para; int ret, i; if (!max_tx_rate) - max_tx_rate = HCLGE_ETHER_MAX_RATE; + max_tx_rate = hdev->ae_dev->dev_specs.max_tm_rate; ret = hclge_shaper_para_calc(max_tx_rate, HCLGE_SHAPER_LVL_QSET, - &ir_b, &ir_u, &ir_s); + &ir_para, + hdev->ae_dev->dev_specs.max_tm_rate); if (ret) return ret; - shaper_para = hclge_tm_get_shapping_para(ir_b, ir_u, ir_s, + shaper_para = hclge_tm_get_shapping_para(ir_para.ir_b, ir_para.ir_u, + ir_para.ir_s, HCLGE_SHAPER_BS_U_DEF, HCLGE_SHAPER_BS_S_DEF); @@ -668,7 +670,8 @@ static void hclge_tm_pg_info_init(struct hclge_dev *hdev) hdev->tm_info.pg_info[i].pg_id = i; hdev->tm_info.pg_info[i].pg_sch_mode = HCLGE_SCH_MODE_DWRR; - hdev->tm_info.pg_info[i].bw_limit = HCLGE_ETHER_MAX_RATE; + hdev->tm_info.pg_info[i].bw_limit = + hdev->ae_dev->dev_specs.max_tm_rate; if (i != 0) continue; @@ -729,7 +732,8 @@ static int hclge_tm_pg_to_pri_map(struct hclge_dev *hdev) static int hclge_tm_pg_shaper_cfg(struct hclge_dev *hdev) { - u8 ir_u, ir_b, ir_s; + u32 max_tm_rate = hdev->ae_dev->dev_specs.max_tm_rate; + struct hclge_shaper_ir_para ir_para; u32 shaper_para; int ret; u32 i; @@ -741,10 +745,9 @@ static int hclge_tm_pg_shaper_cfg(struct hclge_dev *hdev) /* Pg to pri */ for (i = 0; i < hdev->tm_info.num_pg; i++) { /* Calc shaper para */ - ret = hclge_shaper_para_calc( - hdev->tm_info.pg_info[i].bw_limit, - HCLGE_SHAPER_LVL_PG, - &ir_b, &ir_u, &ir_s); + ret = hclge_shaper_para_calc(hdev->tm_info.pg_info[i].bw_limit, + HCLGE_SHAPER_LVL_PG, + &ir_para, max_tm_rate); if (ret) return ret; @@ -757,7 +760,9 @@ static int hclge_tm_pg_shaper_cfg(struct hclge_dev *hdev) if (ret) return ret; - shaper_para = hclge_tm_get_shapping_para(ir_b, ir_u, ir_s, + shaper_para = hclge_tm_get_shapping_para(ir_para.ir_b, + ir_para.ir_u, + ir_para.ir_s, HCLGE_SHAPER_BS_U_DEF, HCLGE_SHAPER_BS_S_DEF); ret = hclge_tm_pg_shapping_cfg(hdev, @@ -861,16 +866,16 @@ static int hclge_tm_pri_q_qs_cfg(struct hclge_dev *hdev) static int hclge_tm_pri_tc_base_shaper_cfg(struct hclge_dev *hdev) { - u8 ir_u, ir_b, ir_s; + u32 max_tm_rate = hdev->ae_dev->dev_specs.max_tm_rate; + struct hclge_shaper_ir_para ir_para; u32 shaper_para; int ret; u32 i; for (i = 0; i < hdev->tm_info.num_tc; i++) { - ret = hclge_shaper_para_calc( - hdev->tm_info.tc_info[i].bw_limit, - HCLGE_SHAPER_LVL_PRI, - &ir_b, &ir_u, &ir_s); + ret = hclge_shaper_para_calc(hdev->tm_info.tc_info[i].bw_limit, + HCLGE_SHAPER_LVL_PRI, + &ir_para, max_tm_rate); if (ret) return ret; @@ -882,7 +887,9 @@ static int hclge_tm_pri_tc_base_shaper_cfg(struct hclge_dev *hdev) if (ret) return ret; - shaper_para = hclge_tm_get_shapping_para(ir_b, ir_u, ir_s, + shaper_para = hclge_tm_get_shapping_para(ir_para.ir_b, + ir_para.ir_u, + ir_para.ir_s, HCLGE_SHAPER_BS_U_DEF, HCLGE_SHAPER_BS_S_DEF); ret = hclge_tm_pri_shapping_cfg(hdev, HCLGE_TM_SHAP_P_BUCKET, i, @@ -897,12 +904,13 @@ static int hclge_tm_pri_tc_base_shaper_cfg(struct hclge_dev *hdev) static int hclge_tm_pri_vnet_base_shaper_pri_cfg(struct hclge_vport *vport) { struct hclge_dev *hdev = vport->back; - u8 ir_u, ir_b, ir_s; + struct hclge_shaper_ir_para ir_para; u32 shaper_para; int ret; ret = hclge_shaper_para_calc(vport->bw_limit, HCLGE_SHAPER_LVL_VF, - &ir_b, &ir_u, &ir_s); + &ir_para, + hdev->ae_dev->dev_specs.max_tm_rate); if (ret) return ret; @@ -914,7 +922,8 @@ static int hclge_tm_pri_vnet_base_shaper_pri_cfg(struct hclge_vport *vport) if (ret) return ret; - shaper_para = hclge_tm_get_shapping_para(ir_b, ir_u, ir_s, + shaper_para = hclge_tm_get_shapping_para(ir_para.ir_b, ir_para.ir_u, + ir_para.ir_s, HCLGE_SHAPER_BS_U_DEF, HCLGE_SHAPER_BS_S_DEF); ret = hclge_tm_pri_shapping_cfg(hdev, HCLGE_TM_SHAP_P_BUCKET, @@ -929,15 +938,15 @@ static int hclge_tm_pri_vnet_base_shaper_qs_cfg(struct hclge_vport *vport) { struct hnae3_knic_private_info *kinfo = &vport->nic.kinfo; struct hclge_dev *hdev = vport->back; - u8 ir_u, ir_b, ir_s; + u32 max_tm_rate = hdev->ae_dev->dev_specs.max_tm_rate; + struct hclge_shaper_ir_para ir_para; u32 i; int ret; for (i = 0; i < kinfo->num_tc; i++) { - ret = hclge_shaper_para_calc( - hdev->tm_info.tc_info[i].bw_limit, - HCLGE_SHAPER_LVL_QSET, - &ir_b, &ir_u, &ir_s); + ret = hclge_shaper_para_calc(hdev->tm_info.tc_info[i].bw_limit, + HCLGE_SHAPER_LVL_QSET, + &ir_para, max_tm_rate); if (ret) return ret; } @@ -1355,7 +1364,7 @@ static int hclge_mac_pause_setup_hw(struct hclge_dev *hdev) static int hclge_tm_bp_setup(struct hclge_dev *hdev) { - int ret = 0; + int ret; int i; for (i = 0; i < hdev->tm_info.num_tc; i++) { diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.h index 45bcb67f90fd..bb2a2d8e9259 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.h @@ -19,6 +19,8 @@ #define HCLGE_TM_TX_SCHD_DWRR_MSK BIT(0) #define HCLGE_TM_TX_SCHD_SP_MSK (0xFE) +#define HCLGE_ETHER_MAX_RATE 100000 + struct hclge_pg_to_pri_link_cmd { u8 pg_id; u8 rsvd1[3]; @@ -139,6 +141,12 @@ struct hclge_port_shapping_cmd { __le32 port_shapping_para; }; +struct hclge_shaper_ir_para { + u8 ir_b; /* IR_B parameter of IR shaper */ + u8 ir_u; /* IR_U parameter of IR shaper */ + u8 ir_s; /* IR_S parameter of IR shaper */ +}; + #define hclge_tm_set_field(dest, string, val) \ hnae3_set_field((dest), \ (HCLGE_TM_SHAP_##string##_MSK), \ diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c index fec65239a3c8..66866c1cfb12 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c @@ -313,9 +313,34 @@ int hclgevf_cmd_send(struct hclgevf_hw *hw, struct hclgevf_desc *desc, int num) return status; } -static int hclgevf_cmd_query_firmware_version(struct hclgevf_hw *hw, - u32 *version) +static void hclgevf_set_default_capability(struct hclgevf_dev *hdev) { + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev); + + set_bit(HNAE3_DEV_SUPPORT_FD_B, ae_dev->caps); + set_bit(HNAE3_DEV_SUPPORT_GRO_B, ae_dev->caps); + set_bit(HNAE3_DEV_SUPPORT_FEC_B, ae_dev->caps); +} + +static void hclgevf_parse_capability(struct hclgevf_dev *hdev, + struct hclgevf_query_version_cmd *cmd) +{ + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev); + u32 caps; + + caps = __le32_to_cpu(cmd->caps[0]); + + if (hnae3_get_bit(caps, HCLGEVF_CAP_UDP_GSO_B)) + set_bit(HNAE3_DEV_SUPPORT_UDP_GSO_B, ae_dev->caps); + if (hnae3_get_bit(caps, HCLGEVF_CAP_INT_QL_B)) + set_bit(HNAE3_DEV_SUPPORT_INT_QL_B, ae_dev->caps); + if (hnae3_get_bit(caps, HCLGEVF_CAP_TQP_TXRX_INDEP_B)) + set_bit(HNAE3_DEV_SUPPORT_TQP_TXRX_INDEP_B, ae_dev->caps); +} + +static int hclgevf_cmd_query_version_and_capability(struct hclgevf_dev *hdev) +{ + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev); struct hclgevf_query_version_cmd *resp; struct hclgevf_desc desc; int status; @@ -323,9 +348,20 @@ static int hclgevf_cmd_query_firmware_version(struct hclgevf_hw *hw, resp = (struct hclgevf_query_version_cmd *)desc.data; hclgevf_cmd_setup_basic_desc(&desc, HCLGEVF_OPC_QUERY_FW_VER, 1); - status = hclgevf_cmd_send(hw, &desc, 1); - if (!status) - *version = le32_to_cpu(resp->firmware); + status = hclgevf_cmd_send(&hdev->hw, &desc, 1); + if (status) + return status; + + hdev->fw_version = le32_to_cpu(resp->firmware); + + ae_dev->dev_version = le32_to_cpu(resp->hardware) << + HNAE3_PCI_REVISION_BIT_SIZE; + ae_dev->dev_version |= hdev->pdev->revision; + + if (ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) + hclgevf_set_default_capability(hdev); + + hclgevf_parse_capability(hdev, resp); return status; } @@ -364,7 +400,6 @@ err_csq: int hclgevf_cmd_init(struct hclgevf_dev *hdev) { - u32 version; int ret; spin_lock_bh(&hdev->hw.cmq.csq.lock); @@ -395,23 +430,22 @@ int hclgevf_cmd_init(struct hclgevf_dev *hdev) goto err_cmd_init; } - /* get firmware version */ - ret = hclgevf_cmd_query_firmware_version(&hdev->hw, &version); + /* get version and device capabilities */ + ret = hclgevf_cmd_query_version_and_capability(hdev); if (ret) { dev_err(&hdev->pdev->dev, - "failed(%d) to query firmware version\n", ret); + "failed to query version and capabilities, ret = %d\n", ret); goto err_cmd_init; } - hdev->fw_version = version; dev_info(&hdev->pdev->dev, "The firmware version is %lu.%lu.%lu.%lu\n", - hnae3_get_field(version, HNAE3_FW_VERSION_BYTE3_MASK, + hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE3_MASK, HNAE3_FW_VERSION_BYTE3_SHIFT), - hnae3_get_field(version, HNAE3_FW_VERSION_BYTE2_MASK, + hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE2_MASK, HNAE3_FW_VERSION_BYTE2_SHIFT), - hnae3_get_field(version, HNAE3_FW_VERSION_BYTE1_MASK, + hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE1_MASK, HNAE3_FW_VERSION_BYTE1_SHIFT), - hnae3_get_field(version, HNAE3_FW_VERSION_BYTE0_MASK, + hnae3_get_field(hdev->fw_version, HNAE3_FW_VERSION_BYTE0_MASK, HNAE3_FW_VERSION_BYTE0_SHIFT)); return 0; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.h index 40d6e602ab51..9460c128c095 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.h @@ -91,6 +91,8 @@ enum hclgevf_opcode_type { /* Generic command */ HCLGEVF_OPC_QUERY_FW_VER = 0x0001, HCLGEVF_OPC_QUERY_VF_RSRC = 0x0024, + HCLGEVF_OPC_QUERY_DEV_SPECS = 0x0050, + /* TQP command */ HCLGEVF_OPC_QUERY_TX_STATUS = 0x0B03, HCLGEVF_OPC_QUERY_RX_STATUS = 0x0B13, @@ -141,9 +143,26 @@ struct hclgevf_ctrl_vector_chain { u8 resv; }; +enum HCLGEVF_CAP_BITS { + HCLGEVF_CAP_UDP_GSO_B, + HCLGEVF_CAP_QB_B, + HCLGEVF_CAP_FD_FORWARD_TC_B, + HCLGEVF_CAP_PTP_B, + HCLGEVF_CAP_INT_QL_B, + HCLGEVF_CAP_SIMPLE_BD_B, + HCLGEVF_CAP_TX_PUSH_B, + HCLGEVF_CAP_PHY_IMP_B, + HCLGEVF_CAP_TQP_TXRX_INDEP_B, + HCLGEVF_CAP_HW_PAD_B, + HCLGEVF_CAP_STASH_B, +}; + +#define HCLGEVF_QUERY_CAP_LENGTH 3 struct hclgevf_query_version_cmd { __le32 firmware; - __le32 firmware_rsv[5]; + __le32 hardware; + __le32 rsv; + __le32 caps[HCLGEVF_QUERY_CAP_LENGTH]; /* capabilities of device */ }; #define HCLGEVF_MSIX_OFT_ROCEE_S 0 @@ -253,6 +272,19 @@ struct hclgevf_cfg_tx_queue_pointer_cmd { #define HCLGEVF_NIC_CMQ_DESC_NUM_S 3 #define HCLGEVF_NIC_CMDQ_INT_SRC_REG 0x27100 +#define HCLGEVF_QUERY_DEV_SPECS_BD_NUM 4 + +struct hclgevf_dev_specs_0_cmd { + __le32 rsv0; + __le32 mac_entry_num; + __le32 mng_entry_num; + __le16 rss_ind_tbl_size; + __le16 rss_key_size; + __le16 int_ql_max; + u8 max_non_tso_bd_num; + u8 rsv1[5]; +}; + static inline void hclgevf_write_reg(void __iomem *base, u32 reg, u32 value) { writel(value, base + reg); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c index e972138a14ad..50c84c5e65d2 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c @@ -19,8 +19,9 @@ static struct hnae3_ae_algo ae_algovf; static struct workqueue_struct *hclgevf_wq; static const struct pci_device_id ae_algovf_pci_tbl[] = { - {PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_100G_VF), 0}, - {PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_100G_RDMA_DCB_PFC_VF), 0}, + {PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_VF), 0}, + {PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_RDMA_DCB_PFC_VF), + HNAE3_DEV_SUPPORT_ROCE_DCB_BITS}, /* required last entry */ {0, } }; @@ -171,7 +172,7 @@ static u8 *hclgevf_tqps_get_strings(struct hnae3_handle *handle, u8 *data) { struct hnae3_knic_private_info *kinfo = &handle->kinfo; u8 *buff = data; - int i = 0; + int i; for (i = 0; i < kinfo->num_tqps; i++) { struct hclgevf_tqp *tqp = container_of(kinfo->tqp[i], @@ -745,7 +746,7 @@ static int hclgevf_get_rss(struct hnae3_handle *handle, u32 *indir, u8 *key, struct hclgevf_rss_cfg *rss_cfg = &hdev->rss_cfg; int i, ret; - if (handle->pdev->revision >= 0x21) { + if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) { /* Get hash algorithm */ if (hfunc) { switch (rss_cfg->hash_algo) { @@ -791,7 +792,7 @@ static int hclgevf_set_rss(struct hnae3_handle *handle, const u32 *indir, struct hclgevf_rss_cfg *rss_cfg = &hdev->rss_cfg; int ret, i; - if (handle->pdev->revision >= 0x21) { + if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) { /* Set the RSS Hash Key if specififed by the user */ if (key) { switch (hfunc) { @@ -863,7 +864,7 @@ static int hclgevf_set_rss_tuple(struct hnae3_handle *handle, u8 tuple_sets; int ret; - if (handle->pdev->revision == 0x20) + if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2) return -EOPNOTSUPP; if (nfc->data & @@ -941,7 +942,7 @@ static int hclgevf_get_rss_tuple(struct hnae3_handle *handle, struct hclgevf_rss_cfg *rss_cfg = &hdev->rss_cfg; u8 tuple_sets; - if (handle->pdev->revision == 0x20) + if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2) return -EOPNOTSUPP; nfc->data = 0; @@ -1154,10 +1155,9 @@ static int hclgevf_set_promisc_mode(struct hnae3_handle *handle, bool en_uc_pmc, bool en_mc_pmc) { struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); - struct pci_dev *pdev = hdev->pdev; bool en_bc_pmc; - en_bc_pmc = pdev->revision != 0x20; + en_bc_pmc = hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2; return hclgevf_cmd_set_promisc_mode(hdev, en_uc_pmc, en_mc_pmc, en_bc_pmc); @@ -1702,6 +1702,26 @@ static int hclgevf_notify_client(struct hclgevf_dev *hdev, return ret; } +static int hclgevf_notify_roce_client(struct hclgevf_dev *hdev, + enum hnae3_reset_notify_type type) +{ + struct hnae3_client *client = hdev->roce_client; + struct hnae3_handle *handle = &hdev->roce; + int ret; + + if (!test_bit(HCLGEVF_STATE_ROCE_REGISTERED, &hdev->state) || !client) + return 0; + + if (!client->ops->reset_notify) + return -EOPNOTSUPP; + + ret = client->ops->reset_notify(handle, type); + if (ret) + dev_err(&hdev->pdev->dev, "notify roce client failed %d(%d)", + type, ret); + return ret; +} + static int hclgevf_reset_wait(struct hclgevf_dev *hdev) { #define HCLGEVF_RESET_WAIT_US 20000 @@ -1788,10 +1808,10 @@ static int hclgevf_reset_prepare_wait(struct hclgevf_dev *hdev) { #define HCLGEVF_RESET_SYNC_TIME 100 - struct hclge_vf_to_pf_msg send_msg; - int ret = 0; - if (hdev->reset_type == HNAE3_VF_FUNC_RESET) { + struct hclge_vf_to_pf_msg send_msg; + int ret; + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_RESET, 0); ret = hclgevf_send_mbx_msg(hdev, &send_msg, true, NULL, 0); if (ret) { @@ -1806,10 +1826,10 @@ static int hclgevf_reset_prepare_wait(struct hclgevf_dev *hdev) /* inform hardware that preparatory work is done */ msleep(HCLGEVF_RESET_SYNC_TIME); hclgevf_reset_handshake(hdev, true); - dev_info(&hdev->pdev->dev, "prepare reset(%d) wait done, ret:%d\n", - hdev->reset_type, ret); + dev_info(&hdev->pdev->dev, "prepare reset(%d) wait done\n", + hdev->reset_type); - return ret; + return 0; } static void hclgevf_dump_rst_info(struct hclgevf_dev *hdev) @@ -1865,6 +1885,11 @@ static int hclgevf_reset_prepare(struct hclgevf_dev *hdev) hdev->rst_stats.rst_cnt++; + /* perform reset of the stack & ae device for a client */ + ret = hclgevf_notify_roce_client(hdev, HNAE3_DOWN_CLIENT); + if (ret) + return ret; + rtnl_lock(); /* bring down the nic to stop any ongoing TX/RX */ ret = hclgevf_notify_client(hdev, HNAE3_DOWN_CLIENT); @@ -1880,6 +1905,9 @@ static int hclgevf_reset_rebuild(struct hclgevf_dev *hdev) int ret; hdev->rst_stats.hw_rst_done_cnt++; + ret = hclgevf_notify_roce_client(hdev, HNAE3_UNINIT_CLIENT); + if (ret) + return ret; rtnl_lock(); /* now, re-initialize the nic client and ae device */ @@ -1890,6 +1918,18 @@ static int hclgevf_reset_rebuild(struct hclgevf_dev *hdev) return ret; } + ret = hclgevf_notify_roce_client(hdev, HNAE3_INIT_CLIENT); + /* ignore RoCE notify error if it fails HCLGEVF_RESET_MAX_FAIL_CNT - 1 + * times + */ + if (ret && + hdev->rst_stats.rst_fail_cnt < HCLGEVF_RESET_MAX_FAIL_CNT - 1) + return ret; + + ret = hclgevf_notify_roce_client(hdev, HNAE3_UP_CLIENT); + if (ret) + return ret; + hdev->last_reset_time = jiffies; hdev->rst_stats.rst_done_cnt++; hdev->rst_stats.rst_fail_cnt = 0; @@ -2186,6 +2226,9 @@ static void hclgevf_periodic_service_task(struct hclgevf_dev *hdev) unsigned long delta = round_jiffies_relative(HZ); struct hnae3_handle *handle = &hdev->nic; + if (test_bit(HCLGEVF_STATE_RST_FAIL, &hdev->state)) + return; + if (time_is_after_jiffies(hdev->last_serv_processed + HZ)) { delta = jiffies - hdev->last_serv_processed; @@ -2284,7 +2327,7 @@ static enum hclgevf_evt_cause hclgevf_check_evt_cause(struct hclgevf_dev *hdev, * register, so we should just write 0 to the bit we are * handling, and keep other bits as cmdq_stat_reg. */ - if (hdev->pdev->revision >= 0x21) + if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) *clearval = ~(1U << HCLGEVF_VECTOR0_RX_CMDQ_INT_B); else *clearval = cmdq_stat_reg & @@ -2427,7 +2470,7 @@ static void hclgevf_rss_init_cfg(struct hclgevf_dev *hdev) rss_cfg->hash_algo = HCLGEVF_RSS_HASH_ALGO_TOEPLITZ; rss_cfg->rss_size = hdev->nic.kinfo.rss_size; tuple_sets = &rss_cfg->rss_tuple_sets; - if (hdev->pdev->revision >= 0x21) { + if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) { rss_cfg->hash_algo = HCLGEVF_RSS_HASH_ALGO_SIMPLE; memcpy(rss_cfg->rss_hash_key, hclgevf_hash_key, HCLGEVF_RSS_KEY_SIZE); @@ -2452,7 +2495,7 @@ static int hclgevf_rss_init_hw(struct hclgevf_dev *hdev) struct hclgevf_rss_cfg *rss_cfg = &hdev->rss_cfg; int ret; - if (hdev->pdev->revision >= 0x21) { + if (hdev->ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) { ret = hclgevf_set_rss_algo_key(hdev, rss_cfg->hash_algo, rss_cfg->rss_hash_key); if (ret) @@ -2551,13 +2594,7 @@ static int hclgevf_set_alive(struct hnae3_handle *handle, bool alive) static int hclgevf_client_start(struct hnae3_handle *handle) { - int ret; - - ret = hclgevf_set_alive(handle, true); - if (ret) - return ret; - - return 0; + return hclgevf_set_alive(handle, true); } static void hclgevf_client_stop(struct hnae3_handle *handle) @@ -2760,6 +2797,7 @@ static int hclgevf_init_roce_client_instance(struct hnae3_ae_dev *ae_dev, if (ret) return ret; + set_bit(HCLGEVF_STATE_ROCE_REGISTERED, &hdev->state); hnae3_set_client_init_flag(client, ae_dev, 1); return 0; @@ -2820,6 +2858,7 @@ static void hclgevf_uninit_client_instance(struct hnae3_client *client, /* un-init roce, if it exists */ if (hdev->roce_client) { + clear_bit(HCLGEVF_STATE_ROCE_REGISTERED, &hdev->state); hdev->roce_client->ops->uninit_instance(&hdev->roce, 0); hdev->roce_client = NULL; hdev->roce.client = NULL; @@ -2942,6 +2981,76 @@ static int hclgevf_query_vf_resource(struct hclgevf_dev *hdev) return 0; } +static void hclgevf_set_default_dev_specs(struct hclgevf_dev *hdev) +{ +#define HCLGEVF_MAX_NON_TSO_BD_NUM 8U + + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev); + + ae_dev->dev_specs.max_non_tso_bd_num = + HCLGEVF_MAX_NON_TSO_BD_NUM; + ae_dev->dev_specs.rss_ind_tbl_size = HCLGEVF_RSS_IND_TBL_SIZE; + ae_dev->dev_specs.rss_key_size = HCLGEVF_RSS_KEY_SIZE; +} + +static void hclgevf_parse_dev_specs(struct hclgevf_dev *hdev, + struct hclgevf_desc *desc) +{ + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev); + struct hclgevf_dev_specs_0_cmd *req0; + + req0 = (struct hclgevf_dev_specs_0_cmd *)desc[0].data; + + ae_dev->dev_specs.max_non_tso_bd_num = req0->max_non_tso_bd_num; + ae_dev->dev_specs.rss_ind_tbl_size = + le16_to_cpu(req0->rss_ind_tbl_size); + ae_dev->dev_specs.rss_key_size = le16_to_cpu(req0->rss_key_size); +} + +static void hclgevf_check_dev_specs(struct hclgevf_dev *hdev) +{ + struct hnae3_dev_specs *dev_specs = &hdev->ae_dev->dev_specs; + + if (!dev_specs->max_non_tso_bd_num) + dev_specs->max_non_tso_bd_num = HCLGEVF_MAX_NON_TSO_BD_NUM; + if (!dev_specs->rss_ind_tbl_size) + dev_specs->rss_ind_tbl_size = HCLGEVF_RSS_IND_TBL_SIZE; + if (!dev_specs->rss_key_size) + dev_specs->rss_key_size = HCLGEVF_RSS_KEY_SIZE; +} + +static int hclgevf_query_dev_specs(struct hclgevf_dev *hdev) +{ + struct hclgevf_desc desc[HCLGEVF_QUERY_DEV_SPECS_BD_NUM]; + int ret; + int i; + + /* set default specifications as devices lower than version V3 do not + * support querying specifications from firmware. + */ + if (hdev->ae_dev->dev_version < HNAE3_DEVICE_VERSION_V3) { + hclgevf_set_default_dev_specs(hdev); + return 0; + } + + for (i = 0; i < HCLGEVF_QUERY_DEV_SPECS_BD_NUM - 1; i++) { + hclgevf_cmd_setup_basic_desc(&desc[i], + HCLGEVF_OPC_QUERY_DEV_SPECS, true); + desc[i].flag |= cpu_to_le16(HCLGEVF_CMD_FLAG_NEXT); + } + hclgevf_cmd_setup_basic_desc(&desc[i], HCLGEVF_OPC_QUERY_DEV_SPECS, + true); + + ret = hclgevf_cmd_send(&hdev->hw, desc, HCLGEVF_QUERY_DEV_SPECS_BD_NUM); + if (ret) + return ret; + + hclgevf_parse_dev_specs(hdev, desc); + hclgevf_check_dev_specs(hdev); + + return 0; +} + static int hclgevf_pci_reset(struct hclgevf_dev *hdev) { struct pci_dev *pdev = hdev->pdev; @@ -3050,6 +3159,13 @@ static int hclgevf_init_hdev(struct hclgevf_dev *hdev) if (ret) goto err_cmd_init; + ret = hclgevf_query_dev_specs(hdev); + if (ret) { + dev_err(&pdev->dev, + "failed to query dev specifications, ret = %d\n", ret); + goto err_cmd_init; + } + ret = hclgevf_init_msi(hdev); if (ret) { dev_err(&pdev->dev, "failed(%d) to init MSI/MSI-X\n", ret); @@ -3345,6 +3461,13 @@ static bool hclgevf_get_hw_reset_stat(struct hnae3_handle *handle) return !!hclgevf_read_dev(&hdev->hw, HCLGEVF_RST_ING); } +static bool hclgevf_get_cmdq_stat(struct hnae3_handle *handle) +{ + struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); + + return test_bit(HCLGEVF_STATE_CMD_DISABLE, &hdev->state); +} + static bool hclgevf_ae_dev_resetting(struct hnae3_handle *handle) { struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); @@ -3530,6 +3653,7 @@ static const struct hnae3_ae_ops hclgevf_ops = { .get_link_mode = hclgevf_get_link_mode, .set_promisc_mode = hclgevf_set_promisc_mode, .request_update_promisc_mode = hclgevf_request_update_promisc_mode, + .get_cmdq_stat = hclgevf_get_cmdq_stat, }; static struct hnae3_ae_algo ae_algovf = { diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h index c1fac8920ae3..c5bcc3894fd5 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h @@ -139,6 +139,7 @@ enum hclgevf_states { HCLGEVF_STATE_IRQ_INITED, HCLGEVF_STATE_REMOVING, HCLGEVF_STATE_NIC_REGISTERED, + HCLGEVF_STATE_ROCE_REGISTERED, /* task states */ HCLGEVF_STATE_RST_SERVICE_SCHED, HCLGEVF_STATE_RST_HANDLING, diff --git a/drivers/net/ethernet/hisilicon/hns_mdio.c b/drivers/net/ethernet/hisilicon/hns_mdio.c index 7df5d7d211d4..883d0d7c6858 100644 --- a/drivers/net/ethernet/hisilicon/hns_mdio.c +++ b/drivers/net/ethernet/hisilicon/hns_mdio.c @@ -210,7 +210,7 @@ static void hns_mdio_cmd_write(struct hns_mdio_device *mdio_dev, * @bus: mdio bus * @phy_id: phy id * @regnum: register num - * @value: register value + * @data: register value * * Return 0 on success, negative on failure */ @@ -273,7 +273,6 @@ static int hns_mdio_write(struct mii_bus *bus, * @bus: mdio bus * @phy_id: phy id * @regnum: register num - * @value: register value * * Return phy register value */ diff --git a/drivers/net/ethernet/huawei/hinic/Makefile b/drivers/net/ethernet/huawei/hinic/Makefile index 67b59d0ba769..2f89119c9b69 100644 --- a/drivers/net/ethernet/huawei/hinic/Makefile +++ b/drivers/net/ethernet/huawei/hinic/Makefile @@ -4,4 +4,5 @@ obj-$(CONFIG_HINIC) += hinic.o hinic-y := hinic_main.o hinic_tx.o hinic_rx.o hinic_port.o hinic_hw_dev.o \ hinic_hw_io.o hinic_hw_qp.o hinic_hw_cmdq.o hinic_hw_wq.o \ hinic_hw_mgmt.o hinic_hw_api_cmd.o hinic_hw_eqs.o hinic_hw_if.o \ - hinic_common.o hinic_ethtool.o hinic_devlink.o hinic_hw_mbox.o hinic_sriov.o + hinic_common.o hinic_ethtool.o hinic_devlink.o hinic_hw_mbox.o \ + hinic_sriov.o hinic_debugfs.o diff --git a/drivers/net/ethernet/huawei/hinic/hinic_debugfs.c b/drivers/net/ethernet/huawei/hinic/hinic_debugfs.c new file mode 100644 index 000000000000..19eb839177ec --- /dev/null +++ b/drivers/net/ethernet/huawei/hinic/hinic_debugfs.c @@ -0,0 +1,318 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Huawei HiNIC PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + */ + +#include <linux/debugfs.h> +#include <linux/device.h> + +#include "hinic_debugfs.h" + +static struct dentry *hinic_dbgfs_root; + +enum sq_dbg_info { + GLB_SQ_ID, + SQ_PI, + SQ_CI, + SQ_FI, + SQ_MSIX_ENTRY, +}; + +static char *sq_fields[] = {"glb_sq_id", "sq_pi", "sq_ci", "sq_fi", "sq_msix_entry"}; + +static u64 hinic_dbg_get_sq_info(struct hinic_dev *nic_dev, struct hinic_sq *sq, int idx) +{ + struct hinic_wq *wq = sq->wq; + + switch (idx) { + case GLB_SQ_ID: + return nic_dev->hwdev->func_to_io.global_qpn + sq->qid; + case SQ_PI: + return atomic_read(&wq->prod_idx) & wq->mask; + case SQ_CI: + return atomic_read(&wq->cons_idx) & wq->mask; + case SQ_FI: + return be16_to_cpu(*(__be16 *)(sq->hw_ci_addr)) & wq->mask; + case SQ_MSIX_ENTRY: + return sq->msix_entry; + } + + return 0; +} + +enum rq_dbg_info { + GLB_RQ_ID, + RQ_HW_PI, + RQ_SW_CI, + RQ_SW_PI, + RQ_MSIX_ENTRY, +}; + +static char *rq_fields[] = {"glb_rq_id", "rq_hw_pi", "rq_sw_ci", "rq_sw_pi", "rq_msix_entry"}; + +static u64 hinic_dbg_get_rq_info(struct hinic_dev *nic_dev, struct hinic_rq *rq, int idx) +{ + struct hinic_wq *wq = rq->wq; + + switch (idx) { + case GLB_RQ_ID: + return nic_dev->hwdev->func_to_io.global_qpn + rq->qid; + case RQ_HW_PI: + return be16_to_cpu(*(__be16 *)(rq->pi_virt_addr)) & wq->mask; + case RQ_SW_CI: + return atomic_read(&wq->cons_idx) & wq->mask; + case RQ_SW_PI: + return atomic_read(&wq->prod_idx) & wq->mask; + case RQ_MSIX_ENTRY: + return rq->msix_entry; + } + + return 0; +} + +enum func_tbl_info { + VALID, + RX_MODE, + MTU, + RQ_DEPTH, + QUEUE_NUM, +}; + +static char *func_table_fields[] = {"valid", "rx_mode", "mtu", "rq_depth", "cfg_q_num"}; + +static int hinic_dbg_get_func_table(struct hinic_dev *nic_dev, int idx) +{ + struct tag_sml_funcfg_tbl *funcfg_table_elem; + struct hinic_cmd_lt_rd *read_data; + u16 out_size = sizeof(*read_data); + int err; + + read_data = kzalloc(sizeof(*read_data), GFP_KERNEL); + if (!read_data) + return ~0; + + read_data->node = TBL_ID_FUNC_CFG_SM_NODE; + read_data->inst = TBL_ID_FUNC_CFG_SM_INST; + read_data->entry_size = HINIC_FUNCTION_CONFIGURE_TABLE_SIZE; + read_data->lt_index = HINIC_HWIF_FUNC_IDX(nic_dev->hwdev->hwif); + read_data->len = HINIC_FUNCTION_CONFIGURE_TABLE_SIZE; + + err = hinic_port_msg_cmd(nic_dev->hwdev, HINIC_PORT_CMD_RD_LINE_TBL, read_data, + sizeof(*read_data), read_data, &out_size); + if (err || out_size != sizeof(*read_data) || read_data->status) { + netif_err(nic_dev, drv, nic_dev->netdev, + "Failed to get func table, err: %d, status: 0x%x, out size: 0x%x\n", + err, read_data->status, out_size); + kfree(read_data); + return ~0; + } + + funcfg_table_elem = (struct tag_sml_funcfg_tbl *)read_data->data; + + switch (idx) { + case VALID: + return funcfg_table_elem->dw0.bs.valid; + case RX_MODE: + return funcfg_table_elem->dw0.bs.nic_rx_mode; + case MTU: + return funcfg_table_elem->dw1.bs.mtu; + case RQ_DEPTH: + return funcfg_table_elem->dw13.bs.cfg_rq_depth; + case QUEUE_NUM: + return funcfg_table_elem->dw13.bs.cfg_q_num; + } + + kfree(read_data); + + return ~0; +} + +static ssize_t hinic_dbg_cmd_read(struct file *filp, char __user *buffer, size_t count, + loff_t *ppos) +{ + struct hinic_debug_priv *dbg; + char ret_buf[20]; + int *desc; + u64 out; + int ret; + + desc = filp->private_data; + dbg = container_of(desc, struct hinic_debug_priv, field_id[*desc]); + + switch (dbg->type) { + case HINIC_DBG_SQ_INFO: + out = hinic_dbg_get_sq_info(dbg->dev, dbg->object, *desc); + break; + + case HINIC_DBG_RQ_INFO: + out = hinic_dbg_get_rq_info(dbg->dev, dbg->object, *desc); + break; + + case HINIC_DBG_FUNC_TABLE: + out = hinic_dbg_get_func_table(dbg->dev, *desc); + break; + + default: + netif_warn(dbg->dev, drv, dbg->dev->netdev, "Invalid hinic debug cmd: %d\n", + dbg->type); + return -EINVAL; + } + + ret = snprintf(ret_buf, sizeof(ret_buf), "0x%llx\n", out); + + return simple_read_from_buffer(buffer, count, ppos, ret_buf, ret); +} + +static const struct file_operations hinic_dbg_cmd_fops = { + .owner = THIS_MODULE, + .open = simple_open, + .read = hinic_dbg_cmd_read, +}; + +static int create_dbg_files(struct hinic_dev *dev, enum hinic_dbg_type type, void *data, + struct dentry *root, struct hinic_debug_priv **dbg, char **field, + int nfile) +{ + struct hinic_debug_priv *tmp; + int i; + + tmp = kzalloc(sizeof(*tmp), GFP_KERNEL); + if (!tmp) + return -ENOMEM; + + tmp->dev = dev; + tmp->object = data; + tmp->type = type; + tmp->root = root; + + for (i = 0; i < nfile; i++) { + tmp->field_id[i] = i; + debugfs_create_file(field[i], 0400, root, &tmp->field_id[i], &hinic_dbg_cmd_fops); + } + + *dbg = tmp; + + return 0; +} + +static void rem_dbg_files(struct hinic_debug_priv *dbg) +{ + if (dbg->type != HINIC_DBG_FUNC_TABLE) + debugfs_remove_recursive(dbg->root); + + kfree(dbg); +} + +int hinic_sq_debug_add(struct hinic_dev *dev, u16 sq_id) +{ + struct hinic_sq *sq; + struct dentry *root; + char sub_dir[16]; + + sq = dev->txqs[sq_id].sq; + + sprintf(sub_dir, "0x%x", sq_id); + + root = debugfs_create_dir(sub_dir, dev->sq_dbgfs); + + return create_dbg_files(dev, HINIC_DBG_SQ_INFO, sq, root, &sq->dbg, sq_fields, + ARRAY_SIZE(sq_fields)); +} + +void hinic_sq_debug_rem(struct hinic_sq *sq) +{ + if (sq->dbg) + rem_dbg_files(sq->dbg); +} + +int hinic_rq_debug_add(struct hinic_dev *dev, u16 rq_id) +{ + struct hinic_rq *rq; + struct dentry *root; + char sub_dir[16]; + + rq = dev->rxqs[rq_id].rq; + + sprintf(sub_dir, "0x%x", rq_id); + + root = debugfs_create_dir(sub_dir, dev->rq_dbgfs); + + return create_dbg_files(dev, HINIC_DBG_RQ_INFO, rq, root, &rq->dbg, rq_fields, + ARRAY_SIZE(rq_fields)); +} + +void hinic_rq_debug_rem(struct hinic_rq *rq) +{ + if (rq->dbg) + rem_dbg_files(rq->dbg); +} + +int hinic_func_table_debug_add(struct hinic_dev *dev) +{ + if (HINIC_IS_VF(dev->hwdev->hwif)) + return 0; + + return create_dbg_files(dev, HINIC_DBG_FUNC_TABLE, dev, dev->func_tbl_dbgfs, &dev->dbg, + func_table_fields, ARRAY_SIZE(func_table_fields)); +} + +void hinic_func_table_debug_rem(struct hinic_dev *dev) +{ + if (!HINIC_IS_VF(dev->hwdev->hwif) && dev->dbg) + rem_dbg_files(dev->dbg); +} + +void hinic_sq_dbgfs_init(struct hinic_dev *nic_dev) +{ + nic_dev->sq_dbgfs = debugfs_create_dir("SQs", nic_dev->dbgfs_root); +} + +void hinic_sq_dbgfs_uninit(struct hinic_dev *nic_dev) +{ + debugfs_remove_recursive(nic_dev->sq_dbgfs); +} + +void hinic_rq_dbgfs_init(struct hinic_dev *nic_dev) +{ + nic_dev->rq_dbgfs = debugfs_create_dir("RQs", nic_dev->dbgfs_root); +} + +void hinic_rq_dbgfs_uninit(struct hinic_dev *nic_dev) +{ + debugfs_remove_recursive(nic_dev->rq_dbgfs); +} + +void hinic_func_tbl_dbgfs_init(struct hinic_dev *nic_dev) +{ + if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) + nic_dev->func_tbl_dbgfs = debugfs_create_dir("func_table", nic_dev->dbgfs_root); +} + +void hinic_func_tbl_dbgfs_uninit(struct hinic_dev *nic_dev) +{ + if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) + debugfs_remove_recursive(nic_dev->func_tbl_dbgfs); +} + +void hinic_dbg_init(struct hinic_dev *nic_dev) +{ + nic_dev->dbgfs_root = debugfs_create_dir(pci_name(nic_dev->hwdev->hwif->pdev), + hinic_dbgfs_root); +} + +void hinic_dbg_uninit(struct hinic_dev *nic_dev) +{ + debugfs_remove_recursive(nic_dev->dbgfs_root); + nic_dev->dbgfs_root = NULL; +} + +void hinic_dbg_register_debugfs(const char *debugfs_dir_name) +{ + hinic_dbgfs_root = debugfs_create_dir(debugfs_dir_name, NULL); +} + +void hinic_dbg_unregister_debugfs(void) +{ + debugfs_remove_recursive(hinic_dbgfs_root); + hinic_dbgfs_root = NULL; +} diff --git a/drivers/net/ethernet/huawei/hinic/hinic_debugfs.h b/drivers/net/ethernet/huawei/hinic/hinic_debugfs.h new file mode 100644 index 000000000000..e9e00cfa1329 --- /dev/null +++ b/drivers/net/ethernet/huawei/hinic/hinic_debugfs.h @@ -0,0 +1,114 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Huawei HiNIC PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + */ + +#ifndef HINIC_DEBUGFS_H +#define HINIC_DEBUGFS_H + +#include "hinic_dev.h" + +#define TBL_ID_FUNC_CFG_SM_NODE 11 +#define TBL_ID_FUNC_CFG_SM_INST 1 + +#define HINIC_FUNCTION_CONFIGURE_TABLE_SIZE 64 +#define HINIC_FUNCTION_CONFIGURE_TABLE 1 + +struct hinic_cmd_lt_rd { + u8 status; + u8 version; + u8 rsvd0[6]; + + unsigned char node; + unsigned char inst; + unsigned char entry_size; + unsigned char rsvd; + unsigned int lt_index; + unsigned int offset; + unsigned int len; + unsigned char data[100]; +}; + +struct tag_sml_funcfg_tbl { + union { + struct { + u32 rsvd0 :8; + u32 nic_rx_mode :5; + u32 rsvd1 :18; + u32 valid :1; + } bs; + + u32 value; + } dw0; + + union { + struct { + u32 vlan_id :12; + u32 vlan_mode :3; + u32 fast_recycled_mode :1; + u32 mtu :16; + } bs; + + u32 value; + } dw1; + + u32 dw2; + u32 dw3; + u32 dw4; + u32 dw5; + u32 dw6; + u32 dw7; + u32 dw8; + u32 dw9; + u32 dw10; + u32 dw11; + u32 dw12; + + union { + struct { + u32 rsvd2 :15; + u32 cfg_q_num :9; + u32 cfg_rq_depth :6; + u32 vhd_type :2; + } bs; + + u32 value; + } dw13; + + u32 dw14; + u32 dw15; +}; + +int hinic_sq_debug_add(struct hinic_dev *dev, u16 sq_id); + +void hinic_sq_debug_rem(struct hinic_sq *sq); + +int hinic_rq_debug_add(struct hinic_dev *dev, u16 rq_id); + +void hinic_rq_debug_rem(struct hinic_rq *rq); + +int hinic_func_table_debug_add(struct hinic_dev *dev); + +void hinic_func_table_debug_rem(struct hinic_dev *dev); + +void hinic_sq_dbgfs_init(struct hinic_dev *nic_dev); + +void hinic_sq_dbgfs_uninit(struct hinic_dev *nic_dev); + +void hinic_rq_dbgfs_init(struct hinic_dev *nic_dev); + +void hinic_rq_dbgfs_uninit(struct hinic_dev *nic_dev); + +void hinic_func_tbl_dbgfs_init(struct hinic_dev *nic_dev); + +void hinic_func_tbl_dbgfs_uninit(struct hinic_dev *nic_dev); + +void hinic_dbg_init(struct hinic_dev *nic_dev); + +void hinic_dbg_uninit(struct hinic_dev *nic_dev); + +void hinic_dbg_register_debugfs(const char *debugfs_dir_name); + +void hinic_dbg_unregister_debugfs(void); + +#endif diff --git a/drivers/net/ethernet/huawei/hinic/hinic_dev.h b/drivers/net/ethernet/huawei/hinic/hinic_dev.h index 0a1e20edf7cf..fb3e89141a0d 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_dev.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_dev.h @@ -58,6 +58,20 @@ struct hinic_intr_coal_info { u8 resend_timer_cfg; }; +enum hinic_dbg_type { + HINIC_DBG_SQ_INFO, + HINIC_DBG_RQ_INFO, + HINIC_DBG_FUNC_TABLE, +}; + +struct hinic_debug_priv { + struct hinic_dev *dev; + void *object; + enum hinic_dbg_type type; + struct dentry *root; + int field_id[64]; +}; + struct hinic_dev { struct net_device *netdev; struct hinic_hwdev *hwdev; @@ -97,6 +111,12 @@ struct hinic_dev { int lb_test_rx_idx; int lb_pkt_len; u8 *lb_test_rx_buf; + + struct dentry *dbgfs_root; + struct dentry *sq_dbgfs; + struct dentry *rq_dbgfs; + struct dentry *func_tbl_dbgfs; + struct hinic_debug_priv *dbg; struct devlink *devlink; bool cable_unplugged; bool module_unrecognized; diff --git a/drivers/net/ethernet/huawei/hinic/hinic_devlink.c b/drivers/net/ethernet/huawei/hinic/hinic_devlink.c index 16bda7381ba0..2630d667f393 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_devlink.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_devlink.c @@ -281,18 +281,14 @@ static int hinic_firmware_update(struct hinic_devlink_priv *priv, } static int hinic_devlink_flash_update(struct devlink *devlink, - const char *file_name, - const char *component, + struct devlink_flash_update_params *params, struct netlink_ext_ack *extack) { struct hinic_devlink_priv *priv = devlink_priv(devlink); const struct firmware *fw; int err; - if (component) - return -EOPNOTSUPP; - - err = request_firmware_direct(&fw, file_name, + err = request_firmware_direct(&fw, params->file_name, &priv->hwdev->hwif->pdev->dev); if (err) return err; diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_api_cmd.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_api_cmd.c index 29e88e25a4a4..4e4029d5c8e1 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_api_cmd.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_api_cmd.c @@ -373,7 +373,7 @@ static int wait_for_api_cmd_completion(struct hinic_api_cmd_chain *chain) * @chain: chain for the command * @dest: destination node on the card that will receive the command * @cmd: command data - * @size: the command size + * @cmd_size: the command size * * Return 0 - Success, negative - Failure **/ diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c index e0eb294779ec..5a6bbee819cd 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c @@ -784,7 +784,7 @@ static void free_cmdq(struct hinic_cmdq *cmdq) * init_cmdqs_ctxt - write the cmdq ctxt to HW after init all cmdq * @hwdev: the NIC HW device * @cmdqs: cmdqs to write the ctxts for - * &db_area: db_area for all the cmdqs + * @db_area: db_area for all the cmdqs * * Return 0 - Success, negative - Failure **/ diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c index 0c737765d113..0c74f6674634 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c @@ -437,6 +437,8 @@ static int get_base_qpn(struct hinic_hwdev *hwdev, u16 *base_qpn) /** * hinic_hwdev_ifup - Preparing the HW for passing IO * @hwdev: the NIC HW device + * @sq_depth: the send queue depth + * @rq_depth: the receive queue depth * * Return 0 - Success, negative - Failure **/ @@ -465,6 +467,7 @@ int hinic_hwdev_ifup(struct hinic_hwdev *hwdev, u16 sq_depth, u16 rq_depth) func_to_io->hwdev = hwdev; func_to_io->sq_depth = sq_depth; func_to_io->rq_depth = rq_depth; + func_to_io->global_qpn = base_qpn; err = hinic_io_init(func_to_io, hwif, nic_cap->max_qps, num_ceqs, ceq_msix_entries); @@ -581,6 +584,7 @@ void hinic_hwdev_cb_unregister(struct hinic_hwdev *hwdev, /** * nic_mgmt_msg_handler - nic mgmt event handler * @handle: private data for the handler + * @cmd: message command * @buf_in: input buffer * @in_size: input size * @buf_out: output buffer @@ -908,6 +912,7 @@ int hinic_set_interrupt_cfg(struct hinic_hwdev *hwdev, /** * hinic_init_hwdev - Initialize the NIC HW * @pdev: the NIC pci device + * @devlink: the poniter of hinic devlink * * Return initialized NIC HW device * @@ -1120,7 +1125,7 @@ int hinic_hwdev_msix_cnt_set(struct hinic_hwdev *hwdev, u16 msix_index) * @msix_index: msix_index * @pending_limit: the maximum pending interrupt events (unit 8) * @coalesc_timer: coalesc period for interrupt (unit 8 us) - * @lli_timer: replenishing period for low latency credit (unit 8 us) + * @lli_timer_cfg: replenishing period for low latency credit (unit 8 us) * @lli_credit_limit: maximum credits for low latency msix messages (unit 8) * @resend_timer: maximum wait for resending msix (unit coalesc period) * diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.h index 701eb81e09a7..416492e48274 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.h @@ -96,6 +96,8 @@ enum hinic_port_cmd { HINIC_PORT_CMD_RSS_TEMP_MGR = 49, + HINIC_PORT_CMD_RD_LINE_TBL = 57, + HINIC_PORT_CMD_RSS_CFG = 66, HINIC_PORT_CMD_FWCTXT_INIT = 69, diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.c index ca8cb68a8d20..19942fef99d9 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.c @@ -106,7 +106,7 @@ enum eq_arm_state { * @aeqs: pointer to Async eqs of the chip * @event: aeq event to register callback for it * @handle: private data will be used by the callback - * @hw_handler: callback function + * @hwe_handler: callback function **/ void hinic_aeq_register_hw_cb(struct hinic_aeqs *aeqs, enum hinic_aeq_type event, void *handle, @@ -188,6 +188,7 @@ static u8 eq_cons_idx_checksum_set(u32 val) /** * eq_update_ci - update the HW cons idx of event queue * @eq: the event queue to update the cons idx for + * @arm_state: the arm bit value of eq's interrupt **/ static void eq_update_ci(struct hinic_eq *eq, u32 arm_state) { @@ -368,11 +369,11 @@ static void eq_irq_work(struct work_struct *work) /** * ceq_tasklet - the tasklet of the EQ that received the event - * @ceq_data: the eq + * @t: the tasklet struct pointer **/ -static void ceq_tasklet(unsigned long ceq_data) +static void ceq_tasklet(struct tasklet_struct *t) { - struct hinic_eq *ceq = (struct hinic_eq *)ceq_data; + struct hinic_eq *ceq = from_tasklet(ceq, t, ceq_tasklet); eq_irq_handler(ceq); } @@ -782,8 +783,7 @@ static int init_eq(struct hinic_eq *eq, struct hinic_hwif *hwif, INIT_WORK(&aeq_work->work, eq_irq_work); } else if (type == HINIC_CEQ) { - tasklet_init(&eq->ceq_tasklet, ceq_tasklet, - (unsigned long)eq); + tasklet_setup(&eq->ceq_tasklet, ceq_tasklet); } /* set the attributes of the msix entry */ @@ -794,12 +794,15 @@ static int init_eq(struct hinic_eq *eq, struct hinic_hwif *hwif, HINIC_EQ_MSIX_LLI_CREDIT_LIMIT_DEFAULT, HINIC_EQ_MSIX_RESEND_TIMER_DEFAULT); - if (type == HINIC_AEQ) - err = request_irq(entry.vector, aeq_interrupt, 0, - "hinic_aeq", eq); - else if (type == HINIC_CEQ) - err = request_irq(entry.vector, ceq_interrupt, 0, - "hinic_ceq", eq); + if (type == HINIC_AEQ) { + snprintf(eq->irq_name, sizeof(eq->irq_name), "hinic_aeq%d@pci:%s", eq->q_id, + pci_name(pdev)); + err = request_irq(entry.vector, aeq_interrupt, 0, eq->irq_name, eq); + } else if (type == HINIC_CEQ) { + snprintf(eq->irq_name, sizeof(eq->irq_name), "hinic_ceq%d@pci:%s", eq->q_id, + pci_name(pdev)); + err = request_irq(entry.vector, ceq_interrupt, 0, eq->irq_name, eq); + } if (err) { dev_err(&pdev->dev, "Failed to request irq for the EQ\n"); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.h index 43065fc70869..2f3222174fc7 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.h @@ -186,6 +186,7 @@ struct hinic_eq { int num_elem_in_pg; struct msix_entry msix_entry; + char irq_name[64]; dma_addr_t *dma_addr; void **virt_addr; diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_if.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_if.c index bc8925c0c982..efbaed389440 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_if.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_if.c @@ -230,6 +230,7 @@ static int wait_hwif_ready(struct hinic_hwif *hwif) * @hwif: the HW interface of a pci function device * @attr0: the first attribute that was read from the hw * @attr1: the second attribute that was read from the hw + * @attr2: the third attribute that was read from the hw **/ static void set_hwif_attr(struct hinic_hwif *hwif, u32 attr0, u32 attr1, u32 attr2) diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_io.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_io.c index 3e3fa742e476..4ef4008e65bd 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_io.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_io.c @@ -305,6 +305,7 @@ static int init_qp(struct hinic_func_to_io *func_to_io, func_to_io->sq_db[q_id] = db_base; + qp->sq.qid = q_id; err = hinic_init_sq(&qp->sq, hwif, &func_to_io->sq_wq[q_id], sq_msix_entry, CI_ADDR(func_to_io->ci_addr_base, q_id), @@ -314,6 +315,7 @@ static int init_qp(struct hinic_func_to_io *func_to_io, goto err_sq_init; } + qp->rq.qid = q_id; err = hinic_init_rq(&qp->rq, hwif, &func_to_io->rq_wq[q_id], rq_msix_entry); if (err) { @@ -361,8 +363,8 @@ static void destroy_qp(struct hinic_func_to_io *func_to_io, * @func_to_io: func to io channel that holds the IO components * @base_qpn: base qp number * @num_qps: number queue pairs to create - * @sq_msix_entry: msix entries for sq - * @rq_msix_entry: msix entries for rq + * @sq_msix_entries: msix entries for sq + * @rq_msix_entries: msix entries for rq * * Return 0 - Success, negative - Failure **/ diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_io.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_io.h index ee6d60762d84..52159a90278a 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_io.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_io.h @@ -59,6 +59,7 @@ struct hinic_nic_cfg { struct hinic_func_to_io { struct hinic_hwif *hwif; struct hinic_hwdev *hwdev; + u16 global_qpn; struct hinic_ceqs ceqs; struct hinic_wqs wqs; diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.c index 2ebae6cb5db5..819fa13034c0 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.c @@ -238,6 +238,7 @@ static int send_msg_to_mgmt(struct hinic_pf_to_mgmt *pf_to_mgmt, * @out_size: response length * @direction: the direction of the original message * @resp_msg_id: msg id to response for + * @timeout: time-out period of waiting for response * * Return 0 - Success, negative - Failure **/ diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h index ca3e2d060284..0dfa51ad5855 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h @@ -81,6 +81,8 @@ struct hinic_sq { struct hinic_wq *wq; + u16 qid; + u32 irq; u16 msix_entry; @@ -90,6 +92,7 @@ struct hinic_sq { void __iomem *db_base; struct sk_buff **saved_skb; + struct hinic_debug_priv *dbg; }; struct hinic_rq { @@ -97,6 +100,8 @@ struct hinic_rq { struct hinic_wq *wq; + u16 qid; + struct cpumask affinity_mask; u32 irq; u16 msix_entry; @@ -110,6 +115,7 @@ struct hinic_rq { u16 *pi_virt_addr; dma_addr_t pi_dma_addr; + struct hinic_debug_priv *dbg; }; struct hinic_qp { diff --git a/drivers/net/ethernet/huawei/hinic/hinic_main.c b/drivers/net/ethernet/huawei/hinic/hinic_main.c index 28581bd8ce07..350225bbe0be 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_main.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_main.c @@ -24,6 +24,7 @@ #include <linux/delay.h> #include <linux/err.h> +#include "hinic_debugfs.h" #include "hinic_hw_qp.h" #include "hinic_hw_dev.h" #include "hinic_devlink.h" @@ -153,6 +154,8 @@ static int create_txqs(struct hinic_dev *nic_dev) if (!nic_dev->txqs) return -ENOMEM; + hinic_sq_dbgfs_init(nic_dev); + for (i = 0; i < num_txqs; i++) { struct hinic_sq *sq = hinic_hwdev_get_sq(nic_dev->hwdev, i); @@ -162,13 +165,27 @@ static int create_txqs(struct hinic_dev *nic_dev) "Failed to init Txq\n"); goto err_init_txq; } + + err = hinic_sq_debug_add(nic_dev, i); + if (err) { + netif_err(nic_dev, drv, netdev, + "Failed to add SQ%d debug\n", i); + goto err_add_sq_dbg; + } + } return 0; +err_add_sq_dbg: + hinic_clean_txq(&nic_dev->txqs[i]); err_init_txq: - for (j = 0; j < i; j++) + for (j = 0; j < i; j++) { + hinic_sq_debug_rem(nic_dev->txqs[j].sq); hinic_clean_txq(&nic_dev->txqs[j]); + } + + hinic_sq_dbgfs_uninit(nic_dev); devm_kfree(&netdev->dev, nic_dev->txqs); return err; @@ -204,8 +221,12 @@ static void free_txqs(struct hinic_dev *nic_dev) if (!nic_dev->txqs) return; - for (i = 0; i < num_txqs; i++) + for (i = 0; i < num_txqs; i++) { + hinic_sq_debug_rem(nic_dev->txqs[i].sq); hinic_clean_txq(&nic_dev->txqs[i]); + } + + hinic_sq_dbgfs_uninit(nic_dev); devm_kfree(&netdev->dev, nic_dev->txqs); nic_dev->txqs = NULL; @@ -231,6 +252,8 @@ static int create_rxqs(struct hinic_dev *nic_dev) if (!nic_dev->rxqs) return -ENOMEM; + hinic_rq_dbgfs_init(nic_dev); + for (i = 0; i < num_rxqs; i++) { struct hinic_rq *rq = hinic_hwdev_get_rq(nic_dev->hwdev, i); @@ -240,13 +263,26 @@ static int create_rxqs(struct hinic_dev *nic_dev) "Failed to init rxq\n"); goto err_init_rxq; } + + err = hinic_rq_debug_add(nic_dev, i); + if (err) { + netif_err(nic_dev, drv, netdev, + "Failed to add RQ%d debug\n", i); + goto err_add_rq_dbg; + } } return 0; +err_add_rq_dbg: + hinic_clean_rxq(&nic_dev->rxqs[i]); err_init_rxq: - for (j = 0; j < i; j++) + for (j = 0; j < i; j++) { + hinic_rq_debug_rem(nic_dev->rxqs[j].rq); hinic_clean_rxq(&nic_dev->rxqs[j]); + } + + hinic_rq_dbgfs_uninit(nic_dev); devm_kfree(&netdev->dev, nic_dev->rxqs); return err; @@ -264,8 +300,12 @@ static void free_rxqs(struct hinic_dev *nic_dev) if (!nic_dev->rxqs) return; - for (i = 0; i < num_rxqs; i++) + for (i = 0; i < num_rxqs; i++) { + hinic_rq_debug_rem(nic_dev->rxqs[i].rq); hinic_clean_rxq(&nic_dev->rxqs[i]); + } + + hinic_rq_dbgfs_uninit(nic_dev); devm_kfree(&netdev->dev, nic_dev->rxqs); nic_dev->rxqs = NULL; @@ -913,11 +953,16 @@ static void netdev_features_init(struct net_device *netdev) netdev->hw_features = NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_RXCSUM | NETIF_F_LRO | - NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX; + NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX | + NETIF_F_GSO_UDP_TUNNEL | NETIF_F_GSO_UDP_TUNNEL_CSUM; netdev->vlan_features = netdev->hw_features; netdev->features = netdev->hw_features | NETIF_F_HW_VLAN_CTAG_FILTER; + + netdev->hw_enc_features = NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_SCTP_CRC | + NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_TSO_ECN | + NETIF_F_GSO_UDP_TUNNEL_CSUM | NETIF_F_GSO_UDP_TUNNEL; } static void hinic_refresh_nic_cfg(struct hinic_dev *nic_dev) @@ -945,7 +990,7 @@ static void hinic_refresh_nic_cfg(struct hinic_dev *nic_dev) * @handle: nic device for the handler * @buf_in: input buffer * @in_size: input size - * @buf_in: output buffer + * @buf_out: output buffer * @out_size: returned output size * * Return 0 - Success, negative - Failure @@ -1284,6 +1329,16 @@ static int nic_dev_init(struct pci_dev *pdev) goto err_init_intr; } + hinic_dbg_init(nic_dev); + + hinic_func_tbl_dbgfs_init(nic_dev); + + err = hinic_func_table_debug_add(nic_dev); + if (err) { + dev_err(&pdev->dev, "Failed to add func_table debug\n"); + goto err_add_func_table_dbg; + } + err = register_netdev(netdev); if (err) { dev_err(&pdev->dev, "Failed to register netdev\n"); @@ -1293,6 +1348,10 @@ static int nic_dev_init(struct pci_dev *pdev) return 0; err_reg_netdev: + hinic_func_table_debug_rem(nic_dev); +err_add_func_table_dbg: + hinic_func_tbl_dbgfs_uninit(nic_dev); + hinic_dbg_uninit(nic_dev); hinic_free_intr_coalesce(nic_dev); err_init_intr: err_set_pfc: @@ -1415,6 +1474,12 @@ static void hinic_remove(struct pci_dev *pdev) unregister_netdev(netdev); + hinic_func_table_debug_rem(nic_dev); + + hinic_func_tbl_dbgfs_uninit(nic_dev); + + hinic_dbg_uninit(nic_dev); + hinic_free_intr_coalesce(nic_dev); hinic_port_del_mac(nic_dev, netdev->dev_addr, 0); @@ -1469,4 +1534,17 @@ static struct pci_driver hinic_driver = { .sriov_configure = hinic_pci_sriov_configure, }; -module_pci_driver(hinic_driver); +static int __init hinic_module_init(void) +{ + hinic_dbg_register_debugfs(HINIC_DRV_NAME); + return pci_register_driver(&hinic_driver); +} + +static void __exit hinic_module_exit(void) +{ + pci_unregister_driver(&hinic_driver); + hinic_dbg_unregister_debugfs(); +} + +module_init(hinic_module_init); +module_exit(hinic_module_exit); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_rx.c b/drivers/net/ethernet/huawei/hinic/hinic_rx.c index d0072f5e7efc..070a7cc6392e 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_rx.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_rx.c @@ -595,7 +595,7 @@ int hinic_init_rxq(struct hinic_rxq *rxq, struct hinic_rq *rq, rxq_stats_init(rxq); rxq->irq_name = devm_kasprintf(&netdev->dev, GFP_KERNEL, - "hinic_rxq%d", qp->q_id); + "%s_rxq%d", netdev->name, qp->q_id); if (!rxq->irq_name) return -ENOMEM; diff --git a/drivers/net/ethernet/huawei/hinic/hinic_tx.c b/drivers/net/ethernet/huawei/hinic/hinic_tx.c index c1f81e9144a1..8da7d46363b2 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_tx.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_tx.c @@ -357,6 +357,7 @@ static int offload_csum(struct hinic_sq_task *task, u32 *queue_info, enum hinic_l4_offload_type l4_offload; u32 offset, l4_len, network_hdr_len; enum hinic_l3_offload_type l3_type; + u32 tunnel_type = NOT_TUNNEL; union hinic_l3 ip; union hinic_l4 l4; u8 l4_proto; @@ -367,27 +368,55 @@ static int offload_csum(struct hinic_sq_task *task, u32 *queue_info, if (skb->encapsulation) { u32 l4_tunnel_len; + tunnel_type = TUNNEL_UDP_NO_CSUM; ip.hdr = skb_network_header(skb); - if (ip.v4->version == 4) + if (ip.v4->version == 4) { l3_type = IPV4_PKT_NO_CHKSUM_OFFLOAD; - else if (ip.v4->version == 6) + l4_proto = ip.v4->protocol; + } else if (ip.v4->version == 6) { + unsigned char *exthdr; + __be16 frag_off; l3_type = IPV6_PKT; - else + tunnel_type = TUNNEL_UDP_CSUM; + exthdr = ip.hdr + sizeof(*ip.v6); + l4_proto = ip.v6->nexthdr; + l4.hdr = skb_transport_header(skb); + if (l4.hdr != exthdr) + ipv6_skip_exthdr(skb, exthdr - skb->data, + &l4_proto, &frag_off); + } else { l3_type = L3TYPE_UNKNOWN; + l4_proto = IPPROTO_RAW; + } hinic_task_set_outter_l3(task, l3_type, skb_network_header_len(skb)); - l4_tunnel_len = skb_inner_network_offset(skb) - - skb_transport_offset(skb); - - hinic_task_set_tunnel_l4(task, TUNNEL_UDP_NO_CSUM, - l4_tunnel_len); + switch (l4_proto) { + case IPPROTO_UDP: + l4_tunnel_len = skb_inner_network_offset(skb) - + skb_transport_offset(skb); + ip.hdr = skb_inner_network_header(skb); + l4.hdr = skb_inner_transport_header(skb); + network_hdr_len = skb_inner_network_header_len(skb); + break; + case IPPROTO_IPIP: + case IPPROTO_IPV6: + tunnel_type = NOT_TUNNEL; + l4_tunnel_len = 0; + + ip.hdr = skb_inner_network_header(skb); + l4.hdr = skb_transport_header(skb); + network_hdr_len = skb_network_header_len(skb); + break; + default: + /* Unsupported tunnel packet, disable csum offload */ + skb_checksum_help(skb); + return 0; + } - ip.hdr = skb_inner_network_header(skb); - l4.hdr = skb_inner_transport_header(skb); - network_hdr_len = skb_inner_network_header_len(skb); + hinic_task_set_tunnel_l4(task, tunnel_type, l4_tunnel_len); } else { ip.hdr = skb_network_header(skb); l4.hdr = skb_transport_header(skb); @@ -853,14 +882,14 @@ int hinic_init_txq(struct hinic_txq *txq, struct hinic_sq *sq, goto err_alloc_free_sges; } - irqname_len = snprintf(NULL, 0, "hinic_txq%d", qp->q_id) + 1; + irqname_len = snprintf(NULL, 0, "%s_txq%d", netdev->name, qp->q_id) + 1; txq->irq_name = devm_kzalloc(&netdev->dev, irqname_len, GFP_KERNEL); if (!txq->irq_name) { err = -ENOMEM; goto err_alloc_irqname; } - sprintf(txq->irq_name, "hinic_txq%d", qp->q_id); + sprintf(txq->irq_name, "%s_txq%d", netdev->name, qp->q_id); err = hinic_hwdev_hw_ci_addr_set(hwdev, sq, CI_UPDATE_NO_PENDING, CI_UPDATE_NO_COALESC); diff --git a/drivers/net/ethernet/ibm/ehea/ehea_main.c b/drivers/net/ethernet/ibm/ehea/ehea_main.c index 3153d62cc73e..c2e740475786 100644 --- a/drivers/net/ethernet/ibm/ehea/ehea_main.c +++ b/drivers/net/ethernet/ibm/ehea/ehea_main.c @@ -1212,9 +1212,9 @@ static void ehea_parse_eqe(struct ehea_adapter *adapter, u64 eqe) } } -static void ehea_neq_tasklet(unsigned long data) +static void ehea_neq_tasklet(struct tasklet_struct *t) { - struct ehea_adapter *adapter = (struct ehea_adapter *)data; + struct ehea_adapter *adapter = from_tasklet(adapter, t, neq_tasklet); struct ehea_eqe *eqe; u64 event_mask; @@ -3417,8 +3417,7 @@ static int ehea_probe_adapter(struct platform_device *dev) goto out_free_ad; } - tasklet_init(&adapter->neq_tasklet, ehea_neq_tasklet, - (unsigned long)adapter); + tasklet_setup(&adapter->neq_tasklet, ehea_neq_tasklet); ret = ehea_create_device_sysfs(dev); if (ret) diff --git a/drivers/net/ethernet/ibm/ibmveth.c b/drivers/net/ethernet/ibm/ibmveth.c index c5c732601e35..7ef3369953b6 100644 --- a/drivers/net/ethernet/ibm/ibmveth.c +++ b/drivers/net/ethernet/ibm/ibmveth.c @@ -1349,6 +1349,7 @@ static int ibmveth_poll(struct napi_struct *napi, int budget) int offset = ibmveth_rxq_frame_offset(adapter); int csum_good = ibmveth_rxq_csum_good(adapter); int lrg_pkt = ibmveth_rxq_large_packet(adapter); + __sum16 iph_check = 0; skb = ibmveth_rxq_get_buffer(adapter); @@ -1385,16 +1386,26 @@ static int ibmveth_poll(struct napi_struct *napi, int budget) skb_put(skb, length); skb->protocol = eth_type_trans(skb, netdev); - if (csum_good) { - skb->ip_summed = CHECKSUM_UNNECESSARY; - ibmveth_rx_csum_helper(skb, adapter); + /* PHYP without PLSO support places a -1 in the ip + * checksum for large send frames. + */ + if (skb->protocol == cpu_to_be16(ETH_P_IP)) { + struct iphdr *iph = (struct iphdr *)skb->data; + + iph_check = iph->check; } - if (length > netdev->mtu + ETH_HLEN) { + if ((length > netdev->mtu + ETH_HLEN) || + lrg_pkt || iph_check == 0xffff) { ibmveth_rx_mss_helper(skb, mss, lrg_pkt); adapter->rx_large_packets++; } + if (csum_good) { + skb->ip_summed = CHECKSUM_UNNECESSARY; + ibmveth_rx_csum_helper(skb, adapter); + } + napi_gro_receive(napi, skb); /* send it up */ netdev->stats.rx_packets++; diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index 1b702a43a5d0..1f7fe6b3dd5a 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -97,15 +97,14 @@ static int pending_scrq(struct ibmvnic_adapter *, static union sub_crq *ibmvnic_next_scrq(struct ibmvnic_adapter *, struct ibmvnic_sub_crq_queue *); static int ibmvnic_poll(struct napi_struct *napi, int data); -static void send_map_query(struct ibmvnic_adapter *adapter); +static void send_query_map(struct ibmvnic_adapter *adapter); static int send_request_map(struct ibmvnic_adapter *, dma_addr_t, __be32, u8); static int send_request_unmap(struct ibmvnic_adapter *, u8); static int send_login(struct ibmvnic_adapter *adapter); -static void send_cap_queries(struct ibmvnic_adapter *adapter); +static void send_query_cap(struct ibmvnic_adapter *adapter); static int init_sub_crqs(struct ibmvnic_adapter *); static int init_sub_crq_irqs(struct ibmvnic_adapter *adapter); -static int ibmvnic_init(struct ibmvnic_adapter *); -static int ibmvnic_reset_init(struct ibmvnic_adapter *); +static int ibmvnic_reset_init(struct ibmvnic_adapter *, bool reset); static void release_crq_queue(struct ibmvnic_adapter *); static int __ibmvnic_set_mac(struct net_device *, u8 *); static int init_crq_queue(struct ibmvnic_adapter *adapter); @@ -297,8 +296,7 @@ static void deactivate_rx_pools(struct ibmvnic_adapter *adapter) { int i; - for (i = 0; i < be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs); - i++) + for (i = 0; i < adapter->num_active_rx_pools; i++) adapter->rx_pool[i].active = 0; } @@ -306,6 +304,7 @@ static void replenish_rx_pool(struct ibmvnic_adapter *adapter, struct ibmvnic_rx_pool *pool) { int count = pool->size - atomic_read(&pool->available); + u64 handle = adapter->rx_scrq[pool->index]->handle; struct device *dev = &adapter->vdev->dev; int buffers_added = 0; unsigned long lpar_rc; @@ -314,7 +313,6 @@ static void replenish_rx_pool(struct ibmvnic_adapter *adapter, unsigned int offset; dma_addr_t dma_addr; unsigned char *dst; - u64 *handle_array; int shift = 0; int index; int i; @@ -322,10 +320,6 @@ static void replenish_rx_pool(struct ibmvnic_adapter *adapter, if (!pool->active) return; - handle_array = (u64 *)((u8 *)(adapter->login_rsp_buf) + - be32_to_cpu(adapter->login_rsp_buf-> - off_rxadd_subcrqs)); - for (i = 0; i < count; ++i) { skb = alloc_skb(pool->buff_size, GFP_ATOMIC); if (!skb) { @@ -369,8 +363,7 @@ static void replenish_rx_pool(struct ibmvnic_adapter *adapter, #endif sub_crq.rx_add.len = cpu_to_be32(pool->buff_size << shift); - lpar_rc = send_subcrq(adapter, handle_array[pool->index], - &sub_crq); + lpar_rc = send_subcrq(adapter, handle, &sub_crq); if (lpar_rc != H_SUCCESS) goto failure; @@ -407,8 +400,7 @@ static void replenish_pools(struct ibmvnic_adapter *adapter) int i; adapter->replenish_task_cycles++; - for (i = 0; i < be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs); - i++) { + for (i = 0; i < adapter->num_active_rx_pools; i++) { if (adapter->rx_pool[i].active) replenish_rx_pool(adapter, &adapter->rx_pool[i]); } @@ -475,25 +467,23 @@ static int init_stats_token(struct ibmvnic_adapter *adapter) static int reset_rx_pools(struct ibmvnic_adapter *adapter) { struct ibmvnic_rx_pool *rx_pool; + u64 buff_size; int rx_scrqs; int i, j, rc; - u64 *size_array; if (!adapter->rx_pool) return -1; - size_array = (u64 *)((u8 *)(adapter->login_rsp_buf) + - be32_to_cpu(adapter->login_rsp_buf->off_rxadd_buff_size)); - - rx_scrqs = be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs); + buff_size = adapter->cur_rx_buf_sz; + rx_scrqs = adapter->num_active_rx_pools; for (i = 0; i < rx_scrqs; i++) { rx_pool = &adapter->rx_pool[i]; netdev_dbg(adapter->netdev, "Re-setting rx_pool[%d]\n", i); - if (rx_pool->buff_size != be64_to_cpu(size_array[i])) { + if (rx_pool->buff_size != buff_size) { free_long_term_buff(adapter, &rx_pool->long_term_buff); - rx_pool->buff_size = be64_to_cpu(size_array[i]); + rx_pool->buff_size = buff_size; rc = alloc_long_term_buff(adapter, &rx_pool->long_term_buff, rx_pool->size * @@ -561,13 +551,11 @@ static int init_rx_pools(struct net_device *netdev) struct device *dev = &adapter->vdev->dev; struct ibmvnic_rx_pool *rx_pool; int rxadd_subcrqs; - u64 *size_array; + u64 buff_size; int i, j; - rxadd_subcrqs = - be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs); - size_array = (u64 *)((u8 *)(adapter->login_rsp_buf) + - be32_to_cpu(adapter->login_rsp_buf->off_rxadd_buff_size)); + rxadd_subcrqs = adapter->num_active_rx_scrqs; + buff_size = adapter->cur_rx_buf_sz; adapter->rx_pool = kcalloc(rxadd_subcrqs, sizeof(struct ibmvnic_rx_pool), @@ -585,11 +573,11 @@ static int init_rx_pools(struct net_device *netdev) netdev_dbg(adapter->netdev, "Initializing rx_pool[%d], %lld buffs, %lld bytes each\n", i, adapter->req_rx_add_entries_per_subcrq, - be64_to_cpu(size_array[i])); + buff_size); rx_pool->size = adapter->req_rx_add_entries_per_subcrq; rx_pool->index = i; - rx_pool->buff_size = be64_to_cpu(size_array[i]); + rx_pool->buff_size = buff_size; rx_pool->active = 1; rx_pool->free_map = kcalloc(rx_pool->size, sizeof(int), @@ -655,7 +643,7 @@ static int reset_tx_pools(struct ibmvnic_adapter *adapter) if (!adapter->tx_pool) return -1; - tx_scrqs = be32_to_cpu(adapter->login_rsp_buf->num_txsubm_subcrqs); + tx_scrqs = adapter->num_active_tx_pools; for (i = 0; i < tx_scrqs; i++) { rc = reset_one_tx_pool(adapter, &adapter->tso_pool[i]); if (rc) @@ -744,7 +732,7 @@ static int init_tx_pools(struct net_device *netdev) int tx_subcrqs; int i, rc; - tx_subcrqs = be32_to_cpu(adapter->login_rsp_buf->num_txsubm_subcrqs); + tx_subcrqs = adapter->num_active_tx_scrqs; adapter->tx_pool = kcalloc(tx_subcrqs, sizeof(struct ibmvnic_tx_pool), GFP_KERNEL); if (!adapter->tx_pool) @@ -894,7 +882,7 @@ static int ibmvnic_login(struct net_device *netdev) "Received partial success, retrying...\n"); adapter->init_done_rc = 0; reinit_completion(&adapter->init_done); - send_cap_queries(adapter); + send_query_cap(adapter); if (!wait_for_completion_timeout(&adapter->init_done, timeout)) { netdev_warn(netdev, @@ -980,7 +968,7 @@ static int set_link_state(struct ibmvnic_adapter *adapter, u8 link_state) return -1; } - if (adapter->init_done_rc == 1) { + if (adapter->init_done_rc == PARTIALSUCCESS) { /* Partuial success, delay and re-send */ mdelay(1000); resend = true; @@ -1125,7 +1113,7 @@ static int init_resources(struct ibmvnic_adapter *adapter) if (rc) return rc; - send_map_query(adapter); + send_query_map(adapter); rc = init_rx_pools(netdev); if (rc) @@ -1530,9 +1518,9 @@ static netdev_tx_t ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev) unsigned int offset; int num_entries = 1; unsigned char *dst; - u64 *handle_array; int index = 0; u8 proto = 0; + u64 handle; netdev_tx_t ret = NETDEV_TX_OK; if (test_bit(0, &adapter->resetting)) { @@ -1559,8 +1547,7 @@ static netdev_tx_t ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev) tx_scrq = adapter->tx_scrq[queue_num]; txq = netdev_get_tx_queue(netdev, skb_get_queue_mapping(skb)); - handle_array = (u64 *)((u8 *)(adapter->login_rsp_buf) + - be32_to_cpu(adapter->login_rsp_buf->off_txsubm_subcrqs)); + handle = tx_scrq->handle; index = tx_pool->free_map[tx_pool->consumer_index]; @@ -1672,14 +1659,14 @@ static netdev_tx_t ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev) ret = NETDEV_TX_OK; goto tx_err_out; } - lpar_rc = send_subcrq_indirect(adapter, handle_array[queue_num], + lpar_rc = send_subcrq_indirect(adapter, handle, (u64)tx_buff->indir_dma, (u64)num_entries); dma_unmap_single(dev, tx_buff->indir_dma, sizeof(tx_buff->indir_arr), DMA_TO_DEVICE); } else { tx_buff->num_entries = num_entries; - lpar_rc = send_subcrq(adapter, handle_array[queue_num], + lpar_rc = send_subcrq(adapter, handle, &tx_crq); } if (lpar_rc != H_SUCCESS) { @@ -1874,7 +1861,7 @@ static int do_change_param_reset(struct ibmvnic_adapter *adapter, return rc; } - rc = ibmvnic_reset_init(adapter); + rc = ibmvnic_reset_init(adapter, true); if (rc) return IBMVNIC_INIT_FAILED; @@ -1992,7 +1979,7 @@ static int do_reset(struct ibmvnic_adapter *adapter, goto out; } - rc = ibmvnic_reset_init(adapter); + rc = ibmvnic_reset_init(adapter, true); if (rc) { rc = IBMVNIC_INIT_FAILED; goto out; @@ -2108,7 +2095,7 @@ static int do_hard_reset(struct ibmvnic_adapter *adapter, return rc; } - rc = ibmvnic_init(adapter); + rc = ibmvnic_reset_init(adapter, false); if (rc) return rc; @@ -3312,7 +3299,7 @@ tx_failed: return -1; } -static void ibmvnic_send_req_caps(struct ibmvnic_adapter *adapter, int retry) +static void send_request_cap(struct ibmvnic_adapter *adapter, int retry) { struct device *dev = &adapter->vdev->dev; union ibmvnic_crq crq; @@ -3583,8 +3570,7 @@ static int ibmvnic_send_crq(struct ibmvnic_adapter *adapter, if (rc) { if (rc == H_CLOSED) { dev_warn(dev, "CRQ Queue closed\n"); - if (test_bit(0, &adapter->resetting)) - ibmvnic_reset(adapter, VNIC_RESET_FATAL); + /* do not reset, report the fail, wait for passive init from server */ } dev_warn(dev, "Send error (rc=%d)\n", rc); @@ -3595,14 +3581,31 @@ static int ibmvnic_send_crq(struct ibmvnic_adapter *adapter, static int ibmvnic_send_crq_init(struct ibmvnic_adapter *adapter) { + struct device *dev = &adapter->vdev->dev; union ibmvnic_crq crq; + int retries = 100; + int rc; memset(&crq, 0, sizeof(crq)); crq.generic.first = IBMVNIC_CRQ_INIT_CMD; crq.generic.cmd = IBMVNIC_CRQ_INIT; netdev_dbg(adapter->netdev, "Sending CRQ init\n"); - return ibmvnic_send_crq(adapter, &crq); + do { + rc = ibmvnic_send_crq(adapter, &crq); + if (rc != H_CLOSED) + break; + retries--; + msleep(50); + + } while (retries > 0); + + if (rc) { + dev_err(dev, "Failed to send init request, rc = %d\n", rc); + return rc; + } + + return 0; } static int send_version_xchg(struct ibmvnic_adapter *adapter) @@ -3822,7 +3825,7 @@ static int send_request_unmap(struct ibmvnic_adapter *adapter, u8 map_id) return ibmvnic_send_crq(adapter, &crq); } -static void send_map_query(struct ibmvnic_adapter *adapter) +static void send_query_map(struct ibmvnic_adapter *adapter) { union ibmvnic_crq crq; @@ -3833,7 +3836,7 @@ static void send_map_query(struct ibmvnic_adapter *adapter) } /* Send a series of CRQs requesting various capabilities of the VNIC server */ -static void send_cap_queries(struct ibmvnic_adapter *adapter) +static void send_query_cap(struct ibmvnic_adapter *adapter) { union ibmvnic_crq crq; @@ -3950,6 +3953,113 @@ static void send_cap_queries(struct ibmvnic_adapter *adapter) ibmvnic_send_crq(adapter, &crq); } +static void send_query_ip_offload(struct ibmvnic_adapter *adapter) +{ + int buf_sz = sizeof(struct ibmvnic_query_ip_offload_buffer); + struct device *dev = &adapter->vdev->dev; + union ibmvnic_crq crq; + + adapter->ip_offload_tok = + dma_map_single(dev, + &adapter->ip_offload_buf, + buf_sz, + DMA_FROM_DEVICE); + + if (dma_mapping_error(dev, adapter->ip_offload_tok)) { + if (!firmware_has_feature(FW_FEATURE_CMO)) + dev_err(dev, "Couldn't map offload buffer\n"); + return; + } + + memset(&crq, 0, sizeof(crq)); + crq.query_ip_offload.first = IBMVNIC_CRQ_CMD; + crq.query_ip_offload.cmd = QUERY_IP_OFFLOAD; + crq.query_ip_offload.len = cpu_to_be32(buf_sz); + crq.query_ip_offload.ioba = + cpu_to_be32(adapter->ip_offload_tok); + + ibmvnic_send_crq(adapter, &crq); +} + +static void send_control_ip_offload(struct ibmvnic_adapter *adapter) +{ + struct ibmvnic_control_ip_offload_buffer *ctrl_buf = &adapter->ip_offload_ctrl; + struct ibmvnic_query_ip_offload_buffer *buf = &adapter->ip_offload_buf; + struct device *dev = &adapter->vdev->dev; + netdev_features_t old_hw_features = 0; + union ibmvnic_crq crq; + + adapter->ip_offload_ctrl_tok = + dma_map_single(dev, + ctrl_buf, + sizeof(adapter->ip_offload_ctrl), + DMA_TO_DEVICE); + + if (dma_mapping_error(dev, adapter->ip_offload_ctrl_tok)) { + dev_err(dev, "Couldn't map ip offload control buffer\n"); + return; + } + + ctrl_buf->len = cpu_to_be32(sizeof(adapter->ip_offload_ctrl)); + ctrl_buf->version = cpu_to_be32(INITIAL_VERSION_IOB); + ctrl_buf->ipv4_chksum = buf->ipv4_chksum; + ctrl_buf->ipv6_chksum = buf->ipv6_chksum; + ctrl_buf->tcp_ipv4_chksum = buf->tcp_ipv4_chksum; + ctrl_buf->udp_ipv4_chksum = buf->udp_ipv4_chksum; + ctrl_buf->tcp_ipv6_chksum = buf->tcp_ipv6_chksum; + ctrl_buf->udp_ipv6_chksum = buf->udp_ipv6_chksum; + ctrl_buf->large_tx_ipv4 = buf->large_tx_ipv4; + ctrl_buf->large_tx_ipv6 = buf->large_tx_ipv6; + + /* large_rx disabled for now, additional features needed */ + ctrl_buf->large_rx_ipv4 = 0; + ctrl_buf->large_rx_ipv6 = 0; + + if (adapter->state != VNIC_PROBING) { + old_hw_features = adapter->netdev->hw_features; + adapter->netdev->hw_features = 0; + } + + adapter->netdev->hw_features = NETIF_F_SG | NETIF_F_GSO | NETIF_F_GRO; + + if (buf->tcp_ipv4_chksum || buf->udp_ipv4_chksum) + adapter->netdev->hw_features |= NETIF_F_IP_CSUM; + + if (buf->tcp_ipv6_chksum || buf->udp_ipv6_chksum) + adapter->netdev->hw_features |= NETIF_F_IPV6_CSUM; + + if ((adapter->netdev->features & + (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM))) + adapter->netdev->hw_features |= NETIF_F_RXCSUM; + + if (buf->large_tx_ipv4) + adapter->netdev->hw_features |= NETIF_F_TSO; + if (buf->large_tx_ipv6) + adapter->netdev->hw_features |= NETIF_F_TSO6; + + if (adapter->state == VNIC_PROBING) { + adapter->netdev->features |= adapter->netdev->hw_features; + } else if (old_hw_features != adapter->netdev->hw_features) { + netdev_features_t tmp = 0; + + /* disable features no longer supported */ + adapter->netdev->features &= adapter->netdev->hw_features; + /* turn on features now supported if previously enabled */ + tmp = (old_hw_features ^ adapter->netdev->hw_features) & + adapter->netdev->hw_features; + adapter->netdev->features |= + tmp & adapter->netdev->wanted_features; + } + + memset(&crq, 0, sizeof(crq)); + crq.control_ip_offload.first = IBMVNIC_CRQ_CMD; + crq.control_ip_offload.cmd = CONTROL_IP_OFFLOAD; + crq.control_ip_offload.len = + cpu_to_be32(sizeof(adapter->ip_offload_ctrl)); + crq.control_ip_offload.ioba = cpu_to_be32(adapter->ip_offload_ctrl_tok); + ibmvnic_send_crq(adapter, &crq); +} + static void handle_vpd_size_rsp(union ibmvnic_crq *crq, struct ibmvnic_adapter *adapter) { @@ -4019,8 +4129,6 @@ static void handle_query_ip_offload_rsp(struct ibmvnic_adapter *adapter) { struct device *dev = &adapter->vdev->dev; struct ibmvnic_query_ip_offload_buffer *buf = &adapter->ip_offload_buf; - netdev_features_t old_hw_features = 0; - union ibmvnic_crq crq; int i; dma_unmap_single(dev, adapter->ip_offload_tok, @@ -4070,74 +4178,7 @@ static void handle_query_ip_offload_rsp(struct ibmvnic_adapter *adapter) netdev_dbg(adapter->netdev, "off_ipv6_ext_hd = %d\n", buf->off_ipv6_ext_headers); - adapter->ip_offload_ctrl_tok = - dma_map_single(dev, &adapter->ip_offload_ctrl, - sizeof(adapter->ip_offload_ctrl), DMA_TO_DEVICE); - - if (dma_mapping_error(dev, adapter->ip_offload_ctrl_tok)) { - dev_err(dev, "Couldn't map ip offload control buffer\n"); - return; - } - - adapter->ip_offload_ctrl.len = - cpu_to_be32(sizeof(adapter->ip_offload_ctrl)); - adapter->ip_offload_ctrl.version = cpu_to_be32(INITIAL_VERSION_IOB); - adapter->ip_offload_ctrl.ipv4_chksum = buf->ipv4_chksum; - adapter->ip_offload_ctrl.ipv6_chksum = buf->ipv6_chksum; - adapter->ip_offload_ctrl.tcp_ipv4_chksum = buf->tcp_ipv4_chksum; - adapter->ip_offload_ctrl.udp_ipv4_chksum = buf->udp_ipv4_chksum; - adapter->ip_offload_ctrl.tcp_ipv6_chksum = buf->tcp_ipv6_chksum; - adapter->ip_offload_ctrl.udp_ipv6_chksum = buf->udp_ipv6_chksum; - adapter->ip_offload_ctrl.large_tx_ipv4 = buf->large_tx_ipv4; - adapter->ip_offload_ctrl.large_tx_ipv6 = buf->large_tx_ipv6; - - /* large_rx disabled for now, additional features needed */ - adapter->ip_offload_ctrl.large_rx_ipv4 = 0; - adapter->ip_offload_ctrl.large_rx_ipv6 = 0; - - if (adapter->state != VNIC_PROBING) { - old_hw_features = adapter->netdev->hw_features; - adapter->netdev->hw_features = 0; - } - - adapter->netdev->hw_features = NETIF_F_SG | NETIF_F_GSO | NETIF_F_GRO; - - if (buf->tcp_ipv4_chksum || buf->udp_ipv4_chksum) - adapter->netdev->hw_features |= NETIF_F_IP_CSUM; - - if (buf->tcp_ipv6_chksum || buf->udp_ipv6_chksum) - adapter->netdev->hw_features |= NETIF_F_IPV6_CSUM; - - if ((adapter->netdev->features & - (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM))) - adapter->netdev->hw_features |= NETIF_F_RXCSUM; - - if (buf->large_tx_ipv4) - adapter->netdev->hw_features |= NETIF_F_TSO; - if (buf->large_tx_ipv6) - adapter->netdev->hw_features |= NETIF_F_TSO6; - - if (adapter->state == VNIC_PROBING) { - adapter->netdev->features |= adapter->netdev->hw_features; - } else if (old_hw_features != adapter->netdev->hw_features) { - netdev_features_t tmp = 0; - - /* disable features no longer supported */ - adapter->netdev->features &= adapter->netdev->hw_features; - /* turn on features now supported if previously enabled */ - tmp = (old_hw_features ^ adapter->netdev->hw_features) & - adapter->netdev->hw_features; - adapter->netdev->features |= - tmp & adapter->netdev->wanted_features; - } - - memset(&crq, 0, sizeof(crq)); - crq.control_ip_offload.first = IBMVNIC_CRQ_CMD; - crq.control_ip_offload.cmd = CONTROL_IP_OFFLOAD; - crq.control_ip_offload.len = - cpu_to_be32(sizeof(adapter->ip_offload_ctrl)); - crq.control_ip_offload.ioba = cpu_to_be32(adapter->ip_offload_ctrl_tok); - ibmvnic_send_crq(adapter, &crq); + send_control_ip_offload(adapter); } static const char *ibmvnic_fw_err_cause(u16 cause) @@ -4263,7 +4304,7 @@ static void handle_request_cap_rsp(union ibmvnic_crq *crq, be64_to_cpu(crq->request_capability_rsp.number); } - ibmvnic_send_req_caps(adapter, 1); + send_request_cap(adapter, 1); return; default: dev_err(dev, "Error %d in request cap rsp\n", @@ -4273,30 +4314,8 @@ static void handle_request_cap_rsp(union ibmvnic_crq *crq, /* Done receiving requested capabilities, query IP offload support */ if (atomic_read(&adapter->running_cap_crqs) == 0) { - union ibmvnic_crq newcrq; - int buf_sz = sizeof(struct ibmvnic_query_ip_offload_buffer); - struct ibmvnic_query_ip_offload_buffer *ip_offload_buf = - &adapter->ip_offload_buf; - adapter->wait_capability = false; - adapter->ip_offload_tok = dma_map_single(dev, ip_offload_buf, - buf_sz, - DMA_FROM_DEVICE); - - if (dma_mapping_error(dev, adapter->ip_offload_tok)) { - if (!firmware_has_feature(FW_FEATURE_CMO)) - dev_err(dev, "Couldn't map offload buffer\n"); - return; - } - - memset(&newcrq, 0, sizeof(newcrq)); - newcrq.query_ip_offload.first = IBMVNIC_CRQ_CMD; - newcrq.query_ip_offload.cmd = QUERY_IP_OFFLOAD; - newcrq.query_ip_offload.len = cpu_to_be32(buf_sz); - newcrq.query_ip_offload.ioba = - cpu_to_be32(adapter->ip_offload_tok); - - ibmvnic_send_crq(adapter, &newcrq); + send_query_ip_offload(adapter); } } @@ -4307,6 +4326,11 @@ static int handle_login_rsp(union ibmvnic_crq *login_rsp_crq, struct net_device *netdev = adapter->netdev; struct ibmvnic_login_rsp_buffer *login_rsp = adapter->login_rsp_buf; struct ibmvnic_login_buffer *login = adapter->login_buf; + u64 *tx_handle_array; + u64 *rx_handle_array; + int num_tx_pools; + int num_rx_pools; + u64 *size_array; int i; dma_unmap_single(dev, adapter->login_buf_token, adapter->login_buf_sz, @@ -4341,6 +4365,30 @@ static int handle_login_rsp(union ibmvnic_crq *login_rsp_crq, ibmvnic_remove(adapter->vdev); return -EIO; } + size_array = (u64 *)((u8 *)(adapter->login_rsp_buf) + + be32_to_cpu(adapter->login_rsp_buf->off_rxadd_buff_size)); + /* variable buffer sizes are not supported, so just read the + * first entry. + */ + adapter->cur_rx_buf_sz = be64_to_cpu(size_array[0]); + + num_tx_pools = be32_to_cpu(adapter->login_rsp_buf->num_txsubm_subcrqs); + num_rx_pools = be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs); + + tx_handle_array = (u64 *)((u8 *)(adapter->login_rsp_buf) + + be32_to_cpu(adapter->login_rsp_buf->off_txsubm_subcrqs)); + rx_handle_array = (u64 *)((u8 *)(adapter->login_rsp_buf) + + be32_to_cpu(adapter->login_rsp_buf->off_rxadd_subcrqs)); + + for (i = 0; i < num_tx_pools; i++) + adapter->tx_scrq[i]->handle = tx_handle_array[i]; + + for (i = 0; i < num_rx_pools; i++) + adapter->rx_scrq[i]->handle = rx_handle_array[i]; + + adapter->num_active_tx_scrqs = num_tx_pools; + adapter->num_active_rx_scrqs = num_rx_pools; + release_login_rsp_buffer(adapter); release_login_buffer(adapter); complete(&adapter->init_done); @@ -4550,7 +4598,7 @@ static void handle_query_cap_rsp(union ibmvnic_crq *crq, out: if (atomic_read(&adapter->running_cap_crqs) == 0) { adapter->wait_capability = false; - ibmvnic_send_req_caps(adapter, 0); + send_request_cap(adapter, 0); } } @@ -4605,7 +4653,7 @@ static int handle_query_phys_parms_rsp(union ibmvnic_crq *crq, case IBMVNIC_1GBPS: adapter->speed = SPEED_1000; break; - case IBMVNIC_10GBP: + case IBMVNIC_10GBPS: adapter->speed = SPEED_10000; break; case IBMVNIC_25GBPS: @@ -4620,6 +4668,9 @@ static int handle_query_phys_parms_rsp(union ibmvnic_crq *crq, case IBMVNIC_100GBPS: adapter->speed = SPEED_100000; break; + case IBMVNIC_200GBPS: + adapter->speed = SPEED_200000; + break; default: if (netif_carrier_ok(netdev)) netdev_warn(netdev, "Unknown speed 0x%08x\n", rspeed); @@ -4715,7 +4766,7 @@ static void ibmvnic_handle_crq(union ibmvnic_crq *crq, be16_to_cpu(crq->version_exchange_rsp.version); dev_info(dev, "Partner protocol version is %d\n", ibmvnic_version); - send_cap_queries(adapter); + send_query_cap(adapter); break; case QUERY_CAPABILITY_RSP: handle_query_cap_rsp(crq, adapter); @@ -4812,9 +4863,9 @@ static irqreturn_t ibmvnic_interrupt(int irq, void *instance) return IRQ_HANDLED; } -static void ibmvnic_tasklet(void *data) +static void ibmvnic_tasklet(struct tasklet_struct *t) { - struct ibmvnic_adapter *adapter = data; + struct ibmvnic_adapter *adapter = from_tasklet(adapter, t, tasklet); struct ibmvnic_crq_queue *queue = &adapter->crq; union ibmvnic_crq *crq; unsigned long flags; @@ -4949,8 +5000,7 @@ static int init_crq_queue(struct ibmvnic_adapter *adapter) retrc = 0; - tasklet_init(&adapter->tasklet, (void *)ibmvnic_tasklet, - (unsigned long)adapter); + tasklet_setup(&adapter->tasklet, (void *)ibmvnic_tasklet); netdev_dbg(adapter->netdev, "registering irq 0x%x\n", vdev->irq); snprintf(crq->name, sizeof(crq->name), "ibmvnic-%x", @@ -4986,7 +5036,7 @@ map_failed: return retrc; } -static int ibmvnic_reset_init(struct ibmvnic_adapter *adapter) +static int ibmvnic_reset_init(struct ibmvnic_adapter *adapter, bool reset) { struct device *dev = &adapter->vdev->dev; unsigned long timeout = msecs_to_jiffies(30000); @@ -4995,12 +5045,19 @@ static int ibmvnic_reset_init(struct ibmvnic_adapter *adapter) adapter->from_passive_init = false; - old_num_rx_queues = adapter->req_rx_queues; - old_num_tx_queues = adapter->req_tx_queues; + if (reset) { + old_num_rx_queues = adapter->req_rx_queues; + old_num_tx_queues = adapter->req_tx_queues; + reinit_completion(&adapter->init_done); + } - reinit_completion(&adapter->init_done); adapter->init_done_rc = 0; - ibmvnic_send_crq_init(adapter); + rc = ibmvnic_send_crq_init(adapter); + if (rc) { + dev_err(dev, "Send crq init failed with error %d\n", rc); + return rc; + } + if (!wait_for_completion_timeout(&adapter->init_done, timeout)) { dev_err(dev, "Initialization sequence timed out\n"); return -1; @@ -5017,7 +5074,8 @@ static int ibmvnic_reset_init(struct ibmvnic_adapter *adapter) return -1; } - if (test_bit(0, &adapter->resetting) && !adapter->wait_for_reset && + if (reset && + test_bit(0, &adapter->resetting) && !adapter->wait_for_reset && adapter->reset_reason != VNIC_RESET_MOBILITY) { if (adapter->req_rx_queues != old_num_rx_queues || adapter->req_tx_queues != old_num_tx_queues) { @@ -5045,48 +5103,6 @@ static int ibmvnic_reset_init(struct ibmvnic_adapter *adapter) return rc; } -static int ibmvnic_init(struct ibmvnic_adapter *adapter) -{ - struct device *dev = &adapter->vdev->dev; - unsigned long timeout = msecs_to_jiffies(30000); - int rc; - - adapter->from_passive_init = false; - - adapter->init_done_rc = 0; - ibmvnic_send_crq_init(adapter); - if (!wait_for_completion_timeout(&adapter->init_done, timeout)) { - dev_err(dev, "Initialization sequence timed out\n"); - return -1; - } - - if (adapter->init_done_rc) { - release_crq_queue(adapter); - return adapter->init_done_rc; - } - - if (adapter->from_passive_init) { - adapter->state = VNIC_OPEN; - adapter->from_passive_init = false; - return -1; - } - - rc = init_sub_crqs(adapter); - if (rc) { - dev_err(dev, "Initialization of sub crqs failed\n"); - release_crq_queue(adapter); - return rc; - } - - rc = init_sub_crq_irqs(adapter); - if (rc) { - dev_err(dev, "Failed to initialize sub crq irqs\n"); - release_crq_queue(adapter); - } - - return rc; -} - static struct device_attribute dev_attr_failover; static int ibmvnic_probe(struct vio_dev *dev, const struct vio_device_id *id) @@ -5149,7 +5165,7 @@ static int ibmvnic_probe(struct vio_dev *dev, const struct vio_device_id *id) goto ibmvnic_init_fail; } - rc = ibmvnic_init(adapter); + rc = ibmvnic_reset_init(adapter, false); if (rc && rc != EAGAIN) goto ibmvnic_init_fail; } while (rc == EAGAIN); @@ -5299,8 +5315,7 @@ static unsigned long ibmvnic_get_desired_dma(struct vio_dev *vdev) for (i = 0; i < adapter->req_tx_queues + adapter->req_rx_queues; i++) ret += 4 * PAGE_SIZE; /* the scrq message queue */ - for (i = 0; i < be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs); - i++) + for (i = 0; i < adapter->num_active_rx_pools; i++) ret += adapter->rx_pool[i].size * IOMMU_PAGE_ALIGN(adapter->rx_pool[i].buff_size, tbl); diff --git a/drivers/net/ethernet/ibm/ibmvnic.h b/drivers/net/ethernet/ibm/ibmvnic.h index f8416e1d4cf0..217dcc7ded70 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.h +++ b/drivers/net/ethernet/ibm/ibmvnic.h @@ -373,7 +373,7 @@ struct ibmvnic_phys_parms { #define IBMVNIC_10MBPS 0x40000000 #define IBMVNIC_100MBPS 0x20000000 #define IBMVNIC_1GBPS 0x10000000 -#define IBMVNIC_10GBP 0x08000000 +#define IBMVNIC_10GBPS 0x08000000 #define IBMVNIC_40GBPS 0x04000000 #define IBMVNIC_100GBPS 0x02000000 #define IBMVNIC_25GBPS 0x01000000 @@ -875,6 +875,7 @@ struct ibmvnic_sub_crq_queue { struct ibmvnic_adapter *adapter; atomic_t used; char name[32]; + u64 handle; }; struct ibmvnic_long_term_buff { @@ -1075,6 +1076,7 @@ struct ibmvnic_adapter { u32 num_active_rx_napi; u32 num_active_tx_scrqs; u32 num_active_tx_pools; + u32 cur_rx_buf_sz; struct tasklet_struct tasklet; enum vnic_state state; diff --git a/drivers/net/ethernet/intel/e100.c b/drivers/net/ethernet/intel/e100.c index 36da059388dc..8cc651d37a7f 100644 --- a/drivers/net/ethernet/intel/e100.c +++ b/drivers/net/ethernet/intel/e100.c @@ -384,7 +384,7 @@ enum cb_status { cb_ok = 0x2000, }; -/** +/* * cb_command - Command Block flags * @cb_tx_nc: 0: controller does CRC (normal), 1: CRC from skb memory */ @@ -1531,7 +1531,7 @@ static int e100_hw_init(struct nic *nic) e100_hw_reset(nic); netif_err(nic, hw, nic->netdev, "e100_hw_init\n"); - if (!in_interrupt() && (err = e100_self_test(nic))) + if ((err = e100_self_test(nic))) return err; if ((err = e100_phy_init(nic))) @@ -2155,7 +2155,7 @@ static int e100_rx_alloc_list(struct nic *nic) nic->rx_to_use = nic->rx_to_clean = NULL; nic->ru_running = RU_UNINITIALIZED; - if (!(nic->rxs = kcalloc(count, sizeof(struct rx), GFP_ATOMIC))) + if (!(nic->rxs = kcalloc(count, sizeof(struct rx), GFP_KERNEL))) return -ENOMEM; for (rx = nic->rxs, i = 0; i < count; rx++, i++) { @@ -2593,7 +2593,7 @@ static void e100_diag_test(struct net_device *netdev, { struct ethtool_cmd cmd; struct nic *nic = netdev_priv(netdev); - int i, err; + int i; memset(data, 0, E100_TEST_LEN * sizeof(u64)); data[0] = !mii_link_ok(&nic->mii); @@ -2601,7 +2601,7 @@ static void e100_diag_test(struct net_device *netdev, if (test->flags & ETH_TEST_FL_OFFLINE) { /* save speed, duplex & autoneg settings */ - err = mii_ethtool_gset(&nic->mii, &cmd); + mii_ethtool_gset(&nic->mii, &cmd); if (netif_running(netdev)) e100_down(nic); @@ -2610,7 +2610,7 @@ static void e100_diag_test(struct net_device *netdev, data[4] = e100_loopback_test(nic, lb_phy); /* restore speed, duplex & autoneg settings */ - err = mii_ethtool_sset(&nic->mii, &cmd); + mii_ethtool_sset(&nic->mii, &cmd); if (netif_running(netdev)) e100_up(nic); diff --git a/drivers/net/ethernet/intel/e1000/e1000_hw.c b/drivers/net/ethernet/intel/e1000/e1000_hw.c index 4e7a0810eaeb..4c0c9433bd60 100644 --- a/drivers/net/ethernet/intel/e1000/e1000_hw.c +++ b/drivers/net/ethernet/intel/e1000/e1000_hw.c @@ -129,7 +129,6 @@ static s32 e1000_set_phy_type(struct e1000_hw *hw) */ static void e1000_phy_init_script(struct e1000_hw *hw) { - u32 ret_val; u16 phy_saved_data; if (hw->phy_init_script) { @@ -138,7 +137,7 @@ static void e1000_phy_init_script(struct e1000_hw *hw) /* Save off the current value of register 0x2F5B to be restored * at the end of this routine. */ - ret_val = e1000_read_phy_reg(hw, 0x2F5B, &phy_saved_data); + e1000_read_phy_reg(hw, 0x2F5B, &phy_saved_data); /* Disabled the PHY transmitter */ e1000_write_phy_reg(hw, 0x2F5B, 0x0003); @@ -377,7 +376,6 @@ s32 e1000_reset_hw(struct e1000_hw *hw) { u32 ctrl; u32 ctrl_ext; - u32 icr; u32 manc; u32 led_ctrl; s32 ret_val; @@ -502,7 +500,7 @@ s32 e1000_reset_hw(struct e1000_hw *hw) ew32(IMC, 0xffffffff); /* Clear any pending interrupt events. */ - icr = er32(ICR); + er32(ICR); /* If MWI was previously enabled, reenable it. */ if (hw->mac_type == e1000_82542_rev2_0) { @@ -1897,7 +1895,6 @@ void e1000_config_collision_dist(struct e1000_hw *hw) /** * e1000_config_mac_to_phy - sync phy and mac settings * @hw: Struct containing variables accessed by shared code - * @mii_reg: data to write to the MII control register * * Sets MAC speed and duplex settings to reflect the those in the PHY * The contents of the PHY register containing the needed information need to @@ -2370,16 +2367,13 @@ static s32 e1000_check_for_serdes_link_generic(struct e1000_hw *hw) */ s32 e1000_check_for_link(struct e1000_hw *hw) { - u32 rxcw = 0; - u32 ctrl; u32 status; u32 rctl; u32 icr; - u32 signal = 0; s32 ret_val; u16 phy_data; - ctrl = er32(CTRL); + er32(CTRL); status = er32(STATUS); /* On adapters with a MAC newer than 82544, SW Definable pin 1 will be @@ -2388,12 +2382,9 @@ s32 e1000_check_for_link(struct e1000_hw *hw) */ if ((hw->media_type == e1000_media_type_fiber) || (hw->media_type == e1000_media_type_internal_serdes)) { - rxcw = er32(RXCW); + er32(RXCW); if (hw->media_type == e1000_media_type_fiber) { - signal = - (hw->mac_type > - e1000_82544) ? E1000_CTRL_SWDPIN1 : 0; if (status & E1000_STATUS_LU) hw->get_link_status = false; } @@ -2922,7 +2913,7 @@ static s32 e1000_read_phy_reg_ex(struct e1000_hw *hw, u32 reg_addr, * * @hw: Struct containing variables accessed by shared code * @reg_addr: address of the PHY register to write - * @data: data to write to the PHY + * @phy_data: data to write to the PHY * * Writes a value to a PHY register */ @@ -4410,17 +4401,9 @@ void e1000_write_vfta(struct e1000_hw *hw, u32 offset, u32 value) static void e1000_clear_vfta(struct e1000_hw *hw) { u32 offset; - u32 vfta_value = 0; - u32 vfta_offset = 0; - u32 vfta_bit_in_reg = 0; for (offset = 0; offset < E1000_VLAN_FILTER_TBL_SIZE; offset++) { - /* If the offset we want to clear is the same offset of the - * manageability VLAN ID, then clear all bits except that of the - * manageability unit - */ - vfta_value = (offset == vfta_offset) ? vfta_bit_in_reg : 0; - E1000_WRITE_REG_ARRAY(hw, VFTA, offset, vfta_value); + E1000_WRITE_REG_ARRAY(hw, VFTA, offset, 0); E1000_WRITE_FLUSH(); } } @@ -4675,78 +4658,76 @@ s32 e1000_led_off(struct e1000_hw *hw) */ static void e1000_clear_hw_cntrs(struct e1000_hw *hw) { - volatile u32 temp; - - temp = er32(CRCERRS); - temp = er32(SYMERRS); - temp = er32(MPC); - temp = er32(SCC); - temp = er32(ECOL); - temp = er32(MCC); - temp = er32(LATECOL); - temp = er32(COLC); - temp = er32(DC); - temp = er32(SEC); - temp = er32(RLEC); - temp = er32(XONRXC); - temp = er32(XONTXC); - temp = er32(XOFFRXC); - temp = er32(XOFFTXC); - temp = er32(FCRUC); - - temp = er32(PRC64); - temp = er32(PRC127); - temp = er32(PRC255); - temp = er32(PRC511); - temp = er32(PRC1023); - temp = er32(PRC1522); - - temp = er32(GPRC); - temp = er32(BPRC); - temp = er32(MPRC); - temp = er32(GPTC); - temp = er32(GORCL); - temp = er32(GORCH); - temp = er32(GOTCL); - temp = er32(GOTCH); - temp = er32(RNBC); - temp = er32(RUC); - temp = er32(RFC); - temp = er32(ROC); - temp = er32(RJC); - temp = er32(TORL); - temp = er32(TORH); - temp = er32(TOTL); - temp = er32(TOTH); - temp = er32(TPR); - temp = er32(TPT); - - temp = er32(PTC64); - temp = er32(PTC127); - temp = er32(PTC255); - temp = er32(PTC511); - temp = er32(PTC1023); - temp = er32(PTC1522); - - temp = er32(MPTC); - temp = er32(BPTC); + er32(CRCERRS); + er32(SYMERRS); + er32(MPC); + er32(SCC); + er32(ECOL); + er32(MCC); + er32(LATECOL); + er32(COLC); + er32(DC); + er32(SEC); + er32(RLEC); + er32(XONRXC); + er32(XONTXC); + er32(XOFFRXC); + er32(XOFFTXC); + er32(FCRUC); + + er32(PRC64); + er32(PRC127); + er32(PRC255); + er32(PRC511); + er32(PRC1023); + er32(PRC1522); + + er32(GPRC); + er32(BPRC); + er32(MPRC); + er32(GPTC); + er32(GORCL); + er32(GORCH); + er32(GOTCL); + er32(GOTCH); + er32(RNBC); + er32(RUC); + er32(RFC); + er32(ROC); + er32(RJC); + er32(TORL); + er32(TORH); + er32(TOTL); + er32(TOTH); + er32(TPR); + er32(TPT); + + er32(PTC64); + er32(PTC127); + er32(PTC255); + er32(PTC511); + er32(PTC1023); + er32(PTC1522); + + er32(MPTC); + er32(BPTC); if (hw->mac_type < e1000_82543) return; - temp = er32(ALGNERRC); - temp = er32(RXERRC); - temp = er32(TNCRS); - temp = er32(CEXTERR); - temp = er32(TSCTC); - temp = er32(TSCTFC); + er32(ALGNERRC); + er32(RXERRC); + er32(TNCRS); + er32(CEXTERR); + er32(TSCTC); + er32(TSCTFC); if (hw->mac_type <= e1000_82544) return; - temp = er32(MGTPRC); - temp = er32(MGTPDC); - temp = er32(MGTPTC); + er32(MGTPRC); + er32(MGTPDC); + er32(MGTPTC); } /** @@ -4778,8 +4759,6 @@ void e1000_reset_adaptive(struct e1000_hw *hw) /** * e1000_update_adaptive - update adaptive IFS * @hw: Struct containing variables accessed by shared code - * @tx_packets: Number of transmits since last callback - * @total_collisions: Number of collisions since last callback * * Called during the callback/watchdog routine to update IFS value based on * the ratio of transmits to collisions. @@ -5064,8 +5043,6 @@ static s32 e1000_check_polarity(struct e1000_hw *hw, /** * e1000_check_downshift - Check if Downshift occurred * @hw: Struct containing variables accessed by shared code - * @downshift: output parameter : 0 - No Downshift occurred. - * 1 - Downshift occurred. * * returns: - E1000_ERR_XXX * E1000_SUCCESS diff --git a/drivers/net/ethernet/intel/e1000/e1000_main.c b/drivers/net/ethernet/intel/e1000/e1000_main.c index 1e6ec081fd9d..5e28cf4fa2cd 100644 --- a/drivers/net/ethernet/intel/e1000/e1000_main.c +++ b/drivers/net/ethernet/intel/e1000/e1000_main.c @@ -199,8 +199,10 @@ module_param(debug, int, 0); MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)"); /** - * e1000_get_hw_dev - return device - * used by hardware layer to print debugging information + * e1000_get_hw_dev - helper function for getting netdev + * @hw: pointer to HW struct + * + * return device used by hardware layer to print debugging information * **/ struct net_device *e1000_get_hw_dev(struct e1000_hw *hw) @@ -354,7 +356,7 @@ static void e1000_release_manageability(struct e1000_adapter *adapter) /** * e1000_configure - configure the hardware for RX and TX - * @adapter = private board structure + * @adapter: private board structure **/ static void e1000_configure(struct e1000_adapter *adapter) { @@ -534,7 +536,6 @@ void e1000_down(struct e1000_adapter *adapter) void e1000_reinit_locked(struct e1000_adapter *adapter) { - WARN_ON(in_interrupt()); while (test_and_set_bit(__E1000_RESETTING, &adapter->flags)) msleep(1); @@ -3489,8 +3490,9 @@ exit: /** * e1000_tx_timeout - Respond to a Tx Hang * @netdev: network interface device structure + * @txqueue: number of the Tx queue that hung (unused) **/ -static void e1000_tx_timeout(struct net_device *netdev, unsigned int txqueue) +static void e1000_tx_timeout(struct net_device *netdev, unsigned int __always_unused txqueue) { struct e1000_adapter *adapter = netdev_priv(netdev); @@ -3787,7 +3789,8 @@ static irqreturn_t e1000_intr(int irq, void *data) /** * e1000_clean - NAPI Rx polling callback - * @adapter: board private structure + * @napi: napi struct containing references to driver info + * @budget: budget given to driver for receive packets **/ static int e1000_clean(struct napi_struct *napi, int budget) { @@ -3818,6 +3821,7 @@ static int e1000_clean(struct napi_struct *napi, int budget) /** * e1000_clean_tx_irq - Reclaim resources after transmit completes * @adapter: board private structure + * @tx_ring: ring to clean **/ static bool e1000_clean_tx_irq(struct e1000_adapter *adapter, struct e1000_tx_ring *tx_ring) @@ -3933,7 +3937,7 @@ static bool e1000_clean_tx_irq(struct e1000_adapter *adapter, * @adapter: board private structure * @status_err: receive descriptor status and error fields * @csum: receive descriptor csum field - * @sk_buff: socket buffer with received data + * @skb: socket buffer with received data **/ static void e1000_rx_checksum(struct e1000_adapter *adapter, u32 status_err, u32 csum, struct sk_buff *skb) @@ -3970,6 +3974,9 @@ static void e1000_rx_checksum(struct e1000_adapter *adapter, u32 status_err, /** * e1000_consume_page - helper function for jumbo Rx path + * @bi: software descriptor shadow data + * @skb: skb being modified + * @length: length of data being added **/ static void e1000_consume_page(struct e1000_rx_buffer *bi, struct sk_buff *skb, u16 length) @@ -4003,6 +4010,7 @@ static void e1000_receive_skb(struct e1000_adapter *adapter, u8 status, /** * e1000_tbi_adjust_stats * @hw: Struct containing variables accessed by shared code + * @stats: point to stats struct * @frame_len: The length of the frame in question * @mac_addr: The Ethernet destination address of the frame in question * @@ -4548,6 +4556,8 @@ e1000_alloc_jumbo_rx_buffers(struct e1000_adapter *adapter, /** * e1000_alloc_rx_buffers - Replace used receive buffers; legacy & extended * @adapter: address of board private structure + * @rx_ring: pointer to ring struct + * @cleaned_count: number of new Rx buffers to try to allocate **/ static void e1000_alloc_rx_buffers(struct e1000_adapter *adapter, struct e1000_rx_ring *rx_ring, @@ -4662,7 +4672,7 @@ static void e1000_alloc_rx_buffers(struct e1000_adapter *adapter, /** * e1000_smartspeed - Workaround for SmartSpeed on 82541 and 82547 controllers. - * @adapter: + * @adapter: address of board private structure **/ static void e1000_smartspeed(struct e1000_adapter *adapter) { @@ -4718,10 +4728,10 @@ static void e1000_smartspeed(struct e1000_adapter *adapter) } /** - * e1000_ioctl - - * @netdev: - * @ifreq: - * @cmd: + * e1000_ioctl - handle ioctl calls + * @netdev: pointer to our netdev + * @ifr: pointer to interface request structure + * @cmd: ioctl data **/ static int e1000_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) { @@ -4737,9 +4747,9 @@ static int e1000_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) /** * e1000_mii_ioctl - - * @netdev: - * @ifreq: - * @cmd: + * @netdev: pointer to our netdev + * @ifr: pointer to interface request structure + * @cmd: ioctl data **/ static int e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) diff --git a/drivers/net/ethernet/intel/e1000e/80003es2lan.c b/drivers/net/ethernet/intel/e1000e/80003es2lan.c index 4b103cca8a39..be9c695dde12 100644 --- a/drivers/net/ethernet/intel/e1000e/80003es2lan.c +++ b/drivers/net/ethernet/intel/e1000e/80003es2lan.c @@ -1072,7 +1072,6 @@ static s32 e1000_setup_copper_link_80003es2lan(struct e1000_hw *hw) /** * e1000_cfg_on_link_up_80003es2lan - es2 link configuration after link-up * @hw: pointer to the HW structure - * @duplex: current duplex setting * * Configure the KMRN interface by applying last minute quirks for * 10/100 operation. diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c index a8fc9208382c..03215b0aee4b 100644 --- a/drivers/net/ethernet/intel/e1000e/ethtool.c +++ b/drivers/net/ethernet/intel/e1000e/ethtool.c @@ -895,6 +895,7 @@ static int e1000_reg_test(struct e1000_adapter *adapter, u64 *data) case e1000_pch_cnp: case e1000_pch_tgp: case e1000_pch_adp: + case e1000_pch_mtp: mask |= BIT(18); break; default: @@ -1560,6 +1561,7 @@ static void e1000_loopback_cleanup(struct e1000_adapter *adapter) case e1000_pch_cnp: case e1000_pch_tgp: case e1000_pch_adp: + case e1000_pch_mtp: fext_nvm11 = er32(FEXTNVM11); fext_nvm11 &= ~E1000_FEXTNVM11_DISABLE_MULR_FIX; ew32(FEXTNVM11, fext_nvm11); diff --git a/drivers/net/ethernet/intel/e1000e/hw.h b/drivers/net/ethernet/intel/e1000e/hw.h index b1447221669e..69a2329ea463 100644 --- a/drivers/net/ethernet/intel/e1000e/hw.h +++ b/drivers/net/ethernet/intel/e1000e/hw.h @@ -102,6 +102,10 @@ struct e1000_hw; #define E1000_DEV_ID_PCH_ADP_I219_V16 0x1A1F #define E1000_DEV_ID_PCH_ADP_I219_LM17 0x1A1C #define E1000_DEV_ID_PCH_ADP_I219_V17 0x1A1D +#define E1000_DEV_ID_PCH_MTP_I219_LM18 0x550A +#define E1000_DEV_ID_PCH_MTP_I219_V18 0x550B +#define E1000_DEV_ID_PCH_MTP_I219_LM19 0x550C +#define E1000_DEV_ID_PCH_MTP_I219_V19 0x550D #define E1000_REVISION_4 4 @@ -127,6 +131,7 @@ enum e1000_mac_type { e1000_pch_cnp, e1000_pch_tgp, e1000_pch_adp, + e1000_pch_mtp, }; enum e1000_media_type { diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c index b2f2fcfdf732..9aa6fad8ed47 100644 --- a/drivers/net/ethernet/intel/e1000e/ich8lan.c +++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c @@ -320,6 +320,7 @@ static s32 e1000_init_phy_workarounds_pchlan(struct e1000_hw *hw) case e1000_pch_cnp: case e1000_pch_tgp: case e1000_pch_adp: + case e1000_pch_mtp: if (e1000_phy_is_accessible_pchlan(hw)) break; @@ -464,6 +465,7 @@ static s32 e1000_init_phy_params_pchlan(struct e1000_hw *hw) case e1000_pch_cnp: case e1000_pch_tgp: case e1000_pch_adp: + case e1000_pch_mtp: /* In case the PHY needs to be in mdio slow mode, * set slow mode and try to get the PHY id again. */ @@ -708,6 +710,7 @@ static s32 e1000_init_mac_params_ich8lan(struct e1000_hw *hw) case e1000_pch_cnp: case e1000_pch_tgp: case e1000_pch_adp: + case e1000_pch_mtp: case e1000_pchlan: /* check management mode */ mac->ops.check_mng_mode = e1000_check_mng_mode_pchlan; @@ -743,7 +746,7 @@ static s32 e1000_init_mac_params_ich8lan(struct e1000_hw *hw) /** * __e1000_access_emi_reg_locked - Read/write EMI register * @hw: pointer to the HW structure - * @addr: EMI address to program + * @address: EMI address to program * @data: pointer to value to read/write from/to the EMI address * @read: boolean flag to indicate read or write * @@ -1648,6 +1651,7 @@ static s32 e1000_get_variants_ich8lan(struct e1000_adapter *adapter) case e1000_pch_cnp: case e1000_pch_tgp: case e1000_pch_adp: + case e1000_pch_mtp: rc = e1000_init_phy_params_pchlan(hw); break; default: @@ -2102,6 +2106,7 @@ static s32 e1000_sw_lcd_config_ich8lan(struct e1000_hw *hw) case e1000_pch_cnp: case e1000_pch_tgp: case e1000_pch_adp: + case e1000_pch_mtp: sw_cfg_mask = E1000_FEXTNVM_SW_CONFIG_ICH8M; break; default: @@ -2266,7 +2271,7 @@ release: /** * e1000_configure_k1_ich8lan - Configure K1 power state * @hw: pointer to the HW structure - * @enable: K1 state to configure + * @k1_enable: K1 state to configure * * Configure the K1 power state based on the provided parameter. * Assumes semaphore already acquired. @@ -2405,8 +2410,10 @@ static s32 e1000_set_mdio_slow_mode_hv(struct e1000_hw *hw) } /** - * e1000_hv_phy_workarounds_ich8lan - A series of Phy workarounds to be - * done after every PHY reset. + * e1000_hv_phy_workarounds_ich8lan - apply PHY workarounds + * @hw: pointer to the HW structure + * + * A series of PHY workarounds to be done after every PHY reset. **/ static s32 e1000_hv_phy_workarounds_ich8lan(struct e1000_hw *hw) { @@ -2694,8 +2701,10 @@ s32 e1000_lv_jumbo_workaround_ich8lan(struct e1000_hw *hw, bool enable) } /** - * e1000_lv_phy_workarounds_ich8lan - A series of Phy workarounds to be - * done after every PHY reset. + * e1000_lv_phy_workarounds_ich8lan - apply ich8 specific workarounds + * @hw: pointer to the HW structure + * + * A series of PHY workarounds to be done after every PHY reset. **/ static s32 e1000_lv_phy_workarounds_ich8lan(struct e1000_hw *hw) { @@ -3141,6 +3150,7 @@ static s32 e1000_valid_nvm_bank_detect_ich8lan(struct e1000_hw *hw, u32 *bank) case e1000_pch_cnp: case e1000_pch_tgp: case e1000_pch_adp: + case e1000_pch_mtp: bank1_offset = nvm->flash_bank_size; act_offset = E1000_ICH_NVM_SIG_WORD; @@ -4086,6 +4096,7 @@ static s32 e1000_validate_nvm_checksum_ich8lan(struct e1000_hw *hw) case e1000_pch_cnp: case e1000_pch_tgp: case e1000_pch_adp: + case e1000_pch_mtp: word = NVM_COMPAT; valid_csum_mask = NVM_COMPAT_VALID_CSUM; break; diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c index 664e8ccc88d2..b30f00891c03 100644 --- a/drivers/net/ethernet/intel/e1000e/netdev.c +++ b/drivers/net/ethernet/intel/e1000e/netdev.c @@ -501,6 +501,7 @@ rx_ring_summary: /** * e1000_desc_unused - calculate if we have unused descriptors + * @ring: pointer to ring struct to perform calculation on **/ static int e1000_desc_unused(struct e1000_ring *ring) { @@ -577,6 +578,7 @@ static void e1000e_rx_hwtstamp(struct e1000_adapter *adapter, u32 status, /** * e1000_receive_skb - helper function to handle Rx indications * @adapter: board private structure + * @netdev: pointer to netdev struct * @staterr: descriptor extended error and status field as written by hardware * @vlan: descriptor vlan field as written by hardware (no le/be conversion) * @skb: pointer to sk_buff to be indicated to stack @@ -601,8 +603,7 @@ static void e1000_receive_skb(struct e1000_adapter *adapter, * e1000_rx_checksum - Receive Checksum Offload * @adapter: board private structure * @status_err: receive descriptor status and error fields - * @csum: receive descriptor csum field - * @sk_buff: socket buffer with received data + * @skb: socket buffer with received data **/ static void e1000_rx_checksum(struct e1000_adapter *adapter, u32 status_err, struct sk_buff *skb) @@ -673,6 +674,8 @@ static void e1000e_update_tdt_wa(struct e1000_ring *tx_ring, unsigned int i) /** * e1000_alloc_rx_buffers - Replace used receive buffers * @rx_ring: Rx descriptor ring + * @cleaned_count: number to reallocate + * @gfp: flags for allocation **/ static void e1000_alloc_rx_buffers(struct e1000_ring *rx_ring, int cleaned_count, gfp_t gfp) @@ -741,6 +744,8 @@ map_skb: /** * e1000_alloc_rx_buffers_ps - Replace used receive buffers; packet split * @rx_ring: Rx descriptor ring + * @cleaned_count: number to reallocate + * @gfp: flags for allocation **/ static void e1000_alloc_rx_buffers_ps(struct e1000_ring *rx_ring, int cleaned_count, gfp_t gfp) @@ -844,6 +849,7 @@ no_buffers: * e1000_alloc_jumbo_rx_buffers - Replace used jumbo receive buffers * @rx_ring: Rx descriptor ring * @cleaned_count: number of buffers to allocate this pass + * @gfp: flags for allocation **/ static void e1000_alloc_jumbo_rx_buffers(struct e1000_ring *rx_ring, @@ -933,6 +939,8 @@ static inline void e1000_rx_hash(struct net_device *netdev, __le32 rss, /** * e1000_clean_rx_irq - Send received data up the network stack * @rx_ring: Rx descriptor ring + * @work_done: output parameter for indicating completed work + * @work_to_do: how many packets we can clean * * the return value indicates whether actual cleaning was done, there * is no guarantee that everything was cleaned @@ -1327,6 +1335,8 @@ static bool e1000_clean_tx_irq(struct e1000_ring *tx_ring) /** * e1000_clean_rx_irq_ps - Send received data up the network stack; packet split * @rx_ring: Rx descriptor ring + * @work_done: output parameter for indicating completed work + * @work_to_do: how many packets we can clean * * the return value indicates whether actual cleaning was done, there * is no guarantee that everything was cleaned @@ -1517,9 +1527,6 @@ next_desc: return cleaned; } -/** - * e1000_consume_page - helper function - **/ static void e1000_consume_page(struct e1000_buffer *bi, struct sk_buff *skb, u16 length) { @@ -1531,7 +1538,9 @@ static void e1000_consume_page(struct e1000_buffer *bi, struct sk_buff *skb, /** * e1000_clean_jumbo_rx_irq - Send received data up the network stack; legacy - * @adapter: board private structure + * @rx_ring: Rx descriptor ring + * @work_done: output parameter for indicating completed work + * @work_to_do: how many packets we can clean * * the return value indicates whether actual cleaning was done, there * is no guarantee that everything was cleaned @@ -1994,6 +2003,7 @@ static irqreturn_t e1000_intr_msix_rx(int __always_unused irq, void *data) /** * e1000_configure_msix - Configure MSI-X hardware + * @adapter: board private structure * * e1000_configure_msix sets up the hardware to properly * generate MSI-X interrupts. @@ -2072,6 +2082,7 @@ void e1000e_reset_interrupt_capability(struct e1000_adapter *adapter) /** * e1000e_set_interrupt_capability - set MSI or MSI-X if supported + * @adapter: board private structure * * Attempt to configure interrupts using the best available * capabilities of the hardware and kernel. @@ -2127,6 +2138,7 @@ void e1000e_set_interrupt_capability(struct e1000_adapter *adapter) /** * e1000_request_msix - Initialize MSI-X interrupts + * @adapter: board private structure * * e1000_request_msix allocates MSI-X vectors and requests interrupts from the * kernel. @@ -2180,6 +2192,7 @@ static int e1000_request_msix(struct e1000_adapter *adapter) /** * e1000_request_irq - initialize interrupts + * @adapter: board private structure * * Attempts to configure interrupts using the best available * capabilities of the hardware and kernel. @@ -2240,6 +2253,7 @@ static void e1000_free_irq(struct e1000_adapter *adapter) /** * e1000_irq_disable - Mask off interrupt generation on the NIC + * @adapter: board private structure **/ static void e1000_irq_disable(struct e1000_adapter *adapter) { @@ -2262,6 +2276,7 @@ static void e1000_irq_disable(struct e1000_adapter *adapter) /** * e1000_irq_enable - Enable default interrupt generation settings + * @adapter: board private structure **/ static void e1000_irq_enable(struct e1000_adapter *adapter) { @@ -2332,6 +2347,8 @@ void e1000e_release_hw_control(struct e1000_adapter *adapter) /** * e1000_alloc_ring_dma - allocate memory for a ring structure + * @adapter: board private structure + * @ring: ring struct for which to allocate dma **/ static int e1000_alloc_ring_dma(struct e1000_adapter *adapter, struct e1000_ring *ring) @@ -2507,7 +2524,6 @@ void e1000e_free_rx_resources(struct e1000_ring *rx_ring) /** * e1000_update_itr - update the dynamic ITR value based on statistics - * @adapter: pointer to adapter * @itr_setting: current adapter->itr * @packets: the number of packets during this measurement interval * @bytes: the number of bytes during this measurement interval @@ -3049,12 +3065,13 @@ static void e1000_configure_tx(struct e1000_adapter *adapter) } } +#define PAGE_USE_COUNT(S) (((S) >> PAGE_SHIFT) + \ + (((S) & (PAGE_SIZE - 1)) ? 1 : 0)) + /** * e1000_setup_rctl - configure the receive control registers * @adapter: Board private structure **/ -#define PAGE_USE_COUNT(S) (((S) >> PAGE_SHIFT) + \ - (((S) & (PAGE_SIZE - 1)) ? 1 : 0)) static void e1000_setup_rctl(struct e1000_adapter *adapter) { struct e1000_hw *hw = &adapter->hw; @@ -3570,6 +3587,7 @@ s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca) case e1000_pch_cnp: case e1000_pch_tgp: case e1000_pch_adp: + case e1000_pch_mtp: if (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_SYSCFI) { /* Stable 24MHz frequency */ incperiod = INCPERIOD_24MHZ; @@ -3605,6 +3623,7 @@ s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca) /** * e1000e_config_hwtstamp - configure the hwtstamp registers and enable/disable * @adapter: board private structure + * @config: timestamp configuration * * Outgoing time stamping can be enabled and disabled. Play nice and * disable it when requested, although it shouldn't cause any overhead @@ -3808,6 +3827,7 @@ void e1000e_power_up_phy(struct e1000_adapter *adapter) /** * e1000_power_down_phy - Power down the PHY + * @adapter: board private structure * * Power down the PHY so no link is implied when interface is down. * The PHY cannot be powered down if management or WoL is active. @@ -3820,6 +3840,7 @@ static void e1000_power_down_phy(struct e1000_adapter *adapter) /** * e1000_flush_tx_ring - remove all descriptors from the tx_ring + * @adapter: board private structure * * We want to clear all pending descriptors from the TX ring. * zeroing happens when the HW reads the regs. We assign the ring itself as @@ -3854,6 +3875,7 @@ static void e1000_flush_tx_ring(struct e1000_adapter *adapter) /** * e1000_flush_rx_ring - remove all descriptors from the rx_ring + * @adapter: board private structure * * Mark all descriptors in the RX ring as consumed and disable the rx ring */ @@ -3886,6 +3908,7 @@ static void e1000_flush_rx_ring(struct e1000_adapter *adapter) /** * e1000_flush_desc_rings - remove all descriptors from the descriptor rings + * @adapter: board private structure * * In i219, the descriptor rings must be emptied before resetting the HW * or before changing the device state to D3 during runtime (runtime PM). @@ -3968,6 +3991,7 @@ static void e1000e_systim_reset(struct e1000_adapter *adapter) /** * e1000e_reset - bring the hardware into a known good state + * @adapter: board private structure * * This function boots the hardware and enables some settings that * require a configuration cycle of the hardware - those cannot be @@ -4081,6 +4105,7 @@ void e1000e_reset(struct e1000_adapter *adapter) case e1000_pch_cnp: case e1000_pch_tgp: case e1000_pch_adp: + case e1000_pch_mtp: fc->refresh_time = 0xFFFF; fc->pause_time = 0xFFFF; @@ -4847,7 +4872,7 @@ static void e1000e_update_phy_task(struct work_struct *work) /** * e1000_update_phy_info - timre call-back to update PHY info - * @data: pointer to adapter cast into an unsigned long + * @t: pointer to timer_list containing private info adapter * * Need to wait a few seconds after link up to get diagnostic information from * the phy @@ -5187,7 +5212,7 @@ static void e1000e_check_82574_phy_workaround(struct e1000_adapter *adapter) /** * e1000_watchdog - Timer Call-back - * @data: pointer to adapter cast into an unsigned long + * @t: pointer to timer_list containing private info adapter **/ static void e1000_watchdog(struct timer_list *t) { @@ -5972,8 +5997,9 @@ static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb, /** * e1000_tx_timeout - Respond to a Tx Hang * @netdev: network interface device structure + * @txqueue: index of the hung queue (unused) **/ -static void e1000_tx_timeout(struct net_device *netdev, unsigned int txqueue) +static void e1000_tx_timeout(struct net_device *netdev, unsigned int __always_unused txqueue) { struct e1000_adapter *adapter = netdev_priv(netdev); @@ -6174,7 +6200,7 @@ static int e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, /** * e1000e_hwtstamp_ioctl - control hardware time stamping * @netdev: network interface device structure - * @ifreq: interface request + * @ifr: interface request * * Outgoing time stamping can be enabled and disabled. Play nice and * disable it when requested, although it shouldn't cause any overhead @@ -7853,6 +7879,10 @@ static const struct pci_device_id e1000_pci_tbl[] = { { PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_V16), board_pch_cnp }, { PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_LM17), board_pch_cnp }, { PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_V17), board_pch_cnp }, + { PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_MTP_I219_LM18), board_pch_cnp }, + { PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_MTP_I219_V18), board_pch_cnp }, + { PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_MTP_I219_LM19), board_pch_cnp }, + { PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_MTP_I219_V19), board_pch_cnp }, { 0, 0, 0, 0, 0, 0, 0 } /* terminate list */ }; diff --git a/drivers/net/ethernet/intel/e1000e/phy.c b/drivers/net/ethernet/intel/e1000e/phy.c index e11c877595fb..bdd9dc163f15 100644 --- a/drivers/net/ethernet/intel/e1000e/phy.c +++ b/drivers/net/ethernet/intel/e1000e/phy.c @@ -2311,6 +2311,7 @@ s32 e1000e_determine_phy_address(struct e1000_hw *hw) /** * e1000_get_phy_addr_for_bm_page - Retrieve PHY page address * @page: page to access + * @reg: register to check * * Returns the phy address for the page requested. **/ @@ -2728,6 +2729,7 @@ void e1000_power_down_phy_copper(struct e1000_hw *hw) * @offset: register offset to be read * @data: pointer to the read data * @locked: semaphore has already been acquired or not + * @page_set: BM_WUC_PAGE already set and access enabled * * Acquires semaphore, if necessary, then reads the PHY register at offset * and stores the retrieved information in data. Release any acquired @@ -2836,6 +2838,7 @@ s32 e1000_read_phy_reg_page_hv(struct e1000_hw *hw, u32 offset, u16 *data) * @offset: register offset to write to * @data: data to write at register offset * @locked: semaphore has already been acquired or not + * @page_set: BM_WUC_PAGE already set and access enabled * * Acquires semaphore, if necessary, then writes the data to PHY register * at the offset. Release any acquired semaphores before exiting. diff --git a/drivers/net/ethernet/intel/e1000e/ptp.c b/drivers/net/ethernet/intel/e1000e/ptp.c index 34b988d70488..f3f671311855 100644 --- a/drivers/net/ethernet/intel/e1000e/ptp.c +++ b/drivers/net/ethernet/intel/e1000e/ptp.c @@ -144,7 +144,7 @@ static int e1000e_phc_get_syncdevicetime(ktime_t *device, /** * e1000e_phc_getsynctime - Reads the current system/device cross timestamp * @ptp: ptp clock structure - * @cts: structure containing timestamp + * @xtstamp: structure containing timestamp * * Read device and system (ART) clock simultaneously and return the scaled * clock values in ns. @@ -297,6 +297,7 @@ void e1000e_ptp_init(struct e1000_adapter *adapter) case e1000_pch_cnp: case e1000_pch_tgp: case e1000_pch_adp: + case e1000_pch_mtp: if ((hw->mac.type < e1000_pch_lpt) || (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_SYSCFI)) { adapter->ptp_clock_info.max_adj = 24000000 - 1; diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_main.c b/drivers/net/ethernet/intel/fm10k/fm10k_main.c index d88dd41a9442..99b8252eb969 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_main.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_main.c @@ -310,10 +310,7 @@ static struct sk_buff *fm10k_fetch_rx_buffer(struct fm10k_ring *rx_ring, rx_buffer->page_offset; /* prefetch first cache line of first page */ - prefetch(page_addr); -#if L1_CACHE_BYTES < 128 - prefetch((void *)((u8 *)page_addr + L1_CACHE_BYTES)); -#endif + net_prefetch(page_addr); /* allocate a skb to store the frags */ skb = napi_alloc_skb(&rx_ring->q_vector->napi, diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c index 140212bfe08b..9e3103fae723 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c @@ -221,8 +221,6 @@ static bool fm10k_prepare_for_reset(struct fm10k_intfc *interface) { struct net_device *netdev = interface->netdev; - WARN_ON(in_interrupt()); - /* put off any impending NetWatchDogTimeout */ netif_trans_update(netdev); diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h index a7e212d1caa2..537300e762f0 100644 --- a/drivers/net/ethernet/intel/i40e/i40e.h +++ b/drivers/net/ethernet/intel/i40e/i40e.h @@ -35,6 +35,7 @@ #include <net/pkt_cls.h> #include <net/tc_act/tc_gact.h> #include <net/tc_act/tc_mirred.h> +#include <net/udp_tunnel.h> #include <net/xdp_sock.h> #include "i40e_type.h" #include "i40e_prototype.h" @@ -90,7 +91,7 @@ #define I40E_OEM_RELEASE_MASK 0x0000ffff #define I40E_RX_DESC(R, i) \ - (&(((union i40e_32byte_rx_desc *)((R)->desc))[i])) + (&(((union i40e_rx_desc *)((R)->desc))[i])) #define I40E_TX_DESC(R, i) \ (&(((struct i40e_tx_desc *)((R)->desc))[i])) #define I40E_TX_CTXTDESC(R, i) \ @@ -133,7 +134,6 @@ enum i40e_state_t { __I40E_PORT_SUSPENDED, __I40E_VF_DISABLE, __I40E_MACVLAN_SYNC_PENDING, - __I40E_UDP_FILTER_SYNC_PENDING, __I40E_TEMP_LINK_POLLING, __I40E_CLIENT_SERVICE_REQUESTED, __I40E_CLIENT_L2_CHANGE, @@ -478,8 +478,8 @@ struct i40e_pf { struct list_head l3_flex_pit_list; struct list_head l4_flex_pit_list; - struct i40e_udp_port_config udp_ports[I40E_MAX_PF_UDP_OFFLOAD_PORTS]; - u16 pending_udp_bitmap; + struct udp_tunnel_nic_shared udp_tunnel_shared; + struct udp_tunnel_nic_info udp_tunnel_nic; struct hlist_head cloud_filter_list; u16 num_cloud_filters; diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq.c b/drivers/net/ethernet/intel/i40e/i40e_adminq.c index c897a2863e4f..593912b17609 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_adminq.c +++ b/drivers/net/ethernet/intel/i40e/i40e_adminq.c @@ -541,6 +541,12 @@ static void i40e_set_hw_flags(struct i40e_hw *hw) (aq->api_maj_ver == 1 && aq->api_min_ver >= I40E_MINOR_VER_GET_LINK_INFO_X722)) hw->flags |= I40E_HW_FLAG_AQ_PHY_ACCESS_CAPABLE; + + if (aq->api_maj_ver > 1 || + (aq->api_maj_ver == 1 && + aq->api_min_ver >= I40E_MINOR_VER_FW_REQUEST_FEC_X722)) + hw->flags |= I40E_HW_FLAG_X722_FEC_REQUEST_CAPABLE; + fallthrough; default: break; diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq.h b/drivers/net/ethernet/intel/i40e/i40e_adminq.h index edec3df78971..ee394aacef4d 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_adminq.h +++ b/drivers/net/ethernet/intel/i40e/i40e_adminq.h @@ -85,8 +85,8 @@ struct i40e_adminq_info { /** * i40e_aq_rc_to_posix - convert errors to user-land codes - * aq_ret: AdminQ handler error code can override aq_rc - * aq_rc: AdminQ firmware error code to convert + * @aq_ret: AdminQ handler error code can override aq_rc + * @aq_rc: AdminQ firmware error code to convert **/ static inline int i40e_aq_rc_to_posix(int aq_ret, int aq_rc) { diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h index c0c8efe42fce..1e960c3c7ef0 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h +++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h @@ -24,6 +24,8 @@ #define I40E_MINOR_VER_GET_LINK_INFO_X722 0x0009 /* API version 1.6 for X722 devices adds ability to stop FW LLDP agent */ #define I40E_MINOR_VER_FW_LLDP_STOPPABLE_X722 0x0006 +/* API version 1.10 for X722 devices adds ability to request FEC encoding */ +#define I40E_MINOR_VER_FW_REQUEST_FEC_X722 0x000A struct i40e_aq_desc { __le16 flags; diff --git a/drivers/net/ethernet/intel/i40e/i40e_client.c b/drivers/net/ethernet/intel/i40e/i40e_client.c index befd3018183f..a2dba32383f6 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_client.c +++ b/drivers/net/ethernet/intel/i40e/i40e_client.c @@ -278,8 +278,6 @@ void i40e_client_update_msix_info(struct i40e_pf *pf) /** * i40e_client_add_instance - add a client instance struct to the instance list * @pf: pointer to the board struct - * @client: pointer to a client struct in the client list. - * @existing: if there was already an existing instance * **/ static void i40e_client_add_instance(struct i40e_pf *pf) diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c index 6ab52cbd697a..adc9e4fa4789 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_common.c +++ b/drivers/net/ethernet/intel/i40e/i40e_common.c @@ -3766,9 +3766,7 @@ i40e_status i40e_aq_stop_lldp(struct i40e_hw *hw, bool shutdown_agent, /** * i40e_aq_start_lldp * @hw: pointer to the hw struct - * @buff: buffer for result * @persist: True if start of LLDP should be persistent across power cycles - * @buff_size: buffer size * @cmd_details: pointer to command details structure or NULL * * Start the embedded LLDP Agent on all ports. @@ -5395,6 +5393,7 @@ static void i40e_mdio_if_number_selection(struct i40e_hw *hw, bool set_mdio, * @hw: pointer to the hw struct * @phy_select: select which phy should be accessed * @dev_addr: PHY device address + * @page_change: flag to indicate if phy page should be updated * @set_mdio: use MDIO I/F number specified by mdio_num * @mdio_num: MDIO I/F number * @reg_addr: PHY register address @@ -5439,6 +5438,7 @@ enum i40e_status_code i40e_aq_set_phy_register_ext(struct i40e_hw *hw, * @hw: pointer to the hw struct * @phy_select: select which phy should be accessed * @dev_addr: PHY device address + * @page_change: flag to indicate if phy page should be updated * @set_mdio: use MDIO I/F number specified by mdio_num * @mdio_num: MDIO I/F number * @reg_addr: PHY register address diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c index d3ad2e3aa838..d7c13ca9be7d 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c +++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c @@ -604,10 +604,9 @@ static void i40e_dbg_dump_desc(int cnt, int vsi_seid, int ring_id, int desc_n, } else { rxd = I40E_RX_DESC(ring, i); dev_info(&pf->pdev->dev, - " d[%03x] = 0x%016llx 0x%016llx 0x%016llx 0x%016llx\n", + " d[%03x] = 0x%016llx 0x%016llx\n", i, rxd->read.pkt_addr, - rxd->read.hdr_addr, - rxd->read.rsvd1, rxd->read.rsvd2); + rxd->read.hdr_addr); } } } else if (cnt == 3) { @@ -625,10 +624,9 @@ static void i40e_dbg_dump_desc(int cnt, int vsi_seid, int ring_id, int desc_n, } else { rxd = I40E_RX_DESC(ring, desc_n); dev_info(&pf->pdev->dev, - "vsi = %02i rx ring = %02i d[%03x] = 0x%016llx 0x%016llx 0x%016llx 0x%016llx\n", + "vsi = %02i rx ring = %02i d[%03x] = 0x%016llx 0x%016llx\n", vsi_seid, ring_id, desc_n, - rxd->read.pkt_addr, rxd->read.hdr_addr, - rxd->read.rsvd1, rxd->read.rsvd2); + rxd->read.pkt_addr, rxd->read.hdr_addr); } } else { dev_info(&pf->pdev->dev, "dump desc rx/tx/xdp <vsi_seid> <ring_id> [<desc_n>]\n"); diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c index 825c104ecba1..26ba1f3eb2d8 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c +++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c @@ -891,6 +891,7 @@ static void i40e_get_settings_link_up(struct i40e_hw *hw, if (hw_link_info->requested_speeds & I40E_LINK_SPEED_10GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 10000baseT_Full); + i40e_get_settings_link_up_fec(hw_link_info->req_fec_info, ks); break; case I40E_PHY_TYPE_SGMII: ethtool_link_ksettings_add_link_mode(ks, supported, Autoneg); @@ -1481,12 +1482,16 @@ static int i40e_set_fec_param(struct net_device *netdev, struct i40e_pf *pf = np->vsi->back; struct i40e_hw *hw = &pf->hw; u8 fec_cfg = 0; - int err = 0; if (hw->device_id != I40E_DEV_ID_25G_SFP28 && - hw->device_id != I40E_DEV_ID_25G_B) { - err = -EPERM; - goto done; + hw->device_id != I40E_DEV_ID_25G_B && + hw->device_id != I40E_DEV_ID_KX_X722) + return -EPERM; + + if (hw->mac.type == I40E_MAC_X722 && + !(hw->flags & I40E_HW_FLAG_X722_FEC_REQUEST_CAPABLE)) { + netdev_err(netdev, "Setting FEC encoding not supported by firmware. Please update the NVM image.\n"); + return -EOPNOTSUPP; } switch (fecparam->fec) { @@ -1508,14 +1513,10 @@ static int i40e_set_fec_param(struct net_device *netdev, default: dev_warn(&pf->pdev->dev, "Unsupported FEC mode: %d", fecparam->fec); - err = -EINVAL; - goto done; + return -EINVAL; } - err = i40e_set_fec_cfg(netdev, fec_cfg); - -done: - return err; + return i40e_set_fec_cfg(netdev, fec_cfg); } static int i40e_nway_reset(struct net_device *netdev) @@ -1967,7 +1968,7 @@ static int i40e_set_ringparam(struct net_device *netdev, (new_rx_count == vsi->rx_rings[0]->count)) return 0; - /* If there is a AF_XDP UMEM attached to any of Rx rings, + /* If there is a AF_XDP page pool attached to any of Rx rings, * disallow changing the number of descriptors -- regardless * if the netdev is running or not. */ @@ -4951,8 +4952,7 @@ flags_complete: } } - if (((changed_flags & I40E_FLAG_RS_FEC) || - (changed_flags & I40E_FLAG_BASE_R_FEC)) && + if (changed_flags & I40E_FLAG_RS_FEC && pf->hw.device_id != I40E_DEV_ID_25G_SFP28 && pf->hw.device_id != I40E_DEV_ID_25G_B) { dev_warn(&pf->pdev->dev, @@ -4960,6 +4960,15 @@ flags_complete: return -EOPNOTSUPP; } + if (changed_flags & I40E_FLAG_BASE_R_FEC && + pf->hw.device_id != I40E_DEV_ID_25G_SFP28 && + pf->hw.device_id != I40E_DEV_ID_25G_B && + pf->hw.device_id != I40E_DEV_ID_KX_X722) { + dev_warn(&pf->pdev->dev, + "Device does not support changing FEC configuration\n"); + return -EOPNOTSUPP; + } + /* Process any additional changes needed as a result of flag changes. * The changed_flags value reflects the list of bits that were * changed in the code above. diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 2e433fdbf2c3..4f8a2154b93f 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -287,6 +287,7 @@ void i40e_service_event_schedule(struct i40e_pf *pf) /** * i40e_tx_timeout - Respond to a Tx Hang * @netdev: network interface device structure + * @txqueue: queue number timing out * * If any port has noticed a Tx timeout, it is likely that the whole * device is munged, not just the one netdev port, so go for the full @@ -1609,6 +1610,8 @@ static int i40e_set_mac(struct net_device *netdev, void *p) * i40e_config_rss_aq - Prepare for RSS using AQ commands * @vsi: vsi structure * @seed: RSS hash seed + * @lut: pointer to lookup table of lut_size + * @lut_size: size of the lookup table **/ static int i40e_config_rss_aq(struct i40e_vsi *vsi, const u8 *seed, u8 *lut, u16 lut_size) @@ -3122,12 +3125,12 @@ static void i40e_config_xps_tx_ring(struct i40e_ring *ring) } /** - * i40e_xsk_umem - Retrieve the AF_XDP ZC if XDP and ZC is enabled + * i40e_xsk_pool - Retrieve the AF_XDP buffer pool if XDP and ZC is enabled * @ring: The Tx or Rx ring * - * Returns the UMEM or NULL. + * Returns the AF_XDP buffer pool or NULL. **/ -static struct xdp_umem *i40e_xsk_umem(struct i40e_ring *ring) +static struct xsk_buff_pool *i40e_xsk_pool(struct i40e_ring *ring) { bool xdp_on = i40e_enabled_xdp_vsi(ring->vsi); int qid = ring->queue_index; @@ -3138,7 +3141,7 @@ static struct xdp_umem *i40e_xsk_umem(struct i40e_ring *ring) if (!xdp_on || !test_bit(qid, ring->vsi->af_xdp_zc_qps)) return NULL; - return xdp_get_umem_from_qid(ring->vsi->netdev, qid); + return xsk_get_pool_from_qid(ring->vsi->netdev, qid); } /** @@ -3157,7 +3160,7 @@ static int i40e_configure_tx_ring(struct i40e_ring *ring) u32 qtx_ctl = 0; if (ring_is_xdp(ring)) - ring->xsk_umem = i40e_xsk_umem(ring); + ring->xsk_pool = i40e_xsk_pool(ring); /* some ATR related tx ring init */ if (vsi->back->flags & I40E_FLAG_FD_ATR_ENABLED) { @@ -3280,12 +3283,13 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring) xdp_rxq_info_unreg_mem_model(&ring->xdp_rxq); kfree(ring->rx_bi); - ring->xsk_umem = i40e_xsk_umem(ring); - if (ring->xsk_umem) { + ring->xsk_pool = i40e_xsk_pool(ring); + if (ring->xsk_pool) { ret = i40e_alloc_rx_bi_zc(ring); if (ret) return ret; - ring->rx_buf_len = xsk_umem_get_rx_frame_size(ring->xsk_umem); + ring->rx_buf_len = + xsk_pool_get_rx_frame_size(ring->xsk_pool); /* For AF_XDP ZC, we disallow packets to span on * multiple buffers, thus letting us skip that * handling in the fast-path. @@ -3320,8 +3324,8 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring) rx_ctx.base = (ring->dma / 128); rx_ctx.qlen = ring->count; - /* use 32 byte descriptors */ - rx_ctx.dsize = 1; + /* use 16 byte descriptors */ + rx_ctx.dsize = 0; /* descriptor type is always zero * rx_ctx.dtype = 0; @@ -3368,8 +3372,8 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring) ring->tail = hw->hw_addr + I40E_QRX_TAIL(pf_q); writel(0, ring->tail); - if (ring->xsk_umem) { - xsk_buff_set_rxq_info(ring->xsk_umem, &ring->xdp_rxq); + if (ring->xsk_pool) { + xsk_pool_set_rxq_info(ring->xsk_pool, &ring->xdp_rxq); ok = i40e_alloc_rx_buffers_zc(ring, I40E_DESC_UNUSED(ring)); } else { ok = !i40e_alloc_rx_buffers(ring, I40E_DESC_UNUSED(ring)); @@ -3380,7 +3384,7 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring) */ dev_info(&vsi->back->pdev->dev, "Failed to allocate some buffers on %sRx ring %d (pf_q %d)\n", - ring->xsk_umem ? "UMEM enabled " : "", + ring->xsk_pool ? "AF_XDP ZC enabled " : "", ring->queue_index, pf_q); } @@ -5814,7 +5818,6 @@ static int i40e_vsi_reconfig_rss(struct i40e_vsi *vsi, u16 rss_size) /** * i40e_channel_setup_queue_map - Setup a channel queue map * @pf: ptr to PF device - * @vsi: the VSI being setup * @ctxt: VSI context structure * @ch: ptr to channel structure * @@ -6057,8 +6060,7 @@ static inline int i40e_setup_hw_channel(struct i40e_pf *pf, /** * i40e_setup_channel - setup new channel using uplink element * @pf: ptr to PF device - * @type: type of channel to be created (VMDq2/VF) - * @uplink_seid: underlying HW switching element (VEB) ID + * @vsi: pointer to the VSI to set up the channel within * @ch: ptr to channel structure * * Setup new channel (VSI) based on specified type (VMDq2/VF) @@ -6623,6 +6625,25 @@ void i40e_print_link_message(struct i40e_vsi *vsi, bool isup) netdev_info(vsi->netdev, "NIC Link is Up, %sbps Full Duplex, Requested FEC: %s, Negotiated FEC: %s, Autoneg: %s, Flow Control: %s\n", speed, req_fec, fec, an, fc); + } else if (pf->hw.device_id == I40E_DEV_ID_KX_X722) { + req_fec = "None"; + fec = "None"; + an = "False"; + + if (pf->hw.phy.link_info.an_info & I40E_AQ_AN_COMPLETED) + an = "True"; + + if (pf->hw.phy.link_info.fec_info & + I40E_AQ_CONFIG_FEC_KR_ENA) + fec = "CL74 FC-FEC/BASE-R"; + + if (pf->hw.phy.link_info.req_fec_info & + I40E_AQ_REQUEST_FEC_KR) + req_fec = "CL74 FC-FEC/BASE-R"; + + netdev_info(vsi->netdev, + "NIC Link is Up, %sbps Full Duplex, Requested FEC: %s, Negotiated FEC: %s, Autoneg: %s, Flow Control: %s\n", + speed, req_fec, fec, an, fc); } else { netdev_info(vsi->netdev, "NIC Link is Up, %sbps Full Duplex, Flow Control: %s\n", @@ -6689,7 +6710,6 @@ static void i40e_vsi_reinit_locked(struct i40e_vsi *vsi) { struct i40e_pf *pf = vsi->back; - WARN_ON(in_interrupt()); while (test_and_set_bit(__I40E_CONFIG_BUSY, pf->state)) usleep_range(1000, 2000); i40e_down(vsi); @@ -7779,7 +7799,7 @@ int i40e_add_del_cloud_filter_big_buf(struct i40e_vsi *vsi, /** * i40e_parse_cls_flower - Parse tc flower filters provided by kernel * @vsi: Pointer to VSI - * @cls_flower: Pointer to struct flow_cls_offload + * @f: Pointer to struct flow_cls_offload * @filter: Pointer to cloud filter structure * **/ @@ -8160,8 +8180,8 @@ static int i40e_delete_clsflower(struct i40e_vsi *vsi, /** * i40e_setup_tc_cls_flower - flower classifier offloads - * @netdev: net device to configure - * @type_data: offload data + * @np: net device to configure + * @cls_flower: offload data **/ static int i40e_setup_tc_cls_flower(struct i40e_netdev_priv *np, struct flow_cls_offload *cls_flower) @@ -8462,9 +8482,6 @@ void i40e_do_reset(struct i40e_pf *pf, u32 reset_flags, bool lock_acquired) { u32 val; - WARN_ON(in_interrupt()); - - /* do the biggest reset indicated */ if (reset_flags & BIT_ULL(__I40E_GLOBAL_RESET_REQUESTED)) { @@ -9585,6 +9602,7 @@ end_reconstitute: /** * i40e_get_capabilities - get info about the HW * @pf: the PF struct + * @list_type: AQ capability to be queried **/ static int i40e_get_capabilities(struct i40e_pf *pf, enum i40e_admin_queue_opc list_type) @@ -10383,106 +10401,6 @@ static void i40e_handle_mdd_event(struct i40e_pf *pf) i40e_flush(hw); } -static const char *i40e_tunnel_name(u8 type) -{ - switch (type) { - case UDP_TUNNEL_TYPE_VXLAN: - return "vxlan"; - case UDP_TUNNEL_TYPE_GENEVE: - return "geneve"; - default: - return "unknown"; - } -} - -/** - * i40e_sync_udp_filters - Trigger a sync event for existing UDP filters - * @pf: board private structure - **/ -static void i40e_sync_udp_filters(struct i40e_pf *pf) -{ - int i; - - /* loop through and set pending bit for all active UDP filters */ - for (i = 0; i < I40E_MAX_PF_UDP_OFFLOAD_PORTS; i++) { - if (pf->udp_ports[i].port) - pf->pending_udp_bitmap |= BIT_ULL(i); - } - - set_bit(__I40E_UDP_FILTER_SYNC_PENDING, pf->state); -} - -/** - * i40e_sync_udp_filters_subtask - Sync the VSI filter list with HW - * @pf: board private structure - **/ -static void i40e_sync_udp_filters_subtask(struct i40e_pf *pf) -{ - struct i40e_hw *hw = &pf->hw; - u8 filter_index, type; - u16 port; - int i; - - if (!test_and_clear_bit(__I40E_UDP_FILTER_SYNC_PENDING, pf->state)) - return; - - /* acquire RTNL to maintain state of flags and port requests */ - rtnl_lock(); - - for (i = 0; i < I40E_MAX_PF_UDP_OFFLOAD_PORTS; i++) { - if (pf->pending_udp_bitmap & BIT_ULL(i)) { - struct i40e_udp_port_config *udp_port; - i40e_status ret = 0; - - udp_port = &pf->udp_ports[i]; - pf->pending_udp_bitmap &= ~BIT_ULL(i); - - port = READ_ONCE(udp_port->port); - type = READ_ONCE(udp_port->type); - filter_index = READ_ONCE(udp_port->filter_index); - - /* release RTNL while we wait on AQ command */ - rtnl_unlock(); - - if (port) - ret = i40e_aq_add_udp_tunnel(hw, port, - type, - &filter_index, - NULL); - else if (filter_index != I40E_UDP_PORT_INDEX_UNUSED) - ret = i40e_aq_del_udp_tunnel(hw, filter_index, - NULL); - - /* reacquire RTNL so we can update filter_index */ - rtnl_lock(); - - if (ret) { - dev_info(&pf->pdev->dev, - "%s %s port %d, index %d failed, err %s aq_err %s\n", - i40e_tunnel_name(type), - port ? "add" : "delete", - port, - filter_index, - i40e_stat_str(&pf->hw, ret), - i40e_aq_str(&pf->hw, - pf->hw.aq.asq_last_status)); - if (port) { - /* failed to add, just reset port, - * drop pending bit for any deletion - */ - udp_port->port = 0; - pf->pending_udp_bitmap &= ~BIT_ULL(i); - } - } else if (port) { - /* record filter index on success */ - udp_port->filter_index = filter_index; - } - } - } - - rtnl_unlock(); -} - /** * i40e_service_task - Run the driver's async subtasks * @work: pointer to work_struct containing our data @@ -10522,7 +10440,6 @@ static void i40e_service_task(struct work_struct *work) pf->vsi[pf->lan_vsi]); } i40e_sync_filters_subtask(pf); - i40e_sync_udp_filters_subtask(pf); } else { i40e_reset_subtask(pf); } @@ -10546,7 +10463,7 @@ static void i40e_service_task(struct work_struct *work) /** * i40e_service_timer - timer callback - * @data: pointer to PF struct + * @t: timer list pointer **/ static void i40e_service_timer(struct timer_list *t) { @@ -11185,11 +11102,10 @@ static int i40e_init_msix(struct i40e_pf *pf) * i40e_vsi_alloc_q_vector - Allocate memory for a single interrupt vector * @vsi: the VSI being configured * @v_idx: index of the vector in the vsi struct - * @cpu: cpu to be used on affinity_mask * * We allocate one q_vector. If allocation fails we return -ENOMEM. **/ -static int i40e_vsi_alloc_q_vector(struct i40e_vsi *vsi, int v_idx, int cpu) +static int i40e_vsi_alloc_q_vector(struct i40e_vsi *vsi, int v_idx) { struct i40e_q_vector *q_vector; @@ -11222,7 +11138,7 @@ static int i40e_vsi_alloc_q_vector(struct i40e_vsi *vsi, int v_idx, int cpu) static int i40e_vsi_alloc_q_vectors(struct i40e_vsi *vsi) { struct i40e_pf *pf = vsi->back; - int err, v_idx, num_q_vectors, current_cpu; + int err, v_idx, num_q_vectors; /* if not MSIX, give the one vector only to the LAN VSI */ if (pf->flags & I40E_FLAG_MSIX_ENABLED) @@ -11232,15 +11148,10 @@ static int i40e_vsi_alloc_q_vectors(struct i40e_vsi *vsi) else return -EINVAL; - current_cpu = cpumask_first(cpu_online_mask); - for (v_idx = 0; v_idx < num_q_vectors; v_idx++) { - err = i40e_vsi_alloc_q_vector(vsi, v_idx, current_cpu); + err = i40e_vsi_alloc_q_vector(vsi, v_idx); if (err) goto err_out; - current_cpu = cpumask_next(current_cpu, cpu_online_mask); - if (unlikely(current_cpu >= nr_cpu_ids)) - current_cpu = cpumask_first(cpu_online_mask); } return 0; @@ -12228,131 +12139,48 @@ static int i40e_set_features(struct net_device *netdev, return 0; } -/** - * i40e_get_udp_port_idx - Lookup a possibly offloaded for Rx UDP port - * @pf: board private structure - * @port: The UDP port to look up - * - * Returns the index number or I40E_MAX_PF_UDP_OFFLOAD_PORTS if port not found - **/ -static u8 i40e_get_udp_port_idx(struct i40e_pf *pf, u16 port) -{ - u8 i; - - for (i = 0; i < I40E_MAX_PF_UDP_OFFLOAD_PORTS; i++) { - /* Do not report ports with pending deletions as - * being available. - */ - if (!port && (pf->pending_udp_bitmap & BIT_ULL(i))) - continue; - if (pf->udp_ports[i].port == port) - return i; - } - - return i; -} - -/** - * i40e_udp_tunnel_add - Get notifications about UDP tunnel ports that come up - * @netdev: This physical port's netdev - * @ti: Tunnel endpoint information - **/ -static void i40e_udp_tunnel_add(struct net_device *netdev, - struct udp_tunnel_info *ti) +static int i40e_udp_tunnel_set_port(struct net_device *netdev, + unsigned int table, unsigned int idx, + struct udp_tunnel_info *ti) { struct i40e_netdev_priv *np = netdev_priv(netdev); - struct i40e_vsi *vsi = np->vsi; - struct i40e_pf *pf = vsi->back; - u16 port = ntohs(ti->port); - u8 next_idx; - u8 idx; - - idx = i40e_get_udp_port_idx(pf, port); - - /* Check if port already exists */ - if (idx < I40E_MAX_PF_UDP_OFFLOAD_PORTS) { - netdev_info(netdev, "port %d already offloaded\n", port); - return; - } - - /* Now check if there is space to add the new port */ - next_idx = i40e_get_udp_port_idx(pf, 0); + struct i40e_hw *hw = &np->vsi->back->hw; + u8 type, filter_index; + i40e_status ret; - if (next_idx == I40E_MAX_PF_UDP_OFFLOAD_PORTS) { - netdev_info(netdev, "maximum number of offloaded UDP ports reached, not adding port %d\n", - port); - return; - } + type = ti->type == UDP_TUNNEL_TYPE_VXLAN ? I40E_AQC_TUNNEL_TYPE_VXLAN : + I40E_AQC_TUNNEL_TYPE_NGE; - switch (ti->type) { - case UDP_TUNNEL_TYPE_VXLAN: - pf->udp_ports[next_idx].type = I40E_AQC_TUNNEL_TYPE_VXLAN; - break; - case UDP_TUNNEL_TYPE_GENEVE: - if (!(pf->hw_features & I40E_HW_GENEVE_OFFLOAD_CAPABLE)) - return; - pf->udp_ports[next_idx].type = I40E_AQC_TUNNEL_TYPE_NGE; - break; - default: - return; + ret = i40e_aq_add_udp_tunnel(hw, ntohs(ti->port), type, &filter_index, + NULL); + if (ret) { + netdev_info(netdev, "add UDP port failed, err %s aq_err %s\n", + i40e_stat_str(hw, ret), + i40e_aq_str(hw, hw->aq.asq_last_status)); + return -EIO; } - /* New port: add it and mark its index in the bitmap */ - pf->udp_ports[next_idx].port = port; - pf->udp_ports[next_idx].filter_index = I40E_UDP_PORT_INDEX_UNUSED; - pf->pending_udp_bitmap |= BIT_ULL(next_idx); - set_bit(__I40E_UDP_FILTER_SYNC_PENDING, pf->state); + udp_tunnel_nic_set_port_priv(netdev, table, idx, filter_index); + return 0; } -/** - * i40e_udp_tunnel_del - Get notifications about UDP tunnel ports that go away - * @netdev: This physical port's netdev - * @ti: Tunnel endpoint information - **/ -static void i40e_udp_tunnel_del(struct net_device *netdev, - struct udp_tunnel_info *ti) +static int i40e_udp_tunnel_unset_port(struct net_device *netdev, + unsigned int table, unsigned int idx, + struct udp_tunnel_info *ti) { struct i40e_netdev_priv *np = netdev_priv(netdev); - struct i40e_vsi *vsi = np->vsi; - struct i40e_pf *pf = vsi->back; - u16 port = ntohs(ti->port); - u8 idx; - - idx = i40e_get_udp_port_idx(pf, port); - - /* Check if port already exists */ - if (idx >= I40E_MAX_PF_UDP_OFFLOAD_PORTS) - goto not_found; + struct i40e_hw *hw = &np->vsi->back->hw; + i40e_status ret; - switch (ti->type) { - case UDP_TUNNEL_TYPE_VXLAN: - if (pf->udp_ports[idx].type != I40E_AQC_TUNNEL_TYPE_VXLAN) - goto not_found; - break; - case UDP_TUNNEL_TYPE_GENEVE: - if (pf->udp_ports[idx].type != I40E_AQC_TUNNEL_TYPE_NGE) - goto not_found; - break; - default: - goto not_found; + ret = i40e_aq_del_udp_tunnel(hw, ti->hw_priv, NULL); + if (ret) { + netdev_info(netdev, "delete UDP port failed, err %s aq_err %s\n", + i40e_stat_str(hw, ret), + i40e_aq_str(hw, hw->aq.asq_last_status)); + return -EIO; } - /* if port exists, set it to 0 (mark for deletion) - * and make it pending - */ - pf->udp_ports[idx].port = 0; - - /* Toggle pending bit instead of setting it. This way if we are - * deleting a port that has yet to be added we just clear the pending - * bit and don't have to worry about it. - */ - pf->pending_udp_bitmap ^= BIT_ULL(idx); - set_bit(__I40E_UDP_FILTER_SYNC_PENDING, pf->state); - - return; -not_found: - netdev_warn(netdev, "UDP port %d was not found, not deleting\n", - port); + return 0; } static int i40e_get_phys_port_id(struct net_device *netdev, @@ -12379,6 +12207,7 @@ static int i40e_get_phys_port_id(struct net_device *netdev, * @addr: the MAC address entry being added * @vid: VLAN ID * @flags: instructions from stack about fdb operation + * @extack: netlink extended ack, unused currently */ static int i40e_ndo_fdb_add(struct ndmsg *ndm, struct nlattr *tb[], struct net_device *dev, @@ -12644,7 +12473,7 @@ static int i40e_xdp_setup(struct i40e_vsi *vsi, */ if (need_reset && prog) for (i = 0; i < vsi->num_queue_pairs; i++) - if (vsi->xdp_rings[i]->xsk_umem) + if (vsi->xdp_rings[i]->xsk_pool) (void)i40e_xsk_wakeup(vsi->netdev, i, XDP_WAKEUP_RX); @@ -12923,8 +12752,8 @@ static int i40e_xdp(struct net_device *dev, switch (xdp->command) { case XDP_SETUP_PROG: return i40e_xdp_setup(vsi, xdp->prog); - case XDP_SETUP_XSK_UMEM: - return i40e_xsk_umem_setup(vsi, xdp->xsk.umem, + case XDP_SETUP_XSK_POOL: + return i40e_xsk_pool_setup(vsi, xdp->xsk.pool, xdp->xsk.queue_id); default: return -EINVAL; @@ -12957,8 +12786,8 @@ static const struct net_device_ops i40e_netdev_ops = { .ndo_set_vf_link_state = i40e_ndo_set_vf_link_state, .ndo_set_vf_spoofchk = i40e_ndo_set_vf_spoofchk, .ndo_set_vf_trust = i40e_ndo_set_vf_trust, - .ndo_udp_tunnel_add = i40e_udp_tunnel_add, - .ndo_udp_tunnel_del = i40e_udp_tunnel_del, + .ndo_udp_tunnel_add = udp_tunnel_nic_add_port, + .ndo_udp_tunnel_del = udp_tunnel_nic_del_port, .ndo_get_phys_port_id = i40e_get_phys_port_id, .ndo_fdb_add = i40e_ndo_fdb_add, .ndo_features_check = i40e_features_check, @@ -13022,6 +12851,8 @@ static int i40e_config_netdev(struct i40e_vsi *vsi) if (!(pf->hw_features & I40E_HW_OUTER_UDP_CSUM_CAPABLE)) netdev->gso_partial_features |= NETIF_F_GSO_UDP_TUNNEL_CSUM; + netdev->udp_tunnel_nic_info = &pf->udp_tunnel_nic; + netdev->gso_partial_features |= NETIF_F_GSO_GRE_CSUM; netdev->hw_enc_features |= hw_enc_features; @@ -14422,7 +14253,7 @@ static int i40e_setup_pf_switch(struct i40e_pf *pf, bool reinit) i40e_ptp_init(pf); /* repopulate tunnel port filters */ - i40e_sync_udp_filters(pf); + udp_tunnel_nic_reset_ntf(pf->vsi[pf->lan_vsi]->netdev); return ret; } @@ -15151,6 +14982,14 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (err) goto err_switch_setup; + pf->udp_tunnel_nic.set_port = i40e_udp_tunnel_set_port; + pf->udp_tunnel_nic.unset_port = i40e_udp_tunnel_unset_port; + pf->udp_tunnel_nic.flags = UDP_TUNNEL_NIC_INFO_MAY_SLEEP; + pf->udp_tunnel_nic.shared = &pf->udp_tunnel_shared; + pf->udp_tunnel_nic.tables[0].n_entries = I40E_MAX_PF_UDP_OFFLOAD_PORTS; + pf->udp_tunnel_nic.tables[0].tunnel_types = UDP_TUNNEL_TYPE_VXLAN | + UDP_TUNNEL_TYPE_GENEVE; + /* The number of VSIs reported by the FW is the minimum guaranteed * to us; HW supports far more and we share the remaining pool with * the other PFs. We allocate space for more than the guarantee with @@ -15160,6 +14999,12 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent) pf->num_alloc_vsi = I40E_MIN_VSI_ALLOC; else pf->num_alloc_vsi = pf->hw.func_caps.num_vsis; + if (pf->num_alloc_vsi > UDP_TUNNEL_NIC_MAX_SHARING_DEVICES) { + dev_warn(&pf->pdev->dev, + "limiting the VSI count due to UDP tunnel limitation %d > %d\n", + pf->num_alloc_vsi, UDP_TUNNEL_NIC_MAX_SHARING_DEVICES); + pf->num_alloc_vsi = UDP_TUNNEL_NIC_MAX_SHARING_DEVICES; + } /* Set up the *vsi struct and our local tracking of the MAIN PF vsi. */ pf->vsi = kcalloc(pf->num_alloc_vsi, sizeof(struct i40e_vsi *), diff --git a/drivers/net/ethernet/intel/i40e/i40e_ptp.c b/drivers/net/ethernet/intel/i40e/i40e_ptp.c index ff7b19c6bc73..7a879614ca55 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_ptp.c +++ b/drivers/net/ethernet/intel/i40e/i40e_ptp.c @@ -259,7 +259,6 @@ static u32 i40e_ptp_get_rx_events(struct i40e_pf *pf) /** * i40e_ptp_rx_hang - Detect error case when Rx timestamp registers are hung * @pf: The PF private data structure - * @vsi: The VSI with the rings relevant to 1588 * * This watchdog task is scheduled to detect error case where hardware has * dropped an Rx packet that was timestamped when the ring is full. The diff --git a/drivers/net/ethernet/intel/i40e/i40e_trace.h b/drivers/net/ethernet/intel/i40e/i40e_trace.h index 424f02077e2e..b5b12299931f 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_trace.h +++ b/drivers/net/ethernet/intel/i40e/i40e_trace.h @@ -22,7 +22,7 @@ #include <linux/tracepoint.h> -/** +/* * i40e_trace() macro enables shared code to refer to trace points * like: * @@ -112,7 +112,7 @@ DECLARE_EVENT_CLASS( i40e_rx_template, TP_PROTO(struct i40e_ring *ring, - union i40e_32byte_rx_desc *desc, + union i40e_16byte_rx_desc *desc, struct sk_buff *skb), TP_ARGS(ring, desc, skb), @@ -140,7 +140,7 @@ DECLARE_EVENT_CLASS( DEFINE_EVENT( i40e_rx_template, i40e_clean_rx_irq, TP_PROTO(struct i40e_ring *ring, - union i40e_32byte_rx_desc *desc, + union i40e_16byte_rx_desc *desc, struct sk_buff *skb), TP_ARGS(ring, desc, skb)); @@ -148,7 +148,7 @@ DEFINE_EVENT( DEFINE_EVENT( i40e_rx_template, i40e_clean_rx_irq_rx, TP_PROTO(struct i40e_ring *ring, - union i40e_32byte_rx_desc *desc, + union i40e_16byte_rx_desc *desc, struct sk_buff *skb), TP_ARGS(ring, desc, skb)); diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c index 3e5c566ceb01..d43ce13a93c9 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c @@ -533,11 +533,11 @@ static void i40e_fd_handle_status(struct i40e_ring *rx_ring, u64 qword0_raw, { struct i40e_pf *pf = rx_ring->vsi->back; struct pci_dev *pdev = pf->pdev; - struct i40e_32b_rx_wb_qw0 *qw0; + struct i40e_16b_rx_wb_qw0 *qw0; u32 fcnt_prog, fcnt_avail; u32 error; - qw0 = (struct i40e_32b_rx_wb_qw0 *)&qword0_raw; + qw0 = (struct i40e_16b_rx_wb_qw0 *)&qword0_raw; error = (qword1 & I40E_RX_PROG_STATUS_DESC_QW1_ERROR_MASK) >> I40E_RX_PROG_STATUS_DESC_QW1_ERROR_SHIFT; @@ -636,7 +636,7 @@ void i40e_clean_tx_ring(struct i40e_ring *tx_ring) unsigned long bi_size; u16 i; - if (ring_is_xdp(tx_ring) && tx_ring->xsk_umem) { + if (ring_is_xdp(tx_ring) && tx_ring->xsk_pool) { i40e_xsk_clean_tx_ring(tx_ring); } else { /* ring already cleared, nothing to do */ @@ -1335,7 +1335,7 @@ void i40e_clean_rx_ring(struct i40e_ring *rx_ring) rx_ring->skb = NULL; } - if (rx_ring->xsk_umem) { + if (rx_ring->xsk_pool) { i40e_xsk_clean_rx_ring(rx_ring); goto skip_free; } @@ -1369,7 +1369,7 @@ void i40e_clean_rx_ring(struct i40e_ring *rx_ring) } skip_free: - if (rx_ring->xsk_umem) + if (rx_ring->xsk_pool) i40e_clear_rx_bi_zc(rx_ring); else i40e_clear_rx_bi(rx_ring); @@ -1418,7 +1418,7 @@ int i40e_setup_rx_descriptors(struct i40e_ring *rx_ring) u64_stats_init(&rx_ring->syncp); /* Round up to nearest 4K */ - rx_ring->size = rx_ring->count * sizeof(union i40e_32byte_rx_desc); + rx_ring->size = rx_ring->count * sizeof(union i40e_rx_desc); rx_ring->size = ALIGN(rx_ring->size, 4096); rx_ring->desc = dma_alloc_coherent(dev, rx_ring->size, &rx_ring->dma, GFP_KERNEL); @@ -1755,7 +1755,6 @@ static inline void i40e_rx_hash(struct i40e_ring *ring, * @rx_ring: rx descriptor ring packet is being transacted on * @rx_desc: pointer to the EOP Rx descriptor * @skb: pointer to current skb being populated - * @rx_ptype: the packet type decoded by hardware * * This function checks the ring, descriptor, and packet information in * order to populate the hash, checksum, VLAN, protocol, and @@ -1953,7 +1952,7 @@ static struct i40e_rx_buffer *i40e_get_rx_buffer(struct i40e_ring *rx_ring, struct i40e_rx_buffer *rx_buffer; rx_buffer = i40e_rx_bi(rx_ring, rx_ring->next_to_clean); - prefetchw(rx_buffer->page); + prefetch_page_address(rx_buffer->page); /* we are reusing so sync this buffer for CPU use */ dma_sync_single_range_for_cpu(rx_ring->dev, @@ -1992,10 +1991,8 @@ static struct sk_buff *i40e_construct_skb(struct i40e_ring *rx_ring, struct sk_buff *skb; /* prefetch first cache line of first page */ - prefetch(xdp->data); -#if L1_CACHE_BYTES < 128 - prefetch(xdp->data + L1_CACHE_BYTES); -#endif + net_prefetch(xdp->data); + /* Note, we get here by enabling legacy-rx via: * * ethtool --set-priv-flags <dev> legacy-rx on @@ -2078,10 +2075,8 @@ static struct sk_buff *i40e_build_skb(struct i40e_ring *rx_ring, * likely have a consumer accessing first few bytes of meta * data, and then actual data. */ - prefetch(xdp->data_meta); -#if L1_CACHE_BYTES < 128 - prefetch(xdp->data_meta + L1_CACHE_BYTES); -#endif + net_prefetch(xdp->data_meta); + /* build an skb around the page buffer */ skb = build_skb(xdp->data_hard_start, truesize); if (unlikely(!skb)) @@ -2300,6 +2295,19 @@ void i40e_finalize_xdp_rx(struct i40e_ring *rx_ring, unsigned int xdp_res) } /** + * i40e_inc_ntc: Advance the next_to_clean index + * @rx_ring: Rx ring + **/ +static void i40e_inc_ntc(struct i40e_ring *rx_ring) +{ + u32 ntc = rx_ring->next_to_clean + 1; + + ntc = (ntc < rx_ring->count) ? ntc : 0; + rx_ring->next_to_clean = ntc; + prefetch(I40E_RX_DESC(rx_ring, ntc)); +} + +/** * i40e_clean_rx_irq - Clean completed descriptors from Rx ring - bounce buf * @rx_ring: rx descriptor ring to transact packets on * @budget: Total limit on number of packets to process @@ -2579,7 +2587,7 @@ int i40e_napi_poll(struct napi_struct *napi, int budget) * budget and be more aggressive about cleaning up the Tx descriptors. */ i40e_for_each_ring(ring, q_vector->tx) { - bool wd = ring->xsk_umem ? + bool wd = ring->xsk_pool ? i40e_clean_xdp_tx_irq(vsi, ring) : i40e_clean_tx_irq(vsi, ring, budget); @@ -2607,7 +2615,7 @@ int i40e_napi_poll(struct napi_struct *napi, int budget) budget_per_ring = budget; i40e_for_each_ring(ring, q_vector->rx) { - int cleaned = ring->xsk_umem ? + int cleaned = ring->xsk_pool ? i40e_clean_rx_irq_zc(ring, budget_per_ring) : i40e_clean_rx_irq(ring, budget_per_ring); @@ -3503,7 +3511,7 @@ dma_error: /** * i40e_xmit_xdp_ring - transmits an XDP buffer to an XDP Tx ring - * @xdp: data to transmit + * @xdpf: data to transmit * @xdp_ring: XDP Tx ring **/ static int i40e_xmit_xdp_ring(struct xdp_frame *xdpf, @@ -3698,7 +3706,9 @@ netdev_tx_t i40e_lan_xmit_frame(struct sk_buff *skb, struct net_device *netdev) /** * i40e_xdp_xmit - Implements ndo_xdp_xmit * @dev: netdev - * @xdp: XDP buffer + * @n: number of frames + * @frames: array of XDP buffer pointers + * @flags: XDP extra info * * Returns number of frames successfully sent. Frames that fail are * free'ed via XDP return API. diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h index 4036893d6825..2feed920ef8a 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h @@ -110,7 +110,7 @@ enum i40e_dyn_idx_t { */ #define I40E_RX_HDR_SIZE I40E_RXBUFFER_256 #define I40E_PACKET_HDR_PAD (ETH_HLEN + ETH_FCS_LEN + (VLAN_HLEN * 2)) -#define i40e_rx_desc i40e_32byte_rx_desc +#define i40e_rx_desc i40e_16byte_rx_desc #define I40E_RX_DMA_ATTR \ (DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING) @@ -388,7 +388,7 @@ struct i40e_ring { struct i40e_channel *ch; struct xdp_rxq_info xdp_rxq; - struct xdp_umem *xsk_umem; + struct xsk_buff_pool *xsk_pool; } ____cacheline_internodealigned_in_smp; static inline bool ring_uses_build_skb(struct i40e_ring *ring) @@ -482,7 +482,6 @@ static inline u32 i40e_get_head(struct i40e_ring *tx_ring) /** * i40e_xmit_descriptor_count - calculate number of Tx descriptors needed * @skb: send buffer - * @tx_ring: ring to send buffer on * * Returns number of data descriptors needed for this skb. Returns 0 to indicate * there is not enough descriptors available in this ring since we need at least diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h b/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h index 667c4dc4b39f..19da3b22160f 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h @@ -21,9 +21,9 @@ void i40e_release_rx_desc(struct i40e_ring *rx_ring, u32 val); #define I40E_XDP_TX BIT(1) #define I40E_XDP_REDIR BIT(2) -/** +/* * build_ctob - Builds the Tx descriptor (cmd, offset and type) qword - **/ + */ static inline __le64 build_ctob(u32 td_cmd, u32 td_offset, unsigned int size, u32 td_tag) { @@ -37,7 +37,7 @@ static inline __le64 build_ctob(u32 td_cmd, u32 td_offset, unsigned int size, /** * i40e_update_tx_stats - Update the egress statistics for the Tx ring * @tx_ring: Tx ring to update - * @total_packet: total packets sent + * @total_packets: total packets sent * @total_bytes: total bytes sent **/ static inline void i40e_update_tx_stats(struct i40e_ring *tx_ring, @@ -99,19 +99,6 @@ static inline bool i40e_rx_is_programming_status(u64 qword1) return qword1 & I40E_RXD_QW1_LENGTH_SPH_MASK; } -/** - * i40e_inc_ntc: Advance the next_to_clean index - * @rx_ring: Rx ring - **/ -static inline void i40e_inc_ntc(struct i40e_ring *rx_ring) -{ - u32 ntc = rx_ring->next_to_clean + 1; - - ntc = (ntc < rx_ring->count) ? ntc : 0; - rx_ring->next_to_clean = ntc; - prefetch(I40E_RX_DESC(rx_ring, ntc)); -} - void i40e_xsk_clean_rx_ring(struct i40e_ring *rx_ring); void i40e_xsk_clean_tx_ring(struct i40e_ring *tx_ring); bool i40e_xsk_any_rx_ring_enabled(struct i40e_vsi *vsi); diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h index 52410d609ba1..c0bdc666f557 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_type.h +++ b/drivers/net/ethernet/intel/i40e/i40e_type.h @@ -595,6 +595,7 @@ struct i40e_hw { #define I40E_HW_FLAG_FW_LLDP_PERSISTENT BIT_ULL(5) #define I40E_HW_FLAG_AQ_PHY_ACCESS_EXTENDED BIT_ULL(6) #define I40E_HW_FLAG_DROP_MODE BIT_ULL(7) +#define I40E_HW_FLAG_X722_FEC_REQUEST_CAPABLE BIT_ULL(8) u64 flags; /* Used in set switch config AQ command */ @@ -628,7 +629,7 @@ union i40e_16byte_rx_desc { __le64 hdr_addr; /* Header buffer address */ } read; struct { - struct { + struct i40e_16b_rx_wb_qw0 { struct { union { __le16 mirroring_status; @@ -647,6 +648,9 @@ union i40e_16byte_rx_desc { __le64 status_error_len; } qword1; } wb; /* writeback */ + struct { + u64 qword[2]; + } raw; }; union i40e_32byte_rx_desc { diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index 47bfb2e95e2d..c96e2f2d4cba 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -2244,7 +2244,8 @@ error_param: } /** - * i40e_validate_queue_map + * i40e_validate_queue_map - check queue map is valid + * @vf: the VF structure pointer * @vsi_id: vsi id * @queuemap: Tx or Rx queue map * @@ -3160,8 +3161,8 @@ err: /** * i40e_validate_cloud_filter - * @mask: mask for TC filter - * @data: data for TC filter + * @vf: pointer to VF structure + * @tc_filter: pointer to filter requested * * This function validates cloud filter programmed as TC filter for ADq **/ @@ -3294,7 +3295,7 @@ err: /** * i40e_find_vsi_from_seid - searches for the vsi with the given seid * @vf: pointer to the VF info - * @seid - seid of the vsi it is searching for + * @seid: seid of the vsi it is searching for **/ static struct i40e_vsi *i40e_find_vsi_from_seid(struct i40e_vf *vf, u16 seid) { diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.c b/drivers/net/ethernet/intel/i40e/i40e_xsk.c index 8ce57b507a21..6acede0acdca 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_xsk.c +++ b/drivers/net/ethernet/intel/i40e/i40e_xsk.c @@ -29,14 +29,16 @@ static struct xdp_buff **i40e_rx_bi(struct i40e_ring *rx_ring, u32 idx) } /** - * i40e_xsk_umem_enable - Enable/associate a UMEM to a certain ring/qid + * i40e_xsk_pool_enable - Enable/associate an AF_XDP buffer pool to a + * certain ring/qid * @vsi: Current VSI - * @umem: UMEM - * @qid: Rx ring to associate UMEM to + * @pool: buffer pool + * @qid: Rx ring to associate buffer pool with * * Returns 0 on success, <0 on failure **/ -static int i40e_xsk_umem_enable(struct i40e_vsi *vsi, struct xdp_umem *umem, +static int i40e_xsk_pool_enable(struct i40e_vsi *vsi, + struct xsk_buff_pool *pool, u16 qid) { struct net_device *netdev = vsi->netdev; @@ -53,7 +55,7 @@ static int i40e_xsk_umem_enable(struct i40e_vsi *vsi, struct xdp_umem *umem, qid >= netdev->real_num_tx_queues) return -EINVAL; - err = xsk_buff_dma_map(umem, &vsi->back->pdev->dev, I40E_RX_DMA_ATTR); + err = xsk_pool_dma_map(pool, &vsi->back->pdev->dev, I40E_RX_DMA_ATTR); if (err) return err; @@ -80,21 +82,22 @@ static int i40e_xsk_umem_enable(struct i40e_vsi *vsi, struct xdp_umem *umem, } /** - * i40e_xsk_umem_disable - Disassociate a UMEM from a certain ring/qid + * i40e_xsk_pool_disable - Disassociate an AF_XDP buffer pool from a + * certain ring/qid * @vsi: Current VSI - * @qid: Rx ring to associate UMEM to + * @qid: Rx ring to associate buffer pool with * * Returns 0 on success, <0 on failure **/ -static int i40e_xsk_umem_disable(struct i40e_vsi *vsi, u16 qid) +static int i40e_xsk_pool_disable(struct i40e_vsi *vsi, u16 qid) { struct net_device *netdev = vsi->netdev; - struct xdp_umem *umem; + struct xsk_buff_pool *pool; bool if_running; int err; - umem = xdp_get_umem_from_qid(netdev, qid); - if (!umem) + pool = xsk_get_pool_from_qid(netdev, qid); + if (!pool) return -EINVAL; if_running = netif_running(vsi->netdev) && i40e_enabled_xdp_vsi(vsi); @@ -106,7 +109,7 @@ static int i40e_xsk_umem_disable(struct i40e_vsi *vsi, u16 qid) } clear_bit(qid, vsi->af_xdp_zc_qps); - xsk_buff_dma_unmap(umem, I40E_RX_DMA_ATTR); + xsk_pool_dma_unmap(pool, I40E_RX_DMA_ATTR); if (if_running) { err = i40e_queue_pair_enable(vsi, qid); @@ -118,20 +121,21 @@ static int i40e_xsk_umem_disable(struct i40e_vsi *vsi, u16 qid) } /** - * i40e_xsk_umem_setup - Enable/disassociate a UMEM to/from a ring/qid + * i40e_xsk_pool_setup - Enable/disassociate an AF_XDP buffer pool to/from + * a ring/qid * @vsi: Current VSI - * @umem: UMEM to enable/associate to a ring, or NULL to disable - * @qid: Rx ring to (dis)associate UMEM (from)to + * @pool: Buffer pool to enable/associate to a ring, or NULL to disable + * @qid: Rx ring to (dis)associate buffer pool (from)to * - * This function enables or disables a UMEM to a certain ring. + * This function enables or disables a buffer pool to a certain ring. * * Returns 0 on success, <0 on failure **/ -int i40e_xsk_umem_setup(struct i40e_vsi *vsi, struct xdp_umem *umem, +int i40e_xsk_pool_setup(struct i40e_vsi *vsi, struct xsk_buff_pool *pool, u16 qid) { - return umem ? i40e_xsk_umem_enable(vsi, umem, qid) : - i40e_xsk_umem_disable(vsi, qid); + return pool ? i40e_xsk_pool_enable(vsi, pool, qid) : + i40e_xsk_pool_disable(vsi, qid); } /** @@ -191,7 +195,7 @@ bool i40e_alloc_rx_buffers_zc(struct i40e_ring *rx_ring, u16 count) rx_desc = I40E_RX_DESC(rx_ring, ntu); bi = i40e_rx_bi(rx_ring, ntu); do { - xdp = xsk_buff_alloc(rx_ring->xsk_umem); + xdp = xsk_buff_alloc(rx_ring->xsk_pool); if (!xdp) { ok = false; goto no_buffers; @@ -254,6 +258,18 @@ static struct sk_buff *i40e_construct_skb_zc(struct i40e_ring *rx_ring, } /** + * i40e_inc_ntc: Advance the next_to_clean index + * @rx_ring: Rx ring + **/ +static void i40e_inc_ntc(struct i40e_ring *rx_ring) +{ + u32 ntc = rx_ring->next_to_clean + 1; + + ntc = (ntc < rx_ring->count) ? ntc : 0; + rx_ring->next_to_clean = ntc; +} + +/** * i40e_clean_rx_irq_zc - Consumes Rx packets from the hardware ring * @rx_ring: Rx ring * @budget: NAPI budget @@ -265,8 +281,8 @@ int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget) unsigned int total_rx_bytes = 0, total_rx_packets = 0; u16 cleaned_count = I40E_DESC_UNUSED(rx_ring); unsigned int xdp_res, xdp_xmit = 0; - bool failure = false; struct sk_buff *skb; + bool failure; while (likely(total_rx_packets < (unsigned int)budget)) { union i40e_rx_desc *rx_desc; @@ -274,13 +290,6 @@ int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget) unsigned int size; u64 qword; - if (cleaned_count >= I40E_RX_BUFFER_WRITE) { - failure = failure || - !i40e_alloc_rx_buffers_zc(rx_ring, - cleaned_count); - cleaned_count = 0; - } - rx_desc = I40E_RX_DESC(rx_ring, rx_ring->next_to_clean); qword = le64_to_cpu(rx_desc->wb.qword1.status_error_len); @@ -310,7 +319,7 @@ int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget) bi = i40e_rx_bi(rx_ring, rx_ring->next_to_clean); (*bi)->data_end = (*bi)->data + size; - xsk_buff_dma_sync_for_cpu(*bi); + xsk_buff_dma_sync_for_cpu(*bi, rx_ring->xsk_pool); xdp_res = i40e_run_xdp_zc(rx_ring, *bi); if (xdp_res) { @@ -355,14 +364,17 @@ int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget) napi_gro_receive(&rx_ring->q_vector->napi, skb); } + if (cleaned_count >= I40E_RX_BUFFER_WRITE) + failure = !i40e_alloc_rx_buffers_zc(rx_ring, cleaned_count); + i40e_finalize_xdp_rx(rx_ring, xdp_xmit); i40e_update_rx_stats(rx_ring, total_rx_bytes, total_rx_packets); - if (xsk_umem_uses_need_wakeup(rx_ring->xsk_umem)) { + if (xsk_uses_need_wakeup(rx_ring->xsk_pool)) { if (failure || rx_ring->next_to_clean == rx_ring->next_to_use) - xsk_set_rx_need_wakeup(rx_ring->xsk_umem); + xsk_set_rx_need_wakeup(rx_ring->xsk_pool); else - xsk_clear_rx_need_wakeup(rx_ring->xsk_umem); + xsk_clear_rx_need_wakeup(rx_ring->xsk_pool); return (int)total_rx_packets; } @@ -385,11 +397,11 @@ static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget) dma_addr_t dma; while (budget-- > 0) { - if (!xsk_umem_consume_tx(xdp_ring->xsk_umem, &desc)) + if (!xsk_tx_peek_desc(xdp_ring->xsk_pool, &desc)) break; - dma = xsk_buff_raw_get_dma(xdp_ring->xsk_umem, desc.addr); - xsk_buff_raw_dma_sync_for_device(xdp_ring->xsk_umem, dma, + dma = xsk_buff_raw_get_dma(xdp_ring->xsk_pool, desc.addr); + xsk_buff_raw_dma_sync_for_device(xdp_ring->xsk_pool, dma, desc.len); tx_bi = &xdp_ring->tx_bi[xdp_ring->next_to_use]; @@ -416,7 +428,7 @@ static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget) I40E_TXD_QW1_CMD_SHIFT); i40e_xdp_ring_update_tail(xdp_ring); - xsk_umem_consume_tx_done(xdp_ring->xsk_umem); + xsk_tx_release(xdp_ring->xsk_pool); i40e_update_tx_stats(xdp_ring, sent_frames, total_bytes); } @@ -448,7 +460,7 @@ static void i40e_clean_xdp_tx_buffer(struct i40e_ring *tx_ring, **/ bool i40e_clean_xdp_tx_irq(struct i40e_vsi *vsi, struct i40e_ring *tx_ring) { - struct xdp_umem *umem = tx_ring->xsk_umem; + struct xsk_buff_pool *bp = tx_ring->xsk_pool; u32 i, completed_frames, xsk_frames = 0; u32 head_idx = i40e_get_head(tx_ring); struct i40e_tx_buffer *tx_bi; @@ -488,13 +500,13 @@ skip: tx_ring->next_to_clean -= tx_ring->count; if (xsk_frames) - xsk_umem_complete_tx(umem, xsk_frames); + xsk_tx_completed(bp, xsk_frames); i40e_arm_wb(tx_ring, vsi, completed_frames); out_xmit: - if (xsk_umem_uses_need_wakeup(tx_ring->xsk_umem)) - xsk_set_tx_need_wakeup(tx_ring->xsk_umem); + if (xsk_uses_need_wakeup(tx_ring->xsk_pool)) + xsk_set_tx_need_wakeup(tx_ring->xsk_pool); return i40e_xmit_zc(tx_ring, I40E_DESC_UNUSED(tx_ring)); } @@ -526,7 +538,7 @@ int i40e_xsk_wakeup(struct net_device *dev, u32 queue_id, u32 flags) if (queue_id >= vsi->num_queue_pairs) return -ENXIO; - if (!vsi->xdp_rings[queue_id]->xsk_umem) + if (!vsi->xdp_rings[queue_id]->xsk_pool) return -ENXIO; ring = vsi->xdp_rings[queue_id]; @@ -565,7 +577,7 @@ void i40e_xsk_clean_rx_ring(struct i40e_ring *rx_ring) void i40e_xsk_clean_tx_ring(struct i40e_ring *tx_ring) { u16 ntc = tx_ring->next_to_clean, ntu = tx_ring->next_to_use; - struct xdp_umem *umem = tx_ring->xsk_umem; + struct xsk_buff_pool *bp = tx_ring->xsk_pool; struct i40e_tx_buffer *tx_bi; u32 xsk_frames = 0; @@ -585,14 +597,15 @@ void i40e_xsk_clean_tx_ring(struct i40e_ring *tx_ring) } if (xsk_frames) - xsk_umem_complete_tx(umem, xsk_frames); + xsk_tx_completed(bp, xsk_frames); } /** - * i40e_xsk_any_rx_ring_enabled - Checks if Rx rings have AF_XDP UMEM attached + * i40e_xsk_any_rx_ring_enabled - Checks if Rx rings have an AF_XDP + * buffer pool attached * @vsi: vsi * - * Returns true if any of the Rx rings has an AF_XDP UMEM attached + * Returns true if any of the Rx rings has an AF_XDP buffer pool attached **/ bool i40e_xsk_any_rx_ring_enabled(struct i40e_vsi *vsi) { @@ -600,7 +613,7 @@ bool i40e_xsk_any_rx_ring_enabled(struct i40e_vsi *vsi) int i; for (i = 0; i < vsi->num_queue_pairs; i++) { - if (xdp_get_umem_from_qid(netdev, i)) + if (xsk_get_pool_from_qid(netdev, i)) return true; } diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.h b/drivers/net/ethernet/intel/i40e/i40e_xsk.h index c524c142127f..7adfd8539247 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_xsk.h +++ b/drivers/net/ethernet/intel/i40e/i40e_xsk.h @@ -5,12 +5,12 @@ #define _I40E_XSK_H_ struct i40e_vsi; -struct xdp_umem; +struct xsk_buff_pool; struct zero_copy_allocator; int i40e_queue_pair_disable(struct i40e_vsi *vsi, int queue_pair); int i40e_queue_pair_enable(struct i40e_vsi *vsi, int queue_pair); -int i40e_xsk_umem_setup(struct i40e_vsi *vsi, struct xdp_umem *umem, +int i40e_xsk_pool_setup(struct i40e_vsi *vsi, struct xsk_buff_pool *pool, u16 qid); bool i40e_alloc_rx_buffers_zc(struct i40e_ring *rx_ring, u16 cleaned_count); int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget); diff --git a/drivers/net/ethernet/intel/iavf/iavf_adminq.h b/drivers/net/ethernet/intel/iavf/iavf_adminq.h index baf2fe26f302..1f60518eb0e5 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_adminq.h +++ b/drivers/net/ethernet/intel/iavf/iavf_adminq.h @@ -85,8 +85,8 @@ struct iavf_adminq_info { /** * iavf_aq_rc_to_posix - convert errors to user-land codes - * aq_ret: AdminQ handler error code can override aq_rc - * aq_rc: AdminQ firmware error code to convert + * @aq_ret: AdminQ handler error code can override aq_rc + * @aq_rc: AdminQ firmware error code to convert **/ static inline int iavf_aq_rc_to_posix(int aq_ret, int aq_rc) { diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c index cf539db79af9..95543dfd4fe7 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_main.c +++ b/drivers/net/ethernet/intel/iavf/iavf_main.c @@ -147,6 +147,7 @@ void iavf_schedule_reset(struct iavf_adapter *adapter) /** * iavf_tx_timeout - Respond to a Tx Hang * @netdev: network interface device structure + * @txqueue: queue number that is timing out **/ static void iavf_tx_timeout(struct net_device *netdev, unsigned int txqueue) { @@ -2572,8 +2573,8 @@ static int iavf_validate_ch_config(struct iavf_adapter *adapter, } /** - * iavf_del_all_cloud_filters - delete all cloud filters - * on the traffic classes + * iavf_del_all_cloud_filters - delete all cloud filters on the traffic classes + * @adapter: board private structure **/ static void iavf_del_all_cloud_filters(struct iavf_adapter *adapter) { @@ -2592,7 +2593,7 @@ static void iavf_del_all_cloud_filters(struct iavf_adapter *adapter) /** * __iavf_setup_tc - configure multiple traffic classes * @netdev: network interface device structure - * @type_date: tc offload data + * @type_data: tc offload data * * This function processes the config information provided by the * user to configure traffic classes/queue channels and packages the @@ -2690,7 +2691,7 @@ exit: /** * iavf_parse_cls_flower - Parse tc flower filters provided by kernel * @adapter: board private structure - * @cls_flower: pointer to struct flow_cls_offload + * @f: pointer to struct flow_cls_offload * @filter: pointer to cloud filter structure */ static int iavf_parse_cls_flower(struct iavf_adapter *adapter, @@ -3064,8 +3065,8 @@ static int iavf_delete_clsflower(struct iavf_adapter *adapter, /** * iavf_setup_tc_cls_flower - flower classifier offloads - * @netdev: net device to configure - * @type_data: offload data + * @adapter: board private structure + * @cls_flower: pointer to flow_cls_offload struct with flow info */ static int iavf_setup_tc_cls_flower(struct iavf_adapter *adapter, struct flow_cls_offload *cls_flower) @@ -3112,7 +3113,7 @@ static LIST_HEAD(iavf_block_cb_list); * iavf_setup_tc - configure multiple traffic classes * @netdev: network interface device structure * @type: type of offload - * @type_date: tc offload data + * @type_data: tc offload data * * This function is the callback to ndo_setup_tc in the * netdev_ops. @@ -3768,8 +3769,7 @@ err_dma: /** * iavf_suspend - Power management suspend routine - * @pdev: PCI device information struct - * @state: unused + * @dev_d: device info pointer * * Called when the system (VM) is entering sleep/suspend. **/ @@ -3799,7 +3799,7 @@ static int __maybe_unused iavf_suspend(struct device *dev_d) /** * iavf_resume - Power management resume routine - * @pdev: PCI device information struct + * @dev_d: device info pointer * * Called when the system (VM) is resumed from sleep/suspend. **/ diff --git a/drivers/net/ethernet/intel/iavf/iavf_trace.h b/drivers/net/ethernet/intel/iavf/iavf_trace.h index 1058e68a02b4..82fda6f5abf0 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_trace.h +++ b/drivers/net/ethernet/intel/iavf/iavf_trace.h @@ -22,7 +22,7 @@ #include <linux/tracepoint.h> -/** +/* * iavf_trace() macro enables shared code to refer to trace points * like: * diff --git a/drivers/net/ethernet/intel/iavf/iavf_txrx.c b/drivers/net/ethernet/intel/iavf/iavf_txrx.c index ca041b39ffda..256fa07d54d5 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_txrx.c +++ b/drivers/net/ethernet/intel/iavf/iavf_txrx.c @@ -1309,10 +1309,7 @@ static struct sk_buff *iavf_construct_skb(struct iavf_ring *rx_ring, return NULL; /* prefetch first cache line of first page */ va = page_address(rx_buffer->page) + rx_buffer->page_offset; - prefetch(va); -#if L1_CACHE_BYTES < 128 - prefetch(va + L1_CACHE_BYTES); -#endif + net_prefetch(va); /* allocate a skb to store the frags */ skb = __napi_alloc_skb(&rx_ring->q_vector->napi, @@ -1376,10 +1373,8 @@ static struct sk_buff *iavf_build_skb(struct iavf_ring *rx_ring, return NULL; /* prefetch first cache line of first page */ va = page_address(rx_buffer->page) + rx_buffer->page_offset; - prefetch(va); -#if L1_CACHE_BYTES < 128 - prefetch(va + L1_CACHE_BYTES); -#endif + net_prefetch(va); + /* build an skb around the page buffer */ skb = build_skb(va - IAVF_SKB_PAD, truesize); if (unlikely(!skb)) diff --git a/drivers/net/ethernet/intel/iavf/iavf_txrx.h b/drivers/net/ethernet/intel/iavf/iavf_txrx.h index dd3348f9da9d..e5b9ba42dd00 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_txrx.h +++ b/drivers/net/ethernet/intel/iavf/iavf_txrx.h @@ -454,7 +454,6 @@ bool __iavf_chk_linearize(struct sk_buff *skb); /** * iavf_xmit_descriptor_count - calculate number of Tx descriptors needed * @skb: send buffer - * @tx_ring: ring to send buffer on * * Returns number of data descriptors needed for this skb. Returns 0 to indicate * there is not enough descriptors available in this ring since we need at least @@ -514,6 +513,7 @@ static inline bool iavf_chk_linearize(struct sk_buff *skb, int count) return count != IAVF_MAX_BUFFER_TXD; } /** + * txring_txq - helper to convert from a ring to a queue * @ring: Tx ring to find the netdev equivalent of **/ static inline struct netdev_queue *txring_txq(const struct iavf_ring *ring) diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index fe140ff38f74..a0723831c4e4 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -284,6 +284,10 @@ struct ice_vsi { spinlock_t arfs_lock; /* protects aRFS hash table and filter state */ atomic_t *arfs_last_fltr_id; + /* devlink port data */ + struct devlink_port devlink_port; + bool devlink_port_registered; + u16 max_frame; u16 rx_buf_len; @@ -321,9 +325,9 @@ struct ice_vsi { struct ice_ring **xdp_rings; /* XDP ring array */ u16 num_xdp_txq; /* Used XDP queues */ u8 xdp_mapping_mode; /* ICE_MAP_MODE_[CONTIG|SCATTER] */ - struct xdp_umem **xsk_umems; - u16 num_xsk_umems_used; - u16 num_xsk_umems; + struct xsk_buff_pool **xsk_pools; + u16 num_xsk_pools_used; + u16 num_xsk_pools; } ____cacheline_internodealigned_in_smp; /* struct that defines an interrupt vector */ @@ -375,9 +379,6 @@ enum ice_pf_flags { struct ice_pf { struct pci_dev *pdev; - /* devlink port data */ - struct devlink_port devlink_port; - struct devlink_region *nvm_region; struct devlink_region *devcaps_region; @@ -507,25 +508,25 @@ static inline void ice_set_ring_xdp(struct ice_ring *ring) } /** - * ice_xsk_umem - get XDP UMEM bound to a ring - * @ring - ring to use + * ice_xsk_pool - get XSK buffer pool bound to a ring + * @ring: ring to use * - * Returns a pointer to xdp_umem structure if there is an UMEM present, + * Returns a pointer to xdp_umem structure if there is a buffer pool present, * NULL otherwise. */ -static inline struct xdp_umem *ice_xsk_umem(struct ice_ring *ring) +static inline struct xsk_buff_pool *ice_xsk_pool(struct ice_ring *ring) { - struct xdp_umem **umems = ring->vsi->xsk_umems; + struct xsk_buff_pool **pools = ring->vsi->xsk_pools; u16 qid = ring->q_index; if (ice_ring_is_xdp(ring)) qid -= ring->vsi->num_xdp_txq; - if (qid >= ring->vsi->num_xsk_umems || !umems || !umems[qid] || + if (qid >= ring->vsi->num_xsk_pools || !pools || !pools[qid] || !ice_is_xdp_ena_vsi(ring->vsi)) return NULL; - return umems[qid]; + return pools[qid]; } /** diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h index ba9375218fef..b06fbe99d8e9 100644 --- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h +++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h @@ -1422,7 +1422,7 @@ struct ice_aqc_nvm_comp_tbl { u8 cvs[]; /* Component Version String */ } __packed; -/** +/* * Send to PF command (indirect 0x0801) ID is only used by PF * * Send to VF command (indirect 0x0802) ID is only used by PF @@ -1826,8 +1826,8 @@ struct ice_aqc_event_lan_overflow { * @opcode: AQ command opcode * @datalen: length in bytes of indirect/external data buffer * @retval: return value from firmware - * @cookie_h: opaque data high-half - * @cookie_l: opaque data low-half + * @cookie_high: opaque data high-half + * @cookie_low: opaque data low-half * @params: command-specific parameters * * Descriptor format for commands the driver posts on the Admin Transmit Queue diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c index 87008476d8fe..fe4320e2d1f2 100644 --- a/drivers/net/ethernet/intel/ice/ice_base.c +++ b/drivers/net/ethernet/intel/ice/ice_base.c @@ -308,12 +308,12 @@ int ice_setup_rx_ctx(struct ice_ring *ring) xdp_rxq_info_reg(&ring->xdp_rxq, ring->netdev, ring->q_index); - ring->xsk_umem = ice_xsk_umem(ring); - if (ring->xsk_umem) { + ring->xsk_pool = ice_xsk_pool(ring); + if (ring->xsk_pool) { xdp_rxq_info_unreg_mem_model(&ring->xdp_rxq); ring->rx_buf_len = - xsk_umem_get_rx_frame_size(ring->xsk_umem); + xsk_pool_get_rx_frame_size(ring->xsk_pool); /* For AF_XDP ZC, we disallow packets to span on * multiple buffers, thus letting us skip that * handling in the fast-path. @@ -324,7 +324,7 @@ int ice_setup_rx_ctx(struct ice_ring *ring) NULL); if (err) return err; - xsk_buff_set_rxq_info(ring->xsk_umem, &ring->xdp_rxq); + xsk_pool_set_rxq_info(ring->xsk_pool, &ring->xdp_rxq); dev_info(dev, "Registered XDP mem model MEM_TYPE_XSK_BUFF_POOL on Rx ring %d\n", ring->q_index); @@ -417,9 +417,9 @@ int ice_setup_rx_ctx(struct ice_ring *ring) ring->tail = hw->hw_addr + QRX_TAIL(pf_q); writel(0, ring->tail); - if (ring->xsk_umem) { - if (!xsk_buff_can_alloc(ring->xsk_umem, num_bufs)) { - dev_warn(dev, "UMEM does not provide enough addresses to fill %d buffers on Rx ring %d\n", + if (ring->xsk_pool) { + if (!xsk_buff_can_alloc(ring->xsk_pool, num_bufs)) { + dev_warn(dev, "XSK buffer pool does not provide enough addresses to fill %d buffers on Rx ring %d\n", num_bufs, ring->q_index); dev_warn(dev, "Change Rx ring/fill queue size to avoid performance issues\n"); @@ -428,7 +428,7 @@ int ice_setup_rx_ctx(struct ice_ring *ring) err = ice_alloc_rx_bufs_zc(ring, num_bufs); if (err) - dev_info(dev, "Failed to allocate some buffers on UMEM enabled Rx ring %d (pf_q %d)\n", + dev_info(dev, "Failed to allocate some buffers on XSK buffer pool enabled Rx ring %d (pf_q %d)\n", ring->q_index, pf_q); return 0; } diff --git a/drivers/net/ethernet/intel/ice/ice_devlink.c b/drivers/net/ethernet/intel/ice/ice_devlink.c index 111d6bfe4222..511da59bd6f2 100644 --- a/drivers/net/ethernet/intel/ice/ice_devlink.c +++ b/drivers/net/ethernet/intel/ice/ice_devlink.c @@ -6,18 +6,14 @@ #include "ice_devlink.h" #include "ice_fw_update.h" -static int ice_info_get_dsn(struct ice_pf *pf, char *buf, size_t len) +static void ice_info_get_dsn(struct ice_pf *pf, char *buf, size_t len) { u8 dsn[8]; /* Copy the DSN into an array in Big Endian format */ put_unaligned_be64(pci_get_dsn(pf->pdev), dsn); - snprintf(buf, len, "%02x-%02x-%02x-%02x-%02x-%02x-%02x-%02x", - dsn[0], dsn[1], dsn[2], dsn[3], - dsn[4], dsn[5], dsn[6], dsn[7]); - - return 0; + snprintf(buf, len, "%8phD", dsn); } static int ice_info_pba(struct ice_pf *pf, char *buf, size_t len) @@ -106,6 +102,13 @@ static int ice_info_ddp_pkg_version(struct ice_pf *pf, char *buf, size_t len) return 0; } +static int ice_info_ddp_pkg_bundle_id(struct ice_pf *pf, char *buf, size_t len) +{ + snprintf(buf, len, "0x%08x", pf->hw.active_track_id); + + return 0; +} + static int ice_info_netlist_ver(struct ice_pf *pf, char *buf, size_t len) { struct ice_netlist_ver_info *netlist = &pf->hw.netlist_ver; @@ -150,6 +153,7 @@ static const struct ice_devlink_version { running(DEVLINK_INFO_VERSION_GENERIC_FW_BUNDLE_ID, ice_info_eetrack), running("fw.app.name", ice_info_ddp_pkg_name), running(DEVLINK_INFO_VERSION_GENERIC_FW_APP, ice_info_ddp_pkg_version), + running("fw.app.bundle_id", ice_info_ddp_pkg_bundle_id), running("fw.netlist", ice_info_netlist_ver), running("fw.netlist.build", ice_info_netlist_build), }; @@ -180,11 +184,7 @@ static int ice_devlink_info_get(struct devlink *devlink, return err; } - err = ice_info_get_dsn(pf, buf, sizeof(buf)); - if (err) { - NL_SET_ERR_MSG_MOD(extack, "Unable to obtain serial number"); - return err; - } + ice_info_get_dsn(pf, buf, sizeof(buf)); err = devlink_info_serial_number_put(req, buf); if (err) { @@ -233,8 +233,7 @@ static int ice_devlink_info_get(struct devlink *devlink, /** * ice_devlink_flash_update - Update firmware stored in flash on the device * @devlink: pointer to devlink associated with device to update - * @path: the path of the firmware file to use via request_firmware - * @component: name of the component to update, or NULL + * @params: flash update parameters * @extack: netlink extended ACK structure * * Perform a device flash update. The bulk of the update logic is contained @@ -243,38 +242,52 @@ static int ice_devlink_info_get(struct devlink *devlink, * Returns: zero on success, or an error code on failure. */ static int -ice_devlink_flash_update(struct devlink *devlink, const char *path, - const char *component, struct netlink_ext_ack *extack) +ice_devlink_flash_update(struct devlink *devlink, + struct devlink_flash_update_params *params, + struct netlink_ext_ack *extack) { struct ice_pf *pf = devlink_priv(devlink); struct device *dev = &pf->pdev->dev; struct ice_hw *hw = &pf->hw; const struct firmware *fw; + u8 preservation; int err; - /* individual component update is not yet supported */ - if (component) + if (!params->overwrite_mask) { + /* preserve all settings and identifiers */ + preservation = ICE_AQC_NVM_PRESERVE_ALL; + } else if (params->overwrite_mask == DEVLINK_FLASH_OVERWRITE_SETTINGS) { + /* overwrite settings, but preserve the vital device identifiers */ + preservation = ICE_AQC_NVM_PRESERVE_SELECTED; + } else if (params->overwrite_mask == (DEVLINK_FLASH_OVERWRITE_SETTINGS | + DEVLINK_FLASH_OVERWRITE_IDENTIFIERS)) { + /* overwrite both settings and identifiers, preserve nothing */ + preservation = ICE_AQC_NVM_NO_PRESERVATION; + } else { + NL_SET_ERR_MSG_MOD(extack, "Requested overwrite mask is not supported"); return -EOPNOTSUPP; + } if (!hw->dev_caps.common_cap.nvm_unified_update) { NL_SET_ERR_MSG_MOD(extack, "Current firmware does not support unified update"); return -EOPNOTSUPP; } - err = ice_check_for_pending_update(pf, component, extack); + err = ice_check_for_pending_update(pf, NULL, extack); if (err) return err; - err = request_firmware(&fw, path, dev); + err = request_firmware(&fw, params->file_name, dev); if (err) { NL_SET_ERR_MSG_MOD(extack, "Unable to read file from disk"); return err; } + dev_dbg(dev, "Beginning flash update with file '%s'\n", params->file_name); + devlink_flash_update_begin_notify(devlink); - devlink_flash_update_status_notify(devlink, "Preparing to flash", - component, 0, 0); - err = ice_flash_pldm_image(pf, fw, extack); + devlink_flash_update_status_notify(devlink, "Preparing to flash", NULL, 0, 0); + err = ice_flash_pldm_image(pf, fw, preservation, extack); devlink_flash_update_end_notify(devlink); release_firmware(fw); @@ -283,6 +296,7 @@ ice_devlink_flash_update(struct devlink *devlink, const char *path, } static const struct devlink_ops ice_devlink_ops = { + .supported_flash_update_params = DEVLINK_SUPPORT_FLASH_UPDATE_OVERWRITE_MASK, .info_get = ice_devlink_info_get, .flash_update = ice_devlink_flash_update, }; @@ -352,55 +366,66 @@ void ice_devlink_unregister(struct ice_pf *pf) } /** - * ice_devlink_create_port - Create a devlink port for this PF - * @pf: the PF to create a port for + * ice_devlink_create_port - Create a devlink port for this VSI + * @vsi: the VSI to create a port for * - * Create and register a devlink_port for this PF. Note that although each - * physical function is connected to a separate devlink instance, the port - * will still be numbered according to the physical function ID. + * Create and register a devlink_port for this VSI. * * Return: zero on success or an error code on failure. */ -int ice_devlink_create_port(struct ice_pf *pf) +int ice_devlink_create_port(struct ice_vsi *vsi) { - struct devlink *devlink = priv_to_devlink(pf); - struct ice_vsi *vsi = ice_get_main_vsi(pf); - struct device *dev = ice_pf_to_dev(pf); struct devlink_port_attrs attrs = {}; + struct ice_port_info *pi; + struct devlink *devlink; + struct device *dev; + struct ice_pf *pf; int err; - if (!vsi) { - dev_err(dev, "%s: unable to find main VSI\n", __func__); - return -EIO; - } + /* Currently we only create devlink_port instances for PF VSIs */ + if (vsi->type != ICE_VSI_PF) + return -EINVAL; + + pf = vsi->back; + devlink = priv_to_devlink(pf); + dev = ice_pf_to_dev(pf); + pi = pf->hw.port_info; attrs.flavour = DEVLINK_PORT_FLAVOUR_PHYSICAL; - attrs.phys.port_number = pf->hw.pf_id; - devlink_port_attrs_set(&pf->devlink_port, &attrs); - err = devlink_port_register(devlink, &pf->devlink_port, pf->hw.pf_id); + attrs.phys.port_number = pi->lport; + devlink_port_attrs_set(&vsi->devlink_port, &attrs); + err = devlink_port_register(devlink, &vsi->devlink_port, vsi->idx); if (err) { dev_err(dev, "devlink_port_register failed: %d\n", err); return err; } + vsi->devlink_port_registered = true; + return 0; } /** - * ice_devlink_destroy_port - Destroy the devlink_port for this PF - * @pf: the PF to cleanup + * ice_devlink_destroy_port - Destroy the devlink_port for this VSI + * @vsi: the VSI to cleanup * - * Unregisters the devlink_port structure associated with this PF. + * Unregisters the devlink_port structure associated with this VSI. */ -void ice_devlink_destroy_port(struct ice_pf *pf) +void ice_devlink_destroy_port(struct ice_vsi *vsi) { - devlink_port_type_clear(&pf->devlink_port); - devlink_port_unregister(&pf->devlink_port); + if (!vsi->devlink_port_registered) + return; + + devlink_port_type_clear(&vsi->devlink_port); + devlink_port_unregister(&vsi->devlink_port); + + vsi->devlink_port_registered = false; } /** * ice_devlink_nvm_snapshot - Capture a snapshot of the Shadow RAM contents * @devlink: the devlink instance + * @ops: the devlink region being snapshotted * @extack: extended ACK response structure * @data: on exit points to snapshot data buffer * @@ -413,6 +438,7 @@ void ice_devlink_destroy_port(struct ice_pf *pf) * error code on failure. */ static int ice_devlink_nvm_snapshot(struct devlink *devlink, + const struct devlink_region_ops *ops, struct netlink_ext_ack *extack, u8 **data) { struct ice_pf *pf = devlink_priv(devlink); @@ -456,6 +482,7 @@ static int ice_devlink_nvm_snapshot(struct devlink *devlink, /** * ice_devlink_devcaps_snapshot - Capture snapshot of device capabilities * @devlink: the devlink instance + * @ops: the devlink region being snapshotted * @extack: extended ACK response structure * @data: on exit points to snapshot data buffer * @@ -468,6 +495,7 @@ static int ice_devlink_nvm_snapshot(struct devlink *devlink, */ static int ice_devlink_devcaps_snapshot(struct devlink *devlink, + const struct devlink_region_ops *ops, struct netlink_ext_ack *extack, u8 **data) { struct ice_pf *pf = devlink_priv(devlink); diff --git a/drivers/net/ethernet/intel/ice/ice_devlink.h b/drivers/net/ethernet/intel/ice/ice_devlink.h index 6e806a08dc23..e07e74426bde 100644 --- a/drivers/net/ethernet/intel/ice/ice_devlink.h +++ b/drivers/net/ethernet/intel/ice/ice_devlink.h @@ -8,8 +8,8 @@ struct ice_pf *ice_allocate_pf(struct device *dev); int ice_devlink_register(struct ice_pf *pf); void ice_devlink_unregister(struct ice_pf *pf); -int ice_devlink_create_port(struct ice_pf *pf); -void ice_devlink_destroy_port(struct ice_pf *pf); +int ice_devlink_create_port(struct ice_vsi *vsi); +void ice_devlink_destroy_port(struct ice_vsi *vsi); void ice_devlink_init_regions(struct ice_pf *pf); void ice_devlink_destroy_regions(struct ice_pf *pf); diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c b/drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c index d7430ce6af26..2d27f66ac853 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c @@ -1268,8 +1268,7 @@ ice_fdir_write_all_fltr(struct ice_pf *pf, struct ice_fdir_fltr *input, bool is_tun = tun == ICE_FD_HW_SEG_TUN; int err; - if (is_tun && !ice_get_open_tunnel_port(&pf->hw, TNL_ALL, - &port_num)) + if (is_tun && !ice_get_open_tunnel_port(&pf->hw, &port_num)) continue; err = ice_fdir_write_fltr(pf, input, add, is_tun); if (err) @@ -1647,8 +1646,7 @@ int ice_add_fdir_ethtool(struct ice_vsi *vsi, struct ethtool_rxnfc *cmd) } /* return error if not an update and no available filters */ - fltrs_needed = ice_get_open_tunnel_port(hw, TNL_ALL, &tunnel_port) ? - 2 : 1; + fltrs_needed = ice_get_open_tunnel_port(hw, &tunnel_port) ? 2 : 1; if (!ice_fdir_find_fltr_by_idx(hw, fsp->location) && ice_fdir_num_avail_fltr(hw, pf->vsi[vsi->idx]) < fltrs_needed) { dev_err(dev, "Failed to add filter. The maximum number of flow director filters has been reached.\n"); diff --git a/drivers/net/ethernet/intel/ice/ice_fdir.c b/drivers/net/ethernet/intel/ice/ice_fdir.c index 6834df14332f..59c0c6a0f8c5 100644 --- a/drivers/net/ethernet/intel/ice/ice_fdir.c +++ b/drivers/net/ethernet/intel/ice/ice_fdir.c @@ -556,7 +556,7 @@ ice_fdir_get_gen_prgm_pkt(struct ice_hw *hw, struct ice_fdir_fltr *input, memcpy(pkt, ice_fdir_pkt[idx].pkt, ice_fdir_pkt[idx].pkt_len); loc = pkt; } else { - if (!ice_get_open_tunnel_port(hw, TNL_ALL, &tnl_port)) + if (!ice_get_open_tunnel_port(hw, &tnl_port)) return ICE_ERR_DOES_NOT_EXIST; if (!ice_fdir_pkt[idx].tun_pkt) return ICE_ERR_PARAM; diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c index b17ae3e20157..9095b4d274ad 100644 --- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c +++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c @@ -489,8 +489,6 @@ static void ice_init_pkg_hints(struct ice_hw *hw, struct ice_seg *ice_seg) if ((label_name[len] - '0') == hw->pf_id) { hw->tnl.tbl[hw->tnl.count].type = tnls[i].type; hw->tnl.tbl[hw->tnl.count].valid = false; - hw->tnl.tbl[hw->tnl.count].in_use = false; - hw->tnl.tbl[hw->tnl.count].marked = false; hw->tnl.tbl[hw->tnl.count].boost_addr = val; hw->tnl.tbl[hw->tnl.count].port = 0; hw->tnl.count++; @@ -505,8 +503,11 @@ static void ice_init_pkg_hints(struct ice_hw *hw, struct ice_seg *ice_seg) for (i = 0; i < hw->tnl.count; i++) { ice_find_boost_entry(ice_seg, hw->tnl.tbl[i].boost_addr, &hw->tnl.tbl[i].boost_entry); - if (hw->tnl.tbl[i].boost_entry) + if (hw->tnl.tbl[i].boost_entry) { hw->tnl.tbl[i].valid = true; + if (hw->tnl.tbl[i].type < __TNL_TYPE_CNT) + hw->tnl.valid_count[hw->tnl.tbl[i].type]++; + } } } @@ -1626,104 +1627,59 @@ static struct ice_buf *ice_pkg_buf(struct ice_buf_build *bld) } /** - * ice_tunnel_port_in_use_hlpr - helper function to determine tunnel usage + * ice_get_open_tunnel_port - retrieve an open tunnel port * @hw: pointer to the HW structure - * @port: port to search for - * @index: optionally returns index - * - * Returns whether a port is already in use as a tunnel, and optionally its - * index + * @port: returns open port */ -static bool ice_tunnel_port_in_use_hlpr(struct ice_hw *hw, u16 port, u16 *index) +bool +ice_get_open_tunnel_port(struct ice_hw *hw, u16 *port) { + bool res = false; u16 i; + mutex_lock(&hw->tnl_lock); + for (i = 0; i < hw->tnl.count && i < ICE_TUNNEL_MAX_ENTRIES; i++) - if (hw->tnl.tbl[i].in_use && hw->tnl.tbl[i].port == port) { - if (index) - *index = i; - return true; + if (hw->tnl.tbl[i].valid && hw->tnl.tbl[i].port) { + *port = hw->tnl.tbl[i].port; + res = true; + break; } - return false; -} - -/** - * ice_tunnel_port_in_use - * @hw: pointer to the HW structure - * @port: port to search for - * @index: optionally returns index - * - * Returns whether a port is already in use as a tunnel, and optionally its - * index - */ -bool ice_tunnel_port_in_use(struct ice_hw *hw, u16 port, u16 *index) -{ - bool res; - - mutex_lock(&hw->tnl_lock); - res = ice_tunnel_port_in_use_hlpr(hw, port, index); mutex_unlock(&hw->tnl_lock); return res; } /** - * ice_find_free_tunnel_entry + * ice_tunnel_idx_to_entry - convert linear index to the sparse one * @hw: pointer to the HW structure - * @type: tunnel type - * @index: optionally returns index + * @type: type of tunnel + * @idx: linear index * - * Returns whether there is a free tunnel entry, and optionally its index + * Stack assumes we have 2 linear tables with indexes [0, count_valid), + * but really the port table may be sprase, and types are mixed, so convert + * the stack index into the device index. */ -static bool -ice_find_free_tunnel_entry(struct ice_hw *hw, enum ice_tunnel_type type, - u16 *index) +static u16 ice_tunnel_idx_to_entry(struct ice_hw *hw, enum ice_tunnel_type type, + u16 idx) { u16 i; for (i = 0; i < hw->tnl.count && i < ICE_TUNNEL_MAX_ENTRIES; i++) - if (hw->tnl.tbl[i].valid && !hw->tnl.tbl[i].in_use && - hw->tnl.tbl[i].type == type) { - if (index) - *index = i; - return true; - } + if (hw->tnl.tbl[i].valid && + hw->tnl.tbl[i].type == type && + idx--) + return i; - return false; -} - -/** - * ice_get_open_tunnel_port - retrieve an open tunnel port - * @hw: pointer to the HW structure - * @type: tunnel type (TNL_ALL will return any open port) - * @port: returns open port - */ -bool -ice_get_open_tunnel_port(struct ice_hw *hw, enum ice_tunnel_type type, - u16 *port) -{ - bool res = false; - u16 i; - - mutex_lock(&hw->tnl_lock); - - for (i = 0; i < hw->tnl.count && i < ICE_TUNNEL_MAX_ENTRIES; i++) - if (hw->tnl.tbl[i].valid && hw->tnl.tbl[i].in_use && - (type == TNL_ALL || hw->tnl.tbl[i].type == type)) { - *port = hw->tnl.tbl[i].port; - res = true; - break; - } - - mutex_unlock(&hw->tnl_lock); - - return res; + WARN_ON_ONCE(1); + return 0; } /** * ice_create_tunnel * @hw: pointer to the HW structure + * @index: device table entry * @type: type of tunnel * @port: port of tunnel to create * @@ -1731,27 +1687,16 @@ ice_get_open_tunnel_port(struct ice_hw *hw, enum ice_tunnel_type type, * creating a package buffer with the tunnel info and issuing an update package * command. */ -enum ice_status -ice_create_tunnel(struct ice_hw *hw, enum ice_tunnel_type type, u16 port) +static enum ice_status +ice_create_tunnel(struct ice_hw *hw, u16 index, + enum ice_tunnel_type type, u16 port) { struct ice_boost_tcam_section *sect_rx, *sect_tx; enum ice_status status = ICE_ERR_MAX_LIMIT; struct ice_buf_build *bld; - u16 index; mutex_lock(&hw->tnl_lock); - if (ice_tunnel_port_in_use_hlpr(hw, port, &index)) { - hw->tnl.tbl[index].ref++; - status = 0; - goto ice_create_tunnel_end; - } - - if (!ice_find_free_tunnel_entry(hw, type, &index)) { - status = ICE_ERR_OUT_OF_RANGE; - goto ice_create_tunnel_end; - } - bld = ice_pkg_buf_alloc(hw); if (!bld) { status = ICE_ERR_NO_MEMORY; @@ -1790,11 +1735,8 @@ ice_create_tunnel(struct ice_hw *hw, enum ice_tunnel_type type, u16 port) memcpy(sect_tx->tcam, sect_rx->tcam, sizeof(*sect_tx->tcam)); status = ice_update_pkg(hw, ice_pkg_buf(bld), 1); - if (!status) { + if (!status) hw->tnl.tbl[index].port = port; - hw->tnl.tbl[index].in_use = true; - hw->tnl.tbl[index].ref = 1; - } ice_create_tunnel_err: ice_pkg_buf_free(hw, bld); @@ -1808,46 +1750,31 @@ ice_create_tunnel_end: /** * ice_destroy_tunnel * @hw: pointer to the HW structure + * @index: device table entry + * @type: type of tunnel * @port: port of tunnel to destroy (ignored if the all parameter is true) - * @all: flag that states to destroy all tunnels * * Destroys a tunnel or all tunnels by creating an update package buffer * targeting the specific updates requested and then performing an update * package. */ -enum ice_status ice_destroy_tunnel(struct ice_hw *hw, u16 port, bool all) +static enum ice_status +ice_destroy_tunnel(struct ice_hw *hw, u16 index, enum ice_tunnel_type type, + u16 port) { struct ice_boost_tcam_section *sect_rx, *sect_tx; enum ice_status status = ICE_ERR_MAX_LIMIT; struct ice_buf_build *bld; - u16 count = 0; - u16 index; - u16 size; - u16 i; mutex_lock(&hw->tnl_lock); - if (!all && ice_tunnel_port_in_use_hlpr(hw, port, &index)) - if (hw->tnl.tbl[index].ref > 1) { - hw->tnl.tbl[index].ref--; - status = 0; - goto ice_destroy_tunnel_end; - } - - /* determine count */ - for (i = 0; i < hw->tnl.count && i < ICE_TUNNEL_MAX_ENTRIES; i++) - if (hw->tnl.tbl[i].valid && hw->tnl.tbl[i].in_use && - (all || hw->tnl.tbl[i].port == port)) - count++; - - if (!count) { - status = ICE_ERR_PARAM; + if (WARN_ON(!hw->tnl.tbl[index].valid || + hw->tnl.tbl[index].type != type || + hw->tnl.tbl[index].port != port)) { + status = ICE_ERR_OUT_OF_RANGE; goto ice_destroy_tunnel_end; } - /* size of section - there is at least one entry */ - size = struct_size(sect_rx, tcam, count); - bld = ice_pkg_buf_alloc(hw); if (!bld) { status = ICE_ERR_NO_MEMORY; @@ -1859,13 +1786,13 @@ enum ice_status ice_destroy_tunnel(struct ice_hw *hw, u16 port, bool all) goto ice_destroy_tunnel_err; sect_rx = ice_pkg_buf_alloc_section(bld, ICE_SID_RXPARSER_BOOST_TCAM, - size); + struct_size(sect_rx, tcam, 1)); if (!sect_rx) goto ice_destroy_tunnel_err; sect_rx->count = cpu_to_le16(1); sect_tx = ice_pkg_buf_alloc_section(bld, ICE_SID_TXPARSER_BOOST_TCAM, - size); + struct_size(sect_tx, tcam, 1)); if (!sect_tx) goto ice_destroy_tunnel_err; sect_tx->count = cpu_to_le16(1); @@ -1873,26 +1800,14 @@ enum ice_status ice_destroy_tunnel(struct ice_hw *hw, u16 port, bool all) /* copy original boost entry to update package buffer, one copy to Rx * section, another copy to the Tx section */ - for (i = 0; i < hw->tnl.count && i < ICE_TUNNEL_MAX_ENTRIES; i++) - if (hw->tnl.tbl[i].valid && hw->tnl.tbl[i].in_use && - (all || hw->tnl.tbl[i].port == port)) { - memcpy(sect_rx->tcam + i, hw->tnl.tbl[i].boost_entry, - sizeof(*sect_rx->tcam)); - memcpy(sect_tx->tcam + i, hw->tnl.tbl[i].boost_entry, - sizeof(*sect_tx->tcam)); - hw->tnl.tbl[i].marked = true; - } + memcpy(sect_rx->tcam, hw->tnl.tbl[index].boost_entry, + sizeof(*sect_rx->tcam)); + memcpy(sect_tx->tcam, hw->tnl.tbl[index].boost_entry, + sizeof(*sect_tx->tcam)); status = ice_update_pkg(hw, ice_pkg_buf(bld), 1); if (!status) - for (i = 0; i < hw->tnl.count && - i < ICE_TUNNEL_MAX_ENTRIES; i++) - if (hw->tnl.tbl[i].marked) { - hw->tnl.tbl[i].ref = 0; - hw->tnl.tbl[i].port = 0; - hw->tnl.tbl[i].in_use = false; - hw->tnl.tbl[i].marked = false; - } + hw->tnl.tbl[index].port = 0; ice_destroy_tunnel_err: ice_pkg_buf_free(hw, bld); @@ -1903,6 +1818,52 @@ ice_destroy_tunnel_end: return status; } +int ice_udp_tunnel_set_port(struct net_device *netdev, unsigned int table, + unsigned int idx, struct udp_tunnel_info *ti) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_vsi *vsi = np->vsi; + struct ice_pf *pf = vsi->back; + enum ice_tunnel_type tnl_type; + enum ice_status status; + u16 index; + + tnl_type = ti->type == UDP_TUNNEL_TYPE_VXLAN ? TNL_VXLAN : TNL_GENEVE; + index = ice_tunnel_idx_to_entry(&pf->hw, idx, tnl_type); + + status = ice_create_tunnel(&pf->hw, index, tnl_type, ntohs(ti->port)); + if (status) { + netdev_err(netdev, "Error adding UDP tunnel - %s\n", + ice_stat_str(status)); + return -EIO; + } + + udp_tunnel_nic_set_port_priv(netdev, table, idx, index); + return 0; +} + +int ice_udp_tunnel_unset_port(struct net_device *netdev, unsigned int table, + unsigned int idx, struct udp_tunnel_info *ti) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_vsi *vsi = np->vsi; + struct ice_pf *pf = vsi->back; + enum ice_tunnel_type tnl_type; + enum ice_status status; + + tnl_type = ti->type == UDP_TUNNEL_TYPE_VXLAN ? TNL_VXLAN : TNL_GENEVE; + + status = ice_destroy_tunnel(&pf->hw, ti->hw_priv, tnl_type, + ntohs(ti->port)); + if (status) { + netdev_err(netdev, "Error removing UDP tunnel - %s\n", + ice_stat_str(status)); + return -EIO; + } + + return 0; +} + /* PTG Management */ /** @@ -4915,7 +4876,7 @@ ice_rem_prof_id_flow(struct ice_hw *hw, enum ice_block blk, u16 vsi, u64 hdl) if (last_profile) { /* If there are no profiles left for this VSIG, - * then simply remove the the VSIG. + * then simply remove the VSIG. */ status = ice_rem_vsig(hw, blk, vsig, &chg); if (status) diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.h b/drivers/net/ethernet/intel/ice/ice_flex_pipe.h index 568ea519af51..20deddb807c5 100644 --- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.h +++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.h @@ -19,12 +19,11 @@ #define ICE_PKG_CNT 4 bool -ice_get_open_tunnel_port(struct ice_hw *hw, enum ice_tunnel_type type, - u16 *port); -enum ice_status -ice_create_tunnel(struct ice_hw *hw, enum ice_tunnel_type type, u16 port); -enum ice_status ice_destroy_tunnel(struct ice_hw *hw, u16 port, bool all); -bool ice_tunnel_port_in_use(struct ice_hw *hw, u16 port, u16 *index); +ice_get_open_tunnel_port(struct ice_hw *hw, u16 *port); +int ice_udp_tunnel_set_port(struct net_device *netdev, unsigned int table, + unsigned int idx, struct udp_tunnel_info *ti); +int ice_udp_tunnel_unset_port(struct net_device *netdev, unsigned int table, + unsigned int idx, struct udp_tunnel_info *ti); enum ice_status ice_add_prof(struct ice_hw *hw, enum ice_block blk, u64 id, u8 ptypes[], diff --git a/drivers/net/ethernet/intel/ice/ice_flex_type.h b/drivers/net/ethernet/intel/ice/ice_flex_type.h index c1c99a267a98..24063c1351b2 100644 --- a/drivers/net/ethernet/intel/ice/ice_flex_type.h +++ b/drivers/net/ethernet/intel/ice/ice_flex_type.h @@ -298,6 +298,7 @@ struct ice_pkg_enum { enum ice_tunnel_type { TNL_VXLAN = 0, TNL_GENEVE, + __TNL_TYPE_CNT, TNL_LAST = 0xFF, TNL_ALL = 0xFF, }; @@ -311,11 +312,8 @@ struct ice_tunnel_entry { enum ice_tunnel_type type; u16 boost_addr; u16 port; - u16 ref; struct ice_boost_tcam_entry *boost_entry; u8 valid; - u8 in_use; - u8 marked; }; #define ICE_TUNNEL_MAX_ENTRIES 16 @@ -323,6 +321,7 @@ struct ice_tunnel_entry { struct ice_tunnel_table { struct ice_tunnel_entry tbl[ICE_TUNNEL_MAX_ENTRIES]; u16 count; + u16 valid_count[__TNL_TYPE_CNT]; }; struct ice_pkg_es { diff --git a/drivers/net/ethernet/intel/ice/ice_flow.c b/drivers/net/ethernet/intel/ice/ice_flow.c index fe677621dd51..eadc85aee389 100644 --- a/drivers/net/ethernet/intel/ice/ice_flow.c +++ b/drivers/net/ethernet/intel/ice/ice_flow.c @@ -99,6 +99,54 @@ static const u32 ice_ptypes_ipv6_il[] = { 0x00000000, 0x00000000, 0x00000000, 0x00000000, }; +/* Packet types for packets with an Outer/First/Single IPv4 header - no L4 */ +static const u32 ice_ipv4_ofos_no_l4[] = { + 0x10C00000, 0x04000800, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, +}; + +/* Packet types for packets with an Innermost/Last IPv4 header - no L4 */ +static const u32 ice_ipv4_il_no_l4[] = { + 0x60000000, 0x18043008, 0x80000002, 0x6010c021, + 0x00000008, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, +}; + +/* Packet types for packets with an Outer/First/Single IPv6 header - no L4 */ +static const u32 ice_ipv6_ofos_no_l4[] = { + 0x00000000, 0x00000000, 0x43000000, 0x10002000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, +}; + +/* Packet types for packets with an Innermost/Last IPv6 header - no L4 */ +static const u32 ice_ipv6_il_no_l4[] = { + 0x00000000, 0x02180430, 0x0000010c, 0x086010c0, + 0x00000430, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, +}; + /* UDP Packet types for non-tunneled packets or tunneled * packets with inner UDP. */ @@ -250,11 +298,23 @@ ice_flow_proc_seg_hdrs(struct ice_flow_prof_params *params) hdrs = prof->segs[i].hdrs; - if (hdrs & ICE_FLOW_SEG_HDR_IPV4) { + if ((hdrs & ICE_FLOW_SEG_HDR_IPV4) && + !(hdrs & ICE_FLOW_SEG_HDRS_L4_MASK)) { + src = !i ? (const unsigned long *)ice_ipv4_ofos_no_l4 : + (const unsigned long *)ice_ipv4_il_no_l4; + bitmap_and(params->ptypes, params->ptypes, src, + ICE_FLOW_PTYPE_MAX); + } else if (hdrs & ICE_FLOW_SEG_HDR_IPV4) { src = !i ? (const unsigned long *)ice_ptypes_ipv4_ofos : (const unsigned long *)ice_ptypes_ipv4_il; bitmap_and(params->ptypes, params->ptypes, src, ICE_FLOW_PTYPE_MAX); + } else if ((hdrs & ICE_FLOW_SEG_HDR_IPV6) && + !(hdrs & ICE_FLOW_SEG_HDRS_L4_MASK)) { + src = !i ? (const unsigned long *)ice_ipv6_ofos_no_l4 : + (const unsigned long *)ice_ipv6_il_no_l4; + bitmap_and(params->ptypes, params->ptypes, src, + ICE_FLOW_PTYPE_MAX); } else if (hdrs & ICE_FLOW_SEG_HDR_IPV6) { src = !i ? (const unsigned long *)ice_ptypes_ipv6_ofos : (const unsigned long *)ice_ptypes_ipv6_il; @@ -385,7 +445,7 @@ ice_flow_xtract_fld(struct ice_hw *hw, struct ice_flow_prof_params *params, * ice_flow_xtract_raws - Create extract sequence entries for raw bytes * @hw: pointer to the HW struct * @params: information about the flow to be processed - * @seg: index of packet segment whose raw fields are to be be extracted + * @seg: index of packet segment whose raw fields are to be extracted */ static enum ice_status ice_flow_xtract_raws(struct ice_hw *hw, struct ice_flow_prof_params *params, @@ -999,7 +1059,7 @@ enum ice_status ice_flow_rem_entry(struct ice_hw *hw, enum ice_block blk, * * This helper function stores information of a field being matched, including * the type of the field and the locations of the value to match, the mask, and - * and the upper-bound value in the start of the input buffer for a flow entry. + * the upper-bound value in the start of the input buffer for a flow entry. * This function should only be used for fixed-size data structures. * * This function also opportunistically determines the protocol headers to be diff --git a/drivers/net/ethernet/intel/ice/ice_flow.h b/drivers/net/ethernet/intel/ice/ice_flow.h index 3913da2116d2..829f90b1e998 100644 --- a/drivers/net/ethernet/intel/ice/ice_flow.h +++ b/drivers/net/ethernet/intel/ice/ice_flow.h @@ -194,8 +194,8 @@ struct ice_flow_entry { u16 entry_sz; }; -#define ICE_FLOW_ENTRY_HNDL(e) ((u64)e) -#define ICE_FLOW_ENTRY_PTR(h) ((struct ice_flow_entry *)(h)) +#define ICE_FLOW_ENTRY_HNDL(e) ((u64)(uintptr_t)e) +#define ICE_FLOW_ENTRY_PTR(h) ((struct ice_flow_entry *)(uintptr_t)(h)) struct ice_flow_prof { struct list_head l_entry; diff --git a/drivers/net/ethernet/intel/ice/ice_fw_update.c b/drivers/net/ethernet/intel/ice/ice_fw_update.c index 8968fdd4816b..8f81b95e679c 100644 --- a/drivers/net/ethernet/intel/ice/ice_fw_update.c +++ b/drivers/net/ethernet/intel/ice/ice_fw_update.c @@ -43,6 +43,8 @@ ice_send_package_data(struct pldmfw *context, const u8 *data, u16 length) enum ice_status status; u8 *package_data; + dev_dbg(dev, "Sending PLDM record package data to firmware\n"); + package_data = kmemdup(data, length, GFP_KERNEL); if (!package_data) return -ENOMEM; @@ -229,6 +231,8 @@ ice_send_component_table(struct pldmfw *context, struct pldmfw_component *compon comp_tbl->cvs_len = component->version_len; memcpy(comp_tbl->cvs, component->version_string, component->version_len); + dev_dbg(dev, "Sending component table to firmware:\n"); + status = ice_nvm_pass_component_tbl(hw, (u8 *)comp_tbl, length, transfer_flag, &comp_response, &comp_response_code, NULL); @@ -279,11 +283,14 @@ ice_write_one_nvm_block(struct ice_pf *pf, u16 module, u32 offset, memset(&event, 0, sizeof(event)); + dev_dbg(dev, "Writing block of %u bytes for module 0x%02x at offset %u\n", + block_size, module, offset); + status = ice_aq_update_nvm(hw, module, offset, block_size, block, last_cmd, 0, NULL); if (status) { - dev_err(dev, "Failed to program flash module 0x%02x at offset %u, err %s aq_err %s\n", - module, offset, ice_stat_str(status), + dev_err(dev, "Failed to flash module 0x%02x with block of size %u at offset %u, err %s aq_err %s\n", + module, block_size, offset, ice_stat_str(status), ice_aq_str(hw->adminq.sq_last_status)); NL_SET_ERR_MSG_MOD(extack, "Failed to program flash module"); return -EIO; @@ -297,8 +304,8 @@ ice_write_one_nvm_block(struct ice_pf *pf, u16 module, u32 offset, */ err = ice_aq_wait_for_event(pf, ice_aqc_opc_nvm_write, 15 * HZ, &event); if (err) { - dev_err(dev, "Timed out waiting for firmware write completion for module 0x%02x, err %d\n", - module, err); + dev_err(dev, "Timed out while trying to flash module 0x%02x with block of size %u at offset %u, err %d\n", + module, block_size, offset, err); NL_SET_ERR_MSG_MOD(extack, "Timed out waiting for firmware"); return -EIO; } @@ -324,8 +331,8 @@ ice_write_one_nvm_block(struct ice_pf *pf, u16 module, u32 offset, } if (completion_retval) { - dev_err(dev, "Firmware failed to program flash module 0x%02x at offset %u, completion err %s\n", - module, offset, + dev_err(dev, "Firmware failed to flash module 0x%02x with block of size %u at offset %u, err %s\n", + module, block_size, offset, ice_aq_str((enum ice_aq_err)completion_retval)); NL_SET_ERR_MSG_MOD(extack, "Firmware failed to program flash module"); return -EIO; @@ -356,12 +363,15 @@ ice_write_nvm_module(struct ice_pf *pf, u16 module, const char *component, const u8 *image, u32 length, struct netlink_ext_ack *extack) { + struct device *dev = ice_pf_to_dev(pf); struct devlink *devlink; u32 offset = 0; bool last_cmd; u8 *block; int err; + dev_dbg(dev, "Beginning write of flash component '%s', module 0x%02x\n", component, module); + devlink = priv_to_devlink(pf); devlink_flash_update_status_notify(devlink, "Flashing", @@ -394,6 +404,8 @@ ice_write_nvm_module(struct ice_pf *pf, u16 module, const char *component, component, offset, length); } while (!last_cmd); + dev_dbg(dev, "Completed write of flash component '%s', module 0x%02x\n", component, module); + if (err) devlink_flash_update_status_notify(devlink, "Flashing failed", component, length, length); @@ -431,6 +443,8 @@ ice_erase_nvm_module(struct ice_pf *pf, u16 module, const char *component, enum ice_status status; int err; + dev_dbg(dev, "Beginning erase of flash component '%s', module 0x%02x\n", component, module); + memset(&event, 0, sizeof(event)); devlink = priv_to_devlink(pf); @@ -476,6 +490,8 @@ ice_erase_nvm_module(struct ice_pf *pf, u16 module, const char *component, goto out_notify_devlink; } + dev_dbg(dev, "Completed erase of flash component '%s', module 0x%02x\n", component, module); + out_notify_devlink: if (err) devlink_flash_update_status_notify(devlink, "Erasing failed", @@ -614,14 +630,9 @@ static int ice_finalize_update(struct pldmfw *context) struct ice_fwu_priv *priv = container_of(context, struct ice_fwu_priv, context); struct netlink_ext_ack *extack = priv->extack; struct ice_pf *pf = priv->pf; - int err; /* Finally, notify firmware to activate the written NVM banks */ - err = ice_switch_flash_banks(pf, priv->activate_flags, extack); - if (err) - return err; - - return 0; + return ice_switch_flash_banks(pf, priv->activate_flags, extack); } static const struct pldmfw_ops ice_fwu_ops = { @@ -636,6 +647,7 @@ static const struct pldmfw_ops ice_fwu_ops = { * ice_flash_pldm_image - Write a PLDM-formatted firmware image to the device * @pf: private device driver structure * @fw: firmware object pointing to the relevant firmware file + * @preservation: preservation level to request from firmware * @extack: netlink extended ACK structure * * Parse the data for a given firmware file, verifying that it is a valid PLDM @@ -649,7 +661,7 @@ static const struct pldmfw_ops ice_fwu_ops = { * Returns: zero on success or a negative error code on failure. */ int ice_flash_pldm_image(struct ice_pf *pf, const struct firmware *fw, - struct netlink_ext_ack *extack) + u8 preservation, struct netlink_ext_ack *extack) { struct device *dev = ice_pf_to_dev(pf); struct ice_hw *hw = &pf->hw; @@ -657,13 +669,24 @@ int ice_flash_pldm_image(struct ice_pf *pf, const struct firmware *fw, enum ice_status status; int err; + switch (preservation) { + case ICE_AQC_NVM_PRESERVE_ALL: + case ICE_AQC_NVM_PRESERVE_SELECTED: + case ICE_AQC_NVM_NO_PRESERVATION: + case ICE_AQC_NVM_FACTORY_DEFAULT: + break; + default: + WARN(1, "Unexpected preservation level request %u", preservation); + return -EINVAL; + } + memset(&priv, 0, sizeof(priv)); priv.context.ops = &ice_fwu_ops; priv.context.dev = dev; priv.extack = extack; priv.pf = pf; - priv.activate_flags = ICE_AQC_NVM_PRESERVE_ALL; + priv.activate_flags = preservation; status = ice_acquire_nvm(hw, ICE_RES_WRITE); if (status) { diff --git a/drivers/net/ethernet/intel/ice/ice_fw_update.h b/drivers/net/ethernet/intel/ice/ice_fw_update.h index 79472cc618b4..c6390f6851ff 100644 --- a/drivers/net/ethernet/intel/ice/ice_fw_update.h +++ b/drivers/net/ethernet/intel/ice/ice_fw_update.h @@ -5,7 +5,7 @@ #define _ICE_FW_UPDATE_H_ int ice_flash_pldm_image(struct ice_pf *pf, const struct firmware *fw, - struct netlink_ext_ack *extack); + u8 preservation, struct netlink_ext_ack *extack); int ice_check_for_pending_update(struct ice_pf *pf, const char *component, struct netlink_ext_ack *extack); diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c index ebbb8f54871c..3df67486d42d 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_lib.c @@ -7,6 +7,7 @@ #include "ice_lib.h" #include "ice_fltr.h" #include "ice_dcb_lib.h" +#include "ice_devlink.h" /** * ice_vsi_type_str - maps VSI type enum to string equivalents @@ -1755,7 +1756,7 @@ int ice_vsi_cfg_xdp_txqs(struct ice_vsi *vsi) return ret; for (i = 0; i < vsi->num_xdp_txq; i++) - vsi->xdp_rings[i]->xsk_umem = ice_xsk_umem(vsi->xdp_rings[i]); + vsi->xdp_rings[i]->xsk_pool = ice_xsk_pool(vsi->xdp_rings[i]); return ret; } @@ -2616,8 +2617,10 @@ int ice_vsi_release(struct ice_vsi *vsi) * PF that is running the work queue items currently. This is done to * avoid check_flush_dependency() warning on this wq */ - if (vsi->netdev && !ice_is_reset_in_progress(pf->state)) + if (vsi->netdev && !ice_is_reset_in_progress(pf->state)) { unregister_netdev(vsi->netdev); + ice_devlink_destroy_port(vsi); + } if (test_bit(ICE_FLAG_RSS_ENA, pf->flags)) ice_rss_clean(vsi); diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 54a7f55eb8c1..2dea4d0e9415 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -486,7 +486,6 @@ static void ice_do_reset(struct ice_pf *pf, enum ice_reset_req reset_type) struct ice_hw *hw = &pf->hw; dev_dbg(dev, "reset_type 0x%x requested\n", reset_type); - WARN_ON(in_interrupt()); ice_prepare_for_reset(pf); @@ -1057,7 +1056,9 @@ struct ice_aq_task { int ice_aq_wait_for_event(struct ice_pf *pf, u16 opcode, unsigned long timeout, struct ice_rq_event_info *event) { + struct device *dev = ice_pf_to_dev(pf); struct ice_aq_task *task; + unsigned long start; long ret; int err; @@ -1074,6 +1075,8 @@ int ice_aq_wait_for_event(struct ice_pf *pf, u16 opcode, unsigned long timeout, hlist_add_head(&task->entry, &pf->aq_wait_list); spin_unlock_bh(&pf->aq_wait_lock); + start = jiffies; + ret = wait_event_interruptible_timeout(pf->aq_wait_queue, task->state, timeout); switch (task->state) { @@ -1092,6 +1095,11 @@ int ice_aq_wait_for_event(struct ice_pf *pf, u16 opcode, unsigned long timeout, break; } + dev_dbg(dev, "Waited %u msecs (max %u msecs) for firmware response to op 0x%04x\n", + jiffies_to_msecs(jiffies - start), + jiffies_to_msecs(timeout), + opcode); + spin_lock_bh(&pf->aq_wait_lock); hlist_del(&task->entry); spin_unlock_bh(&pf->aq_wait_lock); @@ -2273,7 +2281,7 @@ static int ice_xdp_alloc_setup_rings(struct ice_vsi *vsi) if (ice_setup_tx_ring(xdp_ring)) goto free_xdp_rings; ice_set_ring_xdp(xdp_ring); - xdp_ring->xsk_umem = ice_xsk_umem(xdp_ring); + xdp_ring->xsk_pool = ice_xsk_pool(xdp_ring); } return 0; @@ -2417,7 +2425,7 @@ int ice_destroy_xdp_rings(struct ice_vsi *vsi) int i, v_idx; /* q_vectors are freed in reset path so there's no point in detaching - * rings; in case of rebuild being triggered not from reset reset bits + * rings; in case of rebuild being triggered not from reset bits * in pf->state won't be set, so additionally check first q_vector * against NULL */ @@ -2517,13 +2525,13 @@ ice_xdp_setup_prog(struct ice_vsi *vsi, struct bpf_prog *prog, if (if_running) ret = ice_up(vsi); - if (!ret && prog && vsi->xsk_umems) { + if (!ret && prog && vsi->xsk_pools) { int i; ice_for_each_rxq(vsi, i) { struct ice_ring *rx_ring = vsi->rx_rings[i]; - if (rx_ring->xsk_umem) + if (rx_ring->xsk_pool) napi_schedule(&rx_ring->q_vector->napi); } } @@ -2549,8 +2557,8 @@ static int ice_xdp(struct net_device *dev, struct netdev_bpf *xdp) switch (xdp->command) { case XDP_SETUP_PROG: return ice_xdp_setup_prog(vsi, xdp->prog, xdp->extack); - case XDP_SETUP_XSK_UMEM: - return ice_xsk_umem_setup(vsi, xdp->xsk.umem, + case XDP_SETUP_XSK_POOL: + return ice_xsk_pool_setup(vsi, xdp->xsk.pool, xdp->xsk.queue_id); default: return -EINVAL; @@ -2873,6 +2881,7 @@ static void ice_set_ops(struct net_device *netdev) } netdev->netdev_ops = &ice_netdev_ops; + netdev->udp_tunnel_nic_info = &pf->hw.udp_tunnel_nic; ice_set_ethtool_ops(netdev); } @@ -2953,7 +2962,7 @@ static int ice_cfg_netdev(struct ice_vsi *vsi) u8 mac_addr[ETH_ALEN]; int err; - err = ice_devlink_create_port(pf); + err = ice_devlink_create_port(vsi); if (err) return err; @@ -2994,7 +3003,7 @@ static int ice_cfg_netdev(struct ice_vsi *vsi) if (err) goto err_free_netdev; - devlink_port_type_eth_set(&pf->devlink_port, vsi->netdev); + devlink_port_type_eth_set(&vsi->devlink_port, vsi->netdev); netif_carrier_off(vsi->netdev); @@ -3007,7 +3016,7 @@ err_free_netdev: free_netdev(vsi->netdev); vsi->netdev = NULL; err_destroy_devlink_port: - ice_devlink_destroy_port(pf); + ice_devlink_destroy_port(vsi); return err; } @@ -3971,7 +3980,7 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) struct device *dev = &pdev->dev; struct ice_pf *pf; struct ice_hw *hw; - int err; + int i, err; /* this driver uses devres, see * Documentation/driver-api/driver-model/devres.rst @@ -4066,11 +4075,37 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) ice_devlink_init_regions(pf); + pf->hw.udp_tunnel_nic.set_port = ice_udp_tunnel_set_port; + pf->hw.udp_tunnel_nic.unset_port = ice_udp_tunnel_unset_port; + pf->hw.udp_tunnel_nic.flags = UDP_TUNNEL_NIC_INFO_MAY_SLEEP; + pf->hw.udp_tunnel_nic.shared = &pf->hw.udp_tunnel_shared; + i = 0; + if (pf->hw.tnl.valid_count[TNL_VXLAN]) { + pf->hw.udp_tunnel_nic.tables[i].n_entries = + pf->hw.tnl.valid_count[TNL_VXLAN]; + pf->hw.udp_tunnel_nic.tables[i].tunnel_types = + UDP_TUNNEL_TYPE_VXLAN; + i++; + } + if (pf->hw.tnl.valid_count[TNL_GENEVE]) { + pf->hw.udp_tunnel_nic.tables[i].n_entries = + pf->hw.tnl.valid_count[TNL_GENEVE]; + pf->hw.udp_tunnel_nic.tables[i].tunnel_types = + UDP_TUNNEL_TYPE_GENEVE; + i++; + } + pf->num_alloc_vsi = hw->func_caps.guar_num_vsi; if (!pf->num_alloc_vsi) { err = -EIO; goto err_init_pf_unroll; } + if (pf->num_alloc_vsi > UDP_TUNNEL_NIC_MAX_SHARING_DEVICES) { + dev_warn(&pf->pdev->dev, + "limiting the VSI count due to UDP tunnel limitation %d > %d\n", + pf->num_alloc_vsi, UDP_TUNNEL_NIC_MAX_SHARING_DEVICES); + pf->num_alloc_vsi = UDP_TUNNEL_NIC_MAX_SHARING_DEVICES; + } pf->vsi = devm_kcalloc(dev, pf->num_alloc_vsi, sizeof(*pf->vsi), GFP_KERNEL); @@ -4216,7 +4251,6 @@ probe_done: err_send_version_unroll: ice_vsi_release_all(pf); err_alloc_sw_unroll: - ice_devlink_destroy_port(pf); set_bit(__ICE_SERVICE_DIS, pf->state); set_bit(__ICE_DOWN, pf->state); devm_kfree(dev, pf->first_sw); @@ -4331,7 +4365,6 @@ static void ice_remove(struct pci_dev *pdev) if (!ice_is_safe_mode(pf)) ice_remove_arfs(pf); ice_setup_mc_magic_wake(pf); - ice_devlink_destroy_port(pf); ice_vsi_release_all(pf); ice_set_wake(pf); ice_free_irq_msix_misc(pf); @@ -6569,70 +6602,6 @@ static void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue) } /** - * ice_udp_tunnel_add - Get notifications about UDP tunnel ports that come up - * @netdev: This physical port's netdev - * @ti: Tunnel endpoint information - */ -static void -ice_udp_tunnel_add(struct net_device *netdev, struct udp_tunnel_info *ti) -{ - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_vsi *vsi = np->vsi; - struct ice_pf *pf = vsi->back; - enum ice_tunnel_type tnl_type; - u16 port = ntohs(ti->port); - enum ice_status status; - - switch (ti->type) { - case UDP_TUNNEL_TYPE_VXLAN: - tnl_type = TNL_VXLAN; - break; - case UDP_TUNNEL_TYPE_GENEVE: - tnl_type = TNL_GENEVE; - break; - default: - netdev_err(netdev, "Unknown tunnel type\n"); - return; - } - - status = ice_create_tunnel(&pf->hw, tnl_type, port); - if (status == ICE_ERR_OUT_OF_RANGE) - netdev_info(netdev, "Max tunneled UDP ports reached, port %d not added\n", - port); - else if (status) - netdev_err(netdev, "Error adding UDP tunnel - %s\n", - ice_stat_str(status)); -} - -/** - * ice_udp_tunnel_del - Get notifications about UDP tunnel ports that go away - * @netdev: This physical port's netdev - * @ti: Tunnel endpoint information - */ -static void -ice_udp_tunnel_del(struct net_device *netdev, struct udp_tunnel_info *ti) -{ - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_vsi *vsi = np->vsi; - struct ice_pf *pf = vsi->back; - u16 port = ntohs(ti->port); - enum ice_status status; - bool retval; - - retval = ice_tunnel_port_in_use(&pf->hw, port, NULL); - if (!retval) { - netdev_info(netdev, "port %d not found in UDP tunnels list\n", - port); - return; - } - - status = ice_destroy_tunnel(&pf->hw, port, false); - if (status) - netdev_err(netdev, "error deleting port %d from UDP tunnels list\n", - port); -} - -/** * ice_open - Called when a network interface becomes active * @netdev: network interface device structure * @@ -6824,6 +6793,6 @@ static const struct net_device_ops ice_netdev_ops = { .ndo_bpf = ice_xdp, .ndo_xdp_xmit = ice_xdp_xmit, .ndo_xsk_wakeup = ice_xsk_wakeup, - .ndo_udp_tunnel_add = ice_udp_tunnel_add, - .ndo_udp_tunnel_del = ice_udp_tunnel_del, + .ndo_udp_tunnel_add = udp_tunnel_nic_add_port, + .ndo_udp_tunnel_del = udp_tunnel_nic_del_port, }; diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index 9d0d6b0025cf..eae75260fe20 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -145,7 +145,7 @@ void ice_clean_tx_ring(struct ice_ring *tx_ring) { u16 i; - if (ice_ring_is_xdp(tx_ring) && tx_ring->xsk_umem) { + if (ice_ring_is_xdp(tx_ring) && tx_ring->xsk_pool) { ice_xsk_clean_xdp_ring(tx_ring); goto tx_skip_free; } @@ -375,7 +375,7 @@ void ice_clean_rx_ring(struct ice_ring *rx_ring) if (!rx_ring->rx_buf) return; - if (rx_ring->xsk_umem) { + if (rx_ring->xsk_pool) { ice_xsk_clean_rx_ring(rx_ring); goto rx_skip_free; } @@ -919,10 +919,7 @@ ice_build_skb(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf, * likely have a consumer accessing first few bytes of meta * data, and then actual data. */ - prefetch(xdp->data_meta); -#if L1_CACHE_BYTES < 128 - prefetch((void *)(xdp->data + L1_CACHE_BYTES)); -#endif + net_prefetch(xdp->data_meta); /* build an skb around the page buffer */ skb = build_skb(xdp->data_hard_start, truesize); if (unlikely(!skb)) @@ -964,10 +961,7 @@ ice_construct_skb(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf, struct sk_buff *skb; /* prefetch first cache line of first page */ - prefetch(xdp->data); -#if L1_CACHE_BYTES < 128 - prefetch((void *)(xdp->data + L1_CACHE_BYTES)); -#endif /* L1_CACHE_BYTES */ + net_prefetch(xdp->data); /* allocate a skb to store the frags */ skb = __napi_alloc_skb(&rx_ring->q_vector->napi, ICE_RX_HDR_SIZE, @@ -1616,7 +1610,7 @@ int ice_napi_poll(struct napi_struct *napi, int budget) * budget and be more aggressive about cleaning up the Tx descriptors. */ ice_for_each_ring(ring, q_vector->tx) { - bool wd = ring->xsk_umem ? + bool wd = ring->xsk_pool ? ice_clean_tx_irq_zc(ring, budget) : ice_clean_tx_irq(ring, budget); @@ -1646,7 +1640,7 @@ int ice_napi_poll(struct napi_struct *napi, int budget) * comparison in the irq context instead of many inside the * ice_clean_rx_irq function and makes the codebase cleaner. */ - cleaned = ring->xsk_umem ? + cleaned = ring->xsk_pool ? ice_clean_rx_irq_zc(ring, budget_per_ring) : ice_clean_rx_irq(ring, budget_per_ring); work_done += cleaned; diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h index 51b4df7a59d2..ff1a1cbd078e 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx.h @@ -43,7 +43,7 @@ /** * ice_compute_pad - compute the padding - * rx_buf_len: buffer length + * @rx_buf_len: buffer length * * Figure out the size of half page based on given buffer length and * then subtract the skb_shared_info followed by subtraction of the @@ -295,7 +295,7 @@ struct ice_ring { struct rcu_head rcu; /* to avoid race on free */ struct bpf_prog *xdp_prog; - struct xdp_umem *xsk_umem; + struct xsk_buff_pool *xsk_pool; /* CL3 - 3rd cacheline starts here */ struct xdp_rxq_info xdp_rxq; /* CLX - the below items are only accessed infrequently and should be diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h index 4cdccfadf274..2226a291a394 100644 --- a/drivers/net/ethernet/intel/ice/ice_type.h +++ b/drivers/net/ethernet/intel/ice/ice_type.h @@ -676,6 +676,9 @@ struct ice_hw { struct mutex tnl_lock; struct ice_tunnel_table tnl; + struct udp_tunnel_nic_shared udp_tunnel_shared; + struct udp_tunnel_nic_info udp_tunnel_nic; + /* HW block tables */ struct ice_blk_info blk[ICE_BLK_COUNT]; struct mutex fl_profs_locks[ICE_BLK_COUNT]; /* lock fltr profiles */ diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c index 71497776ac62..ec7f6c64132e 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c @@ -871,7 +871,7 @@ static int ice_get_max_valid_res_idx(struct ice_res_tracker *res) * If there are not enough resources available, return an error. This should * always be caught by ice_set_per_vf_res(). * - * Return 0 on success, and -EINVAL when there are not enough MSIX vectors in + * Return 0 on success, and -EINVAL when there are not enough MSIX vectors * in the PF's space available for SR-IOV. */ static int ice_sriov_set_msix_res(struct ice_pf *pf, u16 num_msix_needed) diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c index 20ac5fca68c6..797886524054 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.c +++ b/drivers/net/ethernet/intel/ice/ice_xsk.c @@ -236,7 +236,7 @@ static int ice_qp_ena(struct ice_vsi *vsi, u16 q_idx) if (err) goto free_buf; ice_set_ring_xdp(xdp_ring); - xdp_ring->xsk_umem = ice_xsk_umem(xdp_ring); + xdp_ring->xsk_pool = ice_xsk_pool(xdp_ring); } err = ice_setup_rx_ctx(rx_ring); @@ -260,21 +260,21 @@ free_buf: } /** - * ice_xsk_alloc_umems - allocate a UMEM region for an XDP socket - * @vsi: VSI to allocate the UMEM on + * ice_xsk_alloc_pools - allocate a buffer pool for an XDP socket + * @vsi: VSI to allocate the buffer pool on * * Returns 0 on success, negative on error */ -static int ice_xsk_alloc_umems(struct ice_vsi *vsi) +static int ice_xsk_alloc_pools(struct ice_vsi *vsi) { - if (vsi->xsk_umems) + if (vsi->xsk_pools) return 0; - vsi->xsk_umems = kcalloc(vsi->num_xsk_umems, sizeof(*vsi->xsk_umems), + vsi->xsk_pools = kcalloc(vsi->num_xsk_pools, sizeof(*vsi->xsk_pools), GFP_KERNEL); - if (!vsi->xsk_umems) { - vsi->num_xsk_umems = 0; + if (!vsi->xsk_pools) { + vsi->num_xsk_pools = 0; return -ENOMEM; } @@ -282,73 +282,73 @@ static int ice_xsk_alloc_umems(struct ice_vsi *vsi) } /** - * ice_xsk_remove_umem - Remove an UMEM for a certain ring/qid + * ice_xsk_remove_pool - Remove an buffer pool for a certain ring/qid * @vsi: VSI from which the VSI will be removed - * @qid: Ring/qid associated with the UMEM + * @qid: Ring/qid associated with the buffer pool */ -static void ice_xsk_remove_umem(struct ice_vsi *vsi, u16 qid) +static void ice_xsk_remove_pool(struct ice_vsi *vsi, u16 qid) { - vsi->xsk_umems[qid] = NULL; - vsi->num_xsk_umems_used--; + vsi->xsk_pools[qid] = NULL; + vsi->num_xsk_pools_used--; - if (vsi->num_xsk_umems_used == 0) { - kfree(vsi->xsk_umems); - vsi->xsk_umems = NULL; - vsi->num_xsk_umems = 0; + if (vsi->num_xsk_pools_used == 0) { + kfree(vsi->xsk_pools); + vsi->xsk_pools = NULL; + vsi->num_xsk_pools = 0; } } /** - * ice_xsk_umem_disable - disable a UMEM region + * ice_xsk_pool_disable - disable a buffer pool region * @vsi: Current VSI * @qid: queue ID * * Returns 0 on success, negative on failure */ -static int ice_xsk_umem_disable(struct ice_vsi *vsi, u16 qid) +static int ice_xsk_pool_disable(struct ice_vsi *vsi, u16 qid) { - if (!vsi->xsk_umems || qid >= vsi->num_xsk_umems || - !vsi->xsk_umems[qid]) + if (!vsi->xsk_pools || qid >= vsi->num_xsk_pools || + !vsi->xsk_pools[qid]) return -EINVAL; - xsk_buff_dma_unmap(vsi->xsk_umems[qid], ICE_RX_DMA_ATTR); - ice_xsk_remove_umem(vsi, qid); + xsk_pool_dma_unmap(vsi->xsk_pools[qid], ICE_RX_DMA_ATTR); + ice_xsk_remove_pool(vsi, qid); return 0; } /** - * ice_xsk_umem_enable - enable a UMEM region + * ice_xsk_pool_enable - enable a buffer pool region * @vsi: Current VSI - * @umem: pointer to a requested UMEM region + * @pool: pointer to a requested buffer pool region * @qid: queue ID * * Returns 0 on success, negative on failure */ static int -ice_xsk_umem_enable(struct ice_vsi *vsi, struct xdp_umem *umem, u16 qid) +ice_xsk_pool_enable(struct ice_vsi *vsi, struct xsk_buff_pool *pool, u16 qid) { int err; if (vsi->type != ICE_VSI_PF) return -EINVAL; - if (!vsi->num_xsk_umems) - vsi->num_xsk_umems = min_t(u16, vsi->num_rxq, vsi->num_txq); - if (qid >= vsi->num_xsk_umems) + if (!vsi->num_xsk_pools) + vsi->num_xsk_pools = min_t(u16, vsi->num_rxq, vsi->num_txq); + if (qid >= vsi->num_xsk_pools) return -EINVAL; - err = ice_xsk_alloc_umems(vsi); + err = ice_xsk_alloc_pools(vsi); if (err) return err; - if (vsi->xsk_umems && vsi->xsk_umems[qid]) + if (vsi->xsk_pools && vsi->xsk_pools[qid]) return -EBUSY; - vsi->xsk_umems[qid] = umem; - vsi->num_xsk_umems_used++; + vsi->xsk_pools[qid] = pool; + vsi->num_xsk_pools_used++; - err = xsk_buff_dma_map(vsi->xsk_umems[qid], ice_pf_to_dev(vsi->back), + err = xsk_pool_dma_map(vsi->xsk_pools[qid], ice_pf_to_dev(vsi->back), ICE_RX_DMA_ATTR); if (err) return err; @@ -357,17 +357,17 @@ ice_xsk_umem_enable(struct ice_vsi *vsi, struct xdp_umem *umem, u16 qid) } /** - * ice_xsk_umem_setup - enable/disable a UMEM region depending on its state + * ice_xsk_pool_setup - enable/disable a buffer pool region depending on its state * @vsi: Current VSI - * @umem: UMEM to enable/associate to a ring, NULL to disable + * @pool: buffer pool to enable/associate to a ring, NULL to disable * @qid: queue ID * * Returns 0 on success, negative on failure */ -int ice_xsk_umem_setup(struct ice_vsi *vsi, struct xdp_umem *umem, u16 qid) +int ice_xsk_pool_setup(struct ice_vsi *vsi, struct xsk_buff_pool *pool, u16 qid) { - bool if_running, umem_present = !!umem; - int ret = 0, umem_failure = 0; + bool if_running, pool_present = !!pool; + int ret = 0, pool_failure = 0; if_running = netif_running(vsi->netdev) && ice_is_xdp_ena_vsi(vsi); @@ -375,26 +375,26 @@ int ice_xsk_umem_setup(struct ice_vsi *vsi, struct xdp_umem *umem, u16 qid) ret = ice_qp_dis(vsi, qid); if (ret) { netdev_err(vsi->netdev, "ice_qp_dis error = %d\n", ret); - goto xsk_umem_if_up; + goto xsk_pool_if_up; } } - umem_failure = umem_present ? ice_xsk_umem_enable(vsi, umem, qid) : - ice_xsk_umem_disable(vsi, qid); + pool_failure = pool_present ? ice_xsk_pool_enable(vsi, pool, qid) : + ice_xsk_pool_disable(vsi, qid); -xsk_umem_if_up: +xsk_pool_if_up: if (if_running) { ret = ice_qp_ena(vsi, qid); - if (!ret && umem_present) + if (!ret && pool_present) napi_schedule(&vsi->xdp_rings[qid]->q_vector->napi); else if (ret) netdev_err(vsi->netdev, "ice_qp_ena error = %d\n", ret); } - if (umem_failure) { - netdev_err(vsi->netdev, "Could not %sable UMEM, error = %d\n", - umem_present ? "en" : "dis", umem_failure); - return umem_failure; + if (pool_failure) { + netdev_err(vsi->netdev, "Could not %sable buffer pool, error = %d\n", + pool_present ? "en" : "dis", pool_failure); + return pool_failure; } return ret; @@ -425,7 +425,7 @@ bool ice_alloc_rx_bufs_zc(struct ice_ring *rx_ring, u16 count) rx_buf = &rx_ring->rx_buf[ntu]; do { - rx_buf->xdp = xsk_buff_alloc(rx_ring->xsk_umem); + rx_buf->xdp = xsk_buff_alloc(rx_ring->xsk_pool); if (!rx_buf->xdp) { ret = true; break; @@ -595,7 +595,7 @@ int ice_clean_rx_irq_zc(struct ice_ring *rx_ring, int budget) rx_buf = &rx_ring->rx_buf[rx_ring->next_to_clean]; rx_buf->xdp->data_end = rx_buf->xdp->data + size; - xsk_buff_dma_sync_for_cpu(rx_buf->xdp); + xsk_buff_dma_sync_for_cpu(rx_buf->xdp, rx_ring->xsk_pool); xdp_res = ice_run_xdp_zc(rx_ring, rx_buf->xdp); if (xdp_res) { @@ -645,11 +645,11 @@ int ice_clean_rx_irq_zc(struct ice_ring *rx_ring, int budget) ice_finalize_xdp_rx(rx_ring, xdp_xmit); ice_update_rx_ring_stats(rx_ring, total_rx_packets, total_rx_bytes); - if (xsk_umem_uses_need_wakeup(rx_ring->xsk_umem)) { + if (xsk_uses_need_wakeup(rx_ring->xsk_pool)) { if (failure || rx_ring->next_to_clean == rx_ring->next_to_use) - xsk_set_rx_need_wakeup(rx_ring->xsk_umem); + xsk_set_rx_need_wakeup(rx_ring->xsk_pool); else - xsk_clear_rx_need_wakeup(rx_ring->xsk_umem); + xsk_clear_rx_need_wakeup(rx_ring->xsk_pool); return (int)total_rx_packets; } @@ -682,11 +682,11 @@ static bool ice_xmit_zc(struct ice_ring *xdp_ring, int budget) tx_buf = &xdp_ring->tx_buf[xdp_ring->next_to_use]; - if (!xsk_umem_consume_tx(xdp_ring->xsk_umem, &desc)) + if (!xsk_tx_peek_desc(xdp_ring->xsk_pool, &desc)) break; - dma = xsk_buff_raw_get_dma(xdp_ring->xsk_umem, desc.addr); - xsk_buff_raw_dma_sync_for_device(xdp_ring->xsk_umem, dma, + dma = xsk_buff_raw_get_dma(xdp_ring->xsk_pool, desc.addr); + xsk_buff_raw_dma_sync_for_device(xdp_ring->xsk_pool, dma, desc.len); tx_buf->bytecount = desc.len; @@ -703,7 +703,7 @@ static bool ice_xmit_zc(struct ice_ring *xdp_ring, int budget) if (tx_desc) { ice_xdp_ring_update_tail(xdp_ring); - xsk_umem_consume_tx_done(xdp_ring->xsk_umem); + xsk_tx_release(xdp_ring->xsk_pool); } return budget > 0 && work_done; @@ -777,10 +777,10 @@ bool ice_clean_tx_irq_zc(struct ice_ring *xdp_ring, int budget) xdp_ring->next_to_clean = ntc; if (xsk_frames) - xsk_umem_complete_tx(xdp_ring->xsk_umem, xsk_frames); + xsk_tx_completed(xdp_ring->xsk_pool, xsk_frames); - if (xsk_umem_uses_need_wakeup(xdp_ring->xsk_umem)) - xsk_set_tx_need_wakeup(xdp_ring->xsk_umem); + if (xsk_uses_need_wakeup(xdp_ring->xsk_pool)) + xsk_set_tx_need_wakeup(xdp_ring->xsk_pool); ice_update_tx_ring_stats(xdp_ring, total_packets, total_bytes); xmit_done = ice_xmit_zc(xdp_ring, ICE_DFLT_IRQ_WORK); @@ -814,7 +814,7 @@ ice_xsk_wakeup(struct net_device *netdev, u32 queue_id, if (queue_id >= vsi->num_txq) return -ENXIO; - if (!vsi->xdp_rings[queue_id]->xsk_umem) + if (!vsi->xdp_rings[queue_id]->xsk_pool) return -ENXIO; ring = vsi->xdp_rings[queue_id]; @@ -833,20 +833,20 @@ ice_xsk_wakeup(struct net_device *netdev, u32 queue_id, } /** - * ice_xsk_any_rx_ring_ena - Checks if Rx rings have AF_XDP UMEM attached + * ice_xsk_any_rx_ring_ena - Checks if Rx rings have AF_XDP buff pool attached * @vsi: VSI to be checked * - * Returns true if any of the Rx rings has an AF_XDP UMEM attached + * Returns true if any of the Rx rings has an AF_XDP buff pool attached */ bool ice_xsk_any_rx_ring_ena(struct ice_vsi *vsi) { int i; - if (!vsi->xsk_umems) + if (!vsi->xsk_pools) return false; - for (i = 0; i < vsi->num_xsk_umems; i++) { - if (vsi->xsk_umems[i]) + for (i = 0; i < vsi->num_xsk_pools; i++) { + if (vsi->xsk_pools[i]) return true; } @@ -854,7 +854,7 @@ bool ice_xsk_any_rx_ring_ena(struct ice_vsi *vsi) } /** - * ice_xsk_clean_rx_ring - clean UMEM queues connected to a given Rx ring + * ice_xsk_clean_rx_ring - clean buffer pool queues connected to a given Rx ring * @rx_ring: ring to be cleaned */ void ice_xsk_clean_rx_ring(struct ice_ring *rx_ring) @@ -872,7 +872,7 @@ void ice_xsk_clean_rx_ring(struct ice_ring *rx_ring) } /** - * ice_xsk_clean_xdp_ring - Clean the XDP Tx ring and its UMEM queues + * ice_xsk_clean_xdp_ring - Clean the XDP Tx ring and its buffer pool queues * @xdp_ring: XDP_Tx ring */ void ice_xsk_clean_xdp_ring(struct ice_ring *xdp_ring) @@ -896,5 +896,5 @@ void ice_xsk_clean_xdp_ring(struct ice_ring *xdp_ring) } if (xsk_frames) - xsk_umem_complete_tx(xdp_ring->xsk_umem, xsk_frames); + xsk_tx_completed(xdp_ring->xsk_pool, xsk_frames); } diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.h b/drivers/net/ethernet/intel/ice/ice_xsk.h index fc1a06b4df36..fad783690134 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.h +++ b/drivers/net/ethernet/intel/ice/ice_xsk.h @@ -9,7 +9,8 @@ struct ice_vsi; #ifdef CONFIG_XDP_SOCKETS -int ice_xsk_umem_setup(struct ice_vsi *vsi, struct xdp_umem *umem, u16 qid); +int ice_xsk_pool_setup(struct ice_vsi *vsi, struct xsk_buff_pool *pool, + u16 qid); int ice_clean_rx_irq_zc(struct ice_ring *rx_ring, int budget); bool ice_clean_tx_irq_zc(struct ice_ring *xdp_ring, int budget); int ice_xsk_wakeup(struct net_device *netdev, u32 queue_id, u32 flags); @@ -19,8 +20,8 @@ void ice_xsk_clean_rx_ring(struct ice_ring *rx_ring); void ice_xsk_clean_xdp_ring(struct ice_ring *xdp_ring); #else static inline int -ice_xsk_umem_setup(struct ice_vsi __always_unused *vsi, - struct xdp_umem __always_unused *umem, +ice_xsk_pool_setup(struct ice_vsi __always_unused *vsi, + struct xsk_buff_pool __always_unused *pool, u16 __always_unused qid) { return -EOPNOTSUPP; diff --git a/drivers/net/ethernet/intel/igb/e1000_82575.c b/drivers/net/ethernet/intel/igb/e1000_82575.c index a32391e82762..50863fd87d53 100644 --- a/drivers/net/ethernet/intel/igb/e1000_82575.c +++ b/drivers/net/ethernet/intel/igb/e1000_82575.c @@ -2554,7 +2554,7 @@ out: /** * __igb_access_emi_reg - Read/write EMI register * @hw: pointer to the HW structure - * @addr: EMI address to program + * @address: EMI address to program * @data: pointer to value to read/write from/to the EMI address * @read: boolean flag to indicate read or write **/ @@ -2590,7 +2590,7 @@ s32 igb_read_emi_reg(struct e1000_hw *hw, u16 addr, u16 *data) * igb_set_eee_i350 - Enable/disable EEE support * @hw: pointer to the HW structure * @adv1G: boolean flag enabling 1G EEE advertisement - * @adv100m: boolean flag enabling 100M EEE advertisement + * @adv100M: boolean flag enabling 100M EEE advertisement * * Enable/disable EEE based on setting in dev_spec structure. * @@ -2646,7 +2646,7 @@ out: * igb_set_eee_i354 - Enable/disable EEE support * @hw: pointer to the HW structure * @adv1G: boolean flag enabling 1G EEE advertisement - * @adv100m: boolean flag enabling 100M EEE advertisement + * @adv100M: boolean flag enabling 100M EEE advertisement * * Enable/disable EEE legacy mode based on setting in dev_spec structure. * diff --git a/drivers/net/ethernet/intel/igb/e1000_i210.c b/drivers/net/ethernet/intel/igb/e1000_i210.c index c393cb2c0f16..9265901455cd 100644 --- a/drivers/net/ethernet/intel/igb/e1000_i210.c +++ b/drivers/net/ethernet/intel/igb/e1000_i210.c @@ -357,13 +357,14 @@ static s32 igb_read_invm_word_i210(struct e1000_hw *hw, u8 address, u16 *data) /** * igb_read_invm_i210 - Read invm wrapper function for I210/I211 * @hw: pointer to the HW structure - * @words: number of words to read + * @offset: offset to read from + * @words: number of words to read (unused) * @data: pointer to the data read * * Wrapper function to return data formerly found in the NVM. **/ static s32 igb_read_invm_i210(struct e1000_hw *hw, u16 offset, - u16 words __always_unused, u16 *data) + u16 __always_unused words, u16 *data) { s32 ret_val = 0; diff --git a/drivers/net/ethernet/intel/igb/e1000_mac.c b/drivers/net/ethernet/intel/igb/e1000_mac.c index 3254737c07a3..fd8eb2f9ab9d 100644 --- a/drivers/net/ethernet/intel/igb/e1000_mac.c +++ b/drivers/net/ethernet/intel/igb/e1000_mac.c @@ -166,6 +166,7 @@ static s32 igb_find_vlvf_slot(struct e1000_hw *hw, u32 vlan, bool vlvf_bypass) * @vlan: VLAN id to add or remove * @vind: VMDq output index that maps queue to VLAN id * @vlan_on: if true add filter, if false remove + * @vlvf_bypass: skip VLVF if no match is found * * Sets or clears a bit in the VLAN filter table array based on VLAN id * and if we are adding or removing the filter diff --git a/drivers/net/ethernet/intel/igb/e1000_mbx.c b/drivers/net/ethernet/intel/igb/e1000_mbx.c index 46debd991bfe..33cceb77e960 100644 --- a/drivers/net/ethernet/intel/igb/e1000_mbx.c +++ b/drivers/net/ethernet/intel/igb/e1000_mbx.c @@ -9,6 +9,7 @@ * @msg: The message buffer * @size: Length of buffer * @mbx_id: id of mailbox to read + * @unlock: skip locking or not * * returns SUCCESS if it successfully read message from buffer **/ diff --git a/drivers/net/ethernet/intel/igb/igb.h b/drivers/net/ethernet/intel/igb/igb.h index 2f015b60a995..0286d2fceee4 100644 --- a/drivers/net/ethernet/intel/igb/igb.h +++ b/drivers/net/ethernet/intel/igb/igb.h @@ -19,6 +19,8 @@ #include <linux/pci.h> #include <linux/mdio.h> +#include <net/xdp.h> + struct igb_adapter; #define E1000_PCS_CFG_IGN_SD 1 @@ -79,6 +81,12 @@ struct igb_adapter; #define IGB_I210_RX_LATENCY_100 2213 #define IGB_I210_RX_LATENCY_1000 448 +/* XDP */ +#define IGB_XDP_PASS 0 +#define IGB_XDP_CONSUMED BIT(0) +#define IGB_XDP_TX BIT(1) +#define IGB_XDP_REDIR BIT(2) + struct vf_data_storage { unsigned char vf_mac_addresses[ETH_ALEN]; u16 vf_mc_hashes[IGB_MAX_VF_MC_ENTRIES]; @@ -132,17 +140,62 @@ struct vf_mac_filter { /* Supported Rx Buffer Sizes */ #define IGB_RXBUFFER_256 256 +#define IGB_RXBUFFER_1536 1536 #define IGB_RXBUFFER_2048 2048 #define IGB_RXBUFFER_3072 3072 #define IGB_RX_HDR_LEN IGB_RXBUFFER_256 #define IGB_TS_HDR_LEN 16 -#define IGB_SKB_PAD (NET_SKB_PAD + NET_IP_ALIGN) +/* Attempt to maximize the headroom available for incoming frames. We + * use a 2K buffer for receives and need 1536/1534 to store the data for + * the frame. This leaves us with 512 bytes of room. From that we need + * to deduct the space needed for the shared info and the padding needed + * to IP align the frame. + * + * Note: For cache line sizes 256 or larger this value is going to end + * up negative. In these cases we should fall back to the 3K + * buffers. + */ #if (PAGE_SIZE < 8192) -#define IGB_MAX_FRAME_BUILD_SKB \ - (SKB_WITH_OVERHEAD(IGB_RXBUFFER_2048) - IGB_SKB_PAD - IGB_TS_HDR_LEN) +#define IGB_MAX_FRAME_BUILD_SKB (IGB_RXBUFFER_1536 - NET_IP_ALIGN) +#define IGB_2K_TOO_SMALL_WITH_PADDING \ +((NET_SKB_PAD + IGB_TS_HDR_LEN + IGB_RXBUFFER_1536) > SKB_WITH_OVERHEAD(IGB_RXBUFFER_2048)) + +static inline int igb_compute_pad(int rx_buf_len) +{ + int page_size, pad_size; + + page_size = ALIGN(rx_buf_len, PAGE_SIZE / 2); + pad_size = SKB_WITH_OVERHEAD(page_size) - rx_buf_len; + + return pad_size; +} + +static inline int igb_skb_pad(void) +{ + int rx_buf_len; + + /* If a 2K buffer cannot handle a standard Ethernet frame then + * optimize padding for a 3K buffer instead of a 1.5K buffer. + * + * For a 3K buffer we need to add enough padding to allow for + * tailroom due to NET_IP_ALIGN possibly shifting us out of + * cache-line alignment. + */ + if (IGB_2K_TOO_SMALL_WITH_PADDING) + rx_buf_len = IGB_RXBUFFER_3072 + SKB_DATA_ALIGN(NET_IP_ALIGN); + else + rx_buf_len = IGB_RXBUFFER_1536; + + /* if needed make room for NET_IP_ALIGN */ + rx_buf_len -= NET_IP_ALIGN; + + return igb_compute_pad(rx_buf_len); +} + +#define IGB_SKB_PAD igb_skb_pad() #else -#define IGB_MAX_FRAME_BUILD_SKB (IGB_RXBUFFER_2048 - IGB_TS_HDR_LEN) +#define IGB_SKB_PAD (NET_SKB_PAD + NET_IP_ALIGN) #endif /* How many Rx Buffers do we bundle into one write to the hardware ? */ @@ -194,13 +247,22 @@ enum igb_tx_flags { #define IGB_SFF_ADDRESSING_MODE 0x4 #define IGB_SFF_8472_UNSUP 0x00 +enum igb_tx_buf_type { + IGB_TYPE_SKB = 0, + IGB_TYPE_XDP, +}; + /* wrapper around a pointer to a socket buffer, * so a DMA handle can be stored along with the buffer */ struct igb_tx_buffer { union e1000_adv_tx_desc *next_to_watch; unsigned long time_stamp; - struct sk_buff *skb; + enum igb_tx_buf_type type; + union { + struct sk_buff *skb; + struct xdp_frame *xdpf; + }; unsigned int bytecount; u16 gso_segs; __be16 protocol; @@ -248,6 +310,7 @@ struct igb_ring_container { struct igb_ring { struct igb_q_vector *q_vector; /* backlink to q_vector */ struct net_device *netdev; /* back pointer to net_device */ + struct bpf_prog *xdp_prog; struct device *dev; /* device pointer for dma mapping */ union { /* array of buffer info structs */ struct igb_tx_buffer *tx_buffer_info; @@ -288,6 +351,7 @@ struct igb_ring { struct u64_stats_sync rx_syncp; }; }; + struct xdp_rxq_info xdp_rxq; } ____cacheline_internodealigned_in_smp; struct igb_q_vector { @@ -339,7 +403,7 @@ static inline unsigned int igb_rx_bufsz(struct igb_ring *ring) return IGB_RXBUFFER_3072; if (ring_uses_build_skb(ring)) - return IGB_MAX_FRAME_BUILD_SKB + IGB_TS_HDR_LEN; + return IGB_MAX_FRAME_BUILD_SKB; #endif return IGB_RXBUFFER_2048; } @@ -467,6 +531,7 @@ struct igb_adapter { unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)]; struct net_device *netdev; + struct bpf_prog *xdp_prog; unsigned long state; unsigned int flags; @@ -643,6 +708,9 @@ enum igb_boards { extern char igb_driver_name[]; +int igb_xmit_xdp_ring(struct igb_adapter *adapter, + struct igb_ring *ring, + struct xdp_frame *xdpf); int igb_open(struct net_device *netdev); int igb_close(struct net_device *netdev); int igb_up(struct igb_adapter *); diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c index 6e8231c1ddf0..28baf203459a 100644 --- a/drivers/net/ethernet/intel/igb/igb_ethtool.c +++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c @@ -961,6 +961,10 @@ static int igb_set_ringparam(struct net_device *netdev, memcpy(&temp_ring[i], adapter->rx_ring[i], sizeof(struct igb_ring)); + /* Clear copied XDP RX-queue info */ + memset(&temp_ring[i].xdp_rxq, 0, + sizeof(temp_ring[i].xdp_rxq)); + temp_ring[i].count = new_rx_count; err = igb_setup_rx_resources(&temp_ring[i]); if (err) { diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index d9c3a6b169f9..5fc2c381da55 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -30,6 +30,8 @@ #include <linux/if_ether.h> #include <linux/aer.h> #include <linux/prefetch.h> +#include <linux/bpf.h> +#include <linux/bpf_trace.h> #include <linux/pm_runtime.h> #include <linux/etherdevice.h> #ifdef CONFIG_IGB_DCA @@ -549,8 +551,7 @@ exit: /** * igb_get_i2c_data - Reads the I2C SDA data bit - * @hw: pointer to hardware structure - * @i2cctl: Current value of I2CCTL register + * @data: opaque pointer to adapter struct * * Returns the I2C data bit value **/ @@ -2220,7 +2221,6 @@ void igb_down(struct igb_adapter *adapter) void igb_reinit_locked(struct igb_adapter *adapter) { - WARN_ON(in_interrupt()); while (test_and_set_bit(__IGB_RESETTING, &adapter->state)) usleep_range(1000, 2000); igb_down(adapter); @@ -2824,6 +2824,147 @@ static int igb_setup_tc(struct net_device *dev, enum tc_setup_type type, } } +static int igb_xdp_setup(struct net_device *dev, struct bpf_prog *prog) +{ + int i, frame_size = dev->mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN; + struct igb_adapter *adapter = netdev_priv(dev); + bool running = netif_running(dev); + struct bpf_prog *old_prog; + bool need_reset; + + /* verify igb ring attributes are sufficient for XDP */ + for (i = 0; i < adapter->num_rx_queues; i++) { + struct igb_ring *ring = adapter->rx_ring[i]; + + if (frame_size > igb_rx_bufsz(ring)) + return -EINVAL; + } + + old_prog = xchg(&adapter->xdp_prog, prog); + need_reset = (!!prog != !!old_prog); + + /* device is up and bpf is added/removed, must setup the RX queues */ + if (need_reset && running) { + igb_close(dev); + } else { + for (i = 0; i < adapter->num_rx_queues; i++) + (void)xchg(&adapter->rx_ring[i]->xdp_prog, + adapter->xdp_prog); + } + + if (old_prog) + bpf_prog_put(old_prog); + + /* bpf is just replaced, RXQ and MTU are already setup */ + if (!need_reset) + return 0; + + if (running) + igb_open(dev); + + return 0; +} + +static int igb_xdp(struct net_device *dev, struct netdev_bpf *xdp) +{ + switch (xdp->command) { + case XDP_SETUP_PROG: + return igb_xdp_setup(dev, xdp->prog); + default: + return -EINVAL; + } +} + +static void igb_xdp_ring_update_tail(struct igb_ring *ring) +{ + /* Force memory writes to complete before letting h/w know there + * are new descriptors to fetch. + */ + wmb(); + writel(ring->next_to_use, ring->tail); +} + +static struct igb_ring *igb_xdp_tx_queue_mapping(struct igb_adapter *adapter) +{ + unsigned int r_idx = smp_processor_id(); + + if (r_idx >= adapter->num_tx_queues) + r_idx = r_idx % adapter->num_tx_queues; + + return adapter->tx_ring[r_idx]; +} + +static int igb_xdp_xmit_back(struct igb_adapter *adapter, struct xdp_buff *xdp) +{ + struct xdp_frame *xdpf = xdp_convert_buff_to_frame(xdp); + int cpu = smp_processor_id(); + struct igb_ring *tx_ring; + struct netdev_queue *nq; + u32 ret; + + if (unlikely(!xdpf)) + return IGB_XDP_CONSUMED; + + /* During program transitions its possible adapter->xdp_prog is assigned + * but ring has not been configured yet. In this case simply abort xmit. + */ + tx_ring = adapter->xdp_prog ? igb_xdp_tx_queue_mapping(adapter) : NULL; + if (unlikely(!tx_ring)) + return -ENXIO; + + nq = txring_txq(tx_ring); + __netif_tx_lock(nq, cpu); + ret = igb_xmit_xdp_ring(adapter, tx_ring, xdpf); + __netif_tx_unlock(nq); + + return ret; +} + +static int igb_xdp_xmit(struct net_device *dev, int n, + struct xdp_frame **frames, u32 flags) +{ + struct igb_adapter *adapter = netdev_priv(dev); + int cpu = smp_processor_id(); + struct igb_ring *tx_ring; + struct netdev_queue *nq; + int drops = 0; + int i; + + if (unlikely(test_bit(__IGB_DOWN, &adapter->state))) + return -ENETDOWN; + + if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK)) + return -EINVAL; + + /* During program transitions its possible adapter->xdp_prog is assigned + * but ring has not been configured yet. In this case simply abort xmit. + */ + tx_ring = adapter->xdp_prog ? igb_xdp_tx_queue_mapping(adapter) : NULL; + if (unlikely(!tx_ring)) + return -ENXIO; + + nq = txring_txq(tx_ring); + __netif_tx_lock(nq, cpu); + + for (i = 0; i < n; i++) { + struct xdp_frame *xdpf = frames[i]; + int err; + + err = igb_xmit_xdp_ring(adapter, tx_ring, xdpf); + if (err != IGB_XDP_TX) { + xdp_return_frame_rx_napi(xdpf); + drops++; + } + } + + __netif_tx_unlock(nq); + + if (unlikely(flags & XDP_XMIT_FLUSH)) + igb_xdp_ring_update_tail(tx_ring); + + return n - drops; +} + static const struct net_device_ops igb_netdev_ops = { .ndo_open = igb_open, .ndo_stop = igb_close, @@ -2848,6 +2989,8 @@ static const struct net_device_ops igb_netdev_ops = { .ndo_fdb_add = igb_ndo_fdb_add, .ndo_features_check = igb_features_check, .ndo_setup_tc = igb_setup_tc, + .ndo_bpf = igb_xdp, + .ndo_xdp_xmit = igb_xdp_xmit, }; /** @@ -3388,7 +3531,9 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent) "Width x1" : "unknown"), netdev->dev_addr); } - if ((hw->mac.type >= e1000_i210 || + if ((hw->mac.type == e1000_82576 && + rd32(E1000_EECD) & E1000_EECD_PRES) || + (hw->mac.type >= e1000_i210 || igb_get_flash_presence_i210(hw))) { ret_val = igb_read_part_string(hw, part_str, E1000_PBANUM_LENGTH); @@ -3868,6 +4013,7 @@ static int igb_sw_init(struct igb_adapter *adapter) /** * igb_open - Called when a network interface is made active * @netdev: network interface device structure + * @resuming: indicates whether we are in a resume call * * Returns 0 on success, negative value on failure * @@ -3985,6 +4131,7 @@ int igb_open(struct net_device *netdev) /** * igb_close - Disables a network interface * @netdev: network interface device structure + * @suspending: indicates we are in a suspend call * * Returns 0, this is not allowed to fail * @@ -4178,6 +4325,7 @@ static void igb_configure_tx(struct igb_adapter *adapter) **/ int igb_setup_rx_resources(struct igb_ring *rx_ring) { + struct igb_adapter *adapter = netdev_priv(rx_ring->netdev); struct device *dev = rx_ring->dev; int size; @@ -4200,6 +4348,13 @@ int igb_setup_rx_resources(struct igb_ring *rx_ring) rx_ring->next_to_clean = 0; rx_ring->next_to_use = 0; + rx_ring->xdp_prog = adapter->xdp_prog; + + /* XDP RX-queue info */ + if (xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev, + rx_ring->queue_index) < 0) + goto err; + return 0; err: @@ -4504,6 +4659,10 @@ void igb_configure_rx_ring(struct igb_adapter *adapter, int reg_idx = ring->reg_idx; u32 rxdctl = 0; + xdp_rxq_info_unreg_mem_model(&ring->xdp_rxq); + WARN_ON(xdp_rxq_info_reg_mem_model(&ring->xdp_rxq, + MEM_TYPE_PAGE_SHARED, NULL)); + /* disable the queue */ wr32(E1000_RXDCTL(reg_idx), 0); @@ -4708,6 +4867,8 @@ void igb_free_rx_resources(struct igb_ring *rx_ring) { igb_clean_rx_ring(rx_ring); + rx_ring->xdp_prog = NULL; + xdp_rxq_info_unreg(&rx_ring->xdp_rxq); vfree(rx_ring->rx_buffer_info); rx_ring->rx_buffer_info = NULL; @@ -5219,7 +5380,7 @@ static void igb_check_lvmmc(struct igb_adapter *adapter) /** * igb_watchdog - Timer Call-back - * @data: pointer to adapter cast into an unsigned long + * @t: pointer to timer_list containing our private info pointer **/ static void igb_watchdog(struct timer_list *t) { @@ -6077,6 +6238,80 @@ dma_error: return -1; } +int igb_xmit_xdp_ring(struct igb_adapter *adapter, + struct igb_ring *tx_ring, + struct xdp_frame *xdpf) +{ + union e1000_adv_tx_desc *tx_desc; + u32 len, cmd_type, olinfo_status; + struct igb_tx_buffer *tx_buffer; + dma_addr_t dma; + u16 i; + + len = xdpf->len; + + if (unlikely(!igb_desc_unused(tx_ring))) + return IGB_XDP_CONSUMED; + + dma = dma_map_single(tx_ring->dev, xdpf->data, len, DMA_TO_DEVICE); + if (dma_mapping_error(tx_ring->dev, dma)) + return IGB_XDP_CONSUMED; + + /* record the location of the first descriptor for this packet */ + tx_buffer = &tx_ring->tx_buffer_info[tx_ring->next_to_use]; + tx_buffer->bytecount = len; + tx_buffer->gso_segs = 1; + tx_buffer->protocol = 0; + + i = tx_ring->next_to_use; + tx_desc = IGB_TX_DESC(tx_ring, i); + + dma_unmap_len_set(tx_buffer, len, len); + dma_unmap_addr_set(tx_buffer, dma, dma); + tx_buffer->type = IGB_TYPE_XDP; + tx_buffer->xdpf = xdpf; + + tx_desc->read.buffer_addr = cpu_to_le64(dma); + + /* put descriptor type bits */ + cmd_type = E1000_ADVTXD_DTYP_DATA | + E1000_ADVTXD_DCMD_DEXT | + E1000_ADVTXD_DCMD_IFCS; + cmd_type |= len | IGB_TXD_DCMD; + tx_desc->read.cmd_type_len = cpu_to_le32(cmd_type); + + olinfo_status = cpu_to_le32(len << E1000_ADVTXD_PAYLEN_SHIFT); + /* 82575 requires a unique index per ring */ + if (test_bit(IGB_RING_FLAG_TX_CTX_IDX, &tx_ring->flags)) + olinfo_status |= tx_ring->reg_idx << 4; + + tx_desc->read.olinfo_status = olinfo_status; + + netdev_tx_sent_queue(txring_txq(tx_ring), tx_buffer->bytecount); + + /* set the timestamp */ + tx_buffer->time_stamp = jiffies; + + /* Avoid any potential race with xdp_xmit and cleanup */ + smp_wmb(); + + /* set next_to_watch value indicating a packet is present */ + i++; + if (i == tx_ring->count) + i = 0; + + tx_buffer->next_to_watch = tx_desc; + tx_ring->next_to_use = i; + + /* Make sure there is space in the ring for the next send. */ + igb_maybe_stop_tx(tx_ring, DESC_NEEDED); + + if (netif_xmit_stopped(txring_txq(tx_ring)) || !netdev_xmit_more()) + writel(i, tx_ring->tail); + + return IGB_XDP_TX; +} + netdev_tx_t igb_xmit_frame_ring(struct sk_buff *skb, struct igb_ring *tx_ring) { @@ -6105,6 +6340,7 @@ netdev_tx_t igb_xmit_frame_ring(struct sk_buff *skb, /* record the location of the first descriptor for this packet */ first = &tx_ring->tx_buffer_info[tx_ring->next_to_use]; + first->type = IGB_TYPE_SKB; first->skb = skb; first->bytecount = skb->len; first->gso_segs = 1; @@ -6192,8 +6428,9 @@ static netdev_tx_t igb_xmit_frame(struct sk_buff *skb, /** * igb_tx_timeout - Respond to a Tx Hang * @netdev: network interface device structure + * @txqueue: number of the Tx queue that hung (unused) **/ -static void igb_tx_timeout(struct net_device *netdev, unsigned int txqueue) +static void igb_tx_timeout(struct net_device *netdev, unsigned int __always_unused txqueue) { struct igb_adapter *adapter = netdev_priv(netdev); struct e1000_hw *hw = &adapter->hw; @@ -6256,6 +6493,19 @@ static int igb_change_mtu(struct net_device *netdev, int new_mtu) struct igb_adapter *adapter = netdev_priv(netdev); int max_frame = new_mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN; + if (adapter->xdp_prog) { + int i; + + for (i = 0; i < adapter->num_rx_queues; i++) { + struct igb_ring *ring = adapter->rx_ring[i]; + + if (max_frame > igb_rx_bufsz(ring)) { + netdev_warn(adapter->netdev, "Requested MTU size is not supported with XDP\n"); + return -EINVAL; + } + } + } + /* adjust max frame to be at least the size of a standard frame */ if (max_frame < (ETH_FRAME_LEN + ETH_FCS_LEN)) max_frame = ETH_FRAME_LEN + ETH_FCS_LEN; @@ -7809,7 +8059,10 @@ static bool igb_clean_tx_irq(struct igb_q_vector *q_vector, int napi_budget) total_packets += tx_buffer->gso_segs; /* free the skb */ - napi_consume_skb(tx_buffer->skb, napi_budget); + if (tx_buffer->type == IGB_TYPE_SKB) + napi_consume_skb(tx_buffer->skb, napi_budget); + else + xdp_return_frame(tx_buffer->xdpf); /* unmap skb header data */ dma_unmap_single(tx_ring->dev, @@ -7993,8 +8246,8 @@ static bool igb_can_reuse_rx_page(struct igb_rx_buffer *rx_buffer) * the pagecnt_bias and page count so that we fully restock the * number of references the driver holds. */ - if (unlikely(!pagecnt_bias)) { - page_ref_add(page, USHRT_MAX); + if (unlikely(pagecnt_bias == 1)) { + page_ref_add(page, USHRT_MAX - 1); rx_buffer->pagecnt_bias = USHRT_MAX; } @@ -8033,23 +8286,21 @@ static void igb_add_rx_frag(struct igb_ring *rx_ring, static struct sk_buff *igb_construct_skb(struct igb_ring *rx_ring, struct igb_rx_buffer *rx_buffer, - union e1000_adv_rx_desc *rx_desc, - unsigned int size) + struct xdp_buff *xdp, + union e1000_adv_rx_desc *rx_desc) { - void *va = page_address(rx_buffer->page) + rx_buffer->page_offset; #if (PAGE_SIZE < 8192) unsigned int truesize = igb_rx_pg_size(rx_ring) / 2; #else - unsigned int truesize = SKB_DATA_ALIGN(size); + unsigned int truesize = SKB_DATA_ALIGN(xdp->data_end - + xdp->data_hard_start); #endif + unsigned int size = xdp->data_end - xdp->data; unsigned int headlen; struct sk_buff *skb; /* prefetch first cache line of first page */ - prefetch(va); -#if L1_CACHE_BYTES < 128 - prefetch(va + L1_CACHE_BYTES); -#endif + net_prefetch(xdp->data); /* allocate a skb to store the frags */ skb = napi_alloc_skb(&rx_ring->q_vector->napi, IGB_RX_HDR_LEN); @@ -8057,24 +8308,24 @@ static struct sk_buff *igb_construct_skb(struct igb_ring *rx_ring, return NULL; if (unlikely(igb_test_staterr(rx_desc, E1000_RXDADV_STAT_TSIP))) { - igb_ptp_rx_pktstamp(rx_ring->q_vector, va, skb); - va += IGB_TS_HDR_LEN; + igb_ptp_rx_pktstamp(rx_ring->q_vector, xdp->data, skb); + xdp->data += IGB_TS_HDR_LEN; size -= IGB_TS_HDR_LEN; } /* Determine available headroom for copy */ headlen = size; if (headlen > IGB_RX_HDR_LEN) - headlen = eth_get_headlen(skb->dev, va, IGB_RX_HDR_LEN); + headlen = eth_get_headlen(skb->dev, xdp->data, IGB_RX_HDR_LEN); /* align pull length to size of long to optimize memcpy performance */ - memcpy(__skb_put(skb, headlen), va, ALIGN(headlen, sizeof(long))); + memcpy(__skb_put(skb, headlen), xdp->data, ALIGN(headlen, sizeof(long))); /* update all of the pointers */ size -= headlen; if (size) { skb_add_rx_frag(skb, 0, rx_buffer->page, - (va + headlen) - page_address(rx_buffer->page), + (xdp->data + headlen) - page_address(rx_buffer->page), size, truesize); #if (PAGE_SIZE < 8192) rx_buffer->page_offset ^= truesize; @@ -8090,32 +8341,29 @@ static struct sk_buff *igb_construct_skb(struct igb_ring *rx_ring, static struct sk_buff *igb_build_skb(struct igb_ring *rx_ring, struct igb_rx_buffer *rx_buffer, - union e1000_adv_rx_desc *rx_desc, - unsigned int size) + struct xdp_buff *xdp, + union e1000_adv_rx_desc *rx_desc) { - void *va = page_address(rx_buffer->page) + rx_buffer->page_offset; #if (PAGE_SIZE < 8192) unsigned int truesize = igb_rx_pg_size(rx_ring) / 2; #else unsigned int truesize = SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) + - SKB_DATA_ALIGN(IGB_SKB_PAD + size); + SKB_DATA_ALIGN(xdp->data_end - + xdp->data_hard_start); #endif struct sk_buff *skb; /* prefetch first cache line of first page */ - prefetch(va); -#if L1_CACHE_BYTES < 128 - prefetch(va + L1_CACHE_BYTES); -#endif + net_prefetch(xdp->data_meta); /* build an skb around the page buffer */ - skb = build_skb(va - IGB_SKB_PAD, truesize); + skb = build_skb(xdp->data_hard_start, truesize); if (unlikely(!skb)) return NULL; /* update pointers within the skb to store the data */ - skb_reserve(skb, IGB_SKB_PAD); - __skb_put(skb, size); + skb_reserve(skb, xdp->data - xdp->data_hard_start); + __skb_put(skb, xdp->data_end - xdp->data); /* pull timestamp out of packet data */ if (igb_test_staterr(rx_desc, E1000_RXDADV_STAT_TSIP)) { @@ -8133,6 +8381,79 @@ static struct sk_buff *igb_build_skb(struct igb_ring *rx_ring, return skb; } +static struct sk_buff *igb_run_xdp(struct igb_adapter *adapter, + struct igb_ring *rx_ring, + struct xdp_buff *xdp) +{ + int err, result = IGB_XDP_PASS; + struct bpf_prog *xdp_prog; + u32 act; + + rcu_read_lock(); + xdp_prog = READ_ONCE(rx_ring->xdp_prog); + + if (!xdp_prog) + goto xdp_out; + + prefetchw(xdp->data_hard_start); /* xdp_frame write */ + + act = bpf_prog_run_xdp(xdp_prog, xdp); + switch (act) { + case XDP_PASS: + break; + case XDP_TX: + result = igb_xdp_xmit_back(adapter, xdp); + break; + case XDP_REDIRECT: + err = xdp_do_redirect(adapter->netdev, xdp, xdp_prog); + if (!err) + result = IGB_XDP_REDIR; + else + result = IGB_XDP_CONSUMED; + break; + default: + bpf_warn_invalid_xdp_action(act); + fallthrough; + case XDP_ABORTED: + trace_xdp_exception(rx_ring->netdev, xdp_prog, act); + fallthrough; + case XDP_DROP: + result = IGB_XDP_CONSUMED; + break; + } +xdp_out: + rcu_read_unlock(); + return ERR_PTR(-result); +} + +static unsigned int igb_rx_frame_truesize(struct igb_ring *rx_ring, + unsigned int size) +{ + unsigned int truesize; + +#if (PAGE_SIZE < 8192) + truesize = igb_rx_pg_size(rx_ring) / 2; /* Must be power-of-2 */ +#else + truesize = ring_uses_build_skb(rx_ring) ? + SKB_DATA_ALIGN(IGB_SKB_PAD + size) + + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) : + SKB_DATA_ALIGN(size); +#endif + return truesize; +} + +static void igb_rx_buffer_flip(struct igb_ring *rx_ring, + struct igb_rx_buffer *rx_buffer, + unsigned int size) +{ + unsigned int truesize = igb_rx_frame_truesize(rx_ring, size); +#if (PAGE_SIZE < 8192) + rx_buffer->page_offset ^= truesize; +#else + rx_buffer->page_offset += truesize; +#endif +} + static inline void igb_rx_checksum(struct igb_ring *ring, union e1000_adv_rx_desc *rx_desc, struct sk_buff *skb) @@ -8187,7 +8508,6 @@ static inline void igb_rx_hash(struct igb_ring *ring, * igb_is_non_eop - process handling of non-EOP buffers * @rx_ring: Rx ring being processed * @rx_desc: Rx descriptor for current buffer - * @skb: current socket buffer containing buffer in progress * * This function updates next to clean. If the buffer is an EOP buffer * this function exits returning false, otherwise it will place the @@ -8229,6 +8549,10 @@ static bool igb_cleanup_headers(struct igb_ring *rx_ring, union e1000_adv_rx_desc *rx_desc, struct sk_buff *skb) { + /* XDP packets use error pointer so abort at this point */ + if (IS_ERR(skb)) + return true; + if (unlikely((igb_test_staterr(rx_desc, E1000_RXDEXT_ERR_FRAME_ERR_MASK)))) { struct net_device *netdev = rx_ring->netdev; @@ -8287,6 +8611,11 @@ static void igb_process_skb_fields(struct igb_ring *rx_ring, skb->protocol = eth_type_trans(skb, rx_ring->netdev); } +static unsigned int igb_rx_offset(struct igb_ring *rx_ring) +{ + return ring_uses_build_skb(rx_ring) ? IGB_SKB_PAD : 0; +} + static struct igb_rx_buffer *igb_get_rx_buffer(struct igb_ring *rx_ring, const unsigned int size) { @@ -8330,10 +8659,20 @@ static void igb_put_rx_buffer(struct igb_ring *rx_ring, static int igb_clean_rx_irq(struct igb_q_vector *q_vector, const int budget) { + struct igb_adapter *adapter = q_vector->adapter; struct igb_ring *rx_ring = q_vector->rx.ring; struct sk_buff *skb = rx_ring->skb; unsigned int total_bytes = 0, total_packets = 0; u16 cleaned_count = igb_desc_unused(rx_ring); + unsigned int xdp_xmit = 0; + struct xdp_buff xdp; + + xdp.rxq = &rx_ring->xdp_rxq; + + /* Frame size depend on rx_ring setup when PAGE_SIZE=4K */ +#if (PAGE_SIZE < 8192) + xdp.frame_sz = igb_rx_frame_truesize(rx_ring, 0); +#endif while (likely(total_packets < budget)) { union e1000_adv_rx_desc *rx_desc; @@ -8360,13 +8699,38 @@ static int igb_clean_rx_irq(struct igb_q_vector *q_vector, const int budget) rx_buffer = igb_get_rx_buffer(rx_ring, size); /* retrieve a buffer from the ring */ - if (skb) + if (!skb) { + xdp.data = page_address(rx_buffer->page) + + rx_buffer->page_offset; + xdp.data_meta = xdp.data; + xdp.data_hard_start = xdp.data - + igb_rx_offset(rx_ring); + xdp.data_end = xdp.data + size; +#if (PAGE_SIZE > 4096) + /* At larger PAGE_SIZE, frame_sz depend on len size */ + xdp.frame_sz = igb_rx_frame_truesize(rx_ring, size); +#endif + skb = igb_run_xdp(adapter, rx_ring, &xdp); + } + + if (IS_ERR(skb)) { + unsigned int xdp_res = -PTR_ERR(skb); + + if (xdp_res & (IGB_XDP_TX | IGB_XDP_REDIR)) { + xdp_xmit |= xdp_res; + igb_rx_buffer_flip(rx_ring, rx_buffer, size); + } else { + rx_buffer->pagecnt_bias++; + } + total_packets++; + total_bytes += size; + } else if (skb) igb_add_rx_frag(rx_ring, rx_buffer, skb, size); else if (ring_uses_build_skb(rx_ring)) - skb = igb_build_skb(rx_ring, rx_buffer, rx_desc, size); + skb = igb_build_skb(rx_ring, rx_buffer, &xdp, rx_desc); else skb = igb_construct_skb(rx_ring, rx_buffer, - rx_desc, size); + &xdp, rx_desc); /* exit if we failed to retrieve a buffer */ if (!skb) { @@ -8406,6 +8770,15 @@ static int igb_clean_rx_irq(struct igb_q_vector *q_vector, const int budget) /* place incomplete frames back on ring for completion */ rx_ring->skb = skb; + if (xdp_xmit & IGB_XDP_REDIR) + xdp_do_flush_map(); + + if (xdp_xmit & IGB_XDP_TX) { + struct igb_ring *tx_ring = igb_xdp_tx_queue_mapping(adapter); + + igb_xdp_ring_update_tail(tx_ring); + } + u64_stats_update_begin(&rx_ring->rx_syncp); rx_ring->rx_stats.packets += total_packets; rx_ring->rx_stats.bytes += total_bytes; @@ -8419,11 +8792,6 @@ static int igb_clean_rx_irq(struct igb_q_vector *q_vector, const int budget) return total_packets; } -static inline unsigned int igb_rx_offset(struct igb_ring *rx_ring) -{ - return ring_uses_build_skb(rx_ring) ? IGB_SKB_PAD : 0; -} - static bool igb_alloc_mapped_page(struct igb_ring *rx_ring, struct igb_rx_buffer *bi) { @@ -8460,14 +8828,16 @@ static bool igb_alloc_mapped_page(struct igb_ring *rx_ring, bi->dma = dma; bi->page = page; bi->page_offset = igb_rx_offset(rx_ring); - bi->pagecnt_bias = 1; + page_ref_add(page, USHRT_MAX - 1); + bi->pagecnt_bias = USHRT_MAX; return true; } /** - * igb_alloc_rx_buffers - Replace used receive buffers; packet split - * @adapter: address of board private structure + * igb_alloc_rx_buffers - Replace used receive buffers + * @rx_ring: rx descriptor ring to allocate new receive buffers + * @cleaned_count: count of buffers to allocate **/ void igb_alloc_rx_buffers(struct igb_ring *rx_ring, u16 cleaned_count) { @@ -8536,9 +8906,9 @@ void igb_alloc_rx_buffers(struct igb_ring *rx_ring, u16 cleaned_count) /** * igb_mii_ioctl - - * @netdev: - * @ifreq: - * @cmd: + * @netdev: pointer to netdev struct + * @ifr: interface structure + * @cmd: ioctl command to execute **/ static int igb_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) { @@ -8566,9 +8936,9 @@ static int igb_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) /** * igb_ioctl - - * @netdev: - * @ifreq: - * @cmd: + * @netdev: pointer to netdev struct + * @ifr: interface structure + * @cmd: ioctl command to execute **/ static int igb_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) { diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c index 490368d3d03c..7cc5428c3b3d 100644 --- a/drivers/net/ethernet/intel/igb/igb_ptp.c +++ b/drivers/net/ethernet/intel/igb/igb_ptp.c @@ -957,8 +957,8 @@ void igb_ptp_rx_rgtstamp(struct igb_q_vector *q_vector, /** * igb_ptp_get_ts_config - get hardware time stamping config - * @netdev: - * @ifreq: + * @netdev: netdev struct + * @ifr: interface struct * * Get the hwtstamp_config settings to return to the user. Rather than attempt * to deconstruct the settings from the registers, just return a shadow copy @@ -1141,8 +1141,8 @@ static int igb_ptp_set_timestamp_mode(struct igb_adapter *adapter, /** * igb_ptp_set_ts_config - set hardware time stamping config - * @netdev: - * @ifreq: + * @netdev: netdev struct + * @ifr: interface struct * **/ int igb_ptp_set_ts_config(struct net_device *netdev, struct ifreq *ifr) diff --git a/drivers/net/ethernet/intel/igbvf/netdev.c b/drivers/net/ethernet/intel/igbvf/netdev.c index 19269f5d52bc..ee9f8c1dca83 100644 --- a/drivers/net/ethernet/intel/igbvf/netdev.c +++ b/drivers/net/ethernet/intel/igbvf/netdev.c @@ -61,7 +61,7 @@ static const struct igbvf_info *igbvf_info_tbl[] = { /** * igbvf_desc_unused - calculate if we have unused descriptors - * @rx_ring: address of receive ring structure + * @ring: address of receive ring structure **/ static int igbvf_desc_unused(struct igbvf_ring *ring) { @@ -74,6 +74,8 @@ static int igbvf_desc_unused(struct igbvf_ring *ring) /** * igbvf_receive_skb - helper function to handle Rx indications * @adapter: board private structure + * @netdev: pointer to netdev struct + * @skb: skb to indicate to stack * @status: descriptor status field as written by hardware * @vlan: descriptor vlan field as written by hardware (no le/be conversion) * @skb: pointer to sk_buff to be indicated to stack @@ -233,6 +235,8 @@ no_buffers: /** * igbvf_clean_rx_irq - Send received data up the network stack; legacy * @adapter: board private structure + * @work_done: output parameter used to indicate completed work + * @work_to_do: input parameter setting limit of work * * the return value indicates whether actual cleaning was done, there * is no guarantee that everything was cleaned @@ -406,6 +410,7 @@ static void igbvf_put_txbuf(struct igbvf_adapter *adapter, /** * igbvf_setup_tx_resources - allocate Tx resources (Descriptors) * @adapter: board private structure + * @tx_ring: ring being initialized * * Return 0 on success, negative on failure **/ @@ -444,6 +449,7 @@ err: /** * igbvf_setup_rx_resources - allocate Rx resources (Descriptors) * @adapter: board private structure + * @rx_ring: ring being initialized * * Returns 0 on success, negative on failure **/ @@ -540,7 +546,7 @@ void igbvf_free_tx_resources(struct igbvf_ring *tx_ring) /** * igbvf_clean_rx_ring - Free Rx Buffers per Queue - * @adapter: board private structure + * @rx_ring: ring structure pointer to free buffers from **/ static void igbvf_clean_rx_ring(struct igbvf_ring *rx_ring) { @@ -760,7 +766,7 @@ static void igbvf_set_itr(struct igbvf_adapter *adapter) /** * igbvf_clean_tx_irq - Reclaim resources after transmit completes - * @adapter: board private structure + * @tx_ring: ring structure to clean descriptors from * * returns true if ring is completely cleaned **/ @@ -1891,7 +1897,7 @@ static bool igbvf_has_link(struct igbvf_adapter *adapter) /** * igbvf_watchdog - Timer Call-back - * @data: pointer to adapter cast into an unsigned long + * @t: timer list pointer containing private struct **/ static void igbvf_watchdog(struct timer_list *t) { @@ -2372,8 +2378,9 @@ static netdev_tx_t igbvf_xmit_frame(struct sk_buff *skb, /** * igbvf_tx_timeout - Respond to a Tx Hang * @netdev: network interface device structure + * @txqueue: queue timing out (unused) **/ -static void igbvf_tx_timeout(struct net_device *netdev, unsigned int txqueue) +static void igbvf_tx_timeout(struct net_device *netdev, unsigned int __always_unused txqueue) { struct igbvf_adapter *adapter = netdev_priv(netdev); diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h index 2d566f3c827b..35baae900c1f 100644 --- a/drivers/net/ethernet/intel/igc/igc.h +++ b/drivers/net/ethernet/intel/igc/igc.h @@ -215,6 +215,8 @@ struct igc_adapter { spinlock_t tmreg_lock; struct cyclecounter cc; struct timecounter tc; + struct timespec64 prev_ptp_time; /* Pre-reset PTP clock */ + ktime_t ptp_reset_start; /* Reset time in clock mono */ }; void igc_up(struct igc_adapter *adapter); @@ -548,6 +550,7 @@ void igc_ptp_rx_pktstamp(struct igc_q_vector *q_vector, void *va, int igc_ptp_set_ts_config(struct net_device *netdev, struct ifreq *ifr); int igc_ptp_get_ts_config(struct net_device *netdev, struct ifreq *ifr); void igc_ptp_tx_hang(struct igc_adapter *adapter); +void igc_ptp_read(struct igc_adapter *adapter, struct timespec64 *ts); #define igc_rx_pg_size(_ring) (PAGE_SIZE << igc_rx_pg_order(_ring)) diff --git a/drivers/net/ethernet/intel/igc/igc_base.c b/drivers/net/ethernet/intel/igc/igc_base.c index cc5a6cf531c7..fd37d2c203af 100644 --- a/drivers/net/ethernet/intel/igc/igc_base.c +++ b/drivers/net/ethernet/intel/igc/igc_base.c @@ -215,6 +215,11 @@ static s32 igc_get_invariants_base(struct igc_hw *hw) case IGC_DEV_ID_I225_K2: case IGC_DEV_ID_I225_LMVP: case IGC_DEV_ID_I225_IT: + case IGC_DEV_ID_I226_LM: + case IGC_DEV_ID_I226_V: + case IGC_DEV_ID_I226_IT: + case IGC_DEV_ID_I221_V: + case IGC_DEV_ID_I226_BLANK_NVM: case IGC_DEV_ID_I225_BLANK_NVM: mac->type = igc_i225; break; diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h index f1f464967f87..32f5fd684139 100644 --- a/drivers/net/ethernet/intel/igc/igc_defines.h +++ b/drivers/net/ethernet/intel/igc/igc_defines.h @@ -324,22 +324,10 @@ /* Advanced Receive Descriptor bit definitions */ #define IGC_RXDADV_STAT_TSIP 0x08000 /* timestamp in packet */ -#define IGC_RXDEXT_STATERR_CE 0x01000000 -#define IGC_RXDEXT_STATERR_SE 0x02000000 -#define IGC_RXDEXT_STATERR_SEQ 0x04000000 -#define IGC_RXDEXT_STATERR_CXE 0x10000000 -#define IGC_RXDEXT_STATERR_TCPE 0x20000000 +#define IGC_RXDEXT_STATERR_L4E 0x20000000 #define IGC_RXDEXT_STATERR_IPE 0x40000000 #define IGC_RXDEXT_STATERR_RXE 0x80000000 -/* Same mask, but for extended and packet split descriptors */ -#define IGC_RXDEXT_ERR_FRAME_ERR_MASK ( \ - IGC_RXDEXT_STATERR_CE | \ - IGC_RXDEXT_STATERR_SE | \ - IGC_RXDEXT_STATERR_SEQ | \ - IGC_RXDEXT_STATERR_CXE | \ - IGC_RXDEXT_STATERR_RXE) - #define IGC_MRQC_RSS_FIELD_IPV4_TCP 0x00010000 #define IGC_MRQC_RSS_FIELD_IPV4 0x00020000 #define IGC_MRQC_RSS_FIELD_IPV6_TCP_EX 0x00040000 @@ -409,7 +397,7 @@ #define IGC_IMIREXT_SIZE_BP 0x00001000 /* Packet size bypass */ /* Time Sync Transmit Control bit definitions */ -#define IGC_TSYNCTXCTL_VALID 0x00000001 /* Tx timestamp valid */ +#define IGC_TSYNCTXCTL_TXTT_0 0x00000001 /* Tx timestamp reg 0 valid */ #define IGC_TSYNCTXCTL_ENABLED 0x00000010 /* enable Tx timestamping */ #define IGC_TSYNCTXCTL_MAX_ALLOWED_DLY_MASK 0x0000F000 /* max delay */ #define IGC_TSYNCTXCTL_SYNC_COMP_ERR 0x20000000 /* sync err */ diff --git a/drivers/net/ethernet/intel/igc/igc_ethtool.c b/drivers/net/ethernet/intel/igc/igc_ethtool.c index 44410c2265d6..61d331ce38cd 100644 --- a/drivers/net/ethernet/intel/igc/igc_ethtool.c +++ b/drivers/net/ethernet/intel/igc/igc_ethtool.c @@ -321,6 +321,9 @@ static void igc_ethtool_get_regs(struct net_device *netdev, for (i = 0; i < 8; i++) regs_buff[205 + i] = rd32(IGC_ETQF(i)); + + regs_buff[213] = adapter->stats.tlpic; + regs_buff[214] = adapter->stats.rlpic; } static void igc_ethtool_get_wol(struct net_device *netdev, diff --git a/drivers/net/ethernet/intel/igc/igc_hw.h b/drivers/net/ethernet/intel/igc/igc_hw.h index b9fe51b91c47..55dae7c4703f 100644 --- a/drivers/net/ethernet/intel/igc/igc_hw.h +++ b/drivers/net/ethernet/intel/igc/igc_hw.h @@ -24,6 +24,11 @@ #define IGC_DEV_ID_I225_K2 0x3101 #define IGC_DEV_ID_I225_LMVP 0x5502 #define IGC_DEV_ID_I225_IT 0x0D9F +#define IGC_DEV_ID_I226_LM 0x125B +#define IGC_DEV_ID_I226_V 0x125C +#define IGC_DEV_ID_I226_IT 0x125D +#define IGC_DEV_ID_I221_V 0x125E +#define IGC_DEV_ID_I226_BLANK_NVM 0x125F #define IGC_DEV_ID_I225_BLANK_NVM 0x15FD /* Function pointers for the MAC. */ @@ -125,9 +130,6 @@ struct igc_nvm_info { struct igc_nvm_operations ops; enum igc_nvm_type type; - u32 flash_bank_size; - u32 flash_base_addr; - u16 word_size; u16 delay_usec; u16 address_bits; @@ -153,7 +155,6 @@ struct igc_phy_info { u8 mdix; bool is_mdix; - bool reset_disable; bool speed_downgraded; bool autoneg_wait_to_complete; }; @@ -239,6 +240,8 @@ struct igc_hw_stats { u64 prc511; u64 prc1023; u64 prc1522; + u64 tlpic; + u64 rlpic; u64 gprc; u64 bprc; u64 mprc; diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c index 9593aa4eea36..9112dff075cf 100644 --- a/drivers/net/ethernet/intel/igc/igc_main.c +++ b/drivers/net/ethernet/intel/igc/igc_main.c @@ -47,6 +47,11 @@ static const struct pci_device_id igc_pci_tbl[] = { { PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_K2), board_base }, { PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_LMVP), board_base }, { PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_IT), board_base }, + { PCI_VDEVICE(INTEL, IGC_DEV_ID_I226_LM), board_base }, + { PCI_VDEVICE(INTEL, IGC_DEV_ID_I226_V), board_base }, + { PCI_VDEVICE(INTEL, IGC_DEV_ID_I226_IT), board_base }, + { PCI_VDEVICE(INTEL, IGC_DEV_ID_I221_V), board_base }, + { PCI_VDEVICE(INTEL, IGC_DEV_ID_I226_BLANK_NVM), board_base }, { PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_BLANK_NVM), board_base }, /* required last entry */ {0, } @@ -1428,7 +1433,7 @@ static void igc_rx_checksum(struct igc_ring *ring, /* TCP/UDP checksum error bit is set */ if (igc_test_staterr(rx_desc, - IGC_RXDEXT_STATERR_TCPE | + IGC_RXDEXT_STATERR_L4E | IGC_RXDEXT_STATERR_IPE)) { /* work around errata with sctp packets where the TCPE aka * L4E bit is set incorrectly on 64 byte (60 byte w/o crc) @@ -1550,10 +1555,7 @@ static struct sk_buff *igc_build_skb(struct igc_ring *rx_ring, struct sk_buff *skb; /* prefetch first cache line of first page */ - prefetch(va); -#if L1_CACHE_BYTES < 128 - prefetch(va + L1_CACHE_BYTES); -#endif + net_prefetch(va); /* build an skb around the page buffer */ skb = build_skb(va - IGC_SKB_PAD, truesize); @@ -1589,10 +1591,7 @@ static struct sk_buff *igc_construct_skb(struct igc_ring *rx_ring, struct sk_buff *skb; /* prefetch first cache line of first page */ - prefetch(va); -#if L1_CACHE_BYTES < 128 - prefetch(va + L1_CACHE_BYTES); -#endif + net_prefetch(va); /* allocate a skb to store the frags */ skb = napi_alloc_skb(&rx_ring->q_vector->napi, IGC_RX_HDR_LEN); @@ -1743,8 +1742,7 @@ static bool igc_cleanup_headers(struct igc_ring *rx_ring, union igc_adv_rx_desc *rx_desc, struct sk_buff *skb) { - if (unlikely((igc_test_staterr(rx_desc, - IGC_RXDEXT_ERR_FRAME_ERR_MASK)))) { + if (unlikely(igc_test_staterr(rx_desc, IGC_RXDEXT_STATERR_RXE))) { struct net_device *netdev = rx_ring->netdev; if (!(netdev->features & NETIF_F_RXALL)) { @@ -3685,6 +3683,8 @@ void igc_update_stats(struct igc_adapter *adapter) adapter->stats.prc511 += rd32(IGC_PRC511); adapter->stats.prc1023 += rd32(IGC_PRC1023); adapter->stats.prc1522 += rd32(IGC_PRC1522); + adapter->stats.tlpic += rd32(IGC_TLPIC); + adapter->stats.rlpic += rd32(IGC_RLPIC); mpc = rd32(IGC_MPC); adapter->stats.mpc += mpc; @@ -3778,6 +3778,8 @@ void igc_down(struct igc_adapter *adapter) set_bit(__IGC_DOWN, &adapter->state); + igc_ptp_suspend(adapter); + /* disable receives in the hardware */ rctl = rd32(IGC_RCTL); wr32(IGC_RCTL, rctl & ~IGC_RCTL_EN); @@ -3831,7 +3833,6 @@ void igc_down(struct igc_adapter *adapter) void igc_reinit_locked(struct igc_adapter *adapter) { - WARN_ON(in_interrupt()); while (test_and_set_bit(__IGC_RESETTING, &adapter->state)) usleep_range(1000, 2000); igc_down(adapter); @@ -4659,7 +4660,7 @@ int igc_close(struct net_device *netdev) /** * igc_ioctl - Access the hwtstamp interface * @netdev: network interface device structure - * @ifreq: interface request data + * @ifr: interface request data * @cmd: ioctl command **/ static int igc_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) @@ -4700,14 +4701,35 @@ static int igc_save_launchtime_params(struct igc_adapter *adapter, int queue, return 0; } -static bool validate_schedule(const struct tc_taprio_qopt_offload *qopt) +static bool is_base_time_past(ktime_t base_time, const struct timespec64 *now) +{ + struct timespec64 b; + + b = ktime_to_timespec64(base_time); + + return timespec64_compare(now, &b) > 0; +} + +static bool validate_schedule(struct igc_adapter *adapter, + const struct tc_taprio_qopt_offload *qopt) { int queue_uses[IGC_MAX_TX_QUEUES] = { }; + struct timespec64 now; size_t n; if (qopt->cycle_time_extension) return false; + igc_ptp_read(adapter, &now); + + /* If we program the controller's BASET registers with a time + * in the future, it will hold all the packets until that + * time, causing a lot of TX Hangs, so to avoid that, we + * reject schedules that would start in the future. + */ + if (!is_base_time_past(qopt->base_time, &now)) + return false; + for (n = 0; n < qopt->num_entries; n++) { const struct tc_taprio_sched_entry *e; int i; @@ -4762,7 +4784,7 @@ static int igc_save_qbv_schedule(struct igc_adapter *adapter, if (adapter->base_time) return -EALREADY; - if (!validate_schedule(qopt)) + if (!validate_schedule(adapter, qopt)) return -EINVAL; adapter->cycle_time = qopt->cycle_time; diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c index 6a9b5102aa55..ac0b9c85da7c 100644 --- a/drivers/net/ethernet/intel/igc/igc_ptp.c +++ b/drivers/net/ethernet/intel/igc/igc_ptp.c @@ -8,6 +8,7 @@ #include <linux/pci.h> #include <linux/ptp_classify.h> #include <linux/clocksource.h> +#include <linux/ktime.h> #define INCVALUE_MASK 0x7fffffff #define ISGN 0x80000000 @@ -16,17 +17,12 @@ #define IGC_PTP_TX_TIMEOUT (HZ * 15) /* SYSTIM read access for I225 */ -static void igc_ptp_read_i225(struct igc_adapter *adapter, - struct timespec64 *ts) +void igc_ptp_read(struct igc_adapter *adapter, struct timespec64 *ts) { struct igc_hw *hw = &adapter->hw; u32 sec, nsec; - /* The timestamp latches on lowest register read. For I210/I211, the - * lowest register is SYSTIMR. Since we only need to provide nanosecond - * resolution, we can ignore it. - */ - rd32(IGC_SYSTIMR); + /* The timestamp is latched when SYSTIML is read. */ nsec = rd32(IGC_SYSTIML); sec = rd32(IGC_SYSTIMH); @@ -39,9 +35,6 @@ static void igc_ptp_write_i225(struct igc_adapter *adapter, { struct igc_hw *hw = &adapter->hw; - /* Writing the SYSTIMR register is not necessary as it only - * provides sub-nanosecond resolution. - */ wr32(IGC_SYSTIML, ts->tv_nsec); wr32(IGC_SYSTIMH, ts->tv_sec); } @@ -81,7 +74,7 @@ static int igc_ptp_adjtime_i225(struct ptp_clock_info *ptp, s64 delta) spin_lock_irqsave(&igc->tmreg_lock, flags); - igc_ptp_read_i225(igc, &now); + igc_ptp_read(igc, &now); now = timespec64_add(now, then); igc_ptp_write_i225(igc, (const struct timespec64 *)&now); @@ -102,10 +95,9 @@ static int igc_ptp_gettimex64_i225(struct ptp_clock_info *ptp, spin_lock_irqsave(&igc->tmreg_lock, flags); ptp_read_system_prets(sts); - rd32(IGC_SYSTIMR); - ptp_read_system_postts(sts); ts->tv_nsec = rd32(IGC_SYSTIML); ts->tv_sec = rd32(IGC_SYSTIMH); + ptp_read_system_postts(sts); spin_unlock_irqrestore(&igc->tmreg_lock, flags); @@ -422,24 +414,17 @@ static void igc_ptp_tx_work(struct work_struct *work) if (!test_bit(__IGC_PTP_TX_IN_PROGRESS, &adapter->state)) return; - if (time_is_before_jiffies(adapter->ptp_tx_start + - IGC_PTP_TX_TIMEOUT)) { - igc_ptp_tx_timeout(adapter); + tsynctxctl = rd32(IGC_TSYNCTXCTL); + if (WARN_ON_ONCE(!(tsynctxctl & IGC_TSYNCTXCTL_TXTT_0))) return; - } - tsynctxctl = rd32(IGC_TSYNCTXCTL); - if (tsynctxctl & IGC_TSYNCTXCTL_VALID) - igc_ptp_tx_hwtstamp(adapter); - else - /* reschedule to check later */ - schedule_work(&adapter->ptp_tx_work); + igc_ptp_tx_hwtstamp(adapter); } /** * igc_ptp_set_ts_config - set hardware time stamping config * @netdev: network interface device structure - * @ifreq: interface request data + * @ifr: interface request data * **/ int igc_ptp_set_ts_config(struct net_device *netdev, struct ifreq *ifr) @@ -466,7 +451,7 @@ int igc_ptp_set_ts_config(struct net_device *netdev, struct ifreq *ifr) /** * igc_ptp_get_ts_config - get hardware time stamping config * @netdev: network interface device structure - * @ifreq: interface request data + * @ifr: interface request data * * Get the hwtstamp_config settings to return to the user. Rather than attempt * to deconstruct the settings from the registers, just return a shadow copy @@ -515,6 +500,9 @@ void igc_ptp_init(struct igc_adapter *adapter) adapter->tstamp_config.rx_filter = HWTSTAMP_FILTER_NONE; adapter->tstamp_config.tx_type = HWTSTAMP_TX_OFF; + adapter->prev_ptp_time = ktime_to_timespec64(ktime_get_real()); + adapter->ptp_reset_start = ktime_get(); + adapter->ptp_clock = ptp_clock_register(&adapter->ptp_caps, &adapter->pdev->dev); if (IS_ERR(adapter->ptp_clock)) { @@ -526,6 +514,24 @@ void igc_ptp_init(struct igc_adapter *adapter) } } +static void igc_ptp_time_save(struct igc_adapter *adapter) +{ + igc_ptp_read(adapter, &adapter->prev_ptp_time); + adapter->ptp_reset_start = ktime_get(); +} + +static void igc_ptp_time_restore(struct igc_adapter *adapter) +{ + struct timespec64 ts = adapter->prev_ptp_time; + ktime_t delta; + + delta = ktime_sub(ktime_get(), adapter->ptp_reset_start); + + timespec64_add_ns(&ts, ktime_to_ns(delta)); + + igc_ptp_write_i225(adapter, &ts); +} + /** * igc_ptp_suspend - Disable PTP work items and prepare for suspend * @adapter: Board private structure @@ -542,6 +548,8 @@ void igc_ptp_suspend(struct igc_adapter *adapter) dev_kfree_skb_any(adapter->ptp_tx_skb); adapter->ptp_tx_skb = NULL; clear_bit_unlock(__IGC_PTP_TX_IN_PROGRESS, &adapter->state); + + igc_ptp_time_save(adapter); } /** @@ -591,9 +599,7 @@ void igc_ptp_reset(struct igc_adapter *adapter) /* Re-initialize the timer. */ if (hw->mac.type == igc_i225) { - struct timespec64 ts64 = ktime_to_timespec64(ktime_get_real()); - - igc_ptp_write_i225(adapter, &ts64); + igc_ptp_time_restore(adapter); } else { timecounter_init(&adapter->tc, &adapter->cc, ktime_to_ns(ktime_get_real())); diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_hw.c b/drivers/net/ethernet/intel/ixgb/ixgb_hw.c index cbaa933ef30d..a430871d1c27 100644 --- a/drivers/net/ethernet/intel/ixgb/ixgb_hw.c +++ b/drivers/net/ethernet/intel/ixgb/ixgb_hw.c @@ -98,7 +98,6 @@ bool ixgb_adapter_stop(struct ixgb_hw *hw) { u32 ctrl_reg; - u32 icr_reg; ENTER(); @@ -142,7 +141,7 @@ ixgb_adapter_stop(struct ixgb_hw *hw) IXGB_WRITE_REG(hw, IMC, 0xffffffff); /* Clear any pending interrupt events. */ - icr_reg = IXGB_READ_REG(hw, ICR); + IXGB_READ_REG(hw, ICR); return ctrl_reg & IXGB_CTRL0_RST; } @@ -274,7 +273,6 @@ bool ixgb_init_hw(struct ixgb_hw *hw) { u32 i; - u32 ctrl_reg; bool status; ENTER(); @@ -286,7 +284,7 @@ ixgb_init_hw(struct ixgb_hw *hw) */ pr_debug("Issuing a global reset to MAC\n"); - ctrl_reg = ixgb_mac_reset(hw); + ixgb_mac_reset(hw); pr_debug("Issuing an EE reset to MAC\n"); #ifdef HP_ZX1 @@ -949,8 +947,6 @@ bool ixgb_check_for_bad_link(struct ixgb_hw *hw) static void ixgb_clear_hw_cntrs(struct ixgb_hw *hw) { - volatile u32 temp_reg; - ENTER(); /* if we are stopped or resetting exit gracefully */ @@ -959,66 +955,66 @@ ixgb_clear_hw_cntrs(struct ixgb_hw *hw) return; } - temp_reg = IXGB_READ_REG(hw, TPRL); - temp_reg = IXGB_READ_REG(hw, TPRH); - temp_reg = IXGB_READ_REG(hw, GPRCL); - temp_reg = IXGB_READ_REG(hw, GPRCH); - temp_reg = IXGB_READ_REG(hw, BPRCL); - temp_reg = IXGB_READ_REG(hw, BPRCH); - temp_reg = IXGB_READ_REG(hw, MPRCL); - temp_reg = IXGB_READ_REG(hw, MPRCH); - temp_reg = IXGB_READ_REG(hw, UPRCL); - temp_reg = IXGB_READ_REG(hw, UPRCH); - temp_reg = IXGB_READ_REG(hw, VPRCL); - temp_reg = IXGB_READ_REG(hw, VPRCH); - temp_reg = IXGB_READ_REG(hw, JPRCL); - temp_reg = IXGB_READ_REG(hw, JPRCH); - temp_reg = IXGB_READ_REG(hw, GORCL); - temp_reg = IXGB_READ_REG(hw, GORCH); - temp_reg = IXGB_READ_REG(hw, TORL); - temp_reg = IXGB_READ_REG(hw, TORH); - temp_reg = IXGB_READ_REG(hw, RNBC); - temp_reg = IXGB_READ_REG(hw, RUC); - temp_reg = IXGB_READ_REG(hw, ROC); - temp_reg = IXGB_READ_REG(hw, RLEC); - temp_reg = IXGB_READ_REG(hw, CRCERRS); - temp_reg = IXGB_READ_REG(hw, ICBC); - temp_reg = IXGB_READ_REG(hw, ECBC); - temp_reg = IXGB_READ_REG(hw, MPC); - temp_reg = IXGB_READ_REG(hw, TPTL); - temp_reg = IXGB_READ_REG(hw, TPTH); - temp_reg = IXGB_READ_REG(hw, GPTCL); - temp_reg = IXGB_READ_REG(hw, GPTCH); - temp_reg = IXGB_READ_REG(hw, BPTCL); - temp_reg = IXGB_READ_REG(hw, BPTCH); - temp_reg = IXGB_READ_REG(hw, MPTCL); - temp_reg = IXGB_READ_REG(hw, MPTCH); - temp_reg = IXGB_READ_REG(hw, UPTCL); - temp_reg = IXGB_READ_REG(hw, UPTCH); - temp_reg = IXGB_READ_REG(hw, VPTCL); - temp_reg = IXGB_READ_REG(hw, VPTCH); - temp_reg = IXGB_READ_REG(hw, JPTCL); - temp_reg = IXGB_READ_REG(hw, JPTCH); - temp_reg = IXGB_READ_REG(hw, GOTCL); - temp_reg = IXGB_READ_REG(hw, GOTCH); - temp_reg = IXGB_READ_REG(hw, TOTL); - temp_reg = IXGB_READ_REG(hw, TOTH); - temp_reg = IXGB_READ_REG(hw, DC); - temp_reg = IXGB_READ_REG(hw, PLT64C); - temp_reg = IXGB_READ_REG(hw, TSCTC); - temp_reg = IXGB_READ_REG(hw, TSCTFC); - temp_reg = IXGB_READ_REG(hw, IBIC); - temp_reg = IXGB_READ_REG(hw, RFC); - temp_reg = IXGB_READ_REG(hw, LFC); - temp_reg = IXGB_READ_REG(hw, PFRC); - temp_reg = IXGB_READ_REG(hw, PFTC); - temp_reg = IXGB_READ_REG(hw, MCFRC); - temp_reg = IXGB_READ_REG(hw, MCFTC); - temp_reg = IXGB_READ_REG(hw, XONRXC); - temp_reg = IXGB_READ_REG(hw, XONTXC); - temp_reg = IXGB_READ_REG(hw, XOFFRXC); - temp_reg = IXGB_READ_REG(hw, XOFFTXC); - temp_reg = IXGB_READ_REG(hw, RJC); + IXGB_READ_REG(hw, TPRL); + IXGB_READ_REG(hw, TPRH); + IXGB_READ_REG(hw, GPRCL); + IXGB_READ_REG(hw, GPRCH); + IXGB_READ_REG(hw, BPRCL); + IXGB_READ_REG(hw, BPRCH); + IXGB_READ_REG(hw, MPRCL); + IXGB_READ_REG(hw, MPRCH); + IXGB_READ_REG(hw, UPRCL); + IXGB_READ_REG(hw, UPRCH); + IXGB_READ_REG(hw, VPRCL); + IXGB_READ_REG(hw, VPRCH); + IXGB_READ_REG(hw, JPRCL); + IXGB_READ_REG(hw, JPRCH); + IXGB_READ_REG(hw, GORCL); + IXGB_READ_REG(hw, GORCH); + IXGB_READ_REG(hw, TORL); + IXGB_READ_REG(hw, TORH); + IXGB_READ_REG(hw, RNBC); + IXGB_READ_REG(hw, RUC); + IXGB_READ_REG(hw, ROC); + IXGB_READ_REG(hw, RLEC); + IXGB_READ_REG(hw, CRCERRS); + IXGB_READ_REG(hw, ICBC); + IXGB_READ_REG(hw, ECBC); + IXGB_READ_REG(hw, MPC); + IXGB_READ_REG(hw, TPTL); + IXGB_READ_REG(hw, TPTH); + IXGB_READ_REG(hw, GPTCL); + IXGB_READ_REG(hw, GPTCH); + IXGB_READ_REG(hw, BPTCL); + IXGB_READ_REG(hw, BPTCH); + IXGB_READ_REG(hw, MPTCL); + IXGB_READ_REG(hw, MPTCH); + IXGB_READ_REG(hw, UPTCL); + IXGB_READ_REG(hw, UPTCH); + IXGB_READ_REG(hw, VPTCL); + IXGB_READ_REG(hw, VPTCH); + IXGB_READ_REG(hw, JPTCL); + IXGB_READ_REG(hw, JPTCH); + IXGB_READ_REG(hw, GOTCL); + IXGB_READ_REG(hw, GOTCH); + IXGB_READ_REG(hw, TOTL); + IXGB_READ_REG(hw, TOTH); + IXGB_READ_REG(hw, DC); + IXGB_READ_REG(hw, PLT64C); + IXGB_READ_REG(hw, TSCTC); + IXGB_READ_REG(hw, TSCTFC); + IXGB_READ_REG(hw, IBIC); + IXGB_READ_REG(hw, RFC); + IXGB_READ_REG(hw, LFC); + IXGB_READ_REG(hw, PFRC); + IXGB_READ_REG(hw, PFTC); + IXGB_READ_REG(hw, MCFRC); + IXGB_READ_REG(hw, MCFTC); + IXGB_READ_REG(hw, XONRXC); + IXGB_READ_REG(hw, XONTXC); + IXGB_READ_REG(hw, XOFFRXC); + IXGB_READ_REG(hw, XOFFTXC); + IXGB_READ_REG(hw, RJC); } /****************************************************************************** @@ -1161,18 +1157,13 @@ static void ixgb_optics_reset(struct ixgb_hw *hw) { if (hw->phy_type == ixgb_phy_type_txn17401) { - u16 mdio_reg; - ixgb_write_phy_reg(hw, MDIO_CTRL1, IXGB_PHY_ADDRESS, MDIO_MMD_PMAPMD, MDIO_CTRL1_RESET); - mdio_reg = ixgb_read_phy_reg(hw, - MDIO_CTRL1, - IXGB_PHY_ADDRESS, - MDIO_MMD_PMAPMD); + ixgb_read_phy_reg(hw, MDIO_CTRL1, IXGB_PHY_ADDRESS, MDIO_MMD_PMAPMD); } } diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_main.c b/drivers/net/ethernet/intel/ixgb/ixgb_main.c index 048351cf0e4a..1588376d4c67 100644 --- a/drivers/net/ethernet/intel/ixgb/ixgb_main.c +++ b/drivers/net/ethernet/intel/ixgb/ixgb_main.c @@ -1109,7 +1109,7 @@ alloc_failed: /** * ixgb_watchdog - Timer Call-back - * @data: pointer to netdev cast into an unsigned long + * @t: pointer to timer_list containing our private info pointer **/ static void @@ -1531,10 +1531,11 @@ ixgb_xmit_frame(struct sk_buff *skb, struct net_device *netdev) /** * ixgb_tx_timeout - Respond to a Tx Hang * @netdev: network interface device structure + * @txqueue: queue hanging (unused) **/ static void -ixgb_tx_timeout(struct net_device *netdev, unsigned int txqueue) +ixgb_tx_timeout(struct net_device *netdev, unsigned int __always_unused txqueue) { struct ixgb_adapter *adapter = netdev_priv(netdev); @@ -1746,7 +1747,8 @@ ixgb_intr(int irq, void *data) /** * ixgb_clean - NAPI Rx polling callback - * @adapter: board private structure + * @napi: napi struct pointer + * @budget: max number of receives to clean **/ static int @@ -1865,7 +1867,7 @@ ixgb_clean_tx_irq(struct ixgb_adapter *adapter) * ixgb_rx_checksum - Receive Checksum Offload for 82597. * @adapter: board private structure * @rx_desc: receive descriptor - * @sk_buff: socket buffer with received data + * @skb: socket buffer with received data **/ static void @@ -1923,6 +1925,8 @@ static void ixgb_check_copybreak(struct napi_struct *napi, /** * ixgb_clean_rx_irq - Send received data up the network stack, * @adapter: board private structure + * @work_done: output pointer to amount of packets cleaned + * @work_to_do: how much work we can complete **/ static bool @@ -2042,6 +2046,7 @@ rxdesc_done: /** * ixgb_alloc_rx_buffers - Replace used receive buffers * @adapter: address of board private structure + * @cleaned_count: how many buffers to allocate **/ static void @@ -2211,7 +2216,7 @@ static pci_ers_result_t ixgb_io_error_detected(struct pci_dev *pdev, /** * ixgb_io_slot_reset - called after the pci bus has been reset. - * @pdev pointer to pci device with error + * @pdev: pointer to pci device with error * * This callback is called after the PCI bus has been reset. * Basically, this tries to restart the card from scratch. @@ -2259,7 +2264,7 @@ static pci_ers_result_t ixgb_io_slot_reset(struct pci_dev *pdev) /** * ixgb_io_resume - called when its OK to resume normal operations - * @pdev pointer to pci device with error + * @pdev: pointer to pci device with error * * The error recovery driver tells us that its OK to resume * normal operation. Implementation resembles the second-half diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h index 1e8a809233a0..de0fc6ecf491 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h @@ -350,7 +350,7 @@ struct ixgbe_ring { struct ixgbe_rx_queue_stats rx_stats; }; struct xdp_rxq_info xdp_rxq; - struct xdp_umem *xsk_umem; + struct xsk_buff_pool *xsk_pool; u16 ring_idx; /* {rx,tx,xdp}_ring back reference idx */ u16 rx_buf_len; } ____cacheline_internodealigned_in_smp; diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c index 71ec908266a6..a280aa34ca1d 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c @@ -531,6 +531,16 @@ static int ixgbe_set_link_ksettings(struct net_device *netdev, return err; } +static void ixgbe_get_pause_stats(struct net_device *netdev, + struct ethtool_pause_stats *stats) +{ + struct ixgbe_adapter *adapter = netdev_priv(netdev); + struct ixgbe_hw_stats *hwstats = &adapter->stats; + + stats->tx_pause_frames = hwstats->lxontxc + hwstats->lxofftxc; + stats->rx_pause_frames = hwstats->lxonrxc + hwstats->lxoffrxc; +} + static void ixgbe_get_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) { @@ -3546,6 +3556,7 @@ static const struct ethtool_ops ixgbe_ethtool_ops = { .set_eeprom = ixgbe_set_eeprom, .get_ringparam = ixgbe_get_ringparam, .set_ringparam = ixgbe_set_ringparam, + .get_pause_stats = ixgbe_get_pause_stats, .get_pauseparam = ixgbe_get_pauseparam, .set_pauseparam = ixgbe_set_pauseparam, .get_msglevel = ixgbe_get_msglevel, diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c index 2e35c5706cf1..df389a11d3af 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c @@ -1029,10 +1029,10 @@ static void ixgbe_free_q_vector(struct ixgbe_adapter *adapter, int v_idx) WRITE_ONCE(adapter->rx_ring[ring->queue_index], NULL); adapter->q_vector[v_idx] = NULL; - napi_hash_del(&q_vector->napi); - netif_napi_del(&q_vector->napi); + __netif_napi_del(&q_vector->napi); /* + * after a call to __netif_napi_del() napi may still be used and * ixgbe_get_stats64() might access the rings on this vector, * we must wait a grace period before freeing it. */ diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 86ca8b9ea1b8..45ae33e15303 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -2095,10 +2095,8 @@ static struct sk_buff *ixgbe_construct_skb(struct ixgbe_ring *rx_ring, struct sk_buff *skb; /* prefetch first cache line of first page */ - prefetch(xdp->data); -#if L1_CACHE_BYTES < 128 - prefetch(xdp->data + L1_CACHE_BYTES); -#endif + net_prefetch(xdp->data); + /* Note, we get here by enabling legacy-rx via: * * ethtool --set-priv-flags <dev> legacy-rx on @@ -2161,10 +2159,7 @@ static struct sk_buff *ixgbe_build_skb(struct ixgbe_ring *rx_ring, * likely have a consumer accessing first few bytes of meta * data, and then actual data. */ - prefetch(xdp->data_meta); -#if L1_CACHE_BYTES < 128 - prefetch(xdp->data_meta + L1_CACHE_BYTES); -#endif + net_prefetch(xdp->data_meta); /* build an skb to around the page buffer */ skb = build_skb(xdp->data_hard_start, truesize); @@ -3156,7 +3151,7 @@ int ixgbe_poll(struct napi_struct *napi, int budget) #endif ixgbe_for_each_ring(ring, q_vector->tx) { - bool wd = ring->xsk_umem ? + bool wd = ring->xsk_pool ? ixgbe_clean_xdp_tx_irq(q_vector, ring, budget) : ixgbe_clean_tx_irq(q_vector, ring, budget); @@ -3176,7 +3171,7 @@ int ixgbe_poll(struct napi_struct *napi, int budget) per_ring_budget = budget; ixgbe_for_each_ring(ring, q_vector->rx) { - int cleaned = ring->xsk_umem ? + int cleaned = ring->xsk_pool ? ixgbe_clean_rx_irq_zc(q_vector, ring, per_ring_budget) : ixgbe_clean_rx_irq(q_vector, ring, @@ -3471,9 +3466,9 @@ void ixgbe_configure_tx_ring(struct ixgbe_adapter *adapter, u32 txdctl = IXGBE_TXDCTL_ENABLE; u8 reg_idx = ring->reg_idx; - ring->xsk_umem = NULL; + ring->xsk_pool = NULL; if (ring_is_xdp(ring)) - ring->xsk_umem = ixgbe_xsk_umem(adapter, ring); + ring->xsk_pool = ixgbe_xsk_pool(adapter, ring); /* disable queue to avoid issues while updating state */ IXGBE_WRITE_REG(hw, IXGBE_TXDCTL(reg_idx), 0); @@ -3713,8 +3708,8 @@ static void ixgbe_configure_srrctl(struct ixgbe_adapter *adapter, srrctl = IXGBE_RX_HDR_SIZE << IXGBE_SRRCTL_BSIZEHDRSIZE_SHIFT; /* configure the packet buffer length */ - if (rx_ring->xsk_umem) { - u32 xsk_buf_len = xsk_umem_get_rx_frame_size(rx_ring->xsk_umem); + if (rx_ring->xsk_pool) { + u32 xsk_buf_len = xsk_pool_get_rx_frame_size(rx_ring->xsk_pool); /* If the MAC support setting RXDCTL.RLPML, the * SRRCTL[n].BSIZEPKT is set to PAGE_SIZE and @@ -4059,12 +4054,12 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter, u8 reg_idx = ring->reg_idx; xdp_rxq_info_unreg_mem_model(&ring->xdp_rxq); - ring->xsk_umem = ixgbe_xsk_umem(adapter, ring); - if (ring->xsk_umem) { + ring->xsk_pool = ixgbe_xsk_pool(adapter, ring); + if (ring->xsk_pool) { WARN_ON(xdp_rxq_info_reg_mem_model(&ring->xdp_rxq, MEM_TYPE_XSK_BUFF_POOL, NULL)); - xsk_buff_set_rxq_info(ring->xsk_umem, &ring->xdp_rxq); + xsk_pool_set_rxq_info(ring->xsk_pool, &ring->xdp_rxq); } else { WARN_ON(xdp_rxq_info_reg_mem_model(&ring->xdp_rxq, MEM_TYPE_PAGE_SHARED, NULL)); @@ -4119,8 +4114,8 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter, #endif } - if (ring->xsk_umem && hw->mac.type != ixgbe_mac_82599EB) { - u32 xsk_buf_len = xsk_umem_get_rx_frame_size(ring->xsk_umem); + if (ring->xsk_pool && hw->mac.type != ixgbe_mac_82599EB) { + u32 xsk_buf_len = xsk_pool_get_rx_frame_size(ring->xsk_pool); rxdctl &= ~(IXGBE_RXDCTL_RLPMLMASK | IXGBE_RXDCTL_RLPML_EN); @@ -4142,7 +4137,7 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter, IXGBE_WRITE_REG(hw, IXGBE_RXDCTL(reg_idx), rxdctl); ixgbe_rx_desc_queue_enable(adapter, ring); - if (ring->xsk_umem) + if (ring->xsk_pool) ixgbe_alloc_rx_buffers_zc(ring, ixgbe_desc_unused(ring)); else ixgbe_alloc_rx_buffers(ring, ixgbe_desc_unused(ring)); @@ -5292,7 +5287,7 @@ static void ixgbe_clean_rx_ring(struct ixgbe_ring *rx_ring) u16 i = rx_ring->next_to_clean; struct ixgbe_rx_buffer *rx_buffer = &rx_ring->rx_buffer_info[i]; - if (rx_ring->xsk_umem) { + if (rx_ring->xsk_pool) { ixgbe_xsk_clean_rx_ring(rx_ring); goto skip_free; } @@ -5682,7 +5677,6 @@ static void ixgbe_up_complete(struct ixgbe_adapter *adapter) void ixgbe_reinit_locked(struct ixgbe_adapter *adapter) { - WARN_ON(in_interrupt()); /* put off any impending NetWatchDogTimeout */ netif_trans_update(adapter->netdev); @@ -5989,7 +5983,7 @@ static void ixgbe_clean_tx_ring(struct ixgbe_ring *tx_ring) u16 i = tx_ring->next_to_clean; struct ixgbe_tx_buffer *tx_buffer = &tx_ring->tx_buffer_info[i]; - if (tx_ring->xsk_umem) { + if (tx_ring->xsk_pool) { ixgbe_xsk_clean_tx_ring(tx_ring); goto out; } @@ -6185,8 +6179,9 @@ static void ixgbe_set_eee_capable(struct ixgbe_adapter *adapter) /** * ixgbe_tx_timeout - Respond to a Tx Hang * @netdev: network interface device structure + * @txqueue: queue number that timed out **/ -static void ixgbe_tx_timeout(struct net_device *netdev, unsigned int txqueue) +static void ixgbe_tx_timeout(struct net_device *netdev, unsigned int __always_unused txqueue) { struct ixgbe_adapter *adapter = netdev_priv(netdev); @@ -10161,7 +10156,7 @@ static int ixgbe_xdp_setup(struct net_device *dev, struct bpf_prog *prog) */ if (need_reset && prog) for (i = 0; i < adapter->num_rx_queues; i++) - if (adapter->xdp_ring[i]->xsk_umem) + if (adapter->xdp_ring[i]->xsk_pool) (void)ixgbe_xsk_wakeup(adapter->netdev, i, XDP_WAKEUP_RX); @@ -10175,8 +10170,8 @@ static int ixgbe_xdp(struct net_device *dev, struct netdev_bpf *xdp) switch (xdp->command) { case XDP_SETUP_PROG: return ixgbe_xdp_setup(dev, xdp->prog); - case XDP_SETUP_XSK_UMEM: - return ixgbe_xsk_umem_setup(adapter, xdp->xsk.umem, + case XDP_SETUP_XSK_POOL: + return ixgbe_xsk_pool_setup(adapter, xdp->xsk.pool, xdp->xsk.queue_id); default: diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c index 7980d7265e10..f77fa3e4fdd1 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c @@ -771,7 +771,7 @@ mii_bus_write_done: /** * ixgbe_mii_bus_read - Read a clause 22/45 register - * @hw: pointer to hardware structure + * @bus: pointer to mii_bus structure which points to our driver private * @addr: address * @regnum: register number **/ @@ -786,7 +786,7 @@ static s32 ixgbe_mii_bus_read(struct mii_bus *bus, int addr, int regnum) /** * ixgbe_mii_bus_write - Write a clause 22/45 register - * @hw: pointer to hardware structure + * @bus: pointer to mii_bus structure which points to our driver private * @addr: address * @regnum: register number * @val: value to write @@ -803,7 +803,7 @@ static s32 ixgbe_mii_bus_write(struct mii_bus *bus, int addr, int regnum, /** * ixgbe_x550em_a_mii_bus_read - Read a clause 22/45 register on x550em_a - * @hw: pointer to hardware structure + * @bus: pointer to mii_bus structure which points to our driver private * @addr: address * @regnum: register number **/ @@ -820,7 +820,7 @@ static s32 ixgbe_x550em_a_mii_bus_read(struct mii_bus *bus, int addr, /** * ixgbe_x550em_a_mii_bus_write - Write a clause 22/45 register on x550em_a - * @hw: pointer to hardware structure + * @bus: pointer to mii_bus structure which points to our driver private * @addr: address * @regnum: register number * @val: value to write diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_txrx_common.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_txrx_common.h index 7887ae4aaf4f..2aeec78029bc 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_txrx_common.h +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_txrx_common.h @@ -28,9 +28,10 @@ void ixgbe_irq_rearm_queues(struct ixgbe_adapter *adapter, u64 qmask); void ixgbe_txrx_ring_disable(struct ixgbe_adapter *adapter, int ring); void ixgbe_txrx_ring_enable(struct ixgbe_adapter *adapter, int ring); -struct xdp_umem *ixgbe_xsk_umem(struct ixgbe_adapter *adapter, - struct ixgbe_ring *ring); -int ixgbe_xsk_umem_setup(struct ixgbe_adapter *adapter, struct xdp_umem *umem, +struct xsk_buff_pool *ixgbe_xsk_pool(struct ixgbe_adapter *adapter, + struct ixgbe_ring *ring); +int ixgbe_xsk_pool_setup(struct ixgbe_adapter *adapter, + struct xsk_buff_pool *pool, u16 qid); void ixgbe_zca_free(struct zero_copy_allocator *alloc, unsigned long handle); diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c index ec7121f352e2..3771857cf887 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c @@ -8,8 +8,8 @@ #include "ixgbe.h" #include "ixgbe_txrx_common.h" -struct xdp_umem *ixgbe_xsk_umem(struct ixgbe_adapter *adapter, - struct ixgbe_ring *ring) +struct xsk_buff_pool *ixgbe_xsk_pool(struct ixgbe_adapter *adapter, + struct ixgbe_ring *ring) { bool xdp_on = READ_ONCE(adapter->xdp_prog); int qid = ring->ring_idx; @@ -17,11 +17,11 @@ struct xdp_umem *ixgbe_xsk_umem(struct ixgbe_adapter *adapter, if (!xdp_on || !test_bit(qid, adapter->af_xdp_zc_qps)) return NULL; - return xdp_get_umem_from_qid(adapter->netdev, qid); + return xsk_get_pool_from_qid(adapter->netdev, qid); } -static int ixgbe_xsk_umem_enable(struct ixgbe_adapter *adapter, - struct xdp_umem *umem, +static int ixgbe_xsk_pool_enable(struct ixgbe_adapter *adapter, + struct xsk_buff_pool *pool, u16 qid) { struct net_device *netdev = adapter->netdev; @@ -35,7 +35,7 @@ static int ixgbe_xsk_umem_enable(struct ixgbe_adapter *adapter, qid >= netdev->real_num_tx_queues) return -EINVAL; - err = xsk_buff_dma_map(umem, &adapter->pdev->dev, IXGBE_RX_DMA_ATTR); + err = xsk_pool_dma_map(pool, &adapter->pdev->dev, IXGBE_RX_DMA_ATTR); if (err) return err; @@ -59,13 +59,13 @@ static int ixgbe_xsk_umem_enable(struct ixgbe_adapter *adapter, return 0; } -static int ixgbe_xsk_umem_disable(struct ixgbe_adapter *adapter, u16 qid) +static int ixgbe_xsk_pool_disable(struct ixgbe_adapter *adapter, u16 qid) { - struct xdp_umem *umem; + struct xsk_buff_pool *pool; bool if_running; - umem = xdp_get_umem_from_qid(adapter->netdev, qid); - if (!umem) + pool = xsk_get_pool_from_qid(adapter->netdev, qid); + if (!pool) return -EINVAL; if_running = netif_running(adapter->netdev) && @@ -75,7 +75,7 @@ static int ixgbe_xsk_umem_disable(struct ixgbe_adapter *adapter, u16 qid) ixgbe_txrx_ring_disable(adapter, qid); clear_bit(qid, adapter->af_xdp_zc_qps); - xsk_buff_dma_unmap(umem, IXGBE_RX_DMA_ATTR); + xsk_pool_dma_unmap(pool, IXGBE_RX_DMA_ATTR); if (if_running) ixgbe_txrx_ring_enable(adapter, qid); @@ -83,11 +83,12 @@ static int ixgbe_xsk_umem_disable(struct ixgbe_adapter *adapter, u16 qid) return 0; } -int ixgbe_xsk_umem_setup(struct ixgbe_adapter *adapter, struct xdp_umem *umem, +int ixgbe_xsk_pool_setup(struct ixgbe_adapter *adapter, + struct xsk_buff_pool *pool, u16 qid) { - return umem ? ixgbe_xsk_umem_enable(adapter, umem, qid) : - ixgbe_xsk_umem_disable(adapter, qid); + return pool ? ixgbe_xsk_pool_enable(adapter, pool, qid) : + ixgbe_xsk_pool_disable(adapter, qid); } static int ixgbe_run_xdp_zc(struct ixgbe_adapter *adapter, @@ -149,7 +150,7 @@ bool ixgbe_alloc_rx_buffers_zc(struct ixgbe_ring *rx_ring, u16 count) i -= rx_ring->count; do { - bi->xdp = xsk_buff_alloc(rx_ring->xsk_umem); + bi->xdp = xsk_buff_alloc(rx_ring->xsk_pool); if (!bi->xdp) { ok = false; break; @@ -286,7 +287,7 @@ int ixgbe_clean_rx_irq_zc(struct ixgbe_q_vector *q_vector, } bi->xdp->data_end = bi->xdp->data + size; - xsk_buff_dma_sync_for_cpu(bi->xdp); + xsk_buff_dma_sync_for_cpu(bi->xdp, rx_ring->xsk_pool); xdp_res = ixgbe_run_xdp_zc(adapter, rx_ring, bi->xdp); if (xdp_res) { @@ -344,11 +345,11 @@ int ixgbe_clean_rx_irq_zc(struct ixgbe_q_vector *q_vector, q_vector->rx.total_packets += total_rx_packets; q_vector->rx.total_bytes += total_rx_bytes; - if (xsk_umem_uses_need_wakeup(rx_ring->xsk_umem)) { + if (xsk_uses_need_wakeup(rx_ring->xsk_pool)) { if (failure || rx_ring->next_to_clean == rx_ring->next_to_use) - xsk_set_rx_need_wakeup(rx_ring->xsk_umem); + xsk_set_rx_need_wakeup(rx_ring->xsk_pool); else - xsk_clear_rx_need_wakeup(rx_ring->xsk_umem); + xsk_clear_rx_need_wakeup(rx_ring->xsk_pool); return (int)total_rx_packets; } @@ -373,6 +374,7 @@ void ixgbe_xsk_clean_rx_ring(struct ixgbe_ring *rx_ring) static bool ixgbe_xmit_zc(struct ixgbe_ring *xdp_ring, unsigned int budget) { + struct xsk_buff_pool *pool = xdp_ring->xsk_pool; union ixgbe_adv_tx_desc *tx_desc = NULL; struct ixgbe_tx_buffer *tx_bi; bool work_done = true; @@ -387,12 +389,11 @@ static bool ixgbe_xmit_zc(struct ixgbe_ring *xdp_ring, unsigned int budget) break; } - if (!xsk_umem_consume_tx(xdp_ring->xsk_umem, &desc)) + if (!xsk_tx_peek_desc(pool, &desc)) break; - dma = xsk_buff_raw_get_dma(xdp_ring->xsk_umem, desc.addr); - xsk_buff_raw_dma_sync_for_device(xdp_ring->xsk_umem, dma, - desc.len); + dma = xsk_buff_raw_get_dma(pool, desc.addr); + xsk_buff_raw_dma_sync_for_device(pool, dma, desc.len); tx_bi = &xdp_ring->tx_buffer_info[xdp_ring->next_to_use]; tx_bi->bytecount = desc.len; @@ -418,7 +419,7 @@ static bool ixgbe_xmit_zc(struct ixgbe_ring *xdp_ring, unsigned int budget) if (tx_desc) { ixgbe_xdp_ring_update_tail(xdp_ring); - xsk_umem_consume_tx_done(xdp_ring->xsk_umem); + xsk_tx_release(pool); } return !!budget && work_done; @@ -439,7 +440,7 @@ bool ixgbe_clean_xdp_tx_irq(struct ixgbe_q_vector *q_vector, { u16 ntc = tx_ring->next_to_clean, ntu = tx_ring->next_to_use; unsigned int total_packets = 0, total_bytes = 0; - struct xdp_umem *umem = tx_ring->xsk_umem; + struct xsk_buff_pool *pool = tx_ring->xsk_pool; union ixgbe_adv_tx_desc *tx_desc; struct ixgbe_tx_buffer *tx_bi; u32 xsk_frames = 0; @@ -484,10 +485,10 @@ bool ixgbe_clean_xdp_tx_irq(struct ixgbe_q_vector *q_vector, q_vector->tx.total_packets += total_packets; if (xsk_frames) - xsk_umem_complete_tx(umem, xsk_frames); + xsk_tx_completed(pool, xsk_frames); - if (xsk_umem_uses_need_wakeup(tx_ring->xsk_umem)) - xsk_set_tx_need_wakeup(tx_ring->xsk_umem); + if (xsk_uses_need_wakeup(pool)) + xsk_set_tx_need_wakeup(pool); return ixgbe_xmit_zc(tx_ring, q_vector->tx.work_limit); } @@ -511,7 +512,7 @@ int ixgbe_xsk_wakeup(struct net_device *dev, u32 qid, u32 flags) if (test_bit(__IXGBE_TX_DISABLED, &ring->state)) return -ENETDOWN; - if (!ring->xsk_umem) + if (!ring->xsk_pool) return -ENXIO; if (!napi_if_scheduled_mark_missed(&ring->q_vector->napi)) { @@ -526,7 +527,7 @@ int ixgbe_xsk_wakeup(struct net_device *dev, u32 qid, u32 flags) void ixgbe_xsk_clean_tx_ring(struct ixgbe_ring *tx_ring) { u16 ntc = tx_ring->next_to_clean, ntu = tx_ring->next_to_use; - struct xdp_umem *umem = tx_ring->xsk_umem; + struct xsk_buff_pool *pool = tx_ring->xsk_pool; struct ixgbe_tx_buffer *tx_bi; u32 xsk_frames = 0; @@ -546,5 +547,5 @@ void ixgbe_xsk_clean_tx_ring(struct ixgbe_ring *tx_ring) } if (xsk_frames) - xsk_umem_complete_tx(umem, xsk_frames); + xsk_tx_completed(pool, xsk_frames); } diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c index a428113e6d54..82fce27f682b 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c @@ -246,8 +246,9 @@ static void ixgbevf_tx_timeout_reset(struct ixgbevf_adapter *adapter) /** * ixgbevf_tx_timeout - Respond to a Tx Hang * @netdev: network interface device structure + * @txqueue: transmit queue hanging (unused) **/ -static void ixgbevf_tx_timeout(struct net_device *netdev, unsigned int txqueue) +static void ixgbevf_tx_timeout(struct net_device *netdev, unsigned int __always_unused txqueue) { struct ixgbevf_adapter *adapter = netdev_priv(netdev); @@ -866,10 +867,8 @@ struct sk_buff *ixgbevf_construct_skb(struct ixgbevf_ring *rx_ring, struct sk_buff *skb; /* prefetch first cache line of first page */ - prefetch(xdp->data); -#if L1_CACHE_BYTES < 128 - prefetch(xdp->data + L1_CACHE_BYTES); -#endif + net_prefetch(xdp->data); + /* Note, we get here by enabling legacy-rx via: * * ethtool --set-priv-flags <dev> legacy-rx on @@ -947,10 +946,7 @@ static struct sk_buff *ixgbevf_build_skb(struct ixgbevf_ring *rx_ring, * have a consumer accessing first few bytes of meta data, * and then actual data. */ - prefetch(xdp->data_meta); -#if L1_CACHE_BYTES < 128 - prefetch(xdp->data_meta + L1_CACHE_BYTES); -#endif + net_prefetch(xdp->data_meta); /* build an skb around the page buffer */ skb = build_skb(xdp->data_hard_start, truesize); @@ -2526,8 +2522,6 @@ void ixgbevf_down(struct ixgbevf_adapter *adapter) void ixgbevf_reinit_locked(struct ixgbevf_adapter *adapter) { - WARN_ON(in_interrupt()); - while (test_and_set_bit(__IXGBEVF_RESETTING, &adapter->state)) msleep(1); diff --git a/drivers/net/ethernet/jme.c b/drivers/net/ethernet/jme.c index ddc757680089..e9efe074edc1 100644 --- a/drivers/net/ethernet/jme.c +++ b/drivers/net/ethernet/jme.c @@ -1187,9 +1187,9 @@ jme_shutdown_nic(struct jme_adapter *jme) } static void -jme_pcc_tasklet(unsigned long arg) +jme_pcc_tasklet(struct tasklet_struct *t) { - struct jme_adapter *jme = (struct jme_adapter *)arg; + struct jme_adapter *jme = from_tasklet(jme, t, pcc_task); struct net_device *netdev = jme->dev; if (unlikely(test_bit(JME_FLAG_SHUTDOWN, &jme->flags))) { @@ -1265,10 +1265,9 @@ jme_stop_shutdown_timer(struct jme_adapter *jme) jwrite32f(jme, JME_APMC, apmc); } -static void -jme_link_change_tasklet(unsigned long arg) +static void jme_link_change_tasklet(struct tasklet_struct *t) { - struct jme_adapter *jme = (struct jme_adapter *)arg; + struct jme_adapter *jme = from_tasklet(jme, t, linkch_task); struct net_device *netdev = jme->dev; int rc; @@ -1345,9 +1344,9 @@ out: } static void -jme_rx_clean_tasklet(unsigned long arg) +jme_rx_clean_tasklet(struct tasklet_struct *t) { - struct jme_adapter *jme = (struct jme_adapter *)arg; + struct jme_adapter *jme = from_tasklet(jme, t, rxclean_task); struct dynpcc_info *dpi = &(jme->dpi); jme_process_receive(jme, jme->rx_ring_size); @@ -1380,9 +1379,9 @@ jme_poll(JME_NAPI_HOLDER(holder), JME_NAPI_WEIGHT(budget)) } static void -jme_rx_empty_tasklet(unsigned long arg) +jme_rx_empty_tasklet(struct tasklet_struct *t) { - struct jme_adapter *jme = (struct jme_adapter *)arg; + struct jme_adapter *jme = from_tasklet(jme, t, rxempty_task); if (unlikely(atomic_read(&jme->link_changing) != 1)) return; @@ -1392,7 +1391,7 @@ jme_rx_empty_tasklet(unsigned long arg) netif_info(jme, rx_status, jme->dev, "RX Queue Full!\n"); - jme_rx_clean_tasklet(arg); + jme_rx_clean_tasklet(&jme->rxclean_task); while (atomic_read(&jme->rx_empty) > 0) { atomic_dec(&jme->rx_empty); @@ -1416,10 +1415,9 @@ jme_wake_queue_if_stopped(struct jme_adapter *jme) } -static void -jme_tx_clean_tasklet(unsigned long arg) +static void jme_tx_clean_tasklet(struct tasklet_struct *t) { - struct jme_adapter *jme = (struct jme_adapter *)arg; + struct jme_adapter *jme = from_tasklet(jme, t, txclean_task); struct jme_ring *txring = &(jme->txring[0]); struct txdesc *txdesc = txring->desc; struct jme_buffer_info *txbi = txring->bufinf, *ctxbi, *ttxbi; @@ -1834,14 +1832,10 @@ jme_open(struct net_device *netdev) jme_clear_pm_disable_wol(jme); JME_NAPI_ENABLE(jme); - tasklet_init(&jme->linkch_task, jme_link_change_tasklet, - (unsigned long) jme); - tasklet_init(&jme->txclean_task, jme_tx_clean_tasklet, - (unsigned long) jme); - tasklet_init(&jme->rxclean_task, jme_rx_clean_tasklet, - (unsigned long) jme); - tasklet_init(&jme->rxempty_task, jme_rx_empty_tasklet, - (unsigned long) jme); + tasklet_setup(&jme->linkch_task, jme_link_change_tasklet); + tasklet_setup(&jme->txclean_task, jme_tx_clean_tasklet); + tasklet_setup(&jme->rxclean_task, jme_rx_clean_tasklet); + tasklet_setup(&jme->rxempty_task, jme_rx_empty_tasklet); rc = jme_request_irq(jme); if (rc) @@ -3040,9 +3034,7 @@ jme_init_one(struct pci_dev *pdev, atomic_set(&jme->tx_cleaning, 1); atomic_set(&jme->rx_empty, 1); - tasklet_init(&jme->pcc_task, - jme_pcc_tasklet, - (unsigned long) jme); + tasklet_setup(&jme->pcc_task, jme_pcc_tasklet); jme->dpi.cur = PCC_P1; jme->reg_ghc = 0; diff --git a/drivers/net/ethernet/korina.c b/drivers/net/ethernet/korina.c index 03e034918d14..af441d699a57 100644 --- a/drivers/net/ethernet/korina.c +++ b/drivers/net/ethernet/korina.c @@ -1113,7 +1113,7 @@ out: return rc; probe_err_register: - kfree(lp->td_ring); + kfree(KSEG0ADDR(lp->td_ring)); probe_err_td_ring: iounmap(lp->tx_dma_regs); probe_err_dma_tx: @@ -1133,6 +1133,7 @@ static int korina_remove(struct platform_device *pdev) iounmap(lp->eth_regs); iounmap(lp->rx_dma_regs); iounmap(lp->tx_dma_regs); + kfree(KSEG0ADDR(lp->td_ring)); unregister_netdev(bif->dev); free_netdev(bif->dev); diff --git a/drivers/net/ethernet/marvell/Kconfig b/drivers/net/ethernet/marvell/Kconfig index ef4f35ba077d..41815b609569 100644 --- a/drivers/net/ethernet/marvell/Kconfig +++ b/drivers/net/ethernet/marvell/Kconfig @@ -92,6 +92,12 @@ config MVPP2 This driver supports the network interface units in the Marvell ARMADA 375, 7K and 8K SoCs. +config MVPP2_PTP + bool "Marvell Armada 8K Enable PTP support" + depends on NETWORK_PHY_TIMESTAMPING + depends on (PTP_1588_CLOCK = y && MVPP2 = y) || \ + (PTP_1588_CLOCK && MVPP2 = m) + config PXA168_ETH tristate "Marvell pxa168 ethernet support" depends on HAS_IOMEM @@ -172,5 +178,6 @@ config SKY2_DEBUG source "drivers/net/ethernet/marvell/octeontx2/Kconfig" +source "drivers/net/ethernet/marvell/prestera/Kconfig" endif # NET_VENDOR_MARVELL diff --git a/drivers/net/ethernet/marvell/Makefile b/drivers/net/ethernet/marvell/Makefile index 89dea7284d5b..9f88fe822555 100644 --- a/drivers/net/ethernet/marvell/Makefile +++ b/drivers/net/ethernet/marvell/Makefile @@ -12,3 +12,4 @@ obj-$(CONFIG_PXA168_ETH) += pxa168_eth.o obj-$(CONFIG_SKGE) += skge.o obj-$(CONFIG_SKY2) += sky2.o obj-y += octeontx2/ +obj-y += prestera/ diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index 5bf0409f5d42..54b0bf574c05 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -330,7 +330,6 @@ #define MVNETA_SKB_HEADROOM ALIGN(max(NET_SKB_PAD, XDP_PACKET_HEADROOM), 8) #define MVNETA_SKB_PAD (SKB_DATA_ALIGN(sizeof(struct skb_shared_info) + \ MVNETA_SKB_HEADROOM)) -#define MVNETA_SKB_SIZE(len) (SKB_DATA_ALIGN(len) + MVNETA_SKB_PAD) #define MVNETA_MAX_RX_BUF_SIZE (PAGE_SIZE - MVNETA_SKB_PAD) #define IS_TSO_HEADER(txq, addr) \ @@ -752,13 +751,12 @@ static void mvneta_txq_inc_put(struct mvneta_tx_queue *txq) static void mvneta_mib_counters_clear(struct mvneta_port *pp) { int i; - u32 dummy; /* Perform dummy reads from MIB counters */ for (i = 0; i < MVNETA_MIB_LATE_COLLISION; i += 4) - dummy = mvreg_read(pp, (MVNETA_MIB_COUNTERS_BASE + i)); - dummy = mvreg_read(pp, MVNETA_RX_DISCARD_FRAME_COUNT); - dummy = mvreg_read(pp, MVNETA_OVERRUN_FRAME_COUNT); + mvreg_read(pp, (MVNETA_MIB_COUNTERS_BASE + i)); + mvreg_read(pp, MVNETA_RX_DISCARD_FRAME_COUNT); + mvreg_read(pp, MVNETA_OVERRUN_FRAME_COUNT); } /* Get System Network Statistics */ @@ -1833,7 +1831,7 @@ static struct mvneta_tx_queue *mvneta_tx_done_policy(struct mvneta_port *pp, /* Free tx queue skbuffs */ static void mvneta_txq_bufs_free(struct mvneta_port *pp, struct mvneta_tx_queue *txq, int num, - struct netdev_queue *nq) + struct netdev_queue *nq, bool napi) { unsigned int bytes_compl = 0, pkts_compl = 0; int i; @@ -1856,7 +1854,10 @@ static void mvneta_txq_bufs_free(struct mvneta_port *pp, dev_kfree_skb_any(buf->skb); } else if (buf->type == MVNETA_TYPE_XDP_TX || buf->type == MVNETA_TYPE_XDP_NDO) { - xdp_return_frame(buf->xdpf); + if (napi && buf->type == MVNETA_TYPE_XDP_TX) + xdp_return_frame_rx_napi(buf->xdpf); + else + xdp_return_frame(buf->xdpf); } } @@ -1874,7 +1875,7 @@ static void mvneta_txq_done(struct mvneta_port *pp, if (!tx_done) return; - mvneta_txq_bufs_free(pp, txq, tx_done, nq); + mvneta_txq_bufs_free(pp, txq, tx_done, nq, true); txq->count -= tx_done; @@ -2227,8 +2228,7 @@ mvneta_swbm_rx_frame(struct mvneta_port *pp, struct mvneta_rx_desc *rx_desc, struct mvneta_rx_queue *rxq, struct xdp_buff *xdp, int *size, - struct page *page, - struct mvneta_stats *stats) + struct page *page) { unsigned char *data = page_address(page); int data_len = -MVNETA_MH_SIZE, len; @@ -2236,19 +2236,22 @@ mvneta_swbm_rx_frame(struct mvneta_port *pp, enum dma_data_direction dma_dir; struct skb_shared_info *sinfo; - if (MVNETA_SKB_SIZE(rx_desc->data_size) > PAGE_SIZE) { + if (*size > MVNETA_MAX_RX_BUF_SIZE) { len = MVNETA_MAX_RX_BUF_SIZE; data_len += len; } else { - len = rx_desc->data_size; + len = *size; data_len += len - ETH_FCS_LEN; } + *size = *size - len; dma_dir = page_pool_get_dma_dir(rxq->page_pool); dma_sync_single_for_cpu(dev->dev.parent, rx_desc->buf_phys_addr, len, dma_dir); + rx_desc->buf_phys_addr = 0; + /* Prefetch header */ prefetch(data); @@ -2259,9 +2262,6 @@ mvneta_swbm_rx_frame(struct mvneta_port *pp, sinfo = xdp_get_shared_info_from_buff(xdp); sinfo->nr_frags = 0; - - *size = rx_desc->data_size - len; - rx_desc->buf_phys_addr = 0; } static void @@ -2307,11 +2307,8 @@ mvneta_swbm_build_skb(struct mvneta_port *pp, struct mvneta_rx_queue *rxq, { struct skb_shared_info *sinfo = xdp_get_shared_info_from_buff(xdp); int i, num_frags = sinfo->nr_frags; - skb_frag_t frags[MAX_SKB_FRAGS]; struct sk_buff *skb; - memcpy(frags, sinfo->frags, sizeof(skb_frag_t) * num_frags); - skb = build_skb(xdp->data_hard_start, PAGE_SIZE); if (!skb) return ERR_PTR(-ENOMEM); @@ -2323,12 +2320,12 @@ mvneta_swbm_build_skb(struct mvneta_port *pp, struct mvneta_rx_queue *rxq, mvneta_rx_csum(pp, desc_status, skb); for (i = 0; i < num_frags; i++) { - struct page *page = skb_frag_page(&frags[i]); + skb_frag_t *frag = &sinfo->frags[i]; skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, - page, skb_frag_off(&frags[i]), - skb_frag_size(&frags[i]), PAGE_SIZE); - page_pool_release_page(rxq->page_pool, page); + skb_frag_page(frag), skb_frag_off(frag), + skb_frag_size(frag), PAGE_SIZE); + page_pool_release_page(rxq->page_pool, skb_frag_page(frag)); } return skb; @@ -2378,10 +2375,10 @@ static int mvneta_rx_swbm(struct napi_struct *napi, size = rx_desc->data_size; frame_sz = size - ETH_FCS_LEN; - desc_status = rx_desc->status; + desc_status = rx_status; mvneta_swbm_rx_frame(pp, rx_desc, rxq, &xdp_buf, - &size, page, &ps); + &size, page); } else { if (unlikely(!xdp_buf.data_hard_start)) { rx_desc->buf_phys_addr = 0; @@ -2865,7 +2862,7 @@ static void mvneta_txq_done_force(struct mvneta_port *pp, struct netdev_queue *nq = netdev_get_tx_queue(pp->dev, txq->id); int tx_done = txq->count; - mvneta_txq_bufs_free(pp, txq, tx_done, nq); + mvneta_txq_bufs_free(pp, txq, tx_done, nq, false); /* reset txq */ txq->count = 0; diff --git a/drivers/net/ethernet/marvell/mvpp2/Makefile b/drivers/net/ethernet/marvell/mvpp2/Makefile index 51f65a202c6e..9bd8e7964b40 100644 --- a/drivers/net/ethernet/marvell/mvpp2/Makefile +++ b/drivers/net/ethernet/marvell/mvpp2/Makefile @@ -4,4 +4,5 @@ # obj-$(CONFIG_MVPP2) := mvpp2.o -mvpp2-objs := mvpp2_main.o mvpp2_prs.o mvpp2_cls.o mvpp2_debugfs.o +mvpp2-y := mvpp2_main.o mvpp2_prs.o mvpp2_cls.o mvpp2_debugfs.o +mvpp2-$(CONFIG_MVPP2_PTP) += mvpp2_tai.o diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2.h b/drivers/net/ethernet/marvell/mvpp2/mvpp2.h index 32753cc771bf..834775843067 100644 --- a/drivers/net/ethernet/marvell/mvpp2/mvpp2.h +++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2.h @@ -12,6 +12,7 @@ #include <linux/interrupt.h> #include <linux/kernel.h> #include <linux/netdevice.h> +#include <linux/net_tstamp.h> #include <linux/phy.h> #include <linux/phylink.h> #include <net/flow_offload.h> @@ -461,8 +462,12 @@ #define MVPP22_CTRL4_DP_CLK_SEL BIT(5) #define MVPP22_CTRL4_SYNC_BYPASS_DIS BIT(6) #define MVPP22_CTRL4_QSGMII_BYPASS_ACTIVE BIT(7) +#define MVPP22_GMAC_INT_SUM_STAT 0xa0 +#define MVPP22_GMAC_INT_SUM_STAT_INTERNAL BIT(1) +#define MVPP22_GMAC_INT_SUM_STAT_PTP BIT(2) #define MVPP22_GMAC_INT_SUM_MASK 0xa4 #define MVPP22_GMAC_INT_SUM_MASK_LINK_STAT BIT(1) +#define MVPP22_GMAC_INT_SUM_MASK_PTP BIT(2) /* Per-port XGMAC registers. PPv2.2 only, only for GOP port 0, * relative to port->base. @@ -488,9 +493,13 @@ #define MVPP22_XLG_CTRL3_MACMODESELECT_MASK (7 << 13) #define MVPP22_XLG_CTRL3_MACMODESELECT_GMAC (0 << 13) #define MVPP22_XLG_CTRL3_MACMODESELECT_10G (1 << 13) +#define MVPP22_XLG_EXT_INT_STAT 0x158 +#define MVPP22_XLG_EXT_INT_STAT_XLG BIT(1) +#define MVPP22_XLG_EXT_INT_STAT_PTP BIT(7) #define MVPP22_XLG_EXT_INT_MASK 0x15c #define MVPP22_XLG_EXT_INT_MASK_XLG BIT(1) #define MVPP22_XLG_EXT_INT_MASK_GIG BIT(2) +#define MVPP22_XLG_EXT_INT_MASK_PTP BIT(7) #define MVPP22_XLG_CTRL4_REG 0x184 #define MVPP22_XLG_CTRL4_FWD_FC BIT(5) #define MVPP22_XLG_CTRL4_FWD_PFC BIT(6) @@ -501,6 +510,70 @@ #define MVPP22_SMI_MISC_CFG_REG 0x1204 #define MVPP22_SMI_POLLING_EN BIT(10) +/* TAI registers, PPv2.2 only, relative to priv->iface_base */ +#define MVPP22_TAI_INT_CAUSE 0x1400 +#define MVPP22_TAI_INT_MASK 0x1404 +#define MVPP22_TAI_CR0 0x1408 +#define MVPP22_TAI_CR1 0x140c +#define MVPP22_TAI_TCFCR0 0x1410 +#define MVPP22_TAI_TCFCR1 0x1414 +#define MVPP22_TAI_TCFCR2 0x1418 +#define MVPP22_TAI_FATWR 0x141c +#define MVPP22_TAI_TOD_STEP_NANO_CR 0x1420 +#define MVPP22_TAI_TOD_STEP_FRAC_HIGH 0x1424 +#define MVPP22_TAI_TOD_STEP_FRAC_LOW 0x1428 +#define MVPP22_TAI_TAPDC_HIGH 0x142c +#define MVPP22_TAI_TAPDC_LOW 0x1430 +#define MVPP22_TAI_TGTOD_SEC_HIGH 0x1434 +#define MVPP22_TAI_TGTOD_SEC_MED 0x1438 +#define MVPP22_TAI_TGTOD_SEC_LOW 0x143c +#define MVPP22_TAI_TGTOD_NANO_HIGH 0x1440 +#define MVPP22_TAI_TGTOD_NANO_LOW 0x1444 +#define MVPP22_TAI_TGTOD_FRAC_HIGH 0x1448 +#define MVPP22_TAI_TGTOD_FRAC_LOW 0x144c +#define MVPP22_TAI_TLV_SEC_HIGH 0x1450 +#define MVPP22_TAI_TLV_SEC_MED 0x1454 +#define MVPP22_TAI_TLV_SEC_LOW 0x1458 +#define MVPP22_TAI_TLV_NANO_HIGH 0x145c +#define MVPP22_TAI_TLV_NANO_LOW 0x1460 +#define MVPP22_TAI_TLV_FRAC_HIGH 0x1464 +#define MVPP22_TAI_TLV_FRAC_LOW 0x1468 +#define MVPP22_TAI_TCV0_SEC_HIGH 0x146c +#define MVPP22_TAI_TCV0_SEC_MED 0x1470 +#define MVPP22_TAI_TCV0_SEC_LOW 0x1474 +#define MVPP22_TAI_TCV0_NANO_HIGH 0x1478 +#define MVPP22_TAI_TCV0_NANO_LOW 0x147c +#define MVPP22_TAI_TCV0_FRAC_HIGH 0x1480 +#define MVPP22_TAI_TCV0_FRAC_LOW 0x1484 +#define MVPP22_TAI_TCV1_SEC_HIGH 0x1488 +#define MVPP22_TAI_TCV1_SEC_MED 0x148c +#define MVPP22_TAI_TCV1_SEC_LOW 0x1490 +#define MVPP22_TAI_TCV1_NANO_HIGH 0x1494 +#define MVPP22_TAI_TCV1_NANO_LOW 0x1498 +#define MVPP22_TAI_TCV1_FRAC_HIGH 0x149c +#define MVPP22_TAI_TCV1_FRAC_LOW 0x14a0 +#define MVPP22_TAI_TCSR 0x14a4 +#define MVPP22_TAI_TUC_LSB 0x14a8 +#define MVPP22_TAI_GFM_SEC_HIGH 0x14ac +#define MVPP22_TAI_GFM_SEC_MED 0x14b0 +#define MVPP22_TAI_GFM_SEC_LOW 0x14b4 +#define MVPP22_TAI_GFM_NANO_HIGH 0x14b8 +#define MVPP22_TAI_GFM_NANO_LOW 0x14bc +#define MVPP22_TAI_GFM_FRAC_HIGH 0x14c0 +#define MVPP22_TAI_GFM_FRAC_LOW 0x14c4 +#define MVPP22_TAI_PCLK_DA_HIGH 0x14c8 +#define MVPP22_TAI_PCLK_DA_LOW 0x14cc +#define MVPP22_TAI_CTCR 0x14d0 +#define MVPP22_TAI_PCLK_CCC_HIGH 0x14d4 +#define MVPP22_TAI_PCLK_CCC_LOW 0x14d8 +#define MVPP22_TAI_DTC_HIGH 0x14dc +#define MVPP22_TAI_DTC_LOW 0x14e0 +#define MVPP22_TAI_CCC_HIGH 0x14e4 +#define MVPP22_TAI_CCC_LOW 0x14e8 +#define MVPP22_TAI_ICICE 0x14f4 +#define MVPP22_TAI_ICICC_LOW 0x14f8 +#define MVPP22_TAI_TUC_MSB 0x14fc + #define MVPP22_GMAC_BASE(port) (0x7000 + (port) * 0x1000 + 0xe00) #define MVPP2_CAUSE_TXQ_SENT_DESC_ALL_MASK 0xff @@ -527,6 +600,46 @@ #define MVPP22_XPCS_CFG0_PCS_MODE(n) ((n) << 3) #define MVPP22_XPCS_CFG0_ACTIVE_LANE(n) ((n) << 5) +/* PTP registers. PPv2.2 only */ +#define MVPP22_PTP_BASE(port) (0x7800 + (port * 0x1000)) +#define MVPP22_PTP_INT_CAUSE 0x00 +#define MVPP22_PTP_INT_CAUSE_QUEUE1 BIT(6) +#define MVPP22_PTP_INT_CAUSE_QUEUE0 BIT(5) +#define MVPP22_PTP_INT_MASK 0x04 +#define MVPP22_PTP_INT_MASK_QUEUE1 BIT(6) +#define MVPP22_PTP_INT_MASK_QUEUE0 BIT(5) +#define MVPP22_PTP_GCR 0x08 +#define MVPP22_PTP_GCR_RX_RESET BIT(13) +#define MVPP22_PTP_GCR_TX_RESET BIT(1) +#define MVPP22_PTP_GCR_TSU_ENABLE BIT(0) +#define MVPP22_PTP_TX_Q0_R0 0x0c +#define MVPP22_PTP_TX_Q0_R1 0x10 +#define MVPP22_PTP_TX_Q0_R2 0x14 +#define MVPP22_PTP_TX_Q1_R0 0x18 +#define MVPP22_PTP_TX_Q1_R1 0x1c +#define MVPP22_PTP_TX_Q1_R2 0x20 +#define MVPP22_PTP_TPCR 0x24 +#define MVPP22_PTP_V1PCR 0x28 +#define MVPP22_PTP_V2PCR 0x2c +#define MVPP22_PTP_Y1731PCR 0x30 +#define MVPP22_PTP_NTPTSPCR 0x34 +#define MVPP22_PTP_NTPRXPCR 0x38 +#define MVPP22_PTP_NTPTXPCR 0x3c +#define MVPP22_PTP_WAMPPCR 0x40 +#define MVPP22_PTP_NAPCR 0x44 +#define MVPP22_PTP_FAPCR 0x48 +#define MVPP22_PTP_CAPCR 0x50 +#define MVPP22_PTP_ATAPCR 0x54 +#define MVPP22_PTP_ACTAPCR 0x58 +#define MVPP22_PTP_CATAPCR 0x5c +#define MVPP22_PTP_CACTAPCR 0x60 +#define MVPP22_PTP_AITAPCR 0x64 +#define MVPP22_PTP_CAITAPCR 0x68 +#define MVPP22_PTP_CITAPCR 0x6c +#define MVPP22_PTP_NTP_OFF_HIGH 0x70 +#define MVPP22_PTP_NTP_OFF_LOW 0x74 +#define MVPP22_PTP_TX_PIPE_STATUS_DELAY 0x78 + /* System controller registers. Accessed through a regmap. */ #define GENCONF_SOFT_RESET1 0x1108 #define GENCONF_SOFT_RESET1_GOP BIT(6) @@ -692,6 +805,43 @@ enum mvpp2_prs_l3_cast { MVPP2_PRS_L3_BROAD_CAST }; +/* PTP descriptor constants. The low bits of the descriptor are stored + * separately from the high bits. + */ +#define MVPP22_PTP_DESC_MASK_LOW 0xfff + +/* PTPAction */ +enum mvpp22_ptp_action { + MVPP22_PTP_ACTION_NONE = 0, + MVPP22_PTP_ACTION_FORWARD = 1, + MVPP22_PTP_ACTION_CAPTURE = 3, + /* The following have not been verified */ + MVPP22_PTP_ACTION_ADDTIME = 4, + MVPP22_PTP_ACTION_ADDCORRECTEDTIME = 5, + MVPP22_PTP_ACTION_CAPTUREADDTIME = 6, + MVPP22_PTP_ACTION_CAPTUREADDCORRECTEDTIME = 7, + MVPP22_PTP_ACTION_ADDINGRESSTIME = 8, + MVPP22_PTP_ACTION_CAPTUREADDINGRESSTIME = 9, + MVPP22_PTP_ACTION_CAPTUREINGRESSTIME = 10, +}; + +/* PTPPacketFormat */ +enum mvpp22_ptp_packet_format { + MVPP22_PTP_PKT_FMT_PTPV2 = 0, + MVPP22_PTP_PKT_FMT_PTPV1 = 1, + MVPP22_PTP_PKT_FMT_Y1731 = 2, + MVPP22_PTP_PKT_FMT_NTPTS = 3, + MVPP22_PTP_PKT_FMT_NTPRX = 4, + MVPP22_PTP_PKT_FMT_NTPTX = 5, + MVPP22_PTP_PKT_FMT_TWAMP = 6, +}; + +#define MVPP22_PTP_ACTION(x) (((x) & 15) << 0) +#define MVPP22_PTP_PACKETFORMAT(x) (((x) & 7) << 4) +#define MVPP22_PTP_MACTIMESTAMPINGEN BIT(11) +#define MVPP22_PTP_TIMESTAMPENTRYID(x) (((x) & 31) << 12) +#define MVPP22_PTP_TIMESTAMPQUEUESELECT BIT(18) + /* BM constants */ #define MVPP2_BM_JUMBO_BUF_NUM 512 #define MVPP2_BM_LONG_BUF_NUM 1024 @@ -759,6 +909,8 @@ enum mvpp2_prs_l3_cast { #define MVPP2_DESC_DMA_MASK DMA_BIT_MASK(40) +struct mvpp2_tai; + /* Definitions */ struct mvpp2_dbgfs_entries; @@ -794,6 +946,7 @@ struct mvpp2 { /* List of pointers to port structures */ int port_count; struct mvpp2_port *port_list[MVPP2_MAX_PORTS]; + struct mvpp2_tai *tai; /* Number of Tx threads used */ unsigned int nthreads; @@ -907,6 +1060,11 @@ struct mvpp2_ethtool_fs { struct ethtool_rxnfc rxnfc; }; +struct mvpp2_hwtstamp_queue { + struct sk_buff *skb[32]; + u8 next; +}; + struct mvpp2_port { u8 id; @@ -915,7 +1073,7 @@ struct mvpp2_port { */ int gop_id; - int link_irq; + int port_irq; struct mvpp2 *priv; @@ -967,6 +1125,7 @@ struct mvpp2_port { phy_interface_t phy_interface; struct phylink *phylink; struct phylink_config phylink_config; + struct phylink_pcs phylink_pcs; struct phy *comphy; struct mvpp2_bm_pool *pool_long; @@ -989,6 +1148,11 @@ struct mvpp2_port { * them from 0 */ int rss_ctx[MVPP22_N_RSS_TABLES]; + + bool hwtstamp; + bool rx_hwtstamp; + enum hwtstamp_tx_types tx_hwtstamp_type; + struct mvpp2_hwtstamp_queue tx_hwtstamp_queue[2]; }; /* The mvpp2_tx_desc and mvpp2_rx_desc structures describe the @@ -1057,7 +1221,8 @@ struct mvpp22_tx_desc { u8 packet_offset; u8 phys_txq; __le16 data_size; - __le64 reserved1; + __le32 ptp_descriptor; + __le32 reserved2; __le64 buf_dma_addr_ptp; __le64 buf_cookie_misc; }; @@ -1068,7 +1233,7 @@ struct mvpp22_rx_desc { __le16 reserved1; __le16 data_size; __le32 reserved2; - __le32 reserved3; + __le32 timestamp; __le64 buf_dma_addr_key_hash; __le64 buf_cookie_misc; }; @@ -1248,4 +1413,36 @@ void mvpp2_dbgfs_init(struct mvpp2 *priv, const char *name); void mvpp2_dbgfs_cleanup(struct mvpp2 *priv); +#ifdef CONFIG_MVPP2_PTP +int mvpp22_tai_probe(struct device *dev, struct mvpp2 *priv); +void mvpp22_tai_tstamp(struct mvpp2_tai *tai, u32 tstamp, + struct skb_shared_hwtstamps *hwtstamp); +void mvpp22_tai_start(struct mvpp2_tai *tai); +void mvpp22_tai_stop(struct mvpp2_tai *tai); +int mvpp22_tai_ptp_clock_index(struct mvpp2_tai *tai); +#else +static inline int mvpp22_tai_probe(struct device *dev, struct mvpp2 *priv) +{ + return 0; +} +static inline void mvpp22_tai_tstamp(struct mvpp2_tai *tai, u32 tstamp, + struct skb_shared_hwtstamps *hwtstamp) +{ +} +static inline void mvpp22_tai_start(struct mvpp2_tai *tai) +{ +} +static inline void mvpp22_tai_stop(struct mvpp2_tai *tai) +{ +} +static inline int mvpp22_tai_ptp_clock_index(struct mvpp2_tai *tai) +{ + return -1; +} +#endif + +static inline bool mvpp22_rx_hwtstamping(struct mvpp2_port *port) +{ + return IS_ENABLED(CONFIG_MVPP2_PTP) && port->rx_hwtstamp; +} #endif diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c index 6e140d1b8967..f6616c8933ca 100644 --- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c +++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c @@ -28,6 +28,7 @@ #include <linux/phy.h> #include <linux/phylink.h> #include <linux/phy/phy.h> +#include <linux/ptp_classify.h> #include <linux/clk.h> #include <linux/hrtimer.h> #include <linux/ktime.h> @@ -57,13 +58,7 @@ static struct { /* The prototype is added here to be used in start_dev when using ACPI. This * will be removed once phylink is used for all modes (dt+ACPI). */ -static void mvpp2_mac_config(struct phylink_config *config, unsigned int mode, - const struct phylink_link_state *state); -static void mvpp2_mac_link_up(struct phylink_config *config, - struct phy_device *phy, - unsigned int mode, phy_interface_t interface, - int speed, int duplex, - bool tx_pause, bool rx_pause); +static void mvpp2_acpi_start(struct mvpp2_port *port); /* Queue modes */ #define MVPP2_QDIST_SINGLE_MODE 0 @@ -1385,6 +1380,10 @@ static void mvpp22_gop_setup_irq(struct mvpp2_port *port) { u32 val; + mvpp2_modify(port->base + MVPP22_GMAC_INT_SUM_MASK, + MVPP22_GMAC_INT_SUM_MASK_PTP, + MVPP22_GMAC_INT_SUM_MASK_PTP); + if (port->phylink || phy_interface_mode_is_rgmii(port->phy_interface) || phy_interface_mode_is_8023z(port->phy_interface) || @@ -1398,6 +1397,10 @@ static void mvpp22_gop_setup_irq(struct mvpp2_port *port) val = readl(port->base + MVPP22_XLG_INT_MASK); val |= MVPP22_XLG_INT_MASK_LINK; writel(val, port->base + MVPP22_XLG_INT_MASK); + + mvpp2_modify(port->base + MVPP22_XLG_EXT_INT_MASK, + MVPP22_XLG_EXT_INT_MASK_PTP, + MVPP22_XLG_EXT_INT_MASK_PTP); } mvpp22_gop_unmask_irq(port); @@ -1485,8 +1488,8 @@ static void mvpp2_port_loopback_set(struct mvpp2_port *port, else val &= ~MVPP2_GMAC_GMII_LB_EN_MASK; - if (phy_interface_mode_is_8023z(port->phy_interface) || - port->phy_interface == PHY_INTERFACE_MODE_SGMII) + if (phy_interface_mode_is_8023z(state->interface) || + state->interface == PHY_INTERFACE_MODE_SGMII) val |= MVPP2_GMAC_PCS_LB_EN_MASK; else val &= ~MVPP2_GMAC_PCS_LB_EN_MASK; @@ -2980,44 +2983,67 @@ static irqreturn_t mvpp2_isr(int irq, void *dev_id) return IRQ_HANDLED; } -/* Per-port interrupt for link status changes */ -static irqreturn_t mvpp2_link_status_isr(int irq, void *dev_id) +static void mvpp2_isr_handle_ptp_queue(struct mvpp2_port *port, int nq) { - struct mvpp2_port *port = (struct mvpp2_port *)dev_id; - struct net_device *dev = port->dev; - bool event = false, link = false; - u32 val; + struct skb_shared_hwtstamps shhwtstamps; + struct mvpp2_hwtstamp_queue *queue; + struct sk_buff *skb; + void __iomem *ptp_q; + unsigned int id; + u32 r0, r1, r2; - mvpp22_gop_mask_irq(port); + ptp_q = port->priv->iface_base + MVPP22_PTP_BASE(port->gop_id); + if (nq) + ptp_q += MVPP22_PTP_TX_Q1_R0 - MVPP22_PTP_TX_Q0_R0; - if (mvpp2_port_supports_xlg(port) && - mvpp2_is_xlg(port->phy_interface)) { - val = readl(port->base + MVPP22_XLG_INT_STAT); - if (val & MVPP22_XLG_INT_STAT_LINK) { - event = true; - val = readl(port->base + MVPP22_XLG_STATUS); - if (val & MVPP22_XLG_STATUS_LINK_UP) - link = true; - } - } else if (phy_interface_mode_is_rgmii(port->phy_interface) || - phy_interface_mode_is_8023z(port->phy_interface) || - port->phy_interface == PHY_INTERFACE_MODE_SGMII) { - val = readl(port->base + MVPP22_GMAC_INT_STAT); - if (val & MVPP22_GMAC_INT_STAT_LINK) { - event = true; - val = readl(port->base + MVPP2_GMAC_STATUS0); - if (val & MVPP2_GMAC_STATUS0_LINK_UP) - link = true; + queue = &port->tx_hwtstamp_queue[nq]; + + while (1) { + r0 = readl_relaxed(ptp_q + MVPP22_PTP_TX_Q0_R0) & 0xffff; + if (!r0) + break; + + r1 = readl_relaxed(ptp_q + MVPP22_PTP_TX_Q0_R1) & 0xffff; + r2 = readl_relaxed(ptp_q + MVPP22_PTP_TX_Q0_R2) & 0xffff; + + id = (r0 >> 1) & 31; + + skb = queue->skb[id]; + queue->skb[id] = NULL; + if (skb) { + u32 ts = r2 << 19 | r1 << 3 | r0 >> 13; + + mvpp22_tai_tstamp(port->priv->tai, ts, &shhwtstamps); + skb_tstamp_tx(skb, &shhwtstamps); + dev_kfree_skb_any(skb); } } +} + +static void mvpp2_isr_handle_ptp(struct mvpp2_port *port) +{ + void __iomem *ptp; + u32 val; + + ptp = port->priv->iface_base + MVPP22_PTP_BASE(port->gop_id); + val = readl(ptp + MVPP22_PTP_INT_CAUSE); + if (val & MVPP22_PTP_INT_CAUSE_QUEUE0) + mvpp2_isr_handle_ptp_queue(port, 0); + if (val & MVPP22_PTP_INT_CAUSE_QUEUE1) + mvpp2_isr_handle_ptp_queue(port, 1); +} + +static void mvpp2_isr_handle_link(struct mvpp2_port *port, bool link) +{ + struct net_device *dev = port->dev; if (port->phylink) { phylink_mac_change(port->phylink, link); - goto handled; + return; } - if (!netif_running(dev) || !event) - goto handled; + if (!netif_running(dev)) + return; if (link) { mvpp2_interrupts_enable(port); @@ -3034,8 +3060,65 @@ static irqreturn_t mvpp2_link_status_isr(int irq, void *dev_id) mvpp2_interrupts_disable(port); } +} + +static void mvpp2_isr_handle_xlg(struct mvpp2_port *port) +{ + bool link; + u32 val; + + val = readl(port->base + MVPP22_XLG_INT_STAT); + if (val & MVPP22_XLG_INT_STAT_LINK) { + val = readl(port->base + MVPP22_XLG_STATUS); + link = (val & MVPP22_XLG_STATUS_LINK_UP); + mvpp2_isr_handle_link(port, link); + } +} + +static void mvpp2_isr_handle_gmac_internal(struct mvpp2_port *port) +{ + bool link; + u32 val; + + if (phy_interface_mode_is_rgmii(port->phy_interface) || + phy_interface_mode_is_8023z(port->phy_interface) || + port->phy_interface == PHY_INTERFACE_MODE_SGMII) { + val = readl(port->base + MVPP22_GMAC_INT_STAT); + if (val & MVPP22_GMAC_INT_STAT_LINK) { + val = readl(port->base + MVPP2_GMAC_STATUS0); + link = (val & MVPP2_GMAC_STATUS0_LINK_UP); + mvpp2_isr_handle_link(port, link); + } + } +} + +/* Per-port interrupt for link status changes */ +static irqreturn_t mvpp2_port_isr(int irq, void *dev_id) +{ + struct mvpp2_port *port = (struct mvpp2_port *)dev_id; + u32 val; + + mvpp22_gop_mask_irq(port); + + if (mvpp2_port_supports_xlg(port) && + mvpp2_is_xlg(port->phy_interface)) { + /* Check the external status register */ + val = readl(port->base + MVPP22_XLG_EXT_INT_STAT); + if (val & MVPP22_XLG_EXT_INT_STAT_XLG) + mvpp2_isr_handle_xlg(port); + if (val & MVPP22_XLG_EXT_INT_STAT_PTP) + mvpp2_isr_handle_ptp(port); + } else { + /* If it's not the XLG, we must be using the GMAC. + * Check the summary status. + */ + val = readl(port->base + MVPP22_GMAC_INT_SUM_STAT); + if (val & MVPP22_GMAC_INT_SUM_STAT_INTERNAL) + mvpp2_isr_handle_gmac_internal(port); + if (val & MVPP22_GMAC_INT_SUM_STAT_PTP) + mvpp2_isr_handle_ptp(port); + } -handled: mvpp22_gop_unmask_irq(port); return IRQ_HANDLED; } @@ -3427,7 +3510,7 @@ static int mvpp2_rx(struct mvpp2_port *port, struct napi_struct *napi, unsigned int frag_size; dma_addr_t dma_addr; phys_addr_t phys_addr; - u32 rx_status; + u32 rx_status, timestamp; int pool, rx_bytes, err, ret; void *data; @@ -3505,6 +3588,15 @@ static int mvpp2_rx(struct mvpp2_port *port, struct napi_struct *napi, goto err_drop_frame; } + /* If we have RX hardware timestamping enabled, grab the + * timestamp from the queue and convert. + */ + if (mvpp22_rx_hwtstamping(port)) { + timestamp = le32_to_cpu(rx_desc->pp22.timestamp); + mvpp22_tai_tstamp(port->priv->tai, timestamp, + skb_hwtstamps(skb)); + } + err = mvpp2_rx_refill(port, bm_pool, pp, pool); if (err) { netdev_err(port->dev, "failed to refill BM pools\n"); @@ -3579,6 +3671,94 @@ tx_desc_unmap_put(struct mvpp2_port *port, struct mvpp2_tx_queue *txq, mvpp2_txq_desc_put(txq); } +static void mvpp2_txdesc_clear_ptp(struct mvpp2_port *port, + struct mvpp2_tx_desc *desc) +{ + /* We only need to clear the low bits */ + if (port->priv->hw_version != MVPP21) + desc->pp22.ptp_descriptor &= + cpu_to_le32(~MVPP22_PTP_DESC_MASK_LOW); +} + +static bool mvpp2_tx_hw_tstamp(struct mvpp2_port *port, + struct mvpp2_tx_desc *tx_desc, + struct sk_buff *skb) +{ + struct mvpp2_hwtstamp_queue *queue; + unsigned int mtype, type, i; + struct ptp_header *hdr; + u64 ptpdesc; + + if (port->priv->hw_version == MVPP21 || + port->tx_hwtstamp_type == HWTSTAMP_TX_OFF) + return false; + + type = ptp_classify_raw(skb); + if (!type) + return false; + + hdr = ptp_parse_header(skb, type); + if (!hdr) + return false; + + skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; + + ptpdesc = MVPP22_PTP_MACTIMESTAMPINGEN | + MVPP22_PTP_ACTION_CAPTURE; + queue = &port->tx_hwtstamp_queue[0]; + + switch (type & PTP_CLASS_VMASK) { + case PTP_CLASS_V1: + ptpdesc |= MVPP22_PTP_PACKETFORMAT(MVPP22_PTP_PKT_FMT_PTPV1); + break; + + case PTP_CLASS_V2: + ptpdesc |= MVPP22_PTP_PACKETFORMAT(MVPP22_PTP_PKT_FMT_PTPV2); + mtype = hdr->tsmt & 15; + /* Direct PTP Sync messages to queue 1 */ + if (mtype == 0) { + ptpdesc |= MVPP22_PTP_TIMESTAMPQUEUESELECT; + queue = &port->tx_hwtstamp_queue[1]; + } + break; + } + + /* Take a reference on the skb and insert into our queue */ + i = queue->next; + queue->next = (i + 1) & 31; + if (queue->skb[i]) + dev_kfree_skb_any(queue->skb[i]); + queue->skb[i] = skb_get(skb); + + ptpdesc |= MVPP22_PTP_TIMESTAMPENTRYID(i); + + /* + * 3:0 - PTPAction + * 6:4 - PTPPacketFormat + * 7 - PTP_CF_WraparoundCheckEn + * 9:8 - IngressTimestampSeconds[1:0] + * 10 - Reserved + * 11 - MACTimestampingEn + * 17:12 - PTP_TimestampQueueEntryID[5:0] + * 18 - PTPTimestampQueueSelect + * 19 - UDPChecksumUpdateEn + * 27:20 - TimestampOffset + * PTP, NTPTransmit, OWAMP/TWAMP - L3 to PTP header + * NTPTs, Y.1731 - L3 to timestamp entry + * 35:28 - UDP Checksum Offset + * + * stored in tx descriptor bits 75:64 (11:0) and 191:168 (35:12) + */ + tx_desc->pp22.ptp_descriptor &= + cpu_to_le32(~MVPP22_PTP_DESC_MASK_LOW); + tx_desc->pp22.ptp_descriptor |= + cpu_to_le32(ptpdesc & MVPP22_PTP_DESC_MASK_LOW); + tx_desc->pp22.buf_dma_addr_ptp &= cpu_to_le64(~0xffffff0000000000ULL); + tx_desc->pp22.buf_dma_addr_ptp |= cpu_to_le64((ptpdesc >> 12) << 40); + + return true; +} + /* Handle tx fragmentation processing */ static int mvpp2_tx_frag_process(struct mvpp2_port *port, struct sk_buff *skb, struct mvpp2_tx_queue *aggr_txq, @@ -3595,6 +3775,7 @@ static int mvpp2_tx_frag_process(struct mvpp2_port *port, struct sk_buff *skb, void *addr = skb_frag_address(frag); tx_desc = mvpp2_txq_next_desc_get(aggr_txq); + mvpp2_txdesc_clear_ptp(port, tx_desc); mvpp2_txdesc_txq_set(port, tx_desc, txq->id); mvpp2_txdesc_size_set(port, tx_desc, skb_frag_size(frag)); @@ -3644,6 +3825,7 @@ static inline void mvpp2_tso_put_hdr(struct sk_buff *skb, struct mvpp2_tx_desc *tx_desc = mvpp2_txq_next_desc_get(aggr_txq); dma_addr_t addr; + mvpp2_txdesc_clear_ptp(port, tx_desc); mvpp2_txdesc_txq_set(port, tx_desc, txq->id); mvpp2_txdesc_size_set(port, tx_desc, hdr_sz); @@ -3668,6 +3850,7 @@ static inline int mvpp2_tso_put_data(struct sk_buff *skb, struct mvpp2_tx_desc *tx_desc = mvpp2_txq_next_desc_get(aggr_txq); dma_addr_t buf_dma_addr; + mvpp2_txdesc_clear_ptp(port, tx_desc); mvpp2_txdesc_txq_set(port, tx_desc, txq->id); mvpp2_txdesc_size_set(port, tx_desc, sz); @@ -3784,6 +3967,9 @@ static netdev_tx_t mvpp2_tx(struct sk_buff *skb, struct net_device *dev) /* Get a descriptor for the first part of the packet */ tx_desc = mvpp2_txq_next_desc_get(aggr_txq); + if (!(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) || + !mvpp2_tx_hw_tstamp(port, tx_desc, skb)) + mvpp2_txdesc_clear_ptp(port, tx_desc); mvpp2_txdesc_txq_set(port, tx_desc, txq->id); mvpp2_txdesc_size_set(port, tx_desc, skb_headlen(skb)); @@ -4007,17 +4193,7 @@ static void mvpp2_start_dev(struct mvpp2_port *port) if (port->phylink) { phylink_start(port->phylink); } else { - /* Phylink isn't used as of now for ACPI, so the MAC has to be - * configured manually when the interface is started. This will - * be removed as soon as the phylink ACPI support lands in. - */ - struct phylink_link_state state = { - .interface = port->phy_interface, - }; - mvpp2_mac_config(&port->phylink_config, MLO_AN_INBAND, &state); - mvpp2_mac_link_up(&port->phylink_config, NULL, - MLO_AN_INBAND, port->phy_interface, - SPEED_UNKNOWN, DUPLEX_UNKNOWN, false, false); + mvpp2_acpi_start(port); } netif_tx_start_all_queues(port->dev); @@ -4227,12 +4403,13 @@ static int mvpp2_open(struct net_device *dev) valid = true; } - if (priv->hw_version == MVPP22 && port->link_irq) { - err = request_irq(port->link_irq, mvpp2_link_status_isr, 0, + if (priv->hw_version == MVPP22 && port->port_irq) { + err = request_irq(port->port_irq, mvpp2_port_isr, 0, dev->name, port); if (err) { - netdev_err(port->dev, "cannot request link IRQ %d\n", - port->link_irq); + netdev_err(port->dev, + "cannot request port link/ptp IRQ %d\n", + port->port_irq); goto err_free_irq; } @@ -4243,7 +4420,7 @@ static int mvpp2_open(struct net_device *dev) valid = true; } else { - port->link_irq = 0; + port->port_irq = 0; } if (!valid) { @@ -4287,8 +4464,8 @@ static int mvpp2_stop(struct net_device *dev) if (port->phylink) phylink_disconnect_phy(port->phylink); - if (port->link_irq) - free_irq(port->link_irq, port); + if (port->port_irq) + free_irq(port->port_irq, port); mvpp2_irqs_deinit(port); if (!port->has_tx_irqs) { @@ -4548,10 +4725,124 @@ mvpp2_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats) stats->tx_dropped = dev->stats.tx_dropped; } +static int mvpp2_set_ts_config(struct mvpp2_port *port, struct ifreq *ifr) +{ + struct hwtstamp_config config; + void __iomem *ptp; + u32 gcr, int_mask; + + if (copy_from_user(&config, ifr->ifr_data, sizeof(config))) + return -EFAULT; + + if (config.flags) + return -EINVAL; + + if (config.tx_type != HWTSTAMP_TX_OFF && + config.tx_type != HWTSTAMP_TX_ON) + return -ERANGE; + + ptp = port->priv->iface_base + MVPP22_PTP_BASE(port->gop_id); + + int_mask = gcr = 0; + if (config.tx_type != HWTSTAMP_TX_OFF) { + gcr |= MVPP22_PTP_GCR_TSU_ENABLE | MVPP22_PTP_GCR_TX_RESET; + int_mask |= MVPP22_PTP_INT_MASK_QUEUE1 | + MVPP22_PTP_INT_MASK_QUEUE0; + } + + /* It seems we must also release the TX reset when enabling the TSU */ + if (config.rx_filter != HWTSTAMP_FILTER_NONE) + gcr |= MVPP22_PTP_GCR_TSU_ENABLE | MVPP22_PTP_GCR_RX_RESET | + MVPP22_PTP_GCR_TX_RESET; + + if (gcr & MVPP22_PTP_GCR_TSU_ENABLE) + mvpp22_tai_start(port->priv->tai); + + if (config.rx_filter != HWTSTAMP_FILTER_NONE) { + config.rx_filter = HWTSTAMP_FILTER_ALL; + mvpp2_modify(ptp + MVPP22_PTP_GCR, + MVPP22_PTP_GCR_RX_RESET | + MVPP22_PTP_GCR_TX_RESET | + MVPP22_PTP_GCR_TSU_ENABLE, gcr); + port->rx_hwtstamp = true; + } else { + port->rx_hwtstamp = false; + mvpp2_modify(ptp + MVPP22_PTP_GCR, + MVPP22_PTP_GCR_RX_RESET | + MVPP22_PTP_GCR_TX_RESET | + MVPP22_PTP_GCR_TSU_ENABLE, gcr); + } + + mvpp2_modify(ptp + MVPP22_PTP_INT_MASK, + MVPP22_PTP_INT_MASK_QUEUE1 | + MVPP22_PTP_INT_MASK_QUEUE0, int_mask); + + if (!(gcr & MVPP22_PTP_GCR_TSU_ENABLE)) + mvpp22_tai_stop(port->priv->tai); + + port->tx_hwtstamp_type = config.tx_type; + + if (copy_to_user(ifr->ifr_data, &config, sizeof(config))) + return -EFAULT; + + return 0; +} + +static int mvpp2_get_ts_config(struct mvpp2_port *port, struct ifreq *ifr) +{ + struct hwtstamp_config config; + + memset(&config, 0, sizeof(config)); + + config.tx_type = port->tx_hwtstamp_type; + config.rx_filter = port->rx_hwtstamp ? + HWTSTAMP_FILTER_ALL : HWTSTAMP_FILTER_NONE; + + if (copy_to_user(ifr->ifr_data, &config, sizeof(config))) + return -EFAULT; + + return 0; +} + +static int mvpp2_ethtool_get_ts_info(struct net_device *dev, + struct ethtool_ts_info *info) +{ + struct mvpp2_port *port = netdev_priv(dev); + + if (!port->hwtstamp) + return -EOPNOTSUPP; + + info->phc_index = mvpp22_tai_ptp_clock_index(port->priv->tai); + info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE | + SOF_TIMESTAMPING_RX_SOFTWARE | + SOF_TIMESTAMPING_SOFTWARE | + SOF_TIMESTAMPING_TX_HARDWARE | + SOF_TIMESTAMPING_RX_HARDWARE | + SOF_TIMESTAMPING_RAW_HARDWARE; + info->tx_types = BIT(HWTSTAMP_TX_OFF) | + BIT(HWTSTAMP_TX_ON); + info->rx_filters = BIT(HWTSTAMP_FILTER_NONE) | + BIT(HWTSTAMP_FILTER_ALL); + + return 0; +} + static int mvpp2_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { struct mvpp2_port *port = netdev_priv(dev); + switch (cmd) { + case SIOCSHWTSTAMP: + if (port->hwtstamp) + return mvpp2_set_ts_config(port, ifr); + break; + + case SIOCGHWTSTAMP: + if (port->hwtstamp) + return mvpp2_get_ts_config(port, ifr); + break; + } + if (!port->phylink) return -ENOTSUPP; @@ -5021,6 +5312,7 @@ static const struct ethtool_ops mvpp2_eth_tool_ops = { ETHTOOL_COALESCE_MAX_FRAMES, .nway_reset = mvpp2_ethtool_nway_reset, .get_link = ethtool_op_get_link, + .get_ts_info = mvpp2_ethtool_get_ts_info, .set_coalesce = mvpp2_ethtool_set_coalesce, .get_coalesce = mvpp2_ethtool_get_coalesce, .get_drvinfo = mvpp2_ethtool_get_drvinfo, @@ -5392,6 +5684,155 @@ static struct mvpp2_port *mvpp2_phylink_to_port(struct phylink_config *config) return container_of(config, struct mvpp2_port, phylink_config); } +static struct mvpp2_port *mvpp2_pcs_to_port(struct phylink_pcs *pcs) +{ + return container_of(pcs, struct mvpp2_port, phylink_pcs); +} + +static void mvpp2_xlg_pcs_get_state(struct phylink_pcs *pcs, + struct phylink_link_state *state) +{ + struct mvpp2_port *port = mvpp2_pcs_to_port(pcs); + u32 val; + + state->speed = SPEED_10000; + state->duplex = 1; + state->an_complete = 1; + + val = readl(port->base + MVPP22_XLG_STATUS); + state->link = !!(val & MVPP22_XLG_STATUS_LINK_UP); + + state->pause = 0; + val = readl(port->base + MVPP22_XLG_CTRL0_REG); + if (val & MVPP22_XLG_CTRL0_TX_FLOW_CTRL_EN) + state->pause |= MLO_PAUSE_TX; + if (val & MVPP22_XLG_CTRL0_RX_FLOW_CTRL_EN) + state->pause |= MLO_PAUSE_RX; +} + +static int mvpp2_xlg_pcs_config(struct phylink_pcs *pcs, + unsigned int mode, + phy_interface_t interface, + const unsigned long *advertising, + bool permit_pause_to_mac) +{ + return 0; +} + +static const struct phylink_pcs_ops mvpp2_phylink_xlg_pcs_ops = { + .pcs_get_state = mvpp2_xlg_pcs_get_state, + .pcs_config = mvpp2_xlg_pcs_config, +}; + +static void mvpp2_gmac_pcs_get_state(struct phylink_pcs *pcs, + struct phylink_link_state *state) +{ + struct mvpp2_port *port = mvpp2_pcs_to_port(pcs); + u32 val; + + val = readl(port->base + MVPP2_GMAC_STATUS0); + + state->an_complete = !!(val & MVPP2_GMAC_STATUS0_AN_COMPLETE); + state->link = !!(val & MVPP2_GMAC_STATUS0_LINK_UP); + state->duplex = !!(val & MVPP2_GMAC_STATUS0_FULL_DUPLEX); + + switch (port->phy_interface) { + case PHY_INTERFACE_MODE_1000BASEX: + state->speed = SPEED_1000; + break; + case PHY_INTERFACE_MODE_2500BASEX: + state->speed = SPEED_2500; + break; + default: + if (val & MVPP2_GMAC_STATUS0_GMII_SPEED) + state->speed = SPEED_1000; + else if (val & MVPP2_GMAC_STATUS0_MII_SPEED) + state->speed = SPEED_100; + else + state->speed = SPEED_10; + } + + state->pause = 0; + if (val & MVPP2_GMAC_STATUS0_RX_PAUSE) + state->pause |= MLO_PAUSE_RX; + if (val & MVPP2_GMAC_STATUS0_TX_PAUSE) + state->pause |= MLO_PAUSE_TX; +} + +static int mvpp2_gmac_pcs_config(struct phylink_pcs *pcs, unsigned int mode, + phy_interface_t interface, + const unsigned long *advertising, + bool permit_pause_to_mac) +{ + struct mvpp2_port *port = mvpp2_pcs_to_port(pcs); + u32 mask, val, an, old_an, changed; + + mask = MVPP2_GMAC_IN_BAND_AUTONEG_BYPASS | + MVPP2_GMAC_IN_BAND_AUTONEG | + MVPP2_GMAC_AN_SPEED_EN | + MVPP2_GMAC_FLOW_CTRL_AUTONEG | + MVPP2_GMAC_AN_DUPLEX_EN; + + if (phylink_autoneg_inband(mode)) { + mask |= MVPP2_GMAC_CONFIG_MII_SPEED | + MVPP2_GMAC_CONFIG_GMII_SPEED | + MVPP2_GMAC_CONFIG_FULL_DUPLEX; + val = MVPP2_GMAC_IN_BAND_AUTONEG; + + if (interface == PHY_INTERFACE_MODE_SGMII) { + /* SGMII mode receives the speed and duplex from PHY */ + val |= MVPP2_GMAC_AN_SPEED_EN | + MVPP2_GMAC_AN_DUPLEX_EN; + } else { + /* 802.3z mode has fixed speed and duplex */ + val |= MVPP2_GMAC_CONFIG_GMII_SPEED | + MVPP2_GMAC_CONFIG_FULL_DUPLEX; + + /* The FLOW_CTRL_AUTONEG bit selects either the hardware + * automatically or the bits in MVPP22_GMAC_CTRL_4_REG + * manually controls the GMAC pause modes. + */ + if (permit_pause_to_mac) + val |= MVPP2_GMAC_FLOW_CTRL_AUTONEG; + + /* Configure advertisement bits */ + mask |= MVPP2_GMAC_FC_ADV_EN | MVPP2_GMAC_FC_ADV_ASM_EN; + if (phylink_test(advertising, Pause)) + val |= MVPP2_GMAC_FC_ADV_EN; + if (phylink_test(advertising, Asym_Pause)) + val |= MVPP2_GMAC_FC_ADV_ASM_EN; + } + } else { + val = 0; + } + + old_an = an = readl(port->base + MVPP2_GMAC_AUTONEG_CONFIG); + an = (an & ~mask) | val; + changed = an ^ old_an; + if (changed) + writel(an, port->base + MVPP2_GMAC_AUTONEG_CONFIG); + + /* We are only interested in the advertisement bits changing */ + return changed & (MVPP2_GMAC_FC_ADV_EN | MVPP2_GMAC_FC_ADV_ASM_EN); +} + +static void mvpp2_gmac_pcs_an_restart(struct phylink_pcs *pcs) +{ + struct mvpp2_port *port = mvpp2_pcs_to_port(pcs); + u32 val = readl(port->base + MVPP2_GMAC_AUTONEG_CONFIG); + + writel(val | MVPP2_GMAC_IN_BAND_RESTART_AN, + port->base + MVPP2_GMAC_AUTONEG_CONFIG); + writel(val & ~MVPP2_GMAC_IN_BAND_RESTART_AN, + port->base + MVPP2_GMAC_AUTONEG_CONFIG); +} + +static const struct phylink_pcs_ops mvpp2_phylink_gmac_pcs_ops = { + .pcs_get_state = mvpp2_gmac_pcs_get_state, + .pcs_config = mvpp2_gmac_pcs_config, + .pcs_an_restart = mvpp2_gmac_pcs_an_restart, +}; + static void mvpp2_phylink_validate(struct phylink_config *config, unsigned long *supported, struct phylink_link_state *state) @@ -5480,89 +5921,6 @@ empty_set: bitmap_zero(supported, __ETHTOOL_LINK_MODE_MASK_NBITS); } -static void mvpp22_xlg_pcs_get_state(struct mvpp2_port *port, - struct phylink_link_state *state) -{ - u32 val; - - state->speed = SPEED_10000; - state->duplex = 1; - state->an_complete = 1; - - val = readl(port->base + MVPP22_XLG_STATUS); - state->link = !!(val & MVPP22_XLG_STATUS_LINK_UP); - - state->pause = 0; - val = readl(port->base + MVPP22_XLG_CTRL0_REG); - if (val & MVPP22_XLG_CTRL0_TX_FLOW_CTRL_EN) - state->pause |= MLO_PAUSE_TX; - if (val & MVPP22_XLG_CTRL0_RX_FLOW_CTRL_EN) - state->pause |= MLO_PAUSE_RX; -} - -static void mvpp2_gmac_pcs_get_state(struct mvpp2_port *port, - struct phylink_link_state *state) -{ - u32 val; - - val = readl(port->base + MVPP2_GMAC_STATUS0); - - state->an_complete = !!(val & MVPP2_GMAC_STATUS0_AN_COMPLETE); - state->link = !!(val & MVPP2_GMAC_STATUS0_LINK_UP); - state->duplex = !!(val & MVPP2_GMAC_STATUS0_FULL_DUPLEX); - - switch (port->phy_interface) { - case PHY_INTERFACE_MODE_1000BASEX: - state->speed = SPEED_1000; - break; - case PHY_INTERFACE_MODE_2500BASEX: - state->speed = SPEED_2500; - break; - default: - if (val & MVPP2_GMAC_STATUS0_GMII_SPEED) - state->speed = SPEED_1000; - else if (val & MVPP2_GMAC_STATUS0_MII_SPEED) - state->speed = SPEED_100; - else - state->speed = SPEED_10; - } - - state->pause = 0; - if (val & MVPP2_GMAC_STATUS0_RX_PAUSE) - state->pause |= MLO_PAUSE_RX; - if (val & MVPP2_GMAC_STATUS0_TX_PAUSE) - state->pause |= MLO_PAUSE_TX; -} - -static void mvpp2_phylink_mac_pcs_get_state(struct phylink_config *config, - struct phylink_link_state *state) -{ - struct mvpp2_port *port = mvpp2_phylink_to_port(config); - - if (port->priv->hw_version == MVPP22 && port->gop_id == 0) { - u32 mode = readl(port->base + MVPP22_XLG_CTRL3_REG); - mode &= MVPP22_XLG_CTRL3_MACMODESELECT_MASK; - - if (mode == MVPP22_XLG_CTRL3_MACMODESELECT_10G) { - mvpp22_xlg_pcs_get_state(port, state); - return; - } - } - - mvpp2_gmac_pcs_get_state(port, state); -} - -static void mvpp2_mac_an_restart(struct phylink_config *config) -{ - struct mvpp2_port *port = mvpp2_phylink_to_port(config); - u32 val = readl(port->base + MVPP2_GMAC_AUTONEG_CONFIG); - - writel(val | MVPP2_GMAC_IN_BAND_RESTART_AN, - port->base + MVPP2_GMAC_AUTONEG_CONFIG); - writel(val & ~MVPP2_GMAC_IN_BAND_RESTART_AN, - port->base + MVPP2_GMAC_AUTONEG_CONFIG); -} - static void mvpp2_xlg_config(struct mvpp2_port *port, unsigned int mode, const struct phylink_link_state *state) { @@ -5586,23 +5944,16 @@ static void mvpp2_xlg_config(struct mvpp2_port *port, unsigned int mode, static void mvpp2_gmac_config(struct mvpp2_port *port, unsigned int mode, const struct phylink_link_state *state) { - u32 old_an, an; u32 old_ctrl0, ctrl0; u32 old_ctrl2, ctrl2; u32 old_ctrl4, ctrl4; - old_an = an = readl(port->base + MVPP2_GMAC_AUTONEG_CONFIG); old_ctrl0 = ctrl0 = readl(port->base + MVPP2_GMAC_CTRL_0_REG); old_ctrl2 = ctrl2 = readl(port->base + MVPP2_GMAC_CTRL_2_REG); old_ctrl4 = ctrl4 = readl(port->base + MVPP22_GMAC_CTRL_4_REG); - an &= ~(MVPP2_GMAC_AN_SPEED_EN | MVPP2_GMAC_FC_ADV_EN | - MVPP2_GMAC_FC_ADV_ASM_EN | MVPP2_GMAC_FLOW_CTRL_AUTONEG | - MVPP2_GMAC_AN_DUPLEX_EN | MVPP2_GMAC_IN_BAND_AUTONEG | - MVPP2_GMAC_IN_BAND_AUTONEG_BYPASS); ctrl0 &= ~MVPP2_GMAC_PORT_TYPE_MASK; - ctrl2 &= ~(MVPP2_GMAC_INBAND_AN_MASK | MVPP2_GMAC_PORT_RESET_MASK | - MVPP2_GMAC_PCS_ENABLE_MASK); + ctrl2 &= ~(MVPP2_GMAC_INBAND_AN_MASK | MVPP2_GMAC_PCS_ENABLE_MASK); /* Configure port type */ if (phy_interface_mode_is_8023z(state->interface)) { @@ -5624,12 +5975,6 @@ static void mvpp2_gmac_config(struct mvpp2_port *port, unsigned int mode, MVPP22_CTRL4_QSGMII_BYPASS_ACTIVE; } - /* Configure advertisement bits */ - if (phylink_test(state->advertising, Pause)) - an |= MVPP2_GMAC_FC_ADV_EN; - if (phylink_test(state->advertising, Asym_Pause)) - an |= MVPP2_GMAC_FC_ADV_ASM_EN; - /* Configure negotiation style */ if (!phylink_autoneg_inband(mode)) { /* Phy or fixed speed - no in-band AN, nothing to do, leave the @@ -5638,14 +5983,6 @@ static void mvpp2_gmac_config(struct mvpp2_port *port, unsigned int mode, } else if (state->interface == PHY_INTERFACE_MODE_SGMII) { /* SGMII in-band mode receives the speed and duplex from * the PHY. Flow control information is not received. */ - an &= ~(MVPP2_GMAC_FORCE_LINK_DOWN | - MVPP2_GMAC_FORCE_LINK_PASS | - MVPP2_GMAC_CONFIG_MII_SPEED | - MVPP2_GMAC_CONFIG_GMII_SPEED | - MVPP2_GMAC_CONFIG_FULL_DUPLEX); - an |= MVPP2_GMAC_IN_BAND_AUTONEG | - MVPP2_GMAC_AN_SPEED_EN | - MVPP2_GMAC_AN_DUPLEX_EN; } else if (phy_interface_mode_is_8023z(state->interface)) { /* 1000BaseX and 2500BaseX ports cannot negotiate speed nor can * they negotiate duplex: they are always operating with a fixed @@ -5653,42 +5990,6 @@ static void mvpp2_gmac_config(struct mvpp2_port *port, unsigned int mode, * speed and full duplex here. */ ctrl0 |= MVPP2_GMAC_PORT_TYPE_MASK; - an &= ~(MVPP2_GMAC_FORCE_LINK_DOWN | - MVPP2_GMAC_FORCE_LINK_PASS | - MVPP2_GMAC_CONFIG_MII_SPEED | - MVPP2_GMAC_CONFIG_GMII_SPEED | - MVPP2_GMAC_CONFIG_FULL_DUPLEX); - an |= MVPP2_GMAC_IN_BAND_AUTONEG | - MVPP2_GMAC_CONFIG_GMII_SPEED | - MVPP2_GMAC_CONFIG_FULL_DUPLEX; - - if (state->pause & MLO_PAUSE_AN && state->an_enabled) - an |= MVPP2_GMAC_FLOW_CTRL_AUTONEG; - } - -/* Some fields of the auto-negotiation register require the port to be down when - * their value is updated. - */ -#define MVPP2_GMAC_AN_PORT_DOWN_MASK \ - (MVPP2_GMAC_IN_BAND_AUTONEG | \ - MVPP2_GMAC_IN_BAND_AUTONEG_BYPASS | \ - MVPP2_GMAC_CONFIG_MII_SPEED | MVPP2_GMAC_CONFIG_GMII_SPEED | \ - MVPP2_GMAC_AN_SPEED_EN | MVPP2_GMAC_CONFIG_FULL_DUPLEX | \ - MVPP2_GMAC_AN_DUPLEX_EN) - - if ((old_ctrl0 ^ ctrl0) & MVPP2_GMAC_PORT_TYPE_MASK || - (old_ctrl2 ^ ctrl2) & MVPP2_GMAC_INBAND_AN_MASK || - (old_an ^ an) & MVPP2_GMAC_AN_PORT_DOWN_MASK) { - /* Force link down */ - old_an &= ~MVPP2_GMAC_FORCE_LINK_PASS; - old_an |= MVPP2_GMAC_FORCE_LINK_DOWN; - writel(old_an, port->base + MVPP2_GMAC_AUTONEG_CONFIG); - - /* Set the GMAC in a reset state - do this in a way that - * ensures we clear it below. - */ - old_ctrl2 |= MVPP2_GMAC_PORT_RESET_MASK; - writel(old_ctrl2, port->base + MVPP2_GMAC_CTRL_2_REG); } if (old_ctrl0 != ctrl0) @@ -5697,41 +5998,85 @@ static void mvpp2_gmac_config(struct mvpp2_port *port, unsigned int mode, writel(ctrl2, port->base + MVPP2_GMAC_CTRL_2_REG); if (old_ctrl4 != ctrl4) writel(ctrl4, port->base + MVPP22_GMAC_CTRL_4_REG); - if (old_an != an) - writel(an, port->base + MVPP2_GMAC_AUTONEG_CONFIG); - - if (old_ctrl2 & MVPP2_GMAC_PORT_RESET_MASK) { - while (readl(port->base + MVPP2_GMAC_CTRL_2_REG) & - MVPP2_GMAC_PORT_RESET_MASK) - continue; - } } -static void mvpp2_mac_config(struct phylink_config *config, unsigned int mode, - const struct phylink_link_state *state) +static int mvpp2__mac_prepare(struct phylink_config *config, unsigned int mode, + phy_interface_t interface) { struct mvpp2_port *port = mvpp2_phylink_to_port(config); - bool change_interface = port->phy_interface != state->interface; /* Check for invalid configuration */ - if (mvpp2_is_xlg(state->interface) && port->gop_id != 0) { + if (mvpp2_is_xlg(interface) && port->gop_id != 0) { netdev_err(port->dev, "Invalid mode on %s\n", port->dev->name); - return; + return -EINVAL; + } + + if (port->phy_interface != interface || + phylink_autoneg_inband(mode)) { + /* Force the link down when changing the interface or if in + * in-band mode to ensure we do not change the configuration + * while the hardware is indicating link is up. We force both + * XLG and GMAC down to ensure that they're both in a known + * state. + */ + mvpp2_modify(port->base + MVPP2_GMAC_AUTONEG_CONFIG, + MVPP2_GMAC_FORCE_LINK_PASS | + MVPP2_GMAC_FORCE_LINK_DOWN, + MVPP2_GMAC_FORCE_LINK_DOWN); + + if (mvpp2_port_supports_xlg(port)) + mvpp2_modify(port->base + MVPP22_XLG_CTRL0_REG, + MVPP22_XLG_CTRL0_FORCE_LINK_PASS | + MVPP22_XLG_CTRL0_FORCE_LINK_DOWN, + MVPP22_XLG_CTRL0_FORCE_LINK_DOWN); } /* Make sure the port is disabled when reconfiguring the mode */ mvpp2_port_disable(port); - if (port->priv->hw_version == MVPP22 && change_interface) { - mvpp22_gop_mask_irq(port); + if (port->phy_interface != interface) { + /* Place GMAC into reset */ + mvpp2_modify(port->base + MVPP2_GMAC_CTRL_2_REG, + MVPP2_GMAC_PORT_RESET_MASK, + MVPP2_GMAC_PORT_RESET_MASK); - port->phy_interface = state->interface; + if (port->priv->hw_version == MVPP22) { + mvpp22_gop_mask_irq(port); - /* Reconfigure the serdes lanes */ - phy_power_off(port->comphy); - mvpp22_mode_reconfigure(port); + phy_power_off(port->comphy); + } } + /* Select the appropriate PCS operations depending on the + * configured interface mode. We will only switch to a mode + * that the validate() checks have already passed. + */ + if (mvpp2_is_xlg(interface)) + port->phylink_pcs.ops = &mvpp2_phylink_xlg_pcs_ops; + else + port->phylink_pcs.ops = &mvpp2_phylink_gmac_pcs_ops; + + return 0; +} + +static int mvpp2_mac_prepare(struct phylink_config *config, unsigned int mode, + phy_interface_t interface) +{ + struct mvpp2_port *port = mvpp2_phylink_to_port(config); + int ret; + + ret = mvpp2__mac_prepare(config, mode, interface); + if (ret == 0) + phylink_set_pcs(port->phylink, &port->phylink_pcs); + + return ret; +} + +static void mvpp2_mac_config(struct phylink_config *config, unsigned int mode, + const struct phylink_link_state *state) +{ + struct mvpp2_port *port = mvpp2_phylink_to_port(config); + /* mac (re)configuration */ if (mvpp2_is_xlg(state->interface)) mvpp2_xlg_config(port, mode, state); @@ -5742,11 +6087,51 @@ static void mvpp2_mac_config(struct phylink_config *config, unsigned int mode, if (port->priv->hw_version == MVPP21 && port->flags & MVPP2_F_LOOPBACK) mvpp2_port_loopback_set(port, state); +} + +static int mvpp2_mac_finish(struct phylink_config *config, unsigned int mode, + phy_interface_t interface) +{ + struct mvpp2_port *port = mvpp2_phylink_to_port(config); + + if (port->priv->hw_version == MVPP22 && + port->phy_interface != interface) { + port->phy_interface = interface; - if (port->priv->hw_version == MVPP22 && change_interface) + /* Reconfigure the serdes lanes */ + mvpp22_mode_reconfigure(port); + + /* Unmask interrupts */ mvpp22_gop_unmask_irq(port); + } + + if (!mvpp2_is_xlg(interface)) { + /* Release GMAC reset and wait */ + mvpp2_modify(port->base + MVPP2_GMAC_CTRL_2_REG, + MVPP2_GMAC_PORT_RESET_MASK, 0); + + while (readl(port->base + MVPP2_GMAC_CTRL_2_REG) & + MVPP2_GMAC_PORT_RESET_MASK) + continue; + } mvpp2_port_enable(port); + + /* Allow the link to come up if in in-band mode, otherwise the + * link is forced via mac_link_down()/mac_link_up() + */ + if (phylink_autoneg_inband(mode)) { + if (mvpp2_is_xlg(interface)) + mvpp2_modify(port->base + MVPP22_XLG_CTRL0_REG, + MVPP22_XLG_CTRL0_FORCE_LINK_PASS | + MVPP22_XLG_CTRL0_FORCE_LINK_DOWN, 0); + else + mvpp2_modify(port->base + MVPP2_GMAC_AUTONEG_CONFIG, + MVPP2_GMAC_FORCE_LINK_PASS | + MVPP2_GMAC_FORCE_LINK_DOWN, 0); + } + + return 0; } static void mvpp2_mac_link_up(struct phylink_config *config, @@ -5843,13 +6228,36 @@ static void mvpp2_mac_link_down(struct phylink_config *config, static const struct phylink_mac_ops mvpp2_phylink_ops = { .validate = mvpp2_phylink_validate, - .mac_pcs_get_state = mvpp2_phylink_mac_pcs_get_state, - .mac_an_restart = mvpp2_mac_an_restart, + .mac_prepare = mvpp2_mac_prepare, .mac_config = mvpp2_mac_config, + .mac_finish = mvpp2_mac_finish, .mac_link_up = mvpp2_mac_link_up, .mac_link_down = mvpp2_mac_link_down, }; +/* Work-around for ACPI */ +static void mvpp2_acpi_start(struct mvpp2_port *port) +{ + /* Phylink isn't used as of now for ACPI, so the MAC has to be + * configured manually when the interface is started. This will + * be removed as soon as the phylink ACPI support lands in. + */ + struct phylink_link_state state = { + .interface = port->phy_interface, + }; + mvpp2__mac_prepare(&port->phylink_config, MLO_AN_INBAND, + port->phy_interface); + mvpp2_mac_config(&port->phylink_config, MLO_AN_INBAND, &state); + port->phylink_pcs.ops->pcs_config(&port->phylink_pcs, MLO_AN_INBAND, + port->phy_interface, + state.advertising, false); + mvpp2_mac_finish(&port->phylink_config, MLO_AN_INBAND, + port->phy_interface); + mvpp2_mac_link_up(&port->phylink_config, NULL, + MLO_AN_INBAND, port->phy_interface, + SPEED_UNKNOWN, DUPLEX_UNKNOWN, false, false); +} + /* Ports initialization */ static int mvpp2_port_probe(struct platform_device *pdev, struct fwnode_handle *port_fwnode, @@ -5937,16 +6345,16 @@ static int mvpp2_port_probe(struct platform_device *pdev, goto err_free_netdev; if (port_node) - port->link_irq = of_irq_get_byname(port_node, "link"); + port->port_irq = of_irq_get_byname(port_node, "link"); else - port->link_irq = fwnode_irq_get(port_fwnode, port->nqvecs + 1); - if (port->link_irq == -EPROBE_DEFER) { + port->port_irq = fwnode_irq_get(port_fwnode, port->nqvecs + 1); + if (port->port_irq == -EPROBE_DEFER) { err = -EPROBE_DEFER; goto err_deinit_qvecs; } - if (port->link_irq <= 0) + if (port->port_irq <= 0) /* the link irq is optional */ - port->link_irq = 0; + port->port_irq = 0; if (fwnode_property_read_bool(port_fwnode, "marvell,loopback")) port->flags |= MVPP2_F_LOOPBACK; @@ -5983,6 +6391,12 @@ static int mvpp2_port_probe(struct platform_device *pdev, port->stats_base = port->priv->iface_base + MVPP22_MIB_COUNTERS_OFFSET + port->gop_id * MVPP22_MIB_COUNTERS_PORT_SZ; + + /* We may want a property to describe whether we should use + * MAC hardware timestamping. + */ + if (priv->tai) + port->hwtstamp = true; } /* Alloc per-cpu and ethtool stats */ @@ -6110,8 +6524,8 @@ err_free_txq_pcpu: err_free_stats: free_percpu(port->stats); err_free_irq: - if (port->link_irq) - irq_dispose_mapping(port->link_irq); + if (port->port_irq) + irq_dispose_mapping(port->port_irq); err_deinit_qvecs: mvpp2_queue_vectors_deinit(port); err_free_netdev: @@ -6132,8 +6546,8 @@ static void mvpp2_port_remove(struct mvpp2_port *port) for (i = 0; i < port->ntxqs; i++) free_percpu(port->txqs[i]->pcpu); mvpp2_queue_vectors_deinit(port); - if (port->link_irq) - irq_dispose_mapping(port->link_irq); + if (port->port_irq) + irq_dispose_mapping(port->port_irq); free_netdev(port->dev); } @@ -6545,6 +6959,10 @@ static int mvpp2_probe(struct platform_device *pdev) goto err_axi_clk; } + err = mvpp22_tai_probe(&pdev->dev, priv); + if (err < 0) + goto err_axi_clk; + /* Initialize ports */ fwnode_for_each_available_child_node(fwnode, port_fwnode) { err = mvpp2_port_probe(pdev, port_fwnode, priv); @@ -6663,11 +7081,13 @@ static const struct of_device_id mvpp2_match[] = { }; MODULE_DEVICE_TABLE(of, mvpp2_match); +#ifdef CONFIG_ACPI static const struct acpi_device_id mvpp2_acpi_match[] = { { "MRVL0110", MVPP22 }, { }, }; MODULE_DEVICE_TABLE(acpi, mvpp2_acpi_match); +#endif static struct platform_driver mvpp2_driver = { .probe = mvpp2_probe, diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_tai.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_tai.c new file mode 100644 index 000000000000..95862aff49f1 --- /dev/null +++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_tai.c @@ -0,0 +1,457 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Marvell PP2.2 TAI support + * + * Note: + * Do NOT use the event capture support. + * Do Not even set the MPP muxes to allow PTP_EVENT_REQ to be used. + * It will disrupt the operation of this driver, and there is nothing + * that this driver can do to prevent that. Even using PTP_EVENT_REQ + * as an output will be seen as a trigger input, which can't be masked. + * When ever a trigger input is seen, the action in the TCFCR0_TCF + * field will be performed - whether it is a set, increment, decrement + * read, or frequency update. + * + * Other notes (useful, not specified in the documentation): + * - PTP_PULSE_OUT (PTP_EVENT_REQ MPP) + * It looks like the hardware can't generate a pulse at nsec=0. (The + * output doesn't trigger if the nsec field is zero.) + * Note: when configured as an output via the register at 0xfX441120, + * the input is still very much alive, and will trigger the current TCF + * function. + * - PTP_CLK_OUT (PTP_TRIG_GEN MPP) + * This generates a "PPS" signal determined by the CCC registers. It + * seems this is not aligned to the TOD counter in any way (it may be + * initially, but if you specify a non-round second interval, it won't, + * and you can't easily get it back.) + * - PTP_PCLK_OUT + * This generates a 50% duty cycle clock based on the TOD counter, and + * seems it can be set to any period of 1ns resolution. It is probably + * limited by the TOD step size. Its period is defined by the PCLK_CCC + * registers. Again, its alignment to the second is questionable. + * + * Consequently, we support none of these. + */ +#include <linux/io.h> +#include <linux/ptp_clock_kernel.h> +#include <linux/slab.h> + +#include "mvpp2.h" + +#define CR0_SW_NRESET BIT(0) + +#define TCFCR0_PHASE_UPDATE_ENABLE BIT(8) +#define TCFCR0_TCF_MASK (7 << 2) +#define TCFCR0_TCF_UPDATE (0 << 2) +#define TCFCR0_TCF_FREQUPDATE (1 << 2) +#define TCFCR0_TCF_INCREMENT (2 << 2) +#define TCFCR0_TCF_DECREMENT (3 << 2) +#define TCFCR0_TCF_CAPTURE (4 << 2) +#define TCFCR0_TCF_NOP (7 << 2) +#define TCFCR0_TCF_TRIGGER BIT(0) + +#define TCSR_CAPTURE_1_VALID BIT(1) +#define TCSR_CAPTURE_0_VALID BIT(0) + +struct mvpp2_tai { + struct ptp_clock_info caps; + struct ptp_clock *ptp_clock; + void __iomem *base; + spinlock_t lock; + u64 period; // nanosecond period in 32.32 fixed point + /* This timestamp is updated every two seconds */ + struct timespec64 stamp; +}; + +static void mvpp2_tai_modify(void __iomem *reg, u32 mask, u32 set) +{ + u32 val; + + val = readl_relaxed(reg) & ~mask; + val |= set & mask; + writel(val, reg); +} + +static void mvpp2_tai_write(u32 val, void __iomem *reg) +{ + writel_relaxed(val & 0xffff, reg); +} + +static u32 mvpp2_tai_read(void __iomem *reg) +{ + return readl_relaxed(reg) & 0xffff; +} + +static struct mvpp2_tai *ptp_to_tai(struct ptp_clock_info *ptp) +{ + return container_of(ptp, struct mvpp2_tai, caps); +} + +static void mvpp22_tai_read_ts(struct timespec64 *ts, void __iomem *base) +{ + ts->tv_sec = (u64)mvpp2_tai_read(base + 0) << 32 | + mvpp2_tai_read(base + 4) << 16 | + mvpp2_tai_read(base + 8); + + ts->tv_nsec = mvpp2_tai_read(base + 12) << 16 | + mvpp2_tai_read(base + 16); + + /* Read and discard fractional part */ + readl_relaxed(base + 20); + readl_relaxed(base + 24); +} + +static void mvpp2_tai_write_tlv(const struct timespec64 *ts, u32 frac, + void __iomem *base) +{ + mvpp2_tai_write(ts->tv_sec >> 32, base + MVPP22_TAI_TLV_SEC_HIGH); + mvpp2_tai_write(ts->tv_sec >> 16, base + MVPP22_TAI_TLV_SEC_MED); + mvpp2_tai_write(ts->tv_sec, base + MVPP22_TAI_TLV_SEC_LOW); + mvpp2_tai_write(ts->tv_nsec >> 16, base + MVPP22_TAI_TLV_NANO_HIGH); + mvpp2_tai_write(ts->tv_nsec, base + MVPP22_TAI_TLV_NANO_LOW); + mvpp2_tai_write(frac >> 16, base + MVPP22_TAI_TLV_FRAC_HIGH); + mvpp2_tai_write(frac, base + MVPP22_TAI_TLV_FRAC_LOW); +} + +static void mvpp2_tai_op(u32 op, void __iomem *base) +{ + /* Trigger the operation. Note that an external unmaskable + * event on PTP_EVENT_REQ will also trigger this action. + */ + mvpp2_tai_modify(base + MVPP22_TAI_TCFCR0, + TCFCR0_TCF_MASK | TCFCR0_TCF_TRIGGER, + op | TCFCR0_TCF_TRIGGER); + mvpp2_tai_modify(base + MVPP22_TAI_TCFCR0, TCFCR0_TCF_MASK, + TCFCR0_TCF_NOP); +} + +/* The adjustment has a range of +0.5ns to -0.5ns in 2^32 steps, so has units + * of 2^-32 ns. + * + * units(s) = 1 / (2^32 * 10^9) + * fractional = abs_scaled_ppm / (2^16 * 10^6) + * + * What we want to achieve: + * freq_adjusted = freq_nominal * (1 + fractional) + * freq_delta = freq_adjusted - freq_nominal => positive = faster + * freq_delta = freq_nominal * (1 + fractional) - freq_nominal + * So: freq_delta = freq_nominal * fractional + * + * However, we are dealing with periods, so: + * period_adjusted = period_nominal / (1 + fractional) + * period_delta = period_nominal - period_adjusted => positive = faster + * period_delta = period_nominal * fractional / (1 + fractional) + * + * Hence: + * period_delta = period_nominal * abs_scaled_ppm / + * (2^16 * 10^6 + abs_scaled_ppm) + * + * To avoid overflow, we reduce both sides of the divide operation by a factor + * of 16. + */ +static u64 mvpp22_calc_frac_ppm(struct mvpp2_tai *tai, long abs_scaled_ppm) +{ + u64 val = tai->period * abs_scaled_ppm >> 4; + + return div_u64(val, (1000000 << 12) + (abs_scaled_ppm >> 4)); +} + +static s32 mvpp22_calc_max_adj(struct mvpp2_tai *tai) +{ + return 1000000; +} + +static int mvpp22_tai_adjfine(struct ptp_clock_info *ptp, long scaled_ppm) +{ + struct mvpp2_tai *tai = ptp_to_tai(ptp); + unsigned long flags; + void __iomem *base; + bool neg_adj; + s32 frac; + u64 val; + + neg_adj = scaled_ppm < 0; + if (neg_adj) + scaled_ppm = -scaled_ppm; + + val = mvpp22_calc_frac_ppm(tai, scaled_ppm); + + /* Convert to a signed 32-bit adjustment */ + if (neg_adj) { + /* -S32_MIN warns, -val < S32_MIN fails, so go for the easy + * solution. + */ + if (val > 0x80000000) + return -ERANGE; + + frac = -val; + } else { + if (val > S32_MAX) + return -ERANGE; + + frac = val; + } + + base = tai->base; + spin_lock_irqsave(&tai->lock, flags); + mvpp2_tai_write(frac >> 16, base + MVPP22_TAI_TLV_FRAC_HIGH); + mvpp2_tai_write(frac, base + MVPP22_TAI_TLV_FRAC_LOW); + mvpp2_tai_op(TCFCR0_TCF_FREQUPDATE, base); + spin_unlock_irqrestore(&tai->lock, flags); + + return 0; +} + +static int mvpp22_tai_adjtime(struct ptp_clock_info *ptp, s64 delta) +{ + struct mvpp2_tai *tai = ptp_to_tai(ptp); + struct timespec64 ts; + unsigned long flags; + void __iomem *base; + u32 tcf; + + /* We can't deal with S64_MIN */ + if (delta == S64_MIN) + return -ERANGE; + + if (delta < 0) { + delta = -delta; + tcf = TCFCR0_TCF_DECREMENT; + } else { + tcf = TCFCR0_TCF_INCREMENT; + } + + ts = ns_to_timespec64(delta); + + base = tai->base; + spin_lock_irqsave(&tai->lock, flags); + mvpp2_tai_write_tlv(&ts, 0, base); + mvpp2_tai_op(tcf, base); + spin_unlock_irqrestore(&tai->lock, flags); + + return 0; +} + +static int mvpp22_tai_gettimex64(struct ptp_clock_info *ptp, + struct timespec64 *ts, + struct ptp_system_timestamp *sts) +{ + struct mvpp2_tai *tai = ptp_to_tai(ptp); + unsigned long flags; + void __iomem *base; + u32 tcsr; + int ret; + + base = tai->base; + spin_lock_irqsave(&tai->lock, flags); + /* XXX: the only way to read the PTP time is for the CPU to trigger + * an event. However, there is no way to distinguish between the CPU + * triggered event, and an external event on PTP_EVENT_REQ. So this + * is incompatible with external use of PTP_EVENT_REQ. + */ + ptp_read_system_prets(sts); + mvpp2_tai_modify(base + MVPP22_TAI_TCFCR0, + TCFCR0_TCF_MASK | TCFCR0_TCF_TRIGGER, + TCFCR0_TCF_CAPTURE | TCFCR0_TCF_TRIGGER); + ptp_read_system_postts(sts); + mvpp2_tai_modify(base + MVPP22_TAI_TCFCR0, TCFCR0_TCF_MASK, + TCFCR0_TCF_NOP); + + tcsr = readl(base + MVPP22_TAI_TCSR); + if (tcsr & TCSR_CAPTURE_1_VALID) { + mvpp22_tai_read_ts(ts, base + MVPP22_TAI_TCV1_SEC_HIGH); + ret = 0; + } else if (tcsr & TCSR_CAPTURE_0_VALID) { + mvpp22_tai_read_ts(ts, base + MVPP22_TAI_TCV0_SEC_HIGH); + ret = 0; + } else { + /* We don't seem to have a reading... */ + ret = -EBUSY; + } + spin_unlock_irqrestore(&tai->lock, flags); + + return ret; +} + +static int mvpp22_tai_settime64(struct ptp_clock_info *ptp, + const struct timespec64 *ts) +{ + struct mvpp2_tai *tai = ptp_to_tai(ptp); + unsigned long flags; + void __iomem *base; + + base = tai->base; + spin_lock_irqsave(&tai->lock, flags); + mvpp2_tai_write_tlv(ts, 0, base); + + /* Trigger an update to load the value from the TLV registers + * into the TOD counter. Note that an external unmaskable event on + * PTP_EVENT_REQ will also trigger this action. + */ + mvpp2_tai_modify(base + MVPP22_TAI_TCFCR0, + TCFCR0_PHASE_UPDATE_ENABLE | + TCFCR0_TCF_MASK | TCFCR0_TCF_TRIGGER, + TCFCR0_TCF_UPDATE | TCFCR0_TCF_TRIGGER); + mvpp2_tai_modify(base + MVPP22_TAI_TCFCR0, TCFCR0_TCF_MASK, + TCFCR0_TCF_NOP); + spin_unlock_irqrestore(&tai->lock, flags); + + return 0; +} + +static long mvpp22_tai_aux_work(struct ptp_clock_info *ptp) +{ + struct mvpp2_tai *tai = ptp_to_tai(ptp); + + mvpp22_tai_gettimex64(ptp, &tai->stamp, NULL); + + return msecs_to_jiffies(2000); +} + +static void mvpp22_tai_set_step(struct mvpp2_tai *tai) +{ + void __iomem *base = tai->base; + u32 nano, frac; + + nano = upper_32_bits(tai->period); + frac = lower_32_bits(tai->period); + + /* As the fractional nanosecond is a signed offset, if the MSB (sign) + * bit is set, we have to increment the whole nanoseconds. + */ + if (frac >= 0x80000000) + nano += 1; + + mvpp2_tai_write(nano, base + MVPP22_TAI_TOD_STEP_NANO_CR); + mvpp2_tai_write(frac >> 16, base + MVPP22_TAI_TOD_STEP_FRAC_HIGH); + mvpp2_tai_write(frac, base + MVPP22_TAI_TOD_STEP_FRAC_LOW); +} + +static void mvpp22_tai_init(struct mvpp2_tai *tai) +{ + void __iomem *base = tai->base; + + mvpp22_tai_set_step(tai); + + /* Release the TAI reset */ + mvpp2_tai_modify(base + MVPP22_TAI_CR0, CR0_SW_NRESET, CR0_SW_NRESET); +} + +int mvpp22_tai_ptp_clock_index(struct mvpp2_tai *tai) +{ + return ptp_clock_index(tai->ptp_clock); +} + +void mvpp22_tai_tstamp(struct mvpp2_tai *tai, u32 tstamp, + struct skb_shared_hwtstamps *hwtstamp) +{ + struct timespec64 ts; + int delta; + + /* The tstamp consists of 2 bits of seconds and 30 bits of nanoseconds. + * We use our stored timestamp (tai->stamp) to form a full timestamp, + * and we must read the seconds exactly once. + */ + ts.tv_sec = READ_ONCE(tai->stamp.tv_sec); + ts.tv_nsec = tstamp & 0x3fffffff; + + /* Calculate the delta in seconds between our stored timestamp and + * the value read from the queue. Allow timestamps one second in the + * past, otherwise consider them to be in the future. + */ + delta = ((tstamp >> 30) - (ts.tv_sec & 3)) & 3; + if (delta == 3) + delta -= 4; + ts.tv_sec += delta; + + memset(hwtstamp, 0, sizeof(*hwtstamp)); + hwtstamp->hwtstamp = timespec64_to_ktime(ts); +} + +void mvpp22_tai_start(struct mvpp2_tai *tai) +{ + long delay; + + delay = mvpp22_tai_aux_work(&tai->caps); + + ptp_schedule_worker(tai->ptp_clock, delay); +} + +void mvpp22_tai_stop(struct mvpp2_tai *tai) +{ + ptp_cancel_worker_sync(tai->ptp_clock); +} + +static void mvpp22_tai_remove(void *priv) +{ + struct mvpp2_tai *tai = priv; + + if (!IS_ERR(tai->ptp_clock)) + ptp_clock_unregister(tai->ptp_clock); +} + +int mvpp22_tai_probe(struct device *dev, struct mvpp2 *priv) +{ + struct mvpp2_tai *tai; + int ret; + + tai = devm_kzalloc(dev, sizeof(*tai), GFP_KERNEL); + if (!tai) + return -ENOMEM; + + spin_lock_init(&tai->lock); + + tai->base = priv->iface_base; + + /* The step size consists of three registers - a 16-bit nanosecond step + * size, and a 32-bit fractional nanosecond step size split over two + * registers. The fractional nanosecond step size has units of 2^-32ns. + * + * To calculate this, we calculate: + * (10^9 + freq / 2) / (freq * 2^-32) + * which gives us the nanosecond step to the nearest integer in 16.32 + * fixed point format, and the fractional part of the step size with + * the MSB inverted. With rounding of the fractional nanosecond, and + * simplification, this becomes: + * (10^9 << 32 + freq << 31 + (freq + 1) >> 1) / freq + * + * So: + * div = (10^9 << 32 + freq << 31 + (freq + 1) >> 1) / freq + * nano = upper_32_bits(div); + * frac = lower_32_bits(div) ^ 0x80000000; + * Will give the values for the registers. + * + * This is all seems perfect, but alas it is not when considering the + * whole story. The system is clocked from 25MHz, which is multiplied + * by a PLL to 1GHz, and then divided by three, giving 333333333Hz + * (recurring). This gives exactly 3ns, but using 333333333Hz with + * the above gives an error of 13*2^-32ns. + * + * Consequently, we use the period rather than calculating from the + * frequency. + */ + tai->period = 3ULL << 32; + + mvpp22_tai_init(tai); + + tai->caps.owner = THIS_MODULE; + strscpy(tai->caps.name, "Marvell PP2.2", sizeof(tai->caps.name)); + tai->caps.max_adj = mvpp22_calc_max_adj(tai); + tai->caps.adjfine = mvpp22_tai_adjfine; + tai->caps.adjtime = mvpp22_tai_adjtime; + tai->caps.gettimex64 = mvpp22_tai_gettimex64; + tai->caps.settime64 = mvpp22_tai_settime64; + tai->caps.do_aux_work = mvpp22_tai_aux_work; + + ret = devm_add_action(dev, mvpp22_tai_remove, tai); + if (ret) + return ret; + + tai->ptp_clock = ptp_clock_register(&tai->caps, dev); + if (IS_ERR(tai->ptp_clock)) + return PTR_ERR(tai->ptp_clock); + + priv->tai = tai; + + return 0; +} diff --git a/drivers/net/ethernet/marvell/octeontx2/af/Makefile b/drivers/net/ethernet/marvell/octeontx2/af/Makefile index 1b25948c662b..2f7a861d0c7b 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/Makefile +++ b/drivers/net/ethernet/marvell/octeontx2/af/Makefile @@ -3,9 +3,10 @@ # Makefile for Marvell's OcteonTX2 RVU Admin Function driver # +ccflags-y += -I$(src) obj-$(CONFIG_OCTEONTX2_MBOX) += octeontx2_mbox.o obj-$(CONFIG_OCTEONTX2_AF) += octeontx2_af.o -octeontx2_mbox-y := mbox.o +octeontx2_mbox-y := mbox.o rvu_trace.o octeontx2_af-y := cgx.o rvu.o rvu_cgx.o rvu_npa.o rvu_nix.o \ - rvu_reg.o rvu_npc.o rvu_debugfs.o + rvu_reg.o rvu_npc.o rvu_debugfs.o ptp.o diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cgx.c b/drivers/net/ethernet/marvell/octeontx2/af/cgx.c index a4e65da8d95b..8f17e26dca53 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/cgx.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/cgx.c @@ -468,6 +468,35 @@ static void cgx_lmac_pause_frm_config(struct cgx *cgx, int lmac_id, bool enable) } } +void cgx_lmac_ptp_config(void *cgxd, int lmac_id, bool enable) +{ + struct cgx *cgx = cgxd; + u64 cfg; + + if (!cgx) + return; + + if (enable) { + /* Enable inbound PTP timestamping */ + cfg = cgx_read(cgx, lmac_id, CGXX_GMP_GMI_RXX_FRM_CTL); + cfg |= CGX_GMP_GMI_RXX_FRM_CTL_PTP_MODE; + cgx_write(cgx, lmac_id, CGXX_GMP_GMI_RXX_FRM_CTL, cfg); + + cfg = cgx_read(cgx, lmac_id, CGXX_SMUX_RX_FRM_CTL); + cfg |= CGX_SMUX_RX_FRM_CTL_PTP_MODE; + cgx_write(cgx, lmac_id, CGXX_SMUX_RX_FRM_CTL, cfg); + } else { + /* Disable inbound PTP stamping */ + cfg = cgx_read(cgx, lmac_id, CGXX_GMP_GMI_RXX_FRM_CTL); + cfg &= ~CGX_GMP_GMI_RXX_FRM_CTL_PTP_MODE; + cgx_write(cgx, lmac_id, CGXX_GMP_GMI_RXX_FRM_CTL, cfg); + + cfg = cgx_read(cgx, lmac_id, CGXX_SMUX_RX_FRM_CTL); + cfg &= ~CGX_SMUX_RX_FRM_CTL_PTP_MODE; + cgx_write(cgx, lmac_id, CGXX_SMUX_RX_FRM_CTL, cfg); + } +} + /* CGX Firmware interface low level support */ static int cgx_fwi_cmd_send(u64 req, u64 *resp, struct lmac *lmac) { diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cgx.h b/drivers/net/ethernet/marvell/octeontx2/af/cgx.h index 394f96591feb..27ca3291682b 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/cgx.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/cgx.h @@ -58,8 +58,10 @@ #define CGXX_SMUX_RX_FRM_CTL 0x20020 #define CGX_SMUX_RX_FRM_CTL_CTL_BCK BIT_ULL(3) +#define CGX_SMUX_RX_FRM_CTL_PTP_MODE BIT_ULL(12) #define CGXX_GMP_GMI_RXX_FRM_CTL 0x38028 #define CGX_GMP_GMI_RXX_FRM_CTL_CTL_BCK BIT_ULL(3) +#define CGX_GMP_GMI_RXX_FRM_CTL_PTP_MODE BIT_ULL(12) #define CGXX_SMUX_TX_CTL 0x20178 #define CGXX_SMUX_TX_PAUSE_PKT_TIME 0x20110 #define CGXX_SMUX_TX_PAUSE_PKT_INTERVAL 0x20120 @@ -139,4 +141,6 @@ int cgx_lmac_get_pause_frm(void *cgxd, int lmac_id, u8 *tx_pause, u8 *rx_pause); int cgx_lmac_set_pause_frm(void *cgxd, int lmac_id, u8 tx_pause, u8 rx_pause); +void cgx_lmac_ptp_config(void *cgxd, int lmac_id, bool enable); + #endif /* CGX_H */ diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mbox.c b/drivers/net/ethernet/marvell/octeontx2/af/mbox.c index 2718fe201c14..bbabb8e64201 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/mbox.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/mbox.c @@ -14,6 +14,7 @@ #include "rvu_reg.h" #include "mbox.h" +#include "rvu_trace.h" static const u16 msgs_offset = ALIGN(sizeof(struct mbox_hdr), MBOX_MSG_ALIGN); @@ -207,6 +208,9 @@ void otx2_mbox_msg_send(struct otx2_mbox *mbox, int devid) */ tx_hdr->num_msgs = mdev->num_msgs; rx_hdr->num_msgs = 0; + + trace_otx2_msg_send(mbox->pdev, tx_hdr->num_msgs, tx_hdr->msg_size); + spin_unlock(&mdev->mbox_lock); /* The interrupt should be fired after num_msgs is written @@ -303,10 +307,15 @@ int otx2_mbox_check_rsp_msgs(struct otx2_mbox *mbox, int devid) struct mbox_msghdr *preq = mdev->mbase + ireq; struct mbox_msghdr *prsp = mdev->mbase + irsp; - if (preq->id != prsp->id) + if (preq->id != prsp->id) { + trace_otx2_msg_check(mbox->pdev, preq->id, + prsp->id, prsp->rc); goto exit; + } if (prsp->rc) { rc = prsp->rc; + trace_otx2_msg_check(mbox->pdev, preq->id, + prsp->id, prsp->rc); goto exit; } diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h index ab433789d2c3..263a21129416 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h @@ -128,6 +128,7 @@ M(ATTACH_RESOURCES, 0x002, attach_resources, rsrc_attach, msg_rsp) \ M(DETACH_RESOURCES, 0x003, detach_resources, rsrc_detach, msg_rsp) \ M(MSIX_OFFSET, 0x005, msix_offset, msg_req, msix_offset_rsp) \ M(VF_FLR, 0x006, vf_flr, msg_req, msg_rsp) \ +M(PTP_OP, 0x007, ptp_op, ptp_req, ptp_rsp) \ M(GET_HW_CAP, 0x008, get_hw_cap, msg_req, get_hw_cap_rsp) \ /* CGX mbox IDs (range 0x200 - 0x3FF) */ \ M(CGX_START_RXTX, 0x200, cgx_start_rxtx, msg_req, msg_rsp) \ @@ -144,6 +145,8 @@ M(CGX_STOP_LINKEVENTS, 0x208, cgx_stop_linkevents, msg_req, msg_rsp) \ M(CGX_GET_LINKINFO, 0x209, cgx_get_linkinfo, msg_req, cgx_link_info_msg) \ M(CGX_INTLBK_ENABLE, 0x20A, cgx_intlbk_enable, msg_req, msg_rsp) \ M(CGX_INTLBK_DISABLE, 0x20B, cgx_intlbk_disable, msg_req, msg_rsp) \ +M(CGX_PTP_RX_ENABLE, 0x20C, cgx_ptp_rx_enable, msg_req, msg_rsp) \ +M(CGX_PTP_RX_DISABLE, 0x20D, cgx_ptp_rx_disable, msg_req, msg_rsp) \ M(CGX_CFG_PAUSE_FRM, 0x20E, cgx_cfg_pause_frm, cgx_pause_frm_cfg, \ cgx_pause_frm_cfg) \ /* NPA mbox IDs (range 0x400 - 0x5FF) */ \ @@ -214,6 +217,8 @@ M(NIX_LSO_FORMAT_CFG, 0x8011, nix_lso_format_cfg, \ nix_lso_format_cfg, \ nix_lso_format_cfg_rsp) \ M(NIX_RXVLAN_ALLOC, 0x8012, nix_rxvlan_alloc, msg_req, msg_rsp) \ +M(NIX_LF_PTP_TX_ENABLE, 0x8013, nix_lf_ptp_tx_enable, msg_req, msg_rsp) \ +M(NIX_LF_PTP_TX_DISABLE, 0x8014, nix_lf_ptp_tx_disable, msg_req, msg_rsp) \ M(NIX_BP_ENABLE, 0x8016, nix_bp_enable, nix_bp_cfg_req, \ nix_bp_cfg_rsp) \ M(NIX_BP_DISABLE, 0x8017, nix_bp_disable, nix_bp_cfg_req, msg_rsp) \ @@ -621,6 +626,7 @@ struct nix_rss_flowkey_cfg { #define NIX_FLOW_KEY_TYPE_INNR_UDP BIT(15) #define NIX_FLOW_KEY_TYPE_INNR_SCTP BIT(16) #define NIX_FLOW_KEY_TYPE_INNR_ETH_DMAC BIT(17) +#define NIX_FLOW_KEY_TYPE_VLAN BIT(20) u32 flowkey_cfg; /* Flowkey types selected */ u8 group; /* RSS context or group */ }; @@ -859,4 +865,20 @@ struct npc_get_kex_cfg_rsp { u8 mkex_pfl_name[MKEX_NAME_LEN]; }; +enum ptp_op { + PTP_OP_ADJFINE = 0, + PTP_OP_GET_CLOCK = 1, +}; + +struct ptp_req { + struct mbox_msghdr hdr; + u8 op; + s64 scaled_ppm; +}; + +struct ptp_rsp { + struct mbox_msghdr hdr; + u64 clk; +}; + #endif /* MBOX_H */ diff --git a/drivers/net/ethernet/marvell/octeontx2/af/npc.h b/drivers/net/ethernet/marvell/octeontx2/af/npc.h index 3803af9231c6..91a9d00e4fb5 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/npc.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/npc.h @@ -49,6 +49,7 @@ enum npc_kpu_lb_ltype { NPC_LT_LB_EDSA_VLAN, NPC_LT_LB_EXDSA, NPC_LT_LB_EXDSA_VLAN, + NPC_LT_LB_FDSA, NPC_LT_LB_CUSTOM0 = 0xE, NPC_LT_LB_CUSTOM1 = 0xF, }; @@ -77,21 +78,21 @@ enum npc_kpu_ld_ltype { NPC_LT_LD_ICMP, NPC_LT_LD_SCTP, NPC_LT_LD_ICMP6, + NPC_LT_LD_CUSTOM0, + NPC_LT_LD_CUSTOM1, NPC_LT_LD_IGMP = 8, - NPC_LT_LD_ESP, NPC_LT_LD_AH, NPC_LT_LD_GRE, NPC_LT_LD_NVGRE, NPC_LT_LD_NSH, NPC_LT_LD_TU_MPLS_IN_NSH, NPC_LT_LD_TU_MPLS_IN_IP, - NPC_LT_LD_CUSTOM0 = 0xE, - NPC_LT_LD_CUSTOM1 = 0xF, }; enum npc_kpu_le_ltype { NPC_LT_LE_VXLAN = 1, NPC_LT_LE_GENEVE, + NPC_LT_LE_ESP, NPC_LT_LE_GTPU = 4, NPC_LT_LE_VXLANGPE, NPC_LT_LE_GTPC, @@ -173,8 +174,8 @@ struct npc_kpu_profile_action { struct npc_kpu_profile { int cam_entries; int action_entries; - struct npc_kpu_profile_cam *cam; - struct npc_kpu_profile_action *action; + const struct npc_kpu_profile_cam *cam; + const struct npc_kpu_profile_action *action; }; /* NPC KPU register formats */ @@ -296,6 +297,9 @@ struct nix_rx_action { #endif }; +/* NPC_AF_INTFX_KEX_CFG field masks */ +#define NPC_PARSE_NIBBLE GENMASK_ULL(30, 0) + /* NIX Receive Vtag Action Structure */ #define VTAG0_VALID_BIT BIT_ULL(15) #define VTAG0_TYPE_MASK GENMASK_ULL(14, 12) @@ -320,4 +324,37 @@ struct npc_mcam_kex { u64 intf_ld_flags[NPC_MAX_INTF][NPC_MAX_LD][NPC_MAX_LFL]; } __packed; +struct npc_lt_def { + u8 ltype_mask; + u8 ltype_match; + u8 lid; +}; + +struct npc_lt_def_ipsec { + u8 ltype_mask; + u8 ltype_match; + u8 lid; + u8 spi_offset; + u8 spi_nz; +}; + +struct npc_lt_def_cfg { + struct npc_lt_def rx_ol2; + struct npc_lt_def rx_oip4; + struct npc_lt_def rx_iip4; + struct npc_lt_def rx_oip6; + struct npc_lt_def rx_iip6; + struct npc_lt_def rx_otcp; + struct npc_lt_def rx_itcp; + struct npc_lt_def rx_oudp; + struct npc_lt_def rx_iudp; + struct npc_lt_def rx_osctp; + struct npc_lt_def rx_isctp; + struct npc_lt_def_ipsec rx_ipsec[2]; + struct npc_lt_def pck_ol2; + struct npc_lt_def pck_oip4; + struct npc_lt_def pck_oip6; + struct npc_lt_def pck_iip4; +}; + #endif /* NPC_H */ diff --git a/drivers/net/ethernet/marvell/octeontx2/af/npc_profile.h b/drivers/net/ethernet/marvell/octeontx2/af/npc_profile.h index aa2727e6211a..77bb4ed32600 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/npc_profile.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/npc_profile.h @@ -63,6 +63,7 @@ #define NPC_UDP_PORT_VXLANGPE 4790 #define NPC_UDP_PORT_GENEVE 6081 #define NPC_UDP_PORT_MPLS 6635 +#define NPC_UDP_PORT_ESP 4500 #define NPC_VXLANGPE_NP_IP 0x1 #define NPC_VXLANGPE_NP_IP6 0x2 @@ -139,6 +140,13 @@ #define NPC_DSA_EXTEND 0x1000 #define NPC_DSA_EDSA 0x8000 +#define NPC_DSA_FDSA 0xc000 + +#define NPC_KEXOF_DMAC 8 +#define MKEX_SIGN 0x19bbfdbd15f /* strtoull of "mkexprof" with base:36 */ +#define KEX_LD_CFG(bytesm1, hdr_ofs, ena, flags_ena, key_ofs) \ + (((bytesm1) << 16) | ((hdr_ofs) << 8) | ((ena) << 7) | \ + ((flags_ena) << 6) | ((key_ofs) & 0x3F)) enum npc_kpu_parser_state { NPC_S_NA = 0, @@ -166,6 +174,7 @@ enum npc_kpu_parser_state { NPC_S_KPU3_DSA, NPC_S_KPU4_MPLS, NPC_S_KPU4_NSH, + NPC_S_KPU4_FDSA, NPC_S_KPU5_IP, NPC_S_KPU5_IP6, NPC_S_KPU5_ARP, @@ -189,7 +198,6 @@ enum npc_kpu_parser_state { NPC_S_KPU8_IGMP, NPC_S_KPU8_ICMP6, NPC_S_KPU8_GRE, - NPC_S_KPU8_ESP, NPC_S_KPU8_AH, NPC_S_KPU9_TU_MPLS_IN_GRE, NPC_S_KPU9_TU_MPLS_IN_NSH, @@ -201,6 +209,7 @@ enum npc_kpu_parser_state { NPC_S_KPU9_GENEVE, NPC_S_KPU9_GTPC, NPC_S_KPU9_GTPU, + NPC_S_KPU9_ESP, NPC_S_KPU10_TU_MPLS_IN_VXLANGPE, NPC_S_KPU10_TU_MPLS_PL, NPC_S_KPU10_TU_MPLS, @@ -271,6 +280,7 @@ enum npc_kpu_lb_lflag { NPC_F_LB_L_EDSA_VLAN, NPC_F_LB_L_EXDSA, NPC_F_LB_L_EXDSA_VLAN, + NPC_F_LB_L_FDSA, }; enum npc_kpu_lc_uflag { @@ -418,7 +428,7 @@ enum NPC_ERRLEV_E { NPC_ERRLEV_ENUM_LAST = 16, }; -static struct npc_kpu_profile_action ikpu_action_entries[] = { +static const struct npc_kpu_profile_action ikpu_action_entries[] = { { NPC_ERRLEV_RE, NPC_EC_NOERR, 12, 16, 20, 0, 0, @@ -979,7 +989,7 @@ static struct npc_kpu_profile_action ikpu_action_entries[] = { }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 12, 16, 20, 0, 0, + 12, 14, 20, 0, 0, NPC_S_KPU1_EXDSA, 0, 0, NPC_LID_LA, NPC_LT_NA, 0, @@ -997,7 +1007,7 @@ static struct npc_kpu_profile_action ikpu_action_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu1_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu1_cam_entries[] = { { NPC_S_KPU1_ETHER, 0xff, NPC_ETYPE_IP, @@ -1351,10 +1361,19 @@ static struct npc_kpu_profile_cam kpu1_cam_entries[] = { }, { NPC_S_KPU1_EXDSA, 0xff, + 0x0000, + 0x0000, NPC_DSA_EXTEND, NPC_DSA_EXTEND, 0x0000, 0x0000, + }, + { + NPC_S_KPU1_EXDSA, 0xff, + NPC_DSA_FDSA, + NPC_DSA_FDSA, + 0x0000, + 0x0000, 0x0000, 0x0000, }, @@ -1666,7 +1685,7 @@ static struct npc_kpu_profile_cam kpu1_cam_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu2_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu2_cam_entries[] = { { NPC_S_KPU2_CTAG, 0xff, NPC_ETYPE_IP, @@ -2794,7 +2813,7 @@ static struct npc_kpu_profile_cam kpu2_cam_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu3_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu3_cam_entries[] = { { NPC_S_KPU3_CTAG, 0xff, NPC_ETYPE_IP, @@ -3913,7 +3932,7 @@ static struct npc_kpu_profile_cam kpu3_cam_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu4_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu4_cam_entries[] = { { NPC_S_KPU4_MPLS, 0xff, NPC_MPLS_S, @@ -3996,6 +4015,69 @@ static struct npc_kpu_profile_cam kpu4_cam_entries[] = { 0x0000, }, { + NPC_S_KPU4_FDSA, 0xff, + NPC_ETYPE_IP, + 0xffff, + 0x0000, + 0x0000, + 0x0000, + 0x0000, + }, + { + NPC_S_KPU4_FDSA, 0xff, + NPC_ETYPE_IP6, + 0xffff, + 0x0000, + 0x0000, + 0x0000, + 0x0000, + }, + { + NPC_S_KPU4_FDSA, 0xff, + NPC_ETYPE_ARP, + 0xffff, + 0x0000, + 0x0000, + 0x0000, + 0x0000, + }, + { + NPC_S_KPU4_FDSA, 0xff, + NPC_ETYPE_RARP, + 0xffff, + 0x0000, + 0x0000, + 0x0000, + 0x0000, + }, + { + NPC_S_KPU4_FDSA, 0xff, + NPC_ETYPE_PTP, + 0xffff, + 0x0000, + 0x0000, + 0x0000, + 0x0000, + }, + { + NPC_S_KPU4_FDSA, 0xff, + NPC_ETYPE_FCOE, + 0xffff, + 0x0000, + 0x0000, + 0x0000, + 0x0000, + }, + { + NPC_S_KPU4_FDSA, 0xff, + 0x0000, + NPC_DSA_FDSA, + 0x0000, + 0x0000, + 0x0000, + 0x0000, + }, + { NPC_S_NA, 0X00, 0x0000, 0x0000, @@ -4006,7 +4088,7 @@ static struct npc_kpu_profile_cam kpu4_cam_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu5_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu5_cam_entries[] = { { NPC_S_KPU5_IP, 0xff, 0x0000, @@ -4576,7 +4658,7 @@ static struct npc_kpu_profile_cam kpu5_cam_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu6_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu6_cam_entries[] = { { NPC_S_KPU6_IP6_EXT, 0xff, 0x0000, @@ -4921,7 +5003,7 @@ static struct npc_kpu_profile_cam kpu6_cam_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu7_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu7_cam_entries[] = { { NPC_S_KPU7_IP6_EXT, 0xff, 0x0000, @@ -5140,7 +5222,7 @@ static struct npc_kpu_profile_cam kpu7_cam_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu8_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu8_cam_entries[] = { { NPC_S_KPU8_TCP, 0xff, 0x0000, @@ -5341,15 +5423,24 @@ static struct npc_kpu_profile_cam kpu8_cam_entries[] = { }, { NPC_S_KPU8_UDP, 0xff, + NPC_UDP_PORT_ESP, + 0xffff, 0x0000, 0x0000, 0x0000, 0x0000, + }, + { + NPC_S_KPU8_UDP, 0xff, + 0x0000, + 0x0000, + NPC_UDP_PORT_ESP, + 0xffff, 0x0000, 0x0000, }, { - NPC_S_KPU8_SCTP, 0xff, + NPC_S_KPU8_UDP, 0xff, 0x0000, 0x0000, 0x0000, @@ -5358,7 +5449,7 @@ static struct npc_kpu_profile_cam kpu8_cam_entries[] = { 0x0000, }, { - NPC_S_KPU8_ICMP, 0xff, + NPC_S_KPU8_SCTP, 0xff, 0x0000, 0x0000, 0x0000, @@ -5367,7 +5458,7 @@ static struct npc_kpu_profile_cam kpu8_cam_entries[] = { 0x0000, }, { - NPC_S_KPU8_IGMP, 0xff, + NPC_S_KPU8_ICMP, 0xff, 0x0000, 0x0000, 0x0000, @@ -5376,7 +5467,7 @@ static struct npc_kpu_profile_cam kpu8_cam_entries[] = { 0x0000, }, { - NPC_S_KPU8_ICMP6, 0xff, + NPC_S_KPU8_IGMP, 0xff, 0x0000, 0x0000, 0x0000, @@ -5385,7 +5476,7 @@ static struct npc_kpu_profile_cam kpu8_cam_entries[] = { 0x0000, }, { - NPC_S_KPU8_ESP, 0xff, + NPC_S_KPU8_ICMP6, 0xff, 0x0000, 0x0000, 0x0000, @@ -5872,7 +5963,7 @@ static struct npc_kpu_profile_cam kpu8_cam_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu9_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu9_cam_entries[] = { { NPC_S_KPU9_TU_MPLS_IN_GRE, 0xff, NPC_MPLS_S, @@ -6324,6 +6415,15 @@ static struct npc_kpu_profile_cam kpu9_cam_entries[] = { NPC_MPLS_S, }, { + NPC_S_KPU9_ESP, 0xff, + 0x0000, + 0x0000, + 0x0000, + 0x0000, + 0x0000, + 0x0000, + }, + { NPC_S_NA, 0X00, 0x0000, 0x0000, @@ -6334,7 +6434,7 @@ static struct npc_kpu_profile_cam kpu9_cam_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu10_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu10_cam_entries[] = { { NPC_S_KPU10_TU_MPLS, 0xff, NPC_MPLS_S, @@ -6499,7 +6599,7 @@ static struct npc_kpu_profile_cam kpu10_cam_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu11_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu11_cam_entries[] = { { NPC_S_KPU11_TU_ETHER, 0xff, NPC_ETYPE_IP, @@ -6808,7 +6908,7 @@ static struct npc_kpu_profile_cam kpu11_cam_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu12_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu12_cam_entries[] = { { NPC_S_KPU12_TU_IP, 0xff, NPC_IPNH_TCP, @@ -7063,7 +7163,7 @@ static struct npc_kpu_profile_cam kpu12_cam_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu13_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu13_cam_entries[] = { { NPC_S_KPU13_TU_IP6_EXT, 0xff, 0x0000, @@ -7075,7 +7175,7 @@ static struct npc_kpu_profile_cam kpu13_cam_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu14_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu14_cam_entries[] = { { NPC_S_KPU14_TU_IP6_EXT, 0xff, 0x0000, @@ -7087,7 +7187,7 @@ static struct npc_kpu_profile_cam kpu14_cam_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu15_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu15_cam_entries[] = { { NPC_S_KPU15_TU_TCP, 0xff, 0x0000, @@ -7288,7 +7388,7 @@ static struct npc_kpu_profile_cam kpu15_cam_entries[] = { }, }; -static struct npc_kpu_profile_cam kpu16_cam_entries[] = { +static const struct npc_kpu_profile_cam kpu16_cam_entries[] = { { NPC_S_KPU16_TCP_DATA, 0xff, 0x0000, @@ -7345,7 +7445,7 @@ static struct npc_kpu_profile_cam kpu16_cam_entries[] = { }, }; -static struct npc_kpu_profile_action kpu1_action_entries[] = { +static const struct npc_kpu_profile_action kpu1_action_entries[] = { { NPC_ERRLEV_RE, NPC_EC_NOERR, 8, 0, 6, 3, 0, @@ -7673,6 +7773,14 @@ static struct npc_kpu_profile_action kpu1_action_entries[] = { 0, 0, 0, 0, }, { + NPC_ERRLEV_RE, NPC_EC_NOERR, + 4, 8, 16, 2, 0, + NPC_S_KPU4_FDSA, 12, 1, + NPC_LID_LA, NPC_LT_LA_ETHER, + 0, + 0, 0, 0, 0, + }, + { NPC_ERRLEV_LA, NPC_EC_EDSA_UNK, 0, 0, 0, 0, 1, NPC_S_NA, 0, 1, @@ -7962,7 +8070,7 @@ static struct npc_kpu_profile_action kpu1_action_entries[] = { }, }; -static struct npc_kpu_profile_action kpu2_action_entries[] = { +static const struct npc_kpu_profile_action kpu2_action_entries[] = { { NPC_ERRLEV_RE, NPC_EC_NOERR, 8, 0, 6, 2, 0, @@ -8965,7 +9073,7 @@ static struct npc_kpu_profile_action kpu2_action_entries[] = { }, }; -static struct npc_kpu_profile_action kpu3_action_entries[] = { +static const struct npc_kpu_profile_action kpu3_action_entries[] = { { NPC_ERRLEV_RE, NPC_EC_NOERR, 8, 0, 6, 1, 0, @@ -9960,7 +10068,7 @@ static struct npc_kpu_profile_action kpu3_action_entries[] = { }, }; -static struct npc_kpu_profile_action kpu4_action_entries[] = { +static const struct npc_kpu_profile_action kpu4_action_entries[] = { { NPC_ERRLEV_RE, NPC_EC_NOERR, 0, 0, 0, 0, 0, @@ -10034,6 +10142,62 @@ static struct npc_kpu_profile_action kpu4_action_entries[] = { 0, 0, 0, 0, }, { + NPC_ERRLEV_RE, NPC_EC_NOERR, + 8, 0, 6, 0, 0, + NPC_S_KPU5_IP, 6, 1, + NPC_LID_LB, NPC_LT_LB_FDSA, + NPC_F_LB_L_FDSA, + 0, 0, 0, 0, + }, + { + NPC_ERRLEV_RE, NPC_EC_NOERR, + 6, 0, 0, 0, 0, + NPC_S_KPU5_IP6, 6, 1, + NPC_LID_LB, NPC_LT_LB_FDSA, + NPC_F_LB_L_FDSA, + 0, 0, 0, 0, + }, + { + NPC_ERRLEV_RE, NPC_EC_NOERR, + 0, 0, 0, 0, 0, + NPC_S_KPU5_ARP, 6, 1, + NPC_LID_LB, NPC_LT_LB_FDSA, + NPC_F_LB_L_FDSA, + 0, 0, 0, 0, + }, + { + NPC_ERRLEV_RE, NPC_EC_NOERR, + 8, 0, 6, 0, 0, + NPC_S_KPU5_RARP, 6, 1, + NPC_LID_LB, NPC_LT_LB_FDSA, + NPC_F_LB_L_FDSA, + 0, 0, 0, 0, + }, + { + NPC_ERRLEV_RE, NPC_EC_NOERR, + 6, 0, 0, 0, 0, + NPC_S_KPU5_PTP, 6, 1, + NPC_LID_LB, NPC_LT_LB_FDSA, + NPC_F_LB_L_FDSA, + 0, 0, 0, 0, + }, + { + NPC_ERRLEV_RE, NPC_EC_NOERR, + 0, 0, 0, 0, 0, + NPC_S_KPU5_FCOE, 6, 1, + NPC_LID_LB, NPC_LT_LB_FDSA, + NPC_F_LB_L_FDSA, + 0, 0, 0, 0, + }, + { + NPC_ERRLEV_RE, NPC_EC_NOERR, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 1, + NPC_LID_LB, NPC_LT_LB_FDSA, + NPC_F_LB_U_UNK_ETYPE | NPC_F_LB_L_FDSA, + 0, 0, 0, 0, + }, + { NPC_ERRLEV_LB, NPC_EC_L2_K4, 0, 0, 0, 0, 1, NPC_S_NA, 0, 0, @@ -10043,7 +10207,7 @@ static struct npc_kpu_profile_action kpu4_action_entries[] = { }, }; -static struct npc_kpu_profile_action kpu5_action_entries[] = { +static const struct npc_kpu_profile_action kpu5_action_entries[] = { { NPC_ERRLEV_LC, NPC_EC_IP_TTL_0, 0, 0, 0, 0, 1, @@ -10102,8 +10266,8 @@ static struct npc_kpu_profile_action kpu5_action_entries[] = { }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 2, 0, - NPC_S_KPU8_ESP, 20, 1, + 0, 0, 0, 3, 0, + NPC_S_KPU9_ESP, 20, 1, NPC_LID_LC, NPC_LT_LC_IP, 0, 0, 0, 0, 0, @@ -10206,8 +10370,8 @@ static struct npc_kpu_profile_action kpu5_action_entries[] = { }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 2, 0, - NPC_S_KPU8_ESP, 0, 1, + 0, 0, 0, 3, 0, + NPC_S_KPU9_ESP, 0, 1, NPC_LID_LC, NPC_LT_LC_IP_OPT, 0, 0, 0xf, 0, 2, @@ -10414,8 +10578,8 @@ static struct npc_kpu_profile_action kpu5_action_entries[] = { }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 2, 0, - NPC_S_KPU8_ESP, 40, 1, + 0, 0, 0, 3, 0, + NPC_S_KPU9_ESP, 40, 1, NPC_LID_LC, NPC_LT_LC_IP6_EXT, 0, 0, 0, 0, 0, @@ -10550,7 +10714,7 @@ static struct npc_kpu_profile_action kpu5_action_entries[] = { }, }; -static struct npc_kpu_profile_action kpu6_action_entries[] = { +static const struct npc_kpu_profile_action kpu6_action_entries[] = { { NPC_ERRLEV_RE, NPC_EC_NOERR, 0, 0, 0, 0, 1, @@ -10561,80 +10725,80 @@ static struct npc_kpu_profile_action kpu6_action_entries[] = { }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 2, 12, 0, 1, 0, - NPC_S_KPU8_TCP, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 2, 8, 10, 1, 0, - NPC_S_KPU8_UDP, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 1, 0, - NPC_S_KPU8_SCTP, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 1, 0, - NPC_S_KPU8_ICMP, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 1, 0, - NPC_S_KPU8_ICMP6, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 1, 0, - NPC_S_KPU8_ESP, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 1, 0, - NPC_S_KPU8_AH, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 1, 0, - NPC_S_KPU8_GRE, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 6, 0, 0, 5, 0, - NPC_S_KPU12_TU_IP6, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 2, 6, 10, 2, 0, - NPC_S_KPU9_TU_MPLS_IN_IP, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, @@ -10689,8 +10853,8 @@ static struct npc_kpu_profile_action kpu6_action_entries[] = { }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 1, 0, - NPC_S_KPU8_ESP, 8, 0, + 0, 0, 0, 2, 0, + NPC_S_KPU9_ESP, 8, 0, NPC_LID_LC, NPC_LT_NA, 0, 1, 0xff, 0, 3, @@ -10793,8 +10957,8 @@ static struct npc_kpu_profile_action kpu6_action_entries[] = { }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 1, 0, - NPC_S_KPU8_ESP, 8, 0, + 0, 0, 0, 2, 0, + NPC_S_KPU9_ESP, 8, 0, NPC_LID_LC, NPC_LT_NA, 0, 1, 0xff, 0, 3, @@ -10857,7 +11021,7 @@ static struct npc_kpu_profile_action kpu6_action_entries[] = { }, }; -static struct npc_kpu_profile_action kpu7_action_entries[] = { +static const struct npc_kpu_profile_action kpu7_action_entries[] = { { NPC_ERRLEV_RE, NPC_EC_NOERR, 0, 0, 0, 0, 1, @@ -10908,8 +11072,8 @@ static struct npc_kpu_profile_action kpu7_action_entries[] = { }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 0, 0, - NPC_S_KPU8_ESP, 8, 0, + 0, 0, 0, 1, 0, + NPC_S_KPU9_ESP, 8, 0, NPC_LID_LC, NPC_LT_NA, 0, 1, 0xff, 0, 3, @@ -10956,80 +11120,80 @@ static struct npc_kpu_profile_action kpu7_action_entries[] = { }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 2, 12, 0, 0, 0, - NPC_S_KPU8_TCP, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 2, 8, 10, 0, 0, - NPC_S_KPU8_UDP, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 0, 0, - NPC_S_KPU8_SCTP, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 0, 0, - NPC_S_KPU8_ICMP, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 0, 0, - NPC_S_KPU8_ICMP6, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 0, 0, - NPC_S_KPU8_ESP, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 0, 0, - NPC_S_KPU8_AH, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 0, 0, - NPC_S_KPU8_GRE, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 6, 0, 0, 4, 0, - NPC_S_KPU12_TU_IP6, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, }, { NPC_ERRLEV_RE, NPC_EC_NOERR, - 2, 6, 10, 1, 0, - NPC_S_KPU9_TU_MPLS_IN_IP, 8, 0, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 0, NPC_LID_LC, NPC_LT_NA, 0, 0, 0, 0, 0, @@ -11052,7 +11216,7 @@ static struct npc_kpu_profile_action kpu7_action_entries[] = { }, }; -static struct npc_kpu_profile_action kpu8_action_entries[] = { +static const struct npc_kpu_profile_action kpu8_action_entries[] = { { NPC_ERRLEV_LD, NPC_EC_TCP_FLAGS_FIN_ONLY, 0, 0, 0, 0, 1, @@ -11231,6 +11395,22 @@ static struct npc_kpu_profile_action kpu8_action_entries[] = { }, { NPC_ERRLEV_RE, NPC_EC_NOERR, + 0, 0, 0, 0, 0, + NPC_S_KPU9_ESP, 8, 1, + NPC_LID_LD, NPC_LT_LD_UDP, + 0, + 0, 0, 0, 0, + }, + { + NPC_ERRLEV_RE, NPC_EC_NOERR, + 0, 0, 0, 0, 0, + NPC_S_KPU9_ESP, 8, 1, + NPC_LID_LD, NPC_LT_LD_UDP, + 0, + 0, 0, 0, 0, + }, + { + NPC_ERRLEV_RE, NPC_EC_NOERR, 0, 0, 0, 7, 0, NPC_S_KPU16_UDP_DATA, 8, 1, NPC_LID_LD, NPC_LT_LD_UDP, @@ -11273,14 +11453,6 @@ static struct npc_kpu_profile_action kpu8_action_entries[] = { NPC_ERRLEV_RE, NPC_EC_NOERR, 0, 0, 0, 0, 1, NPC_S_NA, 0, 1, - NPC_LID_LD, NPC_LT_LD_ESP, - 0, - 0, 0, 0, 0, - }, - { - NPC_ERRLEV_RE, NPC_EC_NOERR, - 0, 0, 0, 0, 1, - NPC_S_NA, 0, 1, NPC_LID_LD, NPC_LT_LD_AH, 0, 0, 0, 0, 0, @@ -11703,7 +11875,7 @@ static struct npc_kpu_profile_action kpu8_action_entries[] = { }, }; -static struct npc_kpu_profile_action kpu9_action_entries[] = { +static const struct npc_kpu_profile_action kpu9_action_entries[] = { { NPC_ERRLEV_RE, NPC_EC_NOERR, 0, 0, 0, 0, 0, @@ -12105,6 +12277,14 @@ static struct npc_kpu_profile_action kpu9_action_entries[] = { 0, 0, 0, 0, }, { + NPC_ERRLEV_RE, NPC_EC_NOERR, + 0, 0, 0, 0, 1, + NPC_S_NA, 0, 1, + NPC_LID_LE, NPC_LT_LE_ESP, + 0, + 0, 0, 0, 0, + }, + { NPC_ERRLEV_LE, NPC_EC_UNK, 0, 0, 0, 0, 1, NPC_S_NA, 0, 0, @@ -12114,7 +12294,7 @@ static struct npc_kpu_profile_action kpu9_action_entries[] = { }, }; -static struct npc_kpu_profile_action kpu10_action_entries[] = { +static const struct npc_kpu_profile_action kpu10_action_entries[] = { { NPC_ERRLEV_RE, NPC_EC_NOERR, 8, 0, 6, 1, 0, @@ -12261,7 +12441,7 @@ static struct npc_kpu_profile_action kpu10_action_entries[] = { }, }; -static struct npc_kpu_profile_action kpu11_action_entries[] = { +static const struct npc_kpu_profile_action kpu11_action_entries[] = { { NPC_ERRLEV_RE, NPC_EC_NOERR, 8, 0, 6, 0, 0, @@ -12536,7 +12716,7 @@ static struct npc_kpu_profile_action kpu11_action_entries[] = { }, }; -static struct npc_kpu_profile_action kpu12_action_entries[] = { +static const struct npc_kpu_profile_action kpu12_action_entries[] = { { NPC_ERRLEV_RE, NPC_EC_NOERR, 2, 12, 0, 2, 0, @@ -12763,7 +12943,7 @@ static struct npc_kpu_profile_action kpu12_action_entries[] = { }, }; -static struct npc_kpu_profile_action kpu13_action_entries[] = { +static const struct npc_kpu_profile_action kpu13_action_entries[] = { { NPC_ERRLEV_RE, NPC_EC_NOERR, 0, 0, 0, 0, 1, @@ -12774,7 +12954,7 @@ static struct npc_kpu_profile_action kpu13_action_entries[] = { }, }; -static struct npc_kpu_profile_action kpu14_action_entries[] = { +static const struct npc_kpu_profile_action kpu14_action_entries[] = { { NPC_ERRLEV_RE, NPC_EC_NOERR, 0, 0, 0, 0, 1, @@ -12785,7 +12965,7 @@ static struct npc_kpu_profile_action kpu14_action_entries[] = { }, }; -static struct npc_kpu_profile_action kpu15_action_entries[] = { +static const struct npc_kpu_profile_action kpu15_action_entries[] = { { NPC_ERRLEV_LG, NPC_EC_TCP_FLAGS_FIN_ONLY, 0, 0, 0, 0, 1, @@ -12964,7 +13144,7 @@ static struct npc_kpu_profile_action kpu15_action_entries[] = { }, }; -static struct npc_kpu_profile_action kpu16_action_entries[] = { +static const struct npc_kpu_profile_action kpu16_action_entries[] = { { NPC_ERRLEV_RE, NPC_EC_NOERR, 0, 0, 0, 0, 1, @@ -13015,7 +13195,7 @@ static struct npc_kpu_profile_action kpu16_action_entries[] = { }, }; -static struct npc_kpu_profile npc_kpu_profiles[] = { +static const struct npc_kpu_profile npc_kpu_profiles[] = { { ARRAY_SIZE(kpu1_cam_entries), ARRAY_SIZE(kpu1_action_entries), @@ -13114,4 +13294,163 @@ static struct npc_kpu_profile npc_kpu_profiles[] = { }, }; +static const struct npc_lt_def_cfg npc_lt_defaults = { + .rx_ol2 = { + .lid = NPC_LID_LA, + .ltype_match = NPC_LT_LA_ETHER, + .ltype_mask = 0x0F, + }, + .rx_oip4 = { + .lid = NPC_LID_LC, + .ltype_match = NPC_LT_LC_IP, + .ltype_mask = 0x0E, + }, + .rx_iip4 = { + .lid = NPC_LID_LG, + .ltype_match = NPC_LT_LG_TU_IP, + .ltype_mask = 0x0F, + }, + .rx_oip6 = { + .lid = NPC_LID_LC, + .ltype_match = NPC_LT_LC_IP6, + .ltype_mask = 0x0E, + }, + .rx_iip6 = { + .lid = NPC_LID_LG, + .ltype_match = NPC_LT_LG_TU_IP6, + .ltype_mask = 0x0F, + }, + .rx_otcp = { + .lid = NPC_LID_LD, + .ltype_match = NPC_LT_LD_TCP, + .ltype_mask = 0x0F, + }, + .rx_itcp = { + .lid = NPC_LID_LH, + .ltype_match = NPC_LT_LH_TU_TCP, + .ltype_mask = 0x0F, + }, + .rx_oudp = { + .lid = NPC_LID_LD, + .ltype_match = NPC_LT_LD_UDP, + .ltype_mask = 0x0F, + }, + .rx_iudp = { + .lid = NPC_LID_LH, + .ltype_match = NPC_LT_LH_TU_UDP, + .ltype_mask = 0x0F, + }, + .rx_osctp = { + .lid = NPC_LID_LD, + .ltype_match = NPC_LT_LD_SCTP, + .ltype_mask = 0x0F, + }, + .rx_isctp = { + .lid = NPC_LID_LH, + .ltype_match = NPC_LT_LH_TU_SCTP, + .ltype_mask = 0x0F, + }, + .rx_ipsec = { + { + .lid = NPC_LID_LE, + .ltype_match = NPC_LT_LE_ESP, + .ltype_mask = 0x0F, + }, + { + .spi_offset = 8, + .lid = NPC_LID_LH, + .ltype_match = NPC_LT_LH_TU_ESP, + .ltype_mask = 0x0F, + }, + }, + .pck_ol2 = { + .lid = NPC_LID_LA, + .ltype_match = NPC_LT_LA_ETHER, + .ltype_mask = 0x0F, + }, + .pck_oip4 = { + .lid = NPC_LID_LC, + .ltype_match = NPC_LT_LC_IP, + .ltype_mask = 0x0E, + }, + .pck_iip4 = { + .lid = NPC_LID_LG, + .ltype_match = NPC_LT_LG_TU_IP, + .ltype_mask = 0x0F, + }, +}; + +static const struct npc_mcam_kex npc_mkex_default = { + .mkex_sign = MKEX_SIGN, + .name = "default", + .kpu_version = NPC_KPU_PROFILE_VER, + .keyx_cfg = { + /* nibble: LA..LE (ltype only) + Channel */ + [NIX_INTF_RX] = ((u64)NPC_MCAM_KEY_X2 << 32) | 0x49247, + [NIX_INTF_TX] = ((u64)NPC_MCAM_KEY_X2 << 32) | ((1ULL << 19) - 1), + }, + .intf_lid_lt_ld = { + /* Default RX MCAM KEX profile */ + [NIX_INTF_RX] = { + [NPC_LID_LA] = { + /* Layer A: Ethernet: */ + [NPC_LT_LA_ETHER] = { + /* DMAC: 6 bytes, KW1[47:0] */ + KEX_LD_CFG(0x05, 0x0, 0x1, 0x0, NPC_KEXOF_DMAC), + /* Ethertype: 2 bytes, KW0[47:32] */ + KEX_LD_CFG(0x01, 0xc, 0x1, 0x0, 0x4), + }, + }, + [NPC_LID_LB] = { + /* Layer B: Single VLAN (CTAG) */ + /* CTAG VLAN[2..3] + Ethertype, 4 bytes, KW0[63:32] */ + [NPC_LT_LB_CTAG] = { + KEX_LD_CFG(0x03, 0x0, 0x1, 0x0, 0x4), + }, + /* Layer B: Stacked VLAN (STAG|QinQ) */ + [NPC_LT_LB_STAG_QINQ] = { + /* CTAG VLAN[2..3] + Ethertype, 4 bytes, KW0[63:32] */ + KEX_LD_CFG(0x03, 0x4, 0x1, 0x0, 0x4), + }, + [NPC_LT_LB_FDSA] = { + /* SWITCH PORT: 1 byte, KW0[63:48] */ + KEX_LD_CFG(0x0, 0x1, 0x1, 0x0, 0x6), + /* Ethertype: 2 bytes, KW0[47:32] */ + KEX_LD_CFG(0x01, 0x4, 0x1, 0x0, 0x4), + }, + }, + [NPC_LID_LC] = { + /* Layer C: IPv4 */ + [NPC_LT_LC_IP] = { + /* SIP+DIP: 8 bytes, KW2[63:0] */ + KEX_LD_CFG(0x07, 0xc, 0x1, 0x0, 0x10), + /* TOS: 1 byte, KW1[63:56] */ + KEX_LD_CFG(0x0, 0x1, 0x1, 0x0, 0xf), + }, + /* Layer C: IPv6 */ + [NPC_LT_LC_IP6] = { + /* Everything up to SADDR: 8 bytes, KW2[63:0] */ + KEX_LD_CFG(0x07, 0x0, 0x1, 0x0, 0x10), + }, + }, + [NPC_LID_LD] = { + /* Layer D:UDP */ + [NPC_LT_LD_UDP] = { + /* SPORT: 2 bytes, KW3[15:0] */ + KEX_LD_CFG(0x1, 0x0, 0x1, 0x0, 0x18), + /* DPORT: 2 bytes, KW3[31:16] */ + KEX_LD_CFG(0x1, 0x2, 0x1, 0x0, 0x1a), + }, + /* Layer D:TCP */ + [NPC_LT_LD_TCP] = { + /* SPORT: 2 bytes, KW3[15:0] */ + KEX_LD_CFG(0x1, 0x0, 0x1, 0x0, 0x18), + /* DPORT: 2 bytes, KW3[31:16] */ + KEX_LD_CFG(0x1, 0x2, 0x1, 0x0, 0x1a), + }, + }, + }, + }, +}; + #endif /* NPC_PROFILE_H */ diff --git a/drivers/net/ethernet/marvell/octeontx2/af/ptp.c b/drivers/net/ethernet/marvell/octeontx2/af/ptp.c new file mode 100644 index 000000000000..f69f4f35ae48 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeontx2/af/ptp.c @@ -0,0 +1,275 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Marvell PTP driver + * + * Copyright (C) 2020 Marvell International Ltd. + */ + +#include <linux/bitfield.h> +#include <linux/device.h> +#include <linux/module.h> +#include <linux/pci.h> + +#include "ptp.h" +#include "mbox.h" +#include "rvu.h" + +#define DRV_NAME "Marvell PTP Driver" + +#define PCI_DEVID_OCTEONTX2_PTP 0xA00C +#define PCI_SUBSYS_DEVID_OCTX2_98xx_PTP 0xB100 +#define PCI_SUBSYS_DEVID_OCTX2_96XX_PTP 0xB200 +#define PCI_SUBSYS_DEVID_OCTX2_95XX_PTP 0xB300 +#define PCI_SUBSYS_DEVID_OCTX2_LOKI_PTP 0xB400 +#define PCI_SUBSYS_DEVID_OCTX2_95MM_PTP 0xB500 +#define PCI_DEVID_OCTEONTX2_RST 0xA085 + +#define PCI_PTP_BAR_NO 0 +#define PCI_RST_BAR_NO 0 + +#define PTP_CLOCK_CFG 0xF00ULL +#define PTP_CLOCK_CFG_PTP_EN BIT_ULL(0) +#define PTP_CLOCK_LO 0xF08ULL +#define PTP_CLOCK_HI 0xF10ULL +#define PTP_CLOCK_COMP 0xF18ULL + +#define RST_BOOT 0x1600ULL +#define RST_MUL_BITS GENMASK_ULL(38, 33) +#define CLOCK_BASE_RATE 50000000ULL + +static u64 get_clock_rate(void) +{ + u64 cfg, ret = CLOCK_BASE_RATE * 16; + struct pci_dev *pdev; + void __iomem *base; + + /* To get the input clock frequency with which PTP co-processor + * block is running the base frequency(50 MHz) needs to be multiplied + * with multiplier bits present in RST_BOOT register of RESET block. + * Hence below code gets the multiplier bits from the RESET PCI + * device present in the system. + */ + pdev = pci_get_device(PCI_VENDOR_ID_CAVIUM, + PCI_DEVID_OCTEONTX2_RST, NULL); + if (!pdev) + goto error; + + base = pci_ioremap_bar(pdev, PCI_RST_BAR_NO); + if (!base) + goto error_put_pdev; + + cfg = readq(base + RST_BOOT); + ret = CLOCK_BASE_RATE * FIELD_GET(RST_MUL_BITS, cfg); + + iounmap(base); + +error_put_pdev: + pci_dev_put(pdev); + +error: + return ret; +} + +struct ptp *ptp_get(void) +{ + struct pci_dev *pdev; + struct ptp *ptp; + + /* If the PTP pci device is found on the system and ptp + * driver is bound to it then the PTP pci device is returned + * to the caller(rvu driver). + */ + pdev = pci_get_device(PCI_VENDOR_ID_CAVIUM, + PCI_DEVID_OCTEONTX2_PTP, NULL); + if (!pdev) + return ERR_PTR(-ENODEV); + + ptp = pci_get_drvdata(pdev); + if (!ptp) + ptp = ERR_PTR(-EPROBE_DEFER); + if (IS_ERR(ptp)) + pci_dev_put(pdev); + + return ptp; +} + +void ptp_put(struct ptp *ptp) +{ + if (!ptp) + return; + + pci_dev_put(ptp->pdev); +} + +static int ptp_adjfine(struct ptp *ptp, long scaled_ppm) +{ + bool neg_adj = false; + u64 comp; + u64 adj; + s64 ppb; + + if (scaled_ppm < 0) { + neg_adj = true; + scaled_ppm = -scaled_ppm; + } + + /* The hardware adds the clock compensation value to the PTP clock + * on every coprocessor clock cycle. Typical convention is that it + * represent number of nanosecond betwen each cycle. In this + * convention compensation value is in 64 bit fixed-point + * representation where upper 32 bits are number of nanoseconds + * and lower is fractions of nanosecond. + * The scaled_ppm represent the ratio in "parts per million" by which + * the compensation value should be corrected. + * To calculate new compenstation value we use 64bit fixed point + * arithmetic on following formula + * comp = tbase + tbase * scaled_ppm / (1M * 2^16) + * where tbase is the basic compensation value calculated + * initialy in the probe function. + */ + comp = ((u64)1000000000ull << 32) / ptp->clock_rate; + /* convert scaled_ppm to ppb */ + ppb = 1 + scaled_ppm; + ppb *= 125; + ppb >>= 13; + adj = comp * ppb; + adj = div_u64(adj, 1000000000ull); + comp = neg_adj ? comp - adj : comp + adj; + + writeq(comp, ptp->reg_base + PTP_CLOCK_COMP); + + return 0; +} + +static int ptp_get_clock(struct ptp *ptp, u64 *clk) +{ + /* Return the current PTP clock */ + *clk = readq(ptp->reg_base + PTP_CLOCK_HI); + + return 0; +} + +static int ptp_probe(struct pci_dev *pdev, + const struct pci_device_id *ent) +{ + struct device *dev = &pdev->dev; + struct ptp *ptp; + u64 clock_comp; + u64 clock_cfg; + int err; + + ptp = devm_kzalloc(dev, sizeof(*ptp), GFP_KERNEL); + if (!ptp) { + err = -ENOMEM; + goto error; + } + + ptp->pdev = pdev; + + err = pcim_enable_device(pdev); + if (err) + goto error_free; + + err = pcim_iomap_regions(pdev, 1 << PCI_PTP_BAR_NO, pci_name(pdev)); + if (err) + goto error_free; + + ptp->reg_base = pcim_iomap_table(pdev)[PCI_PTP_BAR_NO]; + + ptp->clock_rate = get_clock_rate(); + + /* Enable PTP clock */ + clock_cfg = readq(ptp->reg_base + PTP_CLOCK_CFG); + clock_cfg |= PTP_CLOCK_CFG_PTP_EN; + writeq(clock_cfg, ptp->reg_base + PTP_CLOCK_CFG); + + clock_comp = ((u64)1000000000ull << 32) / ptp->clock_rate; + /* Initial compensation value to start the nanosecs counter */ + writeq(clock_comp, ptp->reg_base + PTP_CLOCK_COMP); + + pci_set_drvdata(pdev, ptp); + + return 0; + +error_free: + devm_kfree(dev, ptp); + +error: + /* For `ptp_get()` we need to differentiate between the case + * when the core has not tried to probe this device and the case when + * the probe failed. In the later case we pretend that the + * initialization was successful and keep the error in + * `dev->driver_data`. + */ + pci_set_drvdata(pdev, ERR_PTR(err)); + return 0; +} + +static void ptp_remove(struct pci_dev *pdev) +{ + struct ptp *ptp = pci_get_drvdata(pdev); + u64 clock_cfg; + + if (IS_ERR_OR_NULL(ptp)) + return; + + /* Disable PTP clock */ + clock_cfg = readq(ptp->reg_base + PTP_CLOCK_CFG); + clock_cfg &= ~PTP_CLOCK_CFG_PTP_EN; + writeq(clock_cfg, ptp->reg_base + PTP_CLOCK_CFG); +} + +static const struct pci_device_id ptp_id_table[] = { + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_OCTEONTX2_PTP, + PCI_VENDOR_ID_CAVIUM, + PCI_SUBSYS_DEVID_OCTX2_98xx_PTP) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_OCTEONTX2_PTP, + PCI_VENDOR_ID_CAVIUM, + PCI_SUBSYS_DEVID_OCTX2_96XX_PTP) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_OCTEONTX2_PTP, + PCI_VENDOR_ID_CAVIUM, + PCI_SUBSYS_DEVID_OCTX2_95XX_PTP) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_OCTEONTX2_PTP, + PCI_VENDOR_ID_CAVIUM, + PCI_SUBSYS_DEVID_OCTX2_LOKI_PTP) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_OCTEONTX2_PTP, + PCI_VENDOR_ID_CAVIUM, + PCI_SUBSYS_DEVID_OCTX2_95MM_PTP) }, + { 0, } +}; + +struct pci_driver ptp_driver = { + .name = DRV_NAME, + .id_table = ptp_id_table, + .probe = ptp_probe, + .remove = ptp_remove, +}; + +int rvu_mbox_handler_ptp_op(struct rvu *rvu, struct ptp_req *req, + struct ptp_rsp *rsp) +{ + int err = 0; + + /* This function is the PTP mailbox handler invoked when + * called by AF consumers/netdev drivers via mailbox mechanism. + * It is used by netdev driver to get the PTP clock and to set + * frequency adjustments. Since mailbox can be called without + * notion of whether the driver is bound to ptp device below + * validation is needed as first step. + */ + if (!rvu->ptp) + return -ENODEV; + + switch (req->op) { + case PTP_OP_ADJFINE: + err = ptp_adjfine(rvu->ptp, req->scaled_ppm); + break; + case PTP_OP_GET_CLOCK: + err = ptp_get_clock(rvu->ptp, &rsp->clk); + break; + default: + err = -EINVAL; + break; + } + + return err; +} diff --git a/drivers/net/ethernet/marvell/octeontx2/af/ptp.h b/drivers/net/ethernet/marvell/octeontx2/af/ptp.h new file mode 100644 index 000000000000..878bc395d28f --- /dev/null +++ b/drivers/net/ethernet/marvell/octeontx2/af/ptp.h @@ -0,0 +1,25 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Marvell PTP driver + * + * Copyright (C) 2020 Marvell International Ltd. + */ + +#ifndef PTP_H +#define PTP_H + +#include <linux/timecounter.h> +#include <linux/time64.h> +#include <linux/spinlock.h> + +struct ptp { + struct pci_dev *pdev; + void __iomem *reg_base; + u32 clock_rate; +}; + +struct ptp *ptp_get(void); +void ptp_put(struct ptp *ptp); + +extern struct pci_driver ptp_driver; + +#endif diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c index 557e4292c846..e1f918960730 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c @@ -18,6 +18,9 @@ #include "cgx.h" #include "rvu.h" #include "rvu_reg.h" +#include "ptp.h" + +#include "rvu_trace.h" #define DRV_NAME "octeontx2-af" #define DRV_STRING "Marvell OcteonTX2 RVU Admin Function Driver" @@ -1548,6 +1551,7 @@ static int rvu_process_mbox_msg(struct otx2_mbox *mbox, int devid, if (rsp && err) \ rsp->hdr.rc = err; \ \ + trace_otx2_msg_process(mbox->pdev, _id, err); \ return rsp ? err : -ENOMEM; \ } MBOX_MESSAGES @@ -1880,6 +1884,8 @@ static irqreturn_t rvu_mbox_intr_handler(int irq, void *rvu_irq) intr = rvu_read64(rvu, BLKADDR_RVUM, RVU_AF_PFAF_MBOX_INT); /* Clear interrupts */ rvu_write64(rvu, BLKADDR_RVUM, RVU_AF_PFAF_MBOX_INT, intr); + if (intr) + trace_otx2_msg_interrupt(rvu->pdev, "PF(s) to AF", intr); /* Sync with mbox memory region */ rmb(); @@ -1897,6 +1903,8 @@ static irqreturn_t rvu_mbox_intr_handler(int irq, void *rvu_irq) intr = rvupf_read64(rvu, RVU_PF_VFPF_MBOX_INTX(0)); rvupf_write64(rvu, RVU_PF_VFPF_MBOX_INTX(0), intr); + if (intr) + trace_otx2_msg_interrupt(rvu->pdev, "VF(s) to AF", intr); rvu_queue_work(&rvu->afvf_wq_info, 0, vfs, intr); @@ -2565,13 +2573,21 @@ static int rvu_probe(struct pci_dev *pdev, const struct pci_device_id *id) pci_set_master(pdev); + rvu->ptp = ptp_get(); + if (IS_ERR(rvu->ptp)) { + err = PTR_ERR(rvu->ptp); + if (err == -EPROBE_DEFER) + goto err_release_regions; + rvu->ptp = NULL; + } + /* Map Admin function CSRs */ rvu->afreg_base = pcim_iomap(pdev, PCI_AF_REG_BAR_NUM, 0); rvu->pfreg_base = pcim_iomap(pdev, PCI_PF_REG_BAR_NUM, 0); if (!rvu->afreg_base || !rvu->pfreg_base) { dev_err(dev, "Unable to map admin function CSRs, aborting\n"); err = -ENOMEM; - goto err_release_regions; + goto err_put_ptp; } /* Store module params in rvu structure */ @@ -2586,7 +2602,7 @@ static int rvu_probe(struct pci_dev *pdev, const struct pci_device_id *id) err = rvu_setup_hw_resources(rvu); if (err) - goto err_release_regions; + goto err_put_ptp; /* Init mailbox btw AF and PFs */ err = rvu_mbox_init(rvu, &rvu->afpf_wq_info, TYPE_AFPF, @@ -2626,6 +2642,8 @@ err_hwsetup: rvu_reset_all_blocks(rvu); rvu_free_hw_resources(rvu); rvu_clear_rvum_blk_revid(rvu); +err_put_ptp: + ptp_put(rvu->ptp); err_release_regions: pci_release_regions(pdev); err_disable_device: @@ -2651,6 +2669,7 @@ static void rvu_remove(struct pci_dev *pdev) rvu_reset_all_blocks(rvu); rvu_free_hw_resources(rvu); rvu_clear_rvum_blk_revid(rvu); + ptp_put(rvu->ptp); pci_release_regions(pdev); pci_disable_device(pdev); pci_set_drvdata(pdev, NULL); @@ -2676,9 +2695,19 @@ static int __init rvu_init_module(void) if (err < 0) return err; + err = pci_register_driver(&ptp_driver); + if (err < 0) + goto ptp_err; + err = pci_register_driver(&rvu_driver); if (err < 0) - pci_unregister_driver(&cgx_driver); + goto rvu_err; + + return 0; +rvu_err: + pci_unregister_driver(&ptp_driver); +ptp_err: + pci_unregister_driver(&cgx_driver); return err; } @@ -2686,6 +2715,7 @@ static int __init rvu_init_module(void) static void __exit rvu_cleanup_module(void) { pci_unregister_driver(&rvu_driver); + pci_unregister_driver(&ptp_driver); pci_unregister_driver(&cgx_driver); } diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h index b89dde2c8b08..90eed3160915 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h @@ -289,6 +289,22 @@ struct rvu_fwdata { u64 reserved[FWDATA_RESERVED_MEM]; }; +struct ptp; + +/* KPU profile adapter structure gathering all KPU configuration data and abstracting out the + * source where it came from. + */ +struct npc_kpu_profile_adapter { + const char *name; + u64 version; + const struct npc_lt_def_cfg *lt_def; + const struct npc_kpu_profile_action *ikpu; /* array[pkinds] */ + const struct npc_kpu_profile *kpu; /* array[kpus] */ + const struct npc_mcam_kex *mkex; + size_t pkinds; + size_t kpus; +}; + struct rvu { void __iomem *afreg_base; void __iomem *pfreg_base; @@ -337,6 +353,11 @@ struct rvu { /* Firmware data */ struct rvu_fwdata *fwdata; + /* NPC KPU data */ + struct npc_kpu_profile_adapter kpu; + + struct ptp *ptp; + #ifdef CONFIG_DEBUG_FS struct rvu_debugfs rvu_dbg; #endif @@ -470,6 +491,7 @@ int rvu_npc_init(struct rvu *rvu); void rvu_npc_freemem(struct rvu *rvu); int rvu_npc_get_pkind(struct rvu *rvu, u16 pf); void rvu_npc_set_pkind(struct rvu *rvu, int pkind, struct rvu_pfvf *pfvf); +int npc_config_ts_kpuaction(struct rvu *rvu, int pf, u16 pcifunc, bool en); void rvu_npc_install_ucast_entry(struct rvu *rvu, u16 pcifunc, int nixlf, u64 chan, u8 *mac_addr); void rvu_npc_install_promisc_entry(struct rvu *rvu, u16 pcifunc, diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c index f3c82e489897..fa9152ff5e2a 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_cgx.c @@ -15,6 +15,7 @@ #include "rvu.h" #include "cgx.h" #include "rvu_reg.h" +#include "rvu_trace.h" struct cgx_evq_entry { struct list_head evq_node; @@ -34,6 +35,7 @@ static struct _req_type __maybe_unused \ return NULL; \ req->hdr.sig = OTX2_MBOX_REQ_SIG; \ req->hdr.id = _id; \ + trace_otx2_msg_alloc(rvu->pdev, _id, sizeof(*req)); \ return req; \ } @@ -509,6 +511,45 @@ int rvu_mbox_handler_cgx_promisc_disable(struct rvu *rvu, struct msg_req *req, return 0; } +static int rvu_cgx_ptp_rx_cfg(struct rvu *rvu, u16 pcifunc, bool enable) +{ + int pf = rvu_get_pf(pcifunc); + u8 cgx_id, lmac_id; + void *cgxd; + + /* This msg is expected only from PFs that are mapped to CGX LMACs, + * if received from other PF/VF simply ACK, nothing to do. + */ + if ((pcifunc & RVU_PFVF_FUNC_MASK) || + !is_pf_cgxmapped(rvu, pf)) + return -ENODEV; + + rvu_get_cgx_lmac_id(rvu->pf2cgxlmac_map[pf], &cgx_id, &lmac_id); + cgxd = rvu_cgx_pdata(cgx_id, rvu); + + cgx_lmac_ptp_config(cgxd, lmac_id, enable); + /* If PTP is enabled then inform NPC that packets to be + * parsed by this PF will have their data shifted by 8 bytes + * and if PTP is disabled then no shift is required + */ + if (npc_config_ts_kpuaction(rvu, pf, pcifunc, enable)) + return -EINVAL; + + return 0; +} + +int rvu_mbox_handler_cgx_ptp_rx_enable(struct rvu *rvu, struct msg_req *req, + struct msg_rsp *rsp) +{ + return rvu_cgx_ptp_rx_cfg(rvu, req->hdr.pcifunc, true); +} + +int rvu_mbox_handler_cgx_ptp_rx_disable(struct rvu *rvu, struct msg_req *req, + struct msg_rsp *rsp) +{ + return rvu_cgx_ptp_rx_cfg(rvu, req->hdr.pcifunc, false); +} + static int rvu_cgx_config_linkevents(struct rvu *rvu, u16 pcifunc, bool en) { int pf = rvu_get_pf(pcifunc); diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c index 0fc70824fd6b..21a89dd76d3c 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c @@ -2508,6 +2508,14 @@ static int set_flowkey_fields(struct nix_rx_flowkey_alg *alg, u32 flow_cfg) field->ltype_match = NPC_LT_LE_GTPU; field->ltype_mask = 0xF; break; + case NIX_FLOW_KEY_TYPE_VLAN: + field->lid = NPC_LID_LB; + field->hdr_offset = 2; /* Skip TPID (2-bytes) */ + field->bytesm1 = 1; /* 2 Bytes (Actually 12 bits) */ + field->ltype_match = NPC_LT_LB_CTAG; + field->ltype_mask = 0xF; + field->fn_mask = 1; /* Mask out the first nibble */ + break; } field->ena = 1; @@ -3103,6 +3111,7 @@ static int nix_aq_init(struct rvu *rvu, struct rvu_block *block) int rvu_nix_init(struct rvu *rvu) { + const struct npc_lt_def_cfg *ltdefs; struct rvu_hwinfo *hw = rvu->hw; struct rvu_block *block; int blkaddr, err; @@ -3133,6 +3142,7 @@ int rvu_nix_init(struct rvu *rvu) rvu_write64(rvu, blkaddr, NIX_AF_SQM_DBG_CTL_STATUS, cfg); } + ltdefs = rvu->kpu.lt_def; /* Calibrate X2P bus to check if CGX/LBK links are fine */ err = nix_calibrate_x2p(rvu, blkaddr); if (err) @@ -3180,28 +3190,38 @@ int rvu_nix_init(struct rvu *rvu) * and validate length and checksums. */ rvu_write64(rvu, blkaddr, NIX_AF_RX_DEF_OL2, - (NPC_LID_LA << 8) | (NPC_LT_LA_ETHER << 4) | 0x0F); + (ltdefs->rx_ol2.lid << 8) | (ltdefs->rx_ol2.ltype_match << 4) | + ltdefs->rx_ol2.ltype_mask); rvu_write64(rvu, blkaddr, NIX_AF_RX_DEF_OIP4, - (NPC_LID_LC << 8) | (NPC_LT_LC_IP << 4) | 0x0F); + (ltdefs->rx_oip4.lid << 8) | (ltdefs->rx_oip4.ltype_match << 4) | + ltdefs->rx_oip4.ltype_mask); rvu_write64(rvu, blkaddr, NIX_AF_RX_DEF_IIP4, - (NPC_LID_LG << 8) | (NPC_LT_LG_TU_IP << 4) | 0x0F); + (ltdefs->rx_iip4.lid << 8) | (ltdefs->rx_iip4.ltype_match << 4) | + ltdefs->rx_iip4.ltype_mask); rvu_write64(rvu, blkaddr, NIX_AF_RX_DEF_OIP6, - (NPC_LID_LC << 8) | (NPC_LT_LC_IP6 << 4) | 0x0F); + (ltdefs->rx_oip6.lid << 8) | (ltdefs->rx_oip6.ltype_match << 4) | + ltdefs->rx_oip6.ltype_mask); rvu_write64(rvu, blkaddr, NIX_AF_RX_DEF_IIP6, - (NPC_LID_LG << 8) | (NPC_LT_LG_TU_IP6 << 4) | 0x0F); + (ltdefs->rx_iip6.lid << 8) | (ltdefs->rx_iip6.ltype_match << 4) | + ltdefs->rx_iip6.ltype_mask); rvu_write64(rvu, blkaddr, NIX_AF_RX_DEF_OTCP, - (NPC_LID_LD << 8) | (NPC_LT_LD_TCP << 4) | 0x0F); + (ltdefs->rx_otcp.lid << 8) | (ltdefs->rx_otcp.ltype_match << 4) | + ltdefs->rx_otcp.ltype_mask); rvu_write64(rvu, blkaddr, NIX_AF_RX_DEF_ITCP, - (NPC_LID_LH << 8) | (NPC_LT_LH_TU_TCP << 4) | 0x0F); + (ltdefs->rx_itcp.lid << 8) | (ltdefs->rx_itcp.ltype_match << 4) | + ltdefs->rx_itcp.ltype_mask); rvu_write64(rvu, blkaddr, NIX_AF_RX_DEF_OUDP, - (NPC_LID_LD << 8) | (NPC_LT_LD_UDP << 4) | 0x0F); + (ltdefs->rx_oudp.lid << 8) | (ltdefs->rx_oudp.ltype_match << 4) | + ltdefs->rx_oudp.ltype_mask); rvu_write64(rvu, blkaddr, NIX_AF_RX_DEF_IUDP, - (NPC_LID_LH << 8) | (NPC_LT_LH_TU_UDP << 4) | 0x0F); + (ltdefs->rx_iudp.lid << 8) | (ltdefs->rx_iudp.ltype_match << 4) | + ltdefs->rx_iudp.ltype_mask); rvu_write64(rvu, blkaddr, NIX_AF_RX_DEF_OSCTP, - (NPC_LID_LD << 8) | (NPC_LT_LD_SCTP << 4) | 0x0F); + (ltdefs->rx_osctp.lid << 8) | (ltdefs->rx_osctp.ltype_match << 4) | + ltdefs->rx_osctp.ltype_mask); rvu_write64(rvu, blkaddr, NIX_AF_RX_DEF_ISCTP, - (NPC_LID_LH << 8) | (NPC_LT_LH_TU_SCTP << 4) | - 0x0F); + (ltdefs->rx_isctp.lid << 8) | (ltdefs->rx_isctp.ltype_match << 4) | + ltdefs->rx_isctp.ltype_mask); err = nix_rx_flowkey_alg_cfg(rvu, blkaddr); if (err) @@ -3318,6 +3338,49 @@ void rvu_nix_lf_teardown(struct rvu *rvu, u16 pcifunc, int blkaddr, int nixlf) nix_ctx_free(rvu, pfvf); } +#define NIX_AF_LFX_TX_CFG_PTP_EN BIT_ULL(32) + +static int rvu_nix_lf_ptp_tx_cfg(struct rvu *rvu, u16 pcifunc, bool enable) +{ + struct rvu_hwinfo *hw = rvu->hw; + struct rvu_block *block; + int blkaddr; + int nixlf; + u64 cfg; + + blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, pcifunc); + if (blkaddr < 0) + return NIX_AF_ERR_AF_LF_INVALID; + + block = &hw->block[blkaddr]; + nixlf = rvu_get_lf(rvu, block, pcifunc, 0); + if (nixlf < 0) + return NIX_AF_ERR_AF_LF_INVALID; + + cfg = rvu_read64(rvu, blkaddr, NIX_AF_LFX_TX_CFG(nixlf)); + + if (enable) + cfg |= NIX_AF_LFX_TX_CFG_PTP_EN; + else + cfg &= ~NIX_AF_LFX_TX_CFG_PTP_EN; + + rvu_write64(rvu, blkaddr, NIX_AF_LFX_TX_CFG(nixlf), cfg); + + return 0; +} + +int rvu_mbox_handler_nix_lf_ptp_tx_enable(struct rvu *rvu, struct msg_req *req, + struct msg_rsp *rsp) +{ + return rvu_nix_lf_ptp_tx_cfg(rvu, req->hdr.pcifunc, true); +} + +int rvu_mbox_handler_nix_lf_ptp_tx_disable(struct rvu *rvu, struct msg_req *req, + struct msg_rsp *rsp) +{ + return rvu_nix_lf_ptp_tx_cfg(rvu, req->hdr.pcifunc, false); +} + int rvu_mbox_handler_nix_lso_format_cfg(struct rvu *rvu, struct nix_lso_format_cfg *req, struct nix_lso_format_cfg_rsp *rsp) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c index fbaf9bcd83f2..511b01dd03ed 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c @@ -27,6 +27,9 @@ #define NIXLF_PROMISC_ENTRY 2 #define NPC_PARSE_RESULT_DMAC_OFFSET 8 +#define NPC_HW_TSTAMP_OFFSET 8 + +static const char def_pfl_name[] = "default"; static void npc_mcam_free_all_entries(struct rvu *rvu, struct npc_mcam *mcam, int blkaddr, u16 pcifunc); @@ -61,6 +64,36 @@ int rvu_npc_get_pkind(struct rvu *rvu, u16 pf) return -1; } +#define NPC_AF_ACTION0_PTR_ADVANCE GENMASK_ULL(27, 20) + +int npc_config_ts_kpuaction(struct rvu *rvu, int pf, u16 pcifunc, bool enable) +{ + int pkind, blkaddr; + u64 val; + + pkind = rvu_npc_get_pkind(rvu, pf); + if (pkind < 0) { + dev_err(rvu->dev, "%s: pkind not mapped\n", __func__); + return -EINVAL; + } + + blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NPC, pcifunc); + if (blkaddr < 0) { + dev_err(rvu->dev, "%s: NPC block not implemented\n", __func__); + return -EINVAL; + } + + val = rvu_read64(rvu, blkaddr, NPC_AF_PKINDX_ACTION0(pkind)); + val &= ~NPC_AF_ACTION0_PTR_ADVANCE; + /* If timestamp is enabled then configure NPC to shift 8 bytes */ + if (enable) + val |= FIELD_PREP(NPC_AF_ACTION0_PTR_ADVANCE, + NPC_HW_TSTAMP_OFFSET); + rvu_write64(rvu, blkaddr, NPC_AF_PKINDX_ACTION0(pkind), val); + + return 0; +} + static int npc_get_nixlf_mcam_index(struct npc_mcam *mcam, u16 pcifunc, int nixlf, int type) { @@ -417,7 +450,7 @@ void rvu_npc_install_promisc_entry(struct rvu *rvu, u16 pcifunc, entry.kw_mask[0] = 0xFFFULL; if (allmulti) { - kwi = NPC_PARSE_RESULT_DMAC_OFFSET / sizeof(u64); + kwi = NPC_KEXOF_DMAC / sizeof(u64); entry.kw[kwi] = BIT_ULL(40); /* LSB bit of 1st byte in DMAC */ entry.kw_mask[kwi] = BIT_ULL(40); } @@ -699,88 +732,8 @@ void rvu_npc_disable_mcam_entries(struct rvu *rvu, u16 pcifunc, int nixlf) rvu_write64(rvu, blkaddr, \ NPC_AF_INTFX_LDATAX_FLAGSX_CFG(intf, ld, flags), cfg) -#define KEX_LD_CFG(bytesm1, hdr_ofs, ena, flags_ena, key_ofs) \ - (((bytesm1) << 16) | ((hdr_ofs) << 8) | ((ena) << 7) | \ - ((flags_ena) << 6) | ((key_ofs) & 0x3F)) - -static void npc_config_ldata_extract(struct rvu *rvu, int blkaddr) -{ - struct npc_mcam *mcam = &rvu->hw->mcam; - int lid, ltype; - int lid_count; - u64 cfg; - - cfg = rvu_read64(rvu, blkaddr, NPC_AF_CONST); - lid_count = (cfg >> 4) & 0xF; - - /* First clear any existing config i.e - * disable LDATA and FLAGS extraction. - */ - for (lid = 0; lid < lid_count; lid++) { - for (ltype = 0; ltype < 16; ltype++) { - SET_KEX_LD(NIX_INTF_RX, lid, ltype, 0, 0ULL); - SET_KEX_LD(NIX_INTF_RX, lid, ltype, 1, 0ULL); - SET_KEX_LD(NIX_INTF_TX, lid, ltype, 0, 0ULL); - SET_KEX_LD(NIX_INTF_TX, lid, ltype, 1, 0ULL); - - SET_KEX_LDFLAGS(NIX_INTF_RX, 0, ltype, 0ULL); - SET_KEX_LDFLAGS(NIX_INTF_RX, 1, ltype, 0ULL); - SET_KEX_LDFLAGS(NIX_INTF_TX, 0, ltype, 0ULL); - SET_KEX_LDFLAGS(NIX_INTF_TX, 1, ltype, 0ULL); - } - } - - if (mcam->keysize != NPC_MCAM_KEY_X2) - return; - - /* Default MCAM KEX profile */ - /* Layer A: Ethernet: */ - - /* DMAC: 6 bytes, KW1[47:0] */ - cfg = KEX_LD_CFG(0x05, 0x0, 0x1, 0x0, NPC_PARSE_RESULT_DMAC_OFFSET); - SET_KEX_LD(NIX_INTF_RX, NPC_LID_LA, NPC_LT_LA_ETHER, 0, cfg); - - /* Ethertype: 2 bytes, KW0[47:32] */ - cfg = KEX_LD_CFG(0x01, 0xc, 0x1, 0x0, 0x4); - SET_KEX_LD(NIX_INTF_RX, NPC_LID_LA, NPC_LT_LA_ETHER, 1, cfg); - - /* Layer B: Single VLAN (CTAG) */ - /* CTAG VLAN[2..3] + Ethertype, 4 bytes, KW0[63:32] */ - cfg = KEX_LD_CFG(0x03, 0x0, 0x1, 0x0, 0x4); - SET_KEX_LD(NIX_INTF_RX, NPC_LID_LB, NPC_LT_LB_CTAG, 0, cfg); - - /* Layer B: Stacked VLAN (STAG|QinQ) */ - /* CTAG VLAN[2..3] + Ethertype, 4 bytes, KW0[63:32] */ - cfg = KEX_LD_CFG(0x03, 0x4, 0x1, 0x0, 0x4); - SET_KEX_LD(NIX_INTF_RX, NPC_LID_LB, NPC_LT_LB_STAG_QINQ, 0, cfg); - - /* Layer C: IPv4 */ - /* SIP+DIP: 8 bytes, KW2[63:0] */ - cfg = KEX_LD_CFG(0x07, 0xc, 0x1, 0x0, 0x10); - SET_KEX_LD(NIX_INTF_RX, NPC_LID_LC, NPC_LT_LC_IP, 0, cfg); - /* TOS: 1 byte, KW1[63:56] */ - cfg = KEX_LD_CFG(0x0, 0x1, 0x1, 0x0, 0xf); - SET_KEX_LD(NIX_INTF_RX, NPC_LID_LC, NPC_LT_LC_IP, 1, cfg); - - /* Layer D:UDP */ - /* SPORT: 2 bytes, KW3[15:0] */ - cfg = KEX_LD_CFG(0x1, 0x0, 0x1, 0x0, 0x18); - SET_KEX_LD(NIX_INTF_RX, NPC_LID_LD, NPC_LT_LD_UDP, 0, cfg); - /* DPORT: 2 bytes, KW3[31:16] */ - cfg = KEX_LD_CFG(0x1, 0x2, 0x1, 0x0, 0x1a); - SET_KEX_LD(NIX_INTF_RX, NPC_LID_LD, NPC_LT_LD_UDP, 1, cfg); - - /* Layer D:TCP */ - /* SPORT: 2 bytes, KW3[15:0] */ - cfg = KEX_LD_CFG(0x1, 0x0, 0x1, 0x0, 0x18); - SET_KEX_LD(NIX_INTF_RX, NPC_LID_LD, NPC_LT_LD_TCP, 0, cfg); - /* DPORT: 2 bytes, KW3[31:16] */ - cfg = KEX_LD_CFG(0x1, 0x2, 0x1, 0x0, 0x1a); - SET_KEX_LD(NIX_INTF_RX, NPC_LID_LD, NPC_LT_LD_TCP, 1, cfg); -} - static void npc_program_mkex_profile(struct rvu *rvu, int blkaddr, - struct npc_mcam_kex *mkex) + const struct npc_mcam_kex *mkex) { int lid, lt, ld, fl; @@ -820,34 +773,31 @@ static void npc_program_mkex_profile(struct rvu *rvu, int blkaddr, } } -/* strtoull of "mkexprof" with base:36 */ -#define MKEX_SIGN 0x19bbfdbd15f #define MKEX_END_SIGN 0xdeadbeef -static void npc_load_mkex_profile(struct rvu *rvu, int blkaddr) +static void npc_load_mkex_profile(struct rvu *rvu, int blkaddr, + const char *mkex_profile) { - const char *mkex_profile = rvu->mkex_pfl_name; struct device *dev = &rvu->pdev->dev; - void __iomem *mkex_prfl_addr = NULL; struct npc_mcam_kex *mcam_kex; - u64 prfl_addr; - u64 prfl_sz; + void *mkex_prfl_addr = NULL; + u64 prfl_addr, prfl_sz; /* If user not selected mkex profile */ - if (!strncmp(mkex_profile, "default", MKEX_NAME_LEN)) - goto load_default; + if (!strncmp(mkex_profile, def_pfl_name, MKEX_NAME_LEN)) + goto program_mkex; if (!rvu->fwdata) - goto load_default; + goto program_mkex; prfl_addr = rvu->fwdata->mcam_addr; prfl_sz = rvu->fwdata->mcam_sz; if (!prfl_addr || !prfl_sz) - goto load_default; + goto program_mkex; - mkex_prfl_addr = ioremap_wc(prfl_addr, prfl_sz); + mkex_prfl_addr = memremap(prfl_addr, prfl_sz, MEMREMAP_WC); if (!mkex_prfl_addr) - goto load_default; + goto program_mkex; mcam_kex = (struct npc_mcam_kex *)mkex_prfl_addr; @@ -859,35 +809,27 @@ static void npc_load_mkex_profile(struct rvu *rvu, int blkaddr) * parse nibble enable configuration has to be * identical for both Rx and Tx interfaces. */ - if (is_rvu_96xx_B0(rvu) && - mcam_kex->keyx_cfg[NIX_INTF_RX] != - mcam_kex->keyx_cfg[NIX_INTF_TX]) - goto load_default; - - /* Program selected mkex profile */ - npc_program_mkex_profile(rvu, blkaddr, mcam_kex); - - goto unmap; + if (!is_rvu_96xx_B0(rvu) || + mcam_kex->keyx_cfg[NIX_INTF_RX] == mcam_kex->keyx_cfg[NIX_INTF_TX]) + rvu->kpu.mkex = mcam_kex; + goto program_mkex; } mcam_kex++; prfl_sz -= sizeof(struct npc_mcam_kex); } - dev_warn(dev, "Failed to load requested profile: %s\n", - rvu->mkex_pfl_name); + dev_warn(dev, "Failed to load requested profile: %s\n", mkex_profile); -load_default: - dev_info(rvu->dev, "Using default mkex profile\n"); - /* Config packet data and flags extraction into PARSE result */ - npc_config_ldata_extract(rvu, blkaddr); - -unmap: +program_mkex: + dev_info(rvu->dev, "Using %s mkex profile\n", rvu->kpu.mkex->name); + /* Program selected mkex profile */ + npc_program_mkex_profile(rvu, blkaddr, rvu->kpu.mkex); if (mkex_prfl_addr) - iounmap(mkex_prfl_addr); + memunmap(mkex_prfl_addr); } static void npc_config_kpuaction(struct rvu *rvu, int blkaddr, - struct npc_kpu_profile_action *kpuaction, + const struct npc_kpu_profile_action *kpuaction, int kpu, int entry, bool pkind) { struct npc_kpu_action0 action0 = {0}; @@ -929,7 +871,7 @@ static void npc_config_kpuaction(struct rvu *rvu, int blkaddr, } static void npc_config_kpucam(struct rvu *rvu, int blkaddr, - struct npc_kpu_profile_cam *kpucam, + const struct npc_kpu_profile_cam *kpucam, int kpu, int entry) { struct npc_kpu_cam cam0 = {0}; @@ -957,7 +899,7 @@ static inline u64 enable_mask(int count) } static void npc_program_kpu_profile(struct rvu *rvu, int blkaddr, int kpu, - struct npc_kpu_profile *profile) + const struct npc_kpu_profile *profile) { int entry, num_entries, max_entries; @@ -995,6 +937,27 @@ static void npc_program_kpu_profile(struct rvu *rvu, int blkaddr, int kpu, rvu_write64(rvu, blkaddr, NPC_AF_KPUX_CFG(kpu), 0x01); } +static int npc_prepare_default_kpu(struct npc_kpu_profile_adapter *profile) +{ + profile->name = def_pfl_name; + profile->version = NPC_KPU_PROFILE_VER; + profile->ikpu = ikpu_action_entries; + profile->pkinds = ARRAY_SIZE(ikpu_action_entries); + profile->kpu = npc_kpu_profiles; + profile->kpus = ARRAY_SIZE(npc_kpu_profiles); + profile->lt_def = &npc_lt_defaults; + profile->mkex = &npc_mkex_default; + + return 0; +} + +static void npc_load_kpu_profile(struct rvu *rvu) +{ + struct npc_kpu_profile_adapter *profile = &rvu->kpu; + + npc_prepare_default_kpu(profile); +} + static void npc_parser_profile_init(struct rvu *rvu, int blkaddr) { struct rvu_hwinfo *hw = rvu->hw; @@ -1013,25 +976,26 @@ static void npc_parser_profile_init(struct rvu *rvu, int blkaddr) rvu_write64(rvu, blkaddr, NPC_AF_KPUX_CFG(idx), 0x00); } + /* Load and customize KPU profile. */ + npc_load_kpu_profile(rvu); + /* First program IKPU profile i.e PKIND configs. * Check HW max count to avoid configuring junk or * writing to unsupported CSR addresses. */ pkind = &hw->pkind; - num_pkinds = ARRAY_SIZE(ikpu_action_entries); + num_pkinds = rvu->kpu.pkinds; num_pkinds = min_t(int, pkind->rsrc.max, num_pkinds); for (idx = 0; idx < num_pkinds; idx++) - npc_config_kpuaction(rvu, blkaddr, - &ikpu_action_entries[idx], 0, idx, true); + npc_config_kpuaction(rvu, blkaddr, &rvu->kpu.ikpu[idx], 0, idx, true); /* Program KPU CAM and Action profiles */ - num_kpus = ARRAY_SIZE(npc_kpu_profiles); + num_kpus = rvu->kpu.kpus; num_kpus = min_t(int, hw->npc_kpus, num_kpus); for (idx = 0; idx < num_kpus; idx++) - npc_program_kpu_profile(rvu, blkaddr, - idx, &npc_kpu_profiles[idx]); + npc_program_kpu_profile(rvu, blkaddr, idx, &rvu->kpu.kpu[idx]); } static int npc_mcam_rsrcs_init(struct rvu *rvu, int blkaddr) @@ -1156,11 +1120,11 @@ free_mem: int rvu_npc_init(struct rvu *rvu) { + struct npc_kpu_profile_adapter *kpu = &rvu->kpu; struct npc_pkind *pkind = &rvu->hw->pkind; struct npc_mcam *mcam = &rvu->hw->mcam; - u64 keyz = NPC_MCAM_KEY_X2; + u64 cfg, nibble_ena, rx_kex, tx_kex; int blkaddr, entry, bank, err; - u64 cfg, nibble_ena; blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NPC, 0); if (blkaddr < 0) { @@ -1194,13 +1158,16 @@ int rvu_npc_init(struct rvu *rvu) /* Config Outer L2, IPv4's NPC layer info */ rvu_write64(rvu, blkaddr, NPC_AF_PCK_DEF_OL2, - (NPC_LID_LA << 8) | (NPC_LT_LA_ETHER << 4) | 0x0F); + (kpu->lt_def->pck_ol2.lid << 8) | (kpu->lt_def->pck_ol2.ltype_match << 4) | + kpu->lt_def->pck_ol2.ltype_mask); rvu_write64(rvu, blkaddr, NPC_AF_PCK_DEF_OIP4, - (NPC_LID_LC << 8) | (NPC_LT_LC_IP << 4) | 0x0F); + (kpu->lt_def->pck_oip4.lid << 8) | (kpu->lt_def->pck_oip4.ltype_match << 4) | + kpu->lt_def->pck_oip4.ltype_mask); /* Config Inner IPV4 NPC layer info */ rvu_write64(rvu, blkaddr, NPC_AF_PCK_DEF_IIP4, - (NPC_LID_LG << 8) | (NPC_LT_LG_TU_IP << 4) | 0x0F); + (kpu->lt_def->pck_iip4.lid << 8) | (kpu->lt_def->pck_iip4.ltype_match << 4) | + kpu->lt_def->pck_iip4.ltype_mask); /* Enable below for Rx pkts. * - Outer IPv4 header checksum validation. @@ -1216,23 +1183,25 @@ int rvu_npc_init(struct rvu *rvu) /* Set RX and TX side MCAM search key size. * LA..LD (ltype only) + Channel */ - nibble_ena = 0x49247; - rvu_write64(rvu, blkaddr, NPC_AF_INTFX_KEX_CFG(NIX_INTF_RX), - ((keyz & 0x3) << 32) | nibble_ena); + rx_kex = npc_mkex_default.keyx_cfg[NIX_INTF_RX]; + tx_kex = npc_mkex_default.keyx_cfg[NIX_INTF_TX]; + nibble_ena = FIELD_GET(NPC_PARSE_NIBBLE, rx_kex); + rvu_write64(rvu, blkaddr, NPC_AF_INTFX_KEX_CFG(NIX_INTF_RX), rx_kex); /* Due to an errata (35786) in A0 pass silicon, parse nibble enable * configuration has to be identical for both Rx and Tx interfaces. */ - if (!is_rvu_96xx_B0(rvu)) - nibble_ena = (1ULL << 19) - 1; - rvu_write64(rvu, blkaddr, NPC_AF_INTFX_KEX_CFG(NIX_INTF_TX), - ((keyz & 0x3) << 32) | nibble_ena); + if (is_rvu_96xx_B0(rvu)) { + tx_kex &= ~NPC_PARSE_NIBBLE; + tx_kex |= FIELD_PREP(NPC_PARSE_NIBBLE, nibble_ena); + } + rvu_write64(rvu, blkaddr, NPC_AF_INTFX_KEX_CFG(NIX_INTF_TX), tx_kex); err = npc_mcam_rsrcs_init(rvu, blkaddr); if (err) return err; /* Configure MKEX profile */ - npc_load_mkex_profile(rvu, blkaddr); + npc_load_mkex_profile(rvu, blkaddr, rvu->mkex_pfl_name); /* Set TX miss action to UCAST_DEFAULT i.e * transmit the packet on NIX LF SQ's default channel. diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_trace.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_trace.c new file mode 100644 index 000000000000..56f90cf9c4c0 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_trace.c @@ -0,0 +1,12 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Marvell OcteonTx2 RVU Admin Function driver tracepoints + * + * Copyright (C) 2020 Marvell International Ltd. + */ + +#define CREATE_TRACE_POINTS +#include "rvu_trace.h" + +EXPORT_TRACEPOINT_SYMBOL(otx2_msg_alloc); +EXPORT_TRACEPOINT_SYMBOL(otx2_msg_interrupt); +EXPORT_TRACEPOINT_SYMBOL(otx2_msg_process); diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_trace.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu_trace.h new file mode 100644 index 000000000000..e6609068e81b --- /dev/null +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_trace.h @@ -0,0 +1,103 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Marvell OcteonTx2 RVU Admin Function driver tracepoints + * + * Copyright (C) 2020 Marvell International Ltd. + */ + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM rvu + +#if !defined(__RVU_TRACE_H) || defined(TRACE_HEADER_MULTI_READ) +#define __RVU_TRACE_H + +#include <linux/types.h> +#include <linux/tracepoint.h> +#include <linux/pci.h> + +TRACE_EVENT(otx2_msg_alloc, + TP_PROTO(const struct pci_dev *pdev, u16 id, u64 size), + TP_ARGS(pdev, id, size), + TP_STRUCT__entry(__string(dev, pci_name(pdev)) + __field(u16, id) + __field(u64, size) + ), + TP_fast_assign(__assign_str(dev, pci_name(pdev)) + __entry->id = id; + __entry->size = size; + ), + TP_printk("[%s] msg:(0x%x) size:%lld\n", __get_str(dev), + __entry->id, __entry->size) +); + +TRACE_EVENT(otx2_msg_send, + TP_PROTO(const struct pci_dev *pdev, u16 num_msgs, u64 msg_size), + TP_ARGS(pdev, num_msgs, msg_size), + TP_STRUCT__entry(__string(dev, pci_name(pdev)) + __field(u16, num_msgs) + __field(u64, msg_size) + ), + TP_fast_assign(__assign_str(dev, pci_name(pdev)) + __entry->num_msgs = num_msgs; + __entry->msg_size = msg_size; + ), + TP_printk("[%s] sent %d msg(s) of size:%lld\n", __get_str(dev), + __entry->num_msgs, __entry->msg_size) +); + +TRACE_EVENT(otx2_msg_check, + TP_PROTO(const struct pci_dev *pdev, u16 reqid, u16 rspid, int rc), + TP_ARGS(pdev, reqid, rspid, rc), + TP_STRUCT__entry(__string(dev, pci_name(pdev)) + __field(u16, reqid) + __field(u16, rspid) + __field(int, rc) + ), + TP_fast_assign(__assign_str(dev, pci_name(pdev)) + __entry->reqid = reqid; + __entry->rspid = rspid; + __entry->rc = rc; + ), + TP_printk("[%s] req->id:0x%x rsp->id:0x%x resp_code:%d\n", + __get_str(dev), __entry->reqid, + __entry->rspid, __entry->rc) +); + +TRACE_EVENT(otx2_msg_interrupt, + TP_PROTO(const struct pci_dev *pdev, const char *msg, u64 intr), + TP_ARGS(pdev, msg, intr), + TP_STRUCT__entry(__string(dev, pci_name(pdev)) + __string(str, msg) + __field(u64, intr) + ), + TP_fast_assign(__assign_str(dev, pci_name(pdev)) + __assign_str(str, msg) + __entry->intr = intr; + ), + TP_printk("[%s] mbox interrupt %s (0x%llx)\n", __get_str(dev), + __get_str(str), __entry->intr) +); + +TRACE_EVENT(otx2_msg_process, + TP_PROTO(const struct pci_dev *pdev, u16 id, int err), + TP_ARGS(pdev, id, err), + TP_STRUCT__entry(__string(dev, pci_name(pdev)) + __field(u16, id) + __field(int, err) + ), + TP_fast_assign(__assign_str(dev, pci_name(pdev)) + __entry->id = id; + __entry->err = err; + ), + TP_printk("[%s] msg:(0x%x) error:%d\n", __get_str(dev), + __entry->id, __entry->err) +); + +#endif /* __RVU_TRACE_H */ + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . + +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE rvu_trace + +#include <trace/define_trace.h> diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile index 778df331c8ac..b2c6385707c9 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile +++ b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile @@ -6,7 +6,8 @@ obj-$(CONFIG_OCTEONTX2_PF) += octeontx2_nicpf.o obj-$(CONFIG_OCTEONTX2_VF) += octeontx2_nicvf.o -octeontx2_nicpf-y := otx2_pf.o otx2_common.o otx2_txrx.o otx2_ethtool.o +octeontx2_nicpf-y := otx2_pf.o otx2_common.o otx2_txrx.o otx2_ethtool.o \ + otx2_ptp.o octeontx2_nicvf-y := otx2_vf.o ccflags-y += -I$(srctree)/drivers/net/ethernet/marvell/octeontx2/af diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c index 93c4cf7fedbf..d2581090f9a4 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c @@ -355,7 +355,7 @@ int otx2_rss_init(struct otx2_nic *pfvf) rss->flowkey_cfg = rss->enable ? rss->flowkey_cfg : NIX_FLOW_KEY_TYPE_IPV4 | NIX_FLOW_KEY_TYPE_IPV6 | NIX_FLOW_KEY_TYPE_TCP | NIX_FLOW_KEY_TYPE_UDP | - NIX_FLOW_KEY_TYPE_SCTP; + NIX_FLOW_KEY_TYPE_SCTP | NIX_FLOW_KEY_TYPE_VLAN; ret = otx2_set_flowkey_cfg(pfvf); if (ret) @@ -365,6 +365,95 @@ int otx2_rss_init(struct otx2_nic *pfvf) return 0; } +/* Setup UDP segmentation algorithm in HW */ +static void otx2_setup_udp_segmentation(struct nix_lso_format_cfg *lso, bool v4) +{ + struct nix_lso_format *field; + + field = (struct nix_lso_format *)&lso->fields[0]; + lso->field_mask = GENMASK(18, 0); + + /* IP's Length field */ + field->layer = NIX_TXLAYER_OL3; + /* In ipv4, length field is at offset 2 bytes, for ipv6 it's 4 */ + field->offset = v4 ? 2 : 4; + field->sizem1 = 1; /* i.e 2 bytes */ + field->alg = NIX_LSOALG_ADD_PAYLEN; + field++; + + /* No ID field in IPv6 header */ + if (v4) { + /* Increment IPID */ + field->layer = NIX_TXLAYER_OL3; + field->offset = 4; + field->sizem1 = 1; /* i.e 2 bytes */ + field->alg = NIX_LSOALG_ADD_SEGNUM; + field++; + } + + /* Update length in UDP header */ + field->layer = NIX_TXLAYER_OL4; + field->offset = 4; + field->sizem1 = 1; + field->alg = NIX_LSOALG_ADD_PAYLEN; +} + +/* Setup segmentation algorithms in HW and retrieve algorithm index */ +void otx2_setup_segmentation(struct otx2_nic *pfvf) +{ + struct nix_lso_format_cfg_rsp *rsp; + struct nix_lso_format_cfg *lso; + struct otx2_hw *hw = &pfvf->hw; + int err; + + mutex_lock(&pfvf->mbox.lock); + + /* UDPv4 segmentation */ + lso = otx2_mbox_alloc_msg_nix_lso_format_cfg(&pfvf->mbox); + if (!lso) + goto fail; + + /* Setup UDP/IP header fields that HW should update per segment */ + otx2_setup_udp_segmentation(lso, true); + + err = otx2_sync_mbox_msg(&pfvf->mbox); + if (err) + goto fail; + + rsp = (struct nix_lso_format_cfg_rsp *) + otx2_mbox_get_rsp(&pfvf->mbox.mbox, 0, &lso->hdr); + if (IS_ERR(rsp)) + goto fail; + + hw->lso_udpv4_idx = rsp->lso_format_idx; + + /* UDPv6 segmentation */ + lso = otx2_mbox_alloc_msg_nix_lso_format_cfg(&pfvf->mbox); + if (!lso) + goto fail; + + /* Setup UDP/IP header fields that HW should update per segment */ + otx2_setup_udp_segmentation(lso, false); + + err = otx2_sync_mbox_msg(&pfvf->mbox); + if (err) + goto fail; + + rsp = (struct nix_lso_format_cfg_rsp *) + otx2_mbox_get_rsp(&pfvf->mbox.mbox, 0, &lso->hdr); + if (IS_ERR(rsp)) + goto fail; + + hw->lso_udpv6_idx = rsp->lso_format_idx; + mutex_unlock(&pfvf->mbox.lock); + return; +fail: + mutex_unlock(&pfvf->mbox.lock); + netdev_info(pfvf->netdev, + "Failed to get LSO index for UDP GSO offload, disabling\n"); + pfvf->netdev->hw_features &= ~NETIF_F_GSO_UDP_L4; +} + void otx2_config_irq_coalescing(struct otx2_nic *pfvf, int qidx) { /* Configure CQE interrupt coalescing parameters @@ -671,6 +760,13 @@ static int otx2_sq_init(struct otx2_nic *pfvf, u16 qidx, u16 sqb_aura) if (!sq->sg) return -ENOMEM; + if (pfvf->ptp) { + err = qmem_alloc(pfvf->dev, &sq->timestamps, qset->sqe_cnt, + sizeof(*sq->timestamps)); + if (err) + return err; + } + sq->head = 0; sq->sqe_per_sqb = (pfvf->hw.sqb_size / sq->sqe_size) - 1; sq->num_sqbs = (qset->sqe_cnt + sq->sqe_per_sqb) / sq->sqe_per_sqb; diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h index 2fa29889522e..d6253f2a414d 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h @@ -13,10 +13,14 @@ #include <linux/pci.h> #include <linux/iommu.h> +#include <linux/net_tstamp.h> +#include <linux/ptp_clock_kernel.h> +#include <linux/timecounter.h> #include <mbox.h> #include "otx2_reg.h" #include "otx2_txrx.h" +#include <rvu_trace.h> /* PCI device IDs */ #define PCI_DEVID_OCTEONTX2_RVU_PF 0xA063 @@ -174,9 +178,11 @@ struct otx2_hw { u16 rq_skid; u8 cq_time_wait; - /* For TSO segmentation */ + /* Segmentation */ u8 lso_tsov4_idx; u8 lso_tsov6_idx; + u8 lso_udpv4_idx; + u8 lso_udpv6_idx; u8 hw_tso; /* MSI-X */ @@ -209,6 +215,17 @@ struct refill_work { struct otx2_nic *pf; }; +struct otx2_ptp { + struct ptp_clock_info ptp_info; + struct ptp_clock *ptp_clock; + struct otx2_nic *nic; + + struct cyclecounter cycle_counter; + struct timecounter time_counter; +}; + +#define OTX2_HW_TIMESTAMP_LEN 8 + struct otx2_nic { void __iomem *reg_base; struct net_device *netdev; @@ -216,6 +233,8 @@ struct otx2_nic { u16 max_frs; u16 rbsize; /* Receive buffer size */ +#define OTX2_FLAG_RX_TSTAMP_ENABLED BIT_ULL(0) +#define OTX2_FLAG_TX_TSTAMP_ENABLED BIT_ULL(1) #define OTX2_FLAG_INTF_DOWN BIT_ULL(2) #define OTX2_FLAG_RX_PAUSE_ENABLED BIT_ULL(9) #define OTX2_FLAG_TX_PAUSE_ENABLED BIT_ULL(10) @@ -251,6 +270,9 @@ struct otx2_nic { /* Block address of NIX either BLKADDR_NIX0 or BLKADDR_NIX1 */ int nix_blkaddr; + + struct otx2_ptp *ptp; + struct hwtstamp_config tstamp; }; static inline bool is_otx2_lbkvf(struct pci_dev *pdev) @@ -502,6 +524,7 @@ static struct _req_type __maybe_unused \ return NULL; \ req->hdr.sig = OTX2_MBOX_REQ_SIG; \ req->hdr.id = _id; \ + trace_otx2_msg_alloc(mbox->mbox.pdev, _id, sizeof(*req)); \ return req; \ } @@ -561,6 +584,7 @@ void otx2_tx_timeout(struct net_device *netdev, unsigned int txq); void otx2_get_mac_from_af(struct net_device *netdev); void otx2_config_irq_coalescing(struct otx2_nic *pfvf, int qidx); int otx2_config_pause_frm(struct otx2_nic *pfvf); +void otx2_setup_segmentation(struct otx2_nic *pfvf); /* RVU block related APIs */ int otx2_attach_npa_nix(struct otx2_nic *pfvf); diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c index d59f5a9c7273..662fb80dbb9d 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c @@ -13,8 +13,10 @@ #include <linux/stddef.h> #include <linux/etherdevice.h> #include <linux/log2.h> +#include <linux/net_tstamp.h> #include "otx2_common.h" +#include "otx2_ptp.h" #define DRV_NAME "octeontx2-nicpf" #define DRV_VF_NAME "octeontx2-nicvf" @@ -426,6 +428,8 @@ static int otx2_get_rss_hash_opts(struct otx2_nic *pfvf, /* Mimimum is IPv4 and IPv6, SIP/DIP */ nfc->data = RXH_IP_SRC | RXH_IP_DST; + if (rss->flowkey_cfg & NIX_FLOW_KEY_TYPE_VLAN) + nfc->data |= RXH_VLAN; switch (nfc->flow_type) { case TCP_V4_FLOW: @@ -475,6 +479,11 @@ static int otx2_set_rss_hash_opts(struct otx2_nic *pfvf, if (!(nfc->data & RXH_IP_SRC) || !(nfc->data & RXH_IP_DST)) return -EINVAL; + if (nfc->data & RXH_VLAN) + rss_cfg |= NIX_FLOW_KEY_TYPE_VLAN; + else + rss_cfg &= ~NIX_FLOW_KEY_TYPE_VLAN; + switch (nfc->flow_type) { case TCP_V4_FLOW: case TCP_V6_FLOW: @@ -663,6 +672,31 @@ static u32 otx2_get_link(struct net_device *netdev) return pfvf->linfo.link_up; } +static int otx2_get_ts_info(struct net_device *netdev, + struct ethtool_ts_info *info) +{ + struct otx2_nic *pfvf = netdev_priv(netdev); + + if (!pfvf->ptp) + return ethtool_op_get_ts_info(netdev, info); + + info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE | + SOF_TIMESTAMPING_RX_SOFTWARE | + SOF_TIMESTAMPING_SOFTWARE | + SOF_TIMESTAMPING_TX_HARDWARE | + SOF_TIMESTAMPING_RX_HARDWARE | + SOF_TIMESTAMPING_RAW_HARDWARE; + + info->phc_index = otx2_ptp_clock_index(pfvf); + + info->tx_types = (1 << HWTSTAMP_TX_OFF) | (1 << HWTSTAMP_TX_ON); + + info->rx_filters = (1 << HWTSTAMP_FILTER_NONE) | + (1 << HWTSTAMP_FILTER_ALL); + + return 0; +} + static const struct ethtool_ops otx2_ethtool_ops = { .supported_coalesce_params = ETHTOOL_COALESCE_USECS | ETHTOOL_COALESCE_MAX_FRAMES, @@ -687,6 +721,7 @@ static const struct ethtool_ops otx2_ethtool_ops = { .set_msglevel = otx2_set_msglevel, .get_pauseparam = otx2_get_pauseparam, .set_pauseparam = otx2_set_pauseparam, + .get_ts_info = otx2_get_ts_info, }; void otx2_set_ethtool_ops(struct net_device *netdev) diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c index 2fb45670aca4..66f1a212f1f4 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c @@ -21,6 +21,8 @@ #include "otx2_common.h" #include "otx2_txrx.h" #include "otx2_struct.h" +#include "otx2_ptp.h" +#include <rvu_trace.h> #define DRV_NAME "octeontx2-nicpf" #define DRV_STRING "Marvell OcteonTX2 NIC Physical Function Driver" @@ -41,6 +43,9 @@ enum { TYPE_PFVF, }; +static int otx2_config_hw_tx_tstamp(struct otx2_nic *pfvf, bool enable); +static int otx2_config_hw_rx_tstamp(struct otx2_nic *pfvf, bool enable); + static int otx2_change_mtu(struct net_device *netdev, int new_mtu) { bool if_up = netif_running(netdev); @@ -554,6 +559,8 @@ static irqreturn_t otx2_pfvf_mbox_intr_handler(int irq, void *pf_irq) otx2_queue_work(mbox, pf->mbox_pfvf_wq, 0, vfs, intr, TYPE_PFVF); + trace_otx2_msg_interrupt(mbox->mbox.pdev, "VF(s) to PF", intr); + return IRQ_HANDLED; } @@ -937,6 +944,9 @@ static irqreturn_t otx2_pfaf_mbox_intr_handler(int irq, void *pf_irq) otx2_write64(pf, RVU_PF_INT, BIT_ULL(0)); mbox = &pf->mbox; + + trace_otx2_msg_interrupt(mbox->mbox.pdev, "AF to PF", BIT_ULL(0)); + otx2_queue_work(mbox, pf->mbox_wq, 0, 1, 1, TYPE_PFAF); return IRQ_HANDLED; @@ -1282,7 +1292,8 @@ static int otx2_init_hw_resources(struct otx2_nic *pf) hw->pool_cnt = hw->rqpool_cnt + hw->sqpool_cnt; /* Get the size of receive buffers to allocate */ - pf->rbsize = RCV_FRAG_LEN(pf->netdev->mtu + OTX2_ETH_HLEN); + pf->rbsize = RCV_FRAG_LEN(OTX2_HW_TIMESTAMP_LEN + pf->netdev->mtu + + OTX2_ETH_HLEN); mutex_lock(&mbox->lock); /* NPA init */ @@ -1497,6 +1508,9 @@ int otx2_open(struct net_device *netdev) if (err) goto err_disable_napi; + /* Setup segmentation algorithms, if failed, clear offload capability */ + otx2_setup_segmentation(pf); + /* Initialize RSS */ err = otx2_rss_init(pf); if (err) @@ -1548,6 +1562,16 @@ int otx2_open(struct net_device *netdev) otx2_set_cints_affinity(pf); + /* When reinitializing enable time stamping if it is enabled before */ + if (pf->flags & OTX2_FLAG_TX_TSTAMP_ENABLED) { + pf->flags &= ~OTX2_FLAG_TX_TSTAMP_ENABLED; + otx2_config_hw_tx_tstamp(pf, true); + } + if (pf->flags & OTX2_FLAG_RX_TSTAMP_ENABLED) { + pf->flags &= ~OTX2_FLAG_RX_TSTAMP_ENABLED; + otx2_config_hw_rx_tstamp(pf, true); + } + pf->flags &= ~OTX2_FLAG_INTF_DOWN; /* 'intf_down' may be checked on any cpu */ smp_wmb(); @@ -1742,6 +1766,143 @@ static void otx2_reset_task(struct work_struct *work) rtnl_unlock(); } +static int otx2_config_hw_rx_tstamp(struct otx2_nic *pfvf, bool enable) +{ + struct msg_req *req; + int err; + + if (pfvf->flags & OTX2_FLAG_RX_TSTAMP_ENABLED && enable) + return 0; + + mutex_lock(&pfvf->mbox.lock); + if (enable) + req = otx2_mbox_alloc_msg_cgx_ptp_rx_enable(&pfvf->mbox); + else + req = otx2_mbox_alloc_msg_cgx_ptp_rx_disable(&pfvf->mbox); + if (!req) { + mutex_unlock(&pfvf->mbox.lock); + return -ENOMEM; + } + + err = otx2_sync_mbox_msg(&pfvf->mbox); + if (err) { + mutex_unlock(&pfvf->mbox.lock); + return err; + } + + mutex_unlock(&pfvf->mbox.lock); + if (enable) + pfvf->flags |= OTX2_FLAG_RX_TSTAMP_ENABLED; + else + pfvf->flags &= ~OTX2_FLAG_RX_TSTAMP_ENABLED; + return 0; +} + +static int otx2_config_hw_tx_tstamp(struct otx2_nic *pfvf, bool enable) +{ + struct msg_req *req; + int err; + + if (pfvf->flags & OTX2_FLAG_TX_TSTAMP_ENABLED && enable) + return 0; + + mutex_lock(&pfvf->mbox.lock); + if (enable) + req = otx2_mbox_alloc_msg_nix_lf_ptp_tx_enable(&pfvf->mbox); + else + req = otx2_mbox_alloc_msg_nix_lf_ptp_tx_disable(&pfvf->mbox); + if (!req) { + mutex_unlock(&pfvf->mbox.lock); + return -ENOMEM; + } + + err = otx2_sync_mbox_msg(&pfvf->mbox); + if (err) { + mutex_unlock(&pfvf->mbox.lock); + return err; + } + + mutex_unlock(&pfvf->mbox.lock); + if (enable) + pfvf->flags |= OTX2_FLAG_TX_TSTAMP_ENABLED; + else + pfvf->flags &= ~OTX2_FLAG_TX_TSTAMP_ENABLED; + return 0; +} + +static int otx2_config_hwtstamp(struct net_device *netdev, struct ifreq *ifr) +{ + struct otx2_nic *pfvf = netdev_priv(netdev); + struct hwtstamp_config config; + + if (!pfvf->ptp) + return -ENODEV; + + if (copy_from_user(&config, ifr->ifr_data, sizeof(config))) + return -EFAULT; + + /* reserved for future extensions */ + if (config.flags) + return -EINVAL; + + switch (config.tx_type) { + case HWTSTAMP_TX_OFF: + otx2_config_hw_tx_tstamp(pfvf, false); + break; + case HWTSTAMP_TX_ON: + otx2_config_hw_tx_tstamp(pfvf, true); + break; + default: + return -ERANGE; + } + + switch (config.rx_filter) { + case HWTSTAMP_FILTER_NONE: + otx2_config_hw_rx_tstamp(pfvf, false); + break; + case HWTSTAMP_FILTER_ALL: + case HWTSTAMP_FILTER_SOME: + case HWTSTAMP_FILTER_PTP_V1_L4_EVENT: + case HWTSTAMP_FILTER_PTP_V1_L4_SYNC: + case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ: + case HWTSTAMP_FILTER_PTP_V2_L4_EVENT: + case HWTSTAMP_FILTER_PTP_V2_L4_SYNC: + case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ: + case HWTSTAMP_FILTER_PTP_V2_L2_EVENT: + case HWTSTAMP_FILTER_PTP_V2_L2_SYNC: + case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ: + case HWTSTAMP_FILTER_PTP_V2_EVENT: + case HWTSTAMP_FILTER_PTP_V2_SYNC: + case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: + otx2_config_hw_rx_tstamp(pfvf, true); + config.rx_filter = HWTSTAMP_FILTER_ALL; + break; + default: + return -ERANGE; + } + + memcpy(&pfvf->tstamp, &config, sizeof(config)); + + return copy_to_user(ifr->ifr_data, &config, + sizeof(config)) ? -EFAULT : 0; +} + +static int otx2_ioctl(struct net_device *netdev, struct ifreq *req, int cmd) +{ + struct otx2_nic *pfvf = netdev_priv(netdev); + struct hwtstamp_config *cfg = &pfvf->tstamp; + + switch (cmd) { + case SIOCSHWTSTAMP: + return otx2_config_hwtstamp(netdev, req); + case SIOCGHWTSTAMP: + return copy_to_user(req->ifr_data, cfg, + sizeof(*cfg)) ? -EFAULT : 0; + default: + return -EOPNOTSUPP; + } +} + static const struct net_device_ops otx2_netdev_ops = { .ndo_open = otx2_open, .ndo_stop = otx2_stop, @@ -1752,6 +1913,7 @@ static const struct net_device_ops otx2_netdev_ops = { .ndo_set_features = otx2_set_features, .ndo_tx_timeout = otx2_tx_timeout, .ndo_get_stats64 = otx2_get_stats64, + .ndo_do_ioctl = otx2_ioctl, }; static int otx2_wq_init(struct otx2_nic *pf) @@ -1924,6 +2086,9 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id) /* Assign default mac address */ otx2_get_mac_from_af(netdev); + /* Don't check for error. Proceed without ptp */ + otx2_ptp_init(pf); + /* NPA's pool is a stack to which SW frees buffer pointers via Aura. * HW allocates buffer pointer from stack and uses it for DMA'ing * ingress packet. In some scenarios HW can free back allocated buffer @@ -1939,7 +2104,8 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id) netdev->hw_features = (NETIF_F_RXCSUM | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_RXHASH | - NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6); + NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6 | + NETIF_F_GSO_UDP_L4); netdev->features |= netdev->hw_features; netdev->hw_features |= NETIF_F_LOOPBACK | NETIF_F_RXALL; @@ -1956,7 +2122,7 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id) err = register_netdev(netdev); if (err) { dev_err(dev, "Failed to register netdevice\n"); - goto err_detach_rsrc; + goto err_ptp_destroy; } err = otx2_wq_init(pf); @@ -1976,6 +2142,8 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id) err_unreg_netdev: unregister_netdev(netdev); +err_ptp_destroy: + otx2_ptp_destroy(pf); err_detach_rsrc: otx2_detach_resources(&pf->mbox); err_disable_mbox_intr: @@ -2117,6 +2285,11 @@ static void otx2_remove(struct pci_dev *pdev) pf = netdev_priv(netdev); + if (pf->flags & OTX2_FLAG_TX_TSTAMP_ENABLED) + otx2_config_hw_tx_tstamp(pf, false); + if (pf->flags & OTX2_FLAG_RX_TSTAMP_ENABLED) + otx2_config_hw_rx_tstamp(pf, false); + cancel_work_sync(&pf->reset_task); /* Disable link notifications */ otx2_cgx_config_linkevents(pf, false); @@ -2126,6 +2299,7 @@ static void otx2_remove(struct pci_dev *pdev) if (pf->otx2_wq) destroy_workqueue(pf->otx2_wq); + otx2_ptp_destroy(pf); otx2_detach_resources(&pf->mbox); otx2_disable_mbox_intr(pf); otx2_pfaf_mbox_destroy(pf); diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ptp.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ptp.c new file mode 100644 index 000000000000..7bcf5246350f --- /dev/null +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ptp.c @@ -0,0 +1,212 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Marvell OcteonTx2 PTP support for ethernet driver + * + * Copyright (C) 2020 Marvell International Ltd. + */ + +#include "otx2_common.h" +#include "otx2_ptp.h" + +static int otx2_ptp_adjfine(struct ptp_clock_info *ptp_info, long scaled_ppm) +{ + struct otx2_ptp *ptp = container_of(ptp_info, struct otx2_ptp, + ptp_info); + struct ptp_req *req; + int err; + + if (!ptp->nic) + return -ENODEV; + + req = otx2_mbox_alloc_msg_ptp_op(&ptp->nic->mbox); + if (!req) + return -ENOMEM; + + req->op = PTP_OP_ADJFINE; + req->scaled_ppm = scaled_ppm; + + err = otx2_sync_mbox_msg(&ptp->nic->mbox); + if (err) + return err; + + return 0; +} + +static u64 ptp_cc_read(const struct cyclecounter *cc) +{ + struct otx2_ptp *ptp = container_of(cc, struct otx2_ptp, cycle_counter); + struct ptp_req *req; + struct ptp_rsp *rsp; + int err; + + if (!ptp->nic) + return 0; + + req = otx2_mbox_alloc_msg_ptp_op(&ptp->nic->mbox); + if (!req) + return 0; + + req->op = PTP_OP_GET_CLOCK; + + err = otx2_sync_mbox_msg(&ptp->nic->mbox); + if (err) + return 0; + + rsp = (struct ptp_rsp *)otx2_mbox_get_rsp(&ptp->nic->mbox.mbox, 0, + &req->hdr); + if (IS_ERR(rsp)) + return 0; + + return rsp->clk; +} + +static int otx2_ptp_adjtime(struct ptp_clock_info *ptp_info, s64 delta) +{ + struct otx2_ptp *ptp = container_of(ptp_info, struct otx2_ptp, + ptp_info); + struct otx2_nic *pfvf = ptp->nic; + + mutex_lock(&pfvf->mbox.lock); + timecounter_adjtime(&ptp->time_counter, delta); + mutex_unlock(&pfvf->mbox.lock); + + return 0; +} + +static int otx2_ptp_gettime(struct ptp_clock_info *ptp_info, + struct timespec64 *ts) +{ + struct otx2_ptp *ptp = container_of(ptp_info, struct otx2_ptp, + ptp_info); + struct otx2_nic *pfvf = ptp->nic; + u64 nsec; + + mutex_lock(&pfvf->mbox.lock); + nsec = timecounter_read(&ptp->time_counter); + mutex_unlock(&pfvf->mbox.lock); + + *ts = ns_to_timespec64(nsec); + + return 0; +} + +static int otx2_ptp_settime(struct ptp_clock_info *ptp_info, + const struct timespec64 *ts) +{ + struct otx2_ptp *ptp = container_of(ptp_info, struct otx2_ptp, + ptp_info); + struct otx2_nic *pfvf = ptp->nic; + u64 nsec; + + nsec = timespec64_to_ns(ts); + + mutex_lock(&pfvf->mbox.lock); + timecounter_init(&ptp->time_counter, &ptp->cycle_counter, nsec); + mutex_unlock(&pfvf->mbox.lock); + + return 0; +} + +static int otx2_ptp_enable(struct ptp_clock_info *ptp_info, + struct ptp_clock_request *rq, int on) +{ + return -EOPNOTSUPP; +} + +int otx2_ptp_init(struct otx2_nic *pfvf) +{ + struct otx2_ptp *ptp_ptr; + struct cyclecounter *cc; + struct ptp_req *req; + int err; + + mutex_lock(&pfvf->mbox.lock); + /* check if PTP block is available */ + req = otx2_mbox_alloc_msg_ptp_op(&pfvf->mbox); + if (!req) { + mutex_unlock(&pfvf->mbox.lock); + return -ENOMEM; + } + + req->op = PTP_OP_GET_CLOCK; + + err = otx2_sync_mbox_msg(&pfvf->mbox); + if (err) { + mutex_unlock(&pfvf->mbox.lock); + return err; + } + mutex_unlock(&pfvf->mbox.lock); + + ptp_ptr = kzalloc(sizeof(*ptp_ptr), GFP_KERNEL); + if (!ptp_ptr) { + err = -ENOMEM; + goto error; + } + + ptp_ptr->nic = pfvf; + + cc = &ptp_ptr->cycle_counter; + cc->read = ptp_cc_read; + cc->mask = CYCLECOUNTER_MASK(64); + cc->mult = 1; + cc->shift = 0; + + timecounter_init(&ptp_ptr->time_counter, &ptp_ptr->cycle_counter, + ktime_to_ns(ktime_get_real())); + + ptp_ptr->ptp_info = (struct ptp_clock_info) { + .owner = THIS_MODULE, + .name = "OcteonTX2 PTP", + .max_adj = 1000000000ull, + .n_ext_ts = 0, + .n_pins = 0, + .pps = 0, + .adjfine = otx2_ptp_adjfine, + .adjtime = otx2_ptp_adjtime, + .gettime64 = otx2_ptp_gettime, + .settime64 = otx2_ptp_settime, + .enable = otx2_ptp_enable, + }; + + ptp_ptr->ptp_clock = ptp_clock_register(&ptp_ptr->ptp_info, pfvf->dev); + if (IS_ERR_OR_NULL(ptp_ptr->ptp_clock)) { + err = ptp_ptr->ptp_clock ? + PTR_ERR(ptp_ptr->ptp_clock) : -ENODEV; + kfree(ptp_ptr); + goto error; + } + + pfvf->ptp = ptp_ptr; + +error: + return err; +} + +void otx2_ptp_destroy(struct otx2_nic *pfvf) +{ + struct otx2_ptp *ptp = pfvf->ptp; + + if (!ptp) + return; + + ptp_clock_unregister(ptp->ptp_clock); + kfree(ptp); + pfvf->ptp = NULL; +} + +int otx2_ptp_clock_index(struct otx2_nic *pfvf) +{ + if (!pfvf->ptp) + return -ENODEV; + + return ptp_clock_index(pfvf->ptp->ptp_clock); +} + +int otx2_ptp_tstamp2time(struct otx2_nic *pfvf, u64 tstamp, u64 *tsns) +{ + if (!pfvf->ptp) + return -ENODEV; + + *tsns = timecounter_cyc2time(&pfvf->ptp->time_counter, tstamp); + + return 0; +} diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ptp.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ptp.h new file mode 100644 index 000000000000..706d63a43ae1 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ptp.h @@ -0,0 +1,13 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Marvell OcteonTx2 PTP support for ethernet driver */ + +#ifndef OTX2_PTP_H +#define OTX2_PTP_H + +int otx2_ptp_init(struct otx2_nic *pfvf); +void otx2_ptp_destroy(struct otx2_nic *pfvf); + +int otx2_ptp_clock_index(struct otx2_nic *pfvf); +int otx2_ptp_tstamp2time(struct otx2_nic *pfvf, u64 tstamp, u64 *tsns); + +#endif diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c index e46834e043be..d5d7a2f37493 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c @@ -16,6 +16,7 @@ #include "otx2_common.h" #include "otx2_struct.h" #include "otx2_txrx.h" +#include "otx2_ptp.h" #define CQE_ADDR(CQ, idx) ((CQ)->cqe_base + ((CQ)->cqe_size * (idx))) @@ -81,8 +82,11 @@ static void otx2_snd_pkt_handler(struct otx2_nic *pfvf, int budget, int *tx_pkts, int *tx_bytes) { struct nix_send_comp_s *snd_comp = &cqe->comp; + struct skb_shared_hwtstamps ts; struct sk_buff *skb = NULL; + u64 timestamp, tsns; struct sg_list *sg; + int err; if (unlikely(snd_comp->status) && netif_msg_tx_err(pfvf)) net_err_ratelimited("%s: TX%d: Error in send CQ status:%x\n", @@ -94,6 +98,18 @@ static void otx2_snd_pkt_handler(struct otx2_nic *pfvf, if (unlikely(!skb)) return; + if (skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS) { + timestamp = ((u64 *)sq->timestamps->base)[snd_comp->sqe_id]; + if (timestamp != 1) { + err = otx2_ptp_tstamp2time(pfvf, timestamp, &tsns); + if (!err) { + memset(&ts, 0, sizeof(ts)); + ts.hwtstamp = ns_to_ktime(tsns); + skb_tstamp_tx(skb, &ts); + } + } + } + *tx_bytes += skb->len; (*tx_pkts)++; otx2_dma_unmap_skb_frags(pfvf, sg); @@ -101,16 +117,47 @@ static void otx2_snd_pkt_handler(struct otx2_nic *pfvf, sg->skb = (u64)NULL; } +static void otx2_set_rxtstamp(struct otx2_nic *pfvf, + struct sk_buff *skb, void *data) +{ + u64 tsns; + int err; + + if (!(pfvf->flags & OTX2_FLAG_RX_TSTAMP_ENABLED)) + return; + + /* The first 8 bytes is the timestamp */ + err = otx2_ptp_tstamp2time(pfvf, be64_to_cpu(*(__be64 *)data), &tsns); + if (err) + return; + + skb_hwtstamps(skb)->hwtstamp = ns_to_ktime(tsns); +} + static void otx2_skb_add_frag(struct otx2_nic *pfvf, struct sk_buff *skb, - u64 iova, int len) + u64 iova, int len, struct nix_rx_parse_s *parse) { struct page *page; + int off = 0; void *va; va = phys_to_virt(otx2_iova_to_phys(pfvf->iommu_domain, iova)); + + if (likely(!skb_shinfo(skb)->nr_frags)) { + /* Check if data starts at some nonzero offset + * from the start of the buffer. For now the + * only possible offset is 8 bytes in the case + * where packet is prepended by a timestamp. + */ + if (parse->laptr) { + otx2_set_rxtstamp(pfvf, skb, va); + off = OTX2_HW_TIMESTAMP_LEN; + } + } + page = virt_to_page(va); skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page, - va - page_address(page), len, pfvf->rbsize); + va - page_address(page) + off, len - off, pfvf->rbsize); otx2_dma_unmap_page(pfvf, iova - OTX2_HEAD_ROOM, pfvf->rbsize, DMA_FROM_DEVICE); @@ -239,7 +286,7 @@ static void otx2_rcv_pkt_handler(struct otx2_nic *pfvf, if (unlikely(!skb)) return; - otx2_skb_add_frag(pfvf, skb, cqe->sg.seg_addr, cqe->sg.seg_size); + otx2_skb_add_frag(pfvf, skb, cqe->sg.seg_addr, cqe->sg.seg_size, parse); cq->pool_ptrs++; otx2_set_rxhash(pfvf, cqe, skb); @@ -477,15 +524,55 @@ static void otx2_sqe_add_ext(struct otx2_nic *pfvf, struct otx2_snd_queue *sq, */ ip_hdr(skb)->tot_len = htons(ext->lso_sb - skb_network_offset(skb)); - } else { + } else if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6) { ext->lso_format = pfvf->hw.lso_tsov6_idx; + ipv6_hdr(skb)->payload_len = htons(ext->lso_sb - skb_network_offset(skb)); + } else if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) { + __be16 l3_proto = vlan_get_protocol(skb); + struct udphdr *udph = udp_hdr(skb); + u16 iplen; + + ext->lso_sb = skb_transport_offset(skb) + + sizeof(struct udphdr); + + /* HW adds payload size to length fields in IP and + * UDP headers while segmentation, hence adjust the + * lengths to just header sizes. + */ + iplen = htons(ext->lso_sb - skb_network_offset(skb)); + if (l3_proto == htons(ETH_P_IP)) { + ip_hdr(skb)->tot_len = iplen; + ext->lso_format = pfvf->hw.lso_udpv4_idx; + } else { + ipv6_hdr(skb)->payload_len = iplen; + ext->lso_format = pfvf->hw.lso_udpv6_idx; + } + + udph->len = htons(sizeof(struct udphdr)); } + } else if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) { + ext->tstmp = 1; } + *offset += sizeof(*ext); } +static void otx2_sqe_add_mem(struct otx2_snd_queue *sq, int *offset, + int alg, u64 iova) +{ + struct nix_sqe_mem_s *mem; + + mem = (struct nix_sqe_mem_s *)(sq->sqe_base + *offset); + mem->subdc = NIX_SUBDC_MEM; + mem->alg = alg; + mem->wmem = 1; /* wait for the memory operation */ + mem->addr = iova; + + *offset += sizeof(*mem); +} + /* Add SQE header subdescriptor structure */ static void otx2_sqe_add_hdr(struct otx2_nic *pfvf, struct otx2_snd_queue *sq, struct nix_sqe_hdr_s *sqe_hdr, @@ -737,6 +824,21 @@ static int otx2_get_sqe_count(struct otx2_nic *pfvf, struct sk_buff *skb) return skb_shinfo(skb)->gso_segs; } +static void otx2_set_txtstamp(struct otx2_nic *pfvf, struct sk_buff *skb, + struct otx2_snd_queue *sq, int *offset) +{ + u64 iova; + + if (!skb_shinfo(skb)->gso_size && + skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) { + skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; + iova = sq->timestamps->iova + (sq->head * sizeof(u64)); + otx2_sqe_add_mem(sq, offset, NIX_SENDMEMALG_E_SETTSTMP, iova); + } else { + skb_tx_timestamp(skb); + } +} + bool otx2_sq_append_skb(struct net_device *netdev, struct otx2_snd_queue *sq, struct sk_buff *skb, u16 qidx) { @@ -790,6 +892,8 @@ bool otx2_sq_append_skb(struct net_device *netdev, struct otx2_snd_queue *sq, return false; } + otx2_set_txtstamp(pfvf, skb, sq, &offset); + sqe_hdr->sizem1 = (offset / 16) - 1; netdev_tx_sent_queue(txq, skb->len); diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h index da97f2d4416f..73af15685657 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h @@ -91,6 +91,7 @@ struct otx2_snd_queue { struct qmem *sqe; struct qmem *tso_hdrs; struct sg_list *sg; + struct qmem *timestamps; struct queue_stats stats; u16 sqb_count; u64 *sqb_ptrs; diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c index 2f90f1721441..67fabf265fe6 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c @@ -187,6 +187,8 @@ static irqreturn_t otx2vf_vfaf_mbox_intr_handler(int irq, void *vf_irq) mdev = &mbox->dev[0]; otx2_sync_mbox_bbuf(mbox, 0); + trace_otx2_msg_interrupt(mbox->pdev, "PF to VF", BIT_ULL(0)); + hdr = (struct mbox_hdr *)(mdev->mbase + mbox->rx_start); if (hdr->num_msgs) { vf->mbox.num_msgs = hdr->num_msgs; @@ -553,7 +555,8 @@ static int otx2vf_probe(struct pci_dev *pdev, const struct pci_device_id *id) netdev->hw_features = NETIF_F_RXCSUM | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_RXHASH | - NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6; + NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6 | + NETIF_F_GSO_UDP_L4; netdev->features = netdev->hw_features; netdev->gso_max_segs = OTX2_MAX_GSO_SEGS; diff --git a/drivers/net/ethernet/marvell/prestera/Kconfig b/drivers/net/ethernet/marvell/prestera/Kconfig new file mode 100644 index 000000000000..b1fcc44f566a --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/Kconfig @@ -0,0 +1,25 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +# Marvell Prestera drivers configuration +# + +config PRESTERA + tristate "Marvell Prestera Switch ASICs support" + depends on NET_SWITCHDEV && VLAN_8021Q + select NET_DEVLINK + help + This driver supports Marvell Prestera Switch ASICs family. + + To compile this driver as a module, choose M here: the + module will be called prestera. + +config PRESTERA_PCI + tristate "PCI interface driver for Marvell Prestera Switch ASICs family" + depends on PCI && HAS_IOMEM && PRESTERA + default PRESTERA + help + This is implementation of PCI interface support for Marvell Prestera + Switch ASICs family. + + To compile this driver as a module, choose M here: the + module will be called prestera_pci. diff --git a/drivers/net/ethernet/marvell/prestera/Makefile b/drivers/net/ethernet/marvell/prestera/Makefile new file mode 100644 index 000000000000..93129e32ebc5 --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/Makefile @@ -0,0 +1,7 @@ +# SPDX-License-Identifier: GPL-2.0 +obj-$(CONFIG_PRESTERA) += prestera.o +prestera-objs := prestera_main.o prestera_hw.o prestera_dsa.o \ + prestera_rxtx.o prestera_devlink.o prestera_ethtool.o \ + prestera_switchdev.o + +obj-$(CONFIG_PRESTERA_PCI) += prestera_pci.o diff --git a/drivers/net/ethernet/marvell/prestera/prestera.h b/drivers/net/ethernet/marvell/prestera/prestera.h new file mode 100644 index 000000000000..55aa4bf8a27c --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/prestera.h @@ -0,0 +1,206 @@ +/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 */ +/* Copyright (c) 2019-2020 Marvell International Ltd. All rights reserved. */ + +#ifndef _PRESTERA_H_ +#define _PRESTERA_H_ + +#include <linux/notifier.h> +#include <linux/skbuff.h> +#include <linux/workqueue.h> +#include <net/devlink.h> +#include <uapi/linux/if_ether.h> + +#define PRESTERA_DRV_NAME "prestera" + +#define PRESTERA_DEFAULT_VID 1 + +struct prestera_fw_rev { + u16 maj; + u16 min; + u16 sub; +}; + +struct prestera_port_stats { + u64 good_octets_received; + u64 bad_octets_received; + u64 mac_trans_error; + u64 broadcast_frames_received; + u64 multicast_frames_received; + u64 frames_64_octets; + u64 frames_65_to_127_octets; + u64 frames_128_to_255_octets; + u64 frames_256_to_511_octets; + u64 frames_512_to_1023_octets; + u64 frames_1024_to_max_octets; + u64 excessive_collision; + u64 multicast_frames_sent; + u64 broadcast_frames_sent; + u64 fc_sent; + u64 fc_received; + u64 buffer_overrun; + u64 undersize; + u64 fragments; + u64 oversize; + u64 jabber; + u64 rx_error_frame_received; + u64 bad_crc; + u64 collisions; + u64 late_collision; + u64 unicast_frames_received; + u64 unicast_frames_sent; + u64 sent_multiple; + u64 sent_deferred; + u64 good_octets_sent; +}; + +struct prestera_port_caps { + u64 supp_link_modes; + u8 supp_fec; + u8 type; + u8 transceiver; +}; + +struct prestera_port { + struct net_device *dev; + struct prestera_switch *sw; + struct devlink_port dl_port; + u32 id; + u32 hw_id; + u32 dev_id; + u16 fp_id; + u16 pvid; + bool autoneg; + u64 adver_link_modes; + u8 adver_fec; + struct prestera_port_caps caps; + struct list_head list; + struct list_head vlans_list; + struct { + struct prestera_port_stats stats; + struct delayed_work caching_dw; + } cached_hw_stats; +}; + +struct prestera_device { + struct device *dev; + u8 __iomem *ctl_regs; + u8 __iomem *pp_regs; + struct prestera_fw_rev fw_rev; + void *priv; + + /* called by device driver to handle received packets */ + void (*recv_pkt)(struct prestera_device *dev); + + /* called by device driver to pass event up to the higher layer */ + int (*recv_msg)(struct prestera_device *dev, void *msg, size_t size); + + /* called by higher layer to send request to the firmware */ + int (*send_req)(struct prestera_device *dev, void *in_msg, + size_t in_size, void *out_msg, size_t out_size, + unsigned int wait); +}; + +enum prestera_event_type { + PRESTERA_EVENT_TYPE_UNSPEC, + + PRESTERA_EVENT_TYPE_PORT, + PRESTERA_EVENT_TYPE_FDB, + PRESTERA_EVENT_TYPE_RXTX, + + PRESTERA_EVENT_TYPE_MAX +}; + +enum prestera_rxtx_event_id { + PRESTERA_RXTX_EVENT_UNSPEC, + PRESTERA_RXTX_EVENT_RCV_PKT, +}; + +enum prestera_port_event_id { + PRESTERA_PORT_EVENT_UNSPEC, + PRESTERA_PORT_EVENT_STATE_CHANGED, +}; + +struct prestera_port_event { + u32 port_id; + union { + u32 oper_state; + } data; +}; + +enum prestera_fdb_event_id { + PRESTERA_FDB_EVENT_UNSPEC, + PRESTERA_FDB_EVENT_LEARNED, + PRESTERA_FDB_EVENT_AGED, +}; + +struct prestera_fdb_event { + u32 port_id; + u32 vid; + union { + u8 mac[ETH_ALEN]; + } data; +}; + +struct prestera_event { + u16 id; + union { + struct prestera_port_event port_evt; + struct prestera_fdb_event fdb_evt; + }; +}; + +struct prestera_switchdev; +struct prestera_rxtx; + +struct prestera_switch { + struct prestera_device *dev; + struct prestera_switchdev *swdev; + struct prestera_rxtx *rxtx; + struct list_head event_handlers; + struct notifier_block netdev_nb; + char base_mac[ETH_ALEN]; + struct list_head port_list; + rwlock_t port_list_lock; + u32 port_count; + u32 mtu_min; + u32 mtu_max; + u8 id; +}; + +struct prestera_rxtx_params { + bool use_sdma; + u32 map_addr; +}; + +#define prestera_dev(sw) ((sw)->dev->dev) + +static inline void prestera_write(const struct prestera_switch *sw, + unsigned int reg, u32 val) +{ + writel(val, sw->dev->pp_regs + reg); +} + +static inline u32 prestera_read(const struct prestera_switch *sw, + unsigned int reg) +{ + return readl(sw->dev->pp_regs + reg); +} + +int prestera_device_register(struct prestera_device *dev); +void prestera_device_unregister(struct prestera_device *dev); + +struct prestera_port *prestera_port_find_by_hwid(struct prestera_switch *sw, + u32 dev_id, u32 hw_id); + +int prestera_port_autoneg_set(struct prestera_port *port, bool enable, + u64 adver_link_modes, u8 adver_fec); + +struct prestera_port *prestera_find_port(struct prestera_switch *sw, u32 id); + +struct prestera_port *prestera_port_dev_lower_find(struct net_device *dev); + +int prestera_port_pvid_set(struct prestera_port *port, u16 vid); + +bool prestera_netdev_check(const struct net_device *dev); + +#endif /* _PRESTERA_H_ */ diff --git a/drivers/net/ethernet/marvell/prestera/prestera_devlink.c b/drivers/net/ethernet/marvell/prestera/prestera_devlink.c new file mode 100644 index 000000000000..94c185a0e2b8 --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/prestera_devlink.c @@ -0,0 +1,112 @@ +// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 +/* Copyright (c) 2019-2020 Marvell International Ltd. All rights reserved */ + +#include <net/devlink.h> + +#include "prestera_devlink.h" + +static int prestera_dl_info_get(struct devlink *dl, + struct devlink_info_req *req, + struct netlink_ext_ack *extack) +{ + struct prestera_switch *sw = devlink_priv(dl); + char buf[16]; + int err; + + err = devlink_info_driver_name_put(req, PRESTERA_DRV_NAME); + if (err) + return err; + + snprintf(buf, sizeof(buf), "%d.%d.%d", + sw->dev->fw_rev.maj, + sw->dev->fw_rev.min, + sw->dev->fw_rev.sub); + + return devlink_info_version_running_put(req, + DEVLINK_INFO_VERSION_GENERIC_FW, + buf); +} + +static const struct devlink_ops prestera_dl_ops = { + .info_get = prestera_dl_info_get, +}; + +struct prestera_switch *prestera_devlink_alloc(void) +{ + struct devlink *dl; + + dl = devlink_alloc(&prestera_dl_ops, sizeof(struct prestera_switch)); + + return devlink_priv(dl); +} + +void prestera_devlink_free(struct prestera_switch *sw) +{ + struct devlink *dl = priv_to_devlink(sw); + + devlink_free(dl); +} + +int prestera_devlink_register(struct prestera_switch *sw) +{ + struct devlink *dl = priv_to_devlink(sw); + int err; + + err = devlink_register(dl, sw->dev->dev); + if (err) + dev_err(prestera_dev(sw), "devlink_register failed: %d\n", err); + + return err; +} + +void prestera_devlink_unregister(struct prestera_switch *sw) +{ + struct devlink *dl = priv_to_devlink(sw); + + devlink_unregister(dl); +} + +int prestera_devlink_port_register(struct prestera_port *port) +{ + struct prestera_switch *sw = port->sw; + struct devlink *dl = priv_to_devlink(sw); + struct devlink_port_attrs attrs = {}; + int err; + + attrs.flavour = DEVLINK_PORT_FLAVOUR_PHYSICAL; + attrs.phys.port_number = port->fp_id; + attrs.switch_id.id_len = sizeof(sw->id); + memcpy(attrs.switch_id.id, &sw->id, attrs.switch_id.id_len); + + devlink_port_attrs_set(&port->dl_port, &attrs); + + err = devlink_port_register(dl, &port->dl_port, port->fp_id); + if (err) { + dev_err(prestera_dev(sw), "devlink_port_register failed: %d\n", err); + return err; + } + + return 0; +} + +void prestera_devlink_port_unregister(struct prestera_port *port) +{ + devlink_port_unregister(&port->dl_port); +} + +void prestera_devlink_port_set(struct prestera_port *port) +{ + devlink_port_type_eth_set(&port->dl_port, port->dev); +} + +void prestera_devlink_port_clear(struct prestera_port *port) +{ + devlink_port_type_clear(&port->dl_port); +} + +struct devlink_port *prestera_devlink_get_port(struct net_device *dev) +{ + struct prestera_port *port = netdev_priv(dev); + + return &port->dl_port; +} diff --git a/drivers/net/ethernet/marvell/prestera/prestera_devlink.h b/drivers/net/ethernet/marvell/prestera/prestera_devlink.h new file mode 100644 index 000000000000..51bee9f75415 --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/prestera_devlink.h @@ -0,0 +1,23 @@ +/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 */ +/* Copyright (c) 2019-2020 Marvell International Ltd. All rights reserved. */ + +#ifndef _PRESTERA_DEVLINK_H_ +#define _PRESTERA_DEVLINK_H_ + +#include "prestera.h" + +struct prestera_switch *prestera_devlink_alloc(void); +void prestera_devlink_free(struct prestera_switch *sw); + +int prestera_devlink_register(struct prestera_switch *sw); +void prestera_devlink_unregister(struct prestera_switch *sw); + +int prestera_devlink_port_register(struct prestera_port *port); +void prestera_devlink_port_unregister(struct prestera_port *port); + +void prestera_devlink_port_set(struct prestera_port *port); +void prestera_devlink_port_clear(struct prestera_port *port); + +struct devlink_port *prestera_devlink_get_port(struct net_device *dev); + +#endif /* _PRESTERA_DEVLINK_H_ */ diff --git a/drivers/net/ethernet/marvell/prestera/prestera_dsa.c b/drivers/net/ethernet/marvell/prestera/prestera_dsa.c new file mode 100644 index 000000000000..a5e01c7a307b --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/prestera_dsa.c @@ -0,0 +1,104 @@ +// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 +/* Copyright (c) 2020 Marvell International Ltd. All rights reserved */ + +#include <linux/bitfield.h> +#include <linux/bitops.h> +#include <linux/errno.h> +#include <linux/string.h> + +#include "prestera_dsa.h" + +#define PRESTERA_DSA_W0_CMD GENMASK(31, 30) +#define PRESTERA_DSA_W0_IS_TAGGED BIT(29) +#define PRESTERA_DSA_W0_DEV_NUM GENMASK(28, 24) +#define PRESTERA_DSA_W0_PORT_NUM GENMASK(23, 19) +#define PRESTERA_DSA_W0_VPT GENMASK(15, 13) +#define PRESTERA_DSA_W0_EXT_BIT BIT(12) +#define PRESTERA_DSA_W0_VID GENMASK(11, 0) + +#define PRESTERA_DSA_W1_EXT_BIT BIT(31) +#define PRESTERA_DSA_W1_CFI_BIT BIT(30) +#define PRESTERA_DSA_W1_PORT_NUM GENMASK(11, 10) + +#define PRESTERA_DSA_W2_EXT_BIT BIT(31) +#define PRESTERA_DSA_W2_PORT_NUM BIT(20) + +#define PRESTERA_DSA_W3_VID GENMASK(30, 27) +#define PRESTERA_DSA_W3_DST_EPORT GENMASK(23, 7) +#define PRESTERA_DSA_W3_DEV_NUM GENMASK(6, 0) + +#define PRESTERA_DSA_VID GENMASK(15, 12) +#define PRESTERA_DSA_DEV_NUM GENMASK(11, 5) + +int prestera_dsa_parse(struct prestera_dsa *dsa, const u8 *dsa_buf) +{ + __be32 *dsa_words = (__be32 *)dsa_buf; + enum prestera_dsa_cmd cmd; + u32 words[4]; + u32 field; + + words[0] = ntohl(dsa_words[0]); + words[1] = ntohl(dsa_words[1]); + words[2] = ntohl(dsa_words[2]); + words[3] = ntohl(dsa_words[3]); + + /* set the common parameters */ + cmd = (enum prestera_dsa_cmd)FIELD_GET(PRESTERA_DSA_W0_CMD, words[0]); + + /* only to CPU is supported */ + if (unlikely(cmd != PRESTERA_DSA_CMD_TO_CPU)) + return -EINVAL; + + if (FIELD_GET(PRESTERA_DSA_W0_EXT_BIT, words[0]) == 0) + return -EINVAL; + if (FIELD_GET(PRESTERA_DSA_W1_EXT_BIT, words[1]) == 0) + return -EINVAL; + if (FIELD_GET(PRESTERA_DSA_W2_EXT_BIT, words[2]) == 0) + return -EINVAL; + + field = FIELD_GET(PRESTERA_DSA_W3_VID, words[3]); + + dsa->vlan.is_tagged = FIELD_GET(PRESTERA_DSA_W0_IS_TAGGED, words[0]); + dsa->vlan.cfi_bit = FIELD_GET(PRESTERA_DSA_W1_CFI_BIT, words[1]); + dsa->vlan.vpt = FIELD_GET(PRESTERA_DSA_W0_VPT, words[0]); + dsa->vlan.vid = FIELD_GET(PRESTERA_DSA_W0_VID, words[0]); + dsa->vlan.vid &= ~PRESTERA_DSA_VID; + dsa->vlan.vid |= FIELD_PREP(PRESTERA_DSA_VID, field); + + field = FIELD_GET(PRESTERA_DSA_W3_DEV_NUM, words[3]); + + dsa->hw_dev_num = FIELD_GET(PRESTERA_DSA_W0_DEV_NUM, words[0]); + dsa->hw_dev_num |= FIELD_PREP(PRESTERA_DSA_DEV_NUM, field); + + dsa->port_num = (FIELD_GET(PRESTERA_DSA_W0_PORT_NUM, words[0]) << 0) | + (FIELD_GET(PRESTERA_DSA_W1_PORT_NUM, words[1]) << 5) | + (FIELD_GET(PRESTERA_DSA_W2_PORT_NUM, words[2]) << 7); + + return 0; +} + +int prestera_dsa_build(const struct prestera_dsa *dsa, u8 *dsa_buf) +{ + __be32 *dsa_words = (__be32 *)dsa_buf; + u32 dev_num = dsa->hw_dev_num; + u32 words[4] = { 0 }; + + words[0] |= FIELD_PREP(PRESTERA_DSA_W0_CMD, PRESTERA_DSA_CMD_FROM_CPU); + + words[0] |= FIELD_PREP(PRESTERA_DSA_W0_DEV_NUM, dev_num); + dev_num = FIELD_GET(PRESTERA_DSA_DEV_NUM, dev_num); + words[3] |= FIELD_PREP(PRESTERA_DSA_W3_DEV_NUM, dev_num); + + words[3] |= FIELD_PREP(PRESTERA_DSA_W3_DST_EPORT, dsa->port_num); + + words[0] |= FIELD_PREP(PRESTERA_DSA_W0_EXT_BIT, 1); + words[1] |= FIELD_PREP(PRESTERA_DSA_W1_EXT_BIT, 1); + words[2] |= FIELD_PREP(PRESTERA_DSA_W2_EXT_BIT, 1); + + dsa_words[0] = htonl(words[0]); + dsa_words[1] = htonl(words[1]); + dsa_words[2] = htonl(words[2]); + dsa_words[3] = htonl(words[3]); + + return 0; +} diff --git a/drivers/net/ethernet/marvell/prestera/prestera_dsa.h b/drivers/net/ethernet/marvell/prestera/prestera_dsa.h new file mode 100644 index 000000000000..67018629bdd2 --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/prestera_dsa.h @@ -0,0 +1,35 @@ +/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 */ +/* Copyright (c) 2020 Marvell International Ltd. All rights reserved. */ + +#ifndef __PRESTERA_DSA_H_ +#define __PRESTERA_DSA_H_ + +#include <linux/types.h> + +#define PRESTERA_DSA_HLEN 16 + +enum prestera_dsa_cmd { + /* DSA command is "To CPU" */ + PRESTERA_DSA_CMD_TO_CPU = 0, + + /* DSA command is "From CPU" */ + PRESTERA_DSA_CMD_FROM_CPU, +}; + +struct prestera_dsa_vlan { + u16 vid; + u8 vpt; + u8 cfi_bit; + bool is_tagged; +}; + +struct prestera_dsa { + struct prestera_dsa_vlan vlan; + u32 hw_dev_num; + u32 port_num; +}; + +int prestera_dsa_parse(struct prestera_dsa *dsa, const u8 *dsa_buf); +int prestera_dsa_build(const struct prestera_dsa *dsa, u8 *dsa_buf); + +#endif /* _PRESTERA_DSA_H_ */ diff --git a/drivers/net/ethernet/marvell/prestera/prestera_ethtool.c b/drivers/net/ethernet/marvell/prestera/prestera_ethtool.c new file mode 100644 index 000000000000..93a5e2baf808 --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/prestera_ethtool.c @@ -0,0 +1,780 @@ +// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 +/* Copyright (c) 2019-2020 Marvell International Ltd. All rights reserved */ + +#include <linux/ethtool.h> +#include <linux/kernel.h> +#include <linux/netdevice.h> + +#include "prestera_ethtool.h" +#include "prestera.h" +#include "prestera_hw.h" + +#define PRESTERA_STATS_CNT \ + (sizeof(struct prestera_port_stats) / sizeof(u64)) +#define PRESTERA_STATS_IDX(name) \ + (offsetof(struct prestera_port_stats, name) / sizeof(u64)) +#define PRESTERA_STATS_FIELD(name) \ + [PRESTERA_STATS_IDX(name)] = __stringify(name) + +static const char driver_kind[] = "prestera"; + +static const struct prestera_link_mode { + enum ethtool_link_mode_bit_indices eth_mode; + u32 speed; + u64 pr_mask; + u8 duplex; + u8 port_type; +} port_link_modes[PRESTERA_LINK_MODE_MAX] = { + [PRESTERA_LINK_MODE_10baseT_Half] = { + .eth_mode = ETHTOOL_LINK_MODE_10baseT_Half_BIT, + .speed = 10, + .pr_mask = 1 << PRESTERA_LINK_MODE_10baseT_Half, + .duplex = PRESTERA_PORT_DUPLEX_HALF, + .port_type = PRESTERA_PORT_TYPE_TP, + }, + [PRESTERA_LINK_MODE_10baseT_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_10baseT_Full_BIT, + .speed = 10, + .pr_mask = 1 << PRESTERA_LINK_MODE_10baseT_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_TP, + }, + [PRESTERA_LINK_MODE_100baseT_Half] = { + .eth_mode = ETHTOOL_LINK_MODE_100baseT_Half_BIT, + .speed = 100, + .pr_mask = 1 << PRESTERA_LINK_MODE_100baseT_Half, + .duplex = PRESTERA_PORT_DUPLEX_HALF, + .port_type = PRESTERA_PORT_TYPE_TP, + }, + [PRESTERA_LINK_MODE_100baseT_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_100baseT_Full_BIT, + .speed = 100, + .pr_mask = 1 << PRESTERA_LINK_MODE_100baseT_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_TP, + }, + [PRESTERA_LINK_MODE_1000baseT_Half] = { + .eth_mode = ETHTOOL_LINK_MODE_1000baseT_Half_BIT, + .speed = 1000, + .pr_mask = 1 << PRESTERA_LINK_MODE_1000baseT_Half, + .duplex = PRESTERA_PORT_DUPLEX_HALF, + .port_type = PRESTERA_PORT_TYPE_TP, + }, + [PRESTERA_LINK_MODE_1000baseT_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_1000baseT_Full_BIT, + .speed = 1000, + .pr_mask = 1 << PRESTERA_LINK_MODE_1000baseT_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_TP, + }, + [PRESTERA_LINK_MODE_1000baseX_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_1000baseX_Full_BIT, + .speed = 1000, + .pr_mask = 1 << PRESTERA_LINK_MODE_1000baseX_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_FIBRE, + }, + [PRESTERA_LINK_MODE_1000baseKX_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_1000baseKX_Full_BIT, + .speed = 1000, + .pr_mask = 1 << PRESTERA_LINK_MODE_1000baseKX_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_TP, + }, + [PRESTERA_LINK_MODE_2500baseX_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_2500baseX_Full_BIT, + .speed = 2500, + .pr_mask = 1 << PRESTERA_LINK_MODE_2500baseX_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + }, + [PRESTERA_LINK_MODE_10GbaseKR_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_10000baseKR_Full_BIT, + .speed = 10000, + .pr_mask = 1 << PRESTERA_LINK_MODE_10GbaseKR_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_TP, + }, + [PRESTERA_LINK_MODE_10GbaseSR_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_10000baseSR_Full_BIT, + .speed = 10000, + .pr_mask = 1 << PRESTERA_LINK_MODE_10GbaseSR_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_FIBRE, + }, + [PRESTERA_LINK_MODE_10GbaseLR_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_10000baseLR_Full_BIT, + .speed = 10000, + .pr_mask = 1 << PRESTERA_LINK_MODE_10GbaseLR_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_FIBRE, + }, + [PRESTERA_LINK_MODE_20GbaseKR2_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_20000baseKR2_Full_BIT, + .speed = 20000, + .pr_mask = 1 << PRESTERA_LINK_MODE_20GbaseKR2_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_TP, + }, + [PRESTERA_LINK_MODE_25GbaseCR_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_25000baseCR_Full_BIT, + .speed = 25000, + .pr_mask = 1 << PRESTERA_LINK_MODE_25GbaseCR_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_DA, + }, + [PRESTERA_LINK_MODE_25GbaseKR_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_25000baseKR_Full_BIT, + .speed = 25000, + .pr_mask = 1 << PRESTERA_LINK_MODE_25GbaseKR_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_TP, + }, + [PRESTERA_LINK_MODE_25GbaseSR_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_25000baseSR_Full_BIT, + .speed = 25000, + .pr_mask = 1 << PRESTERA_LINK_MODE_25GbaseSR_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_FIBRE, + }, + [PRESTERA_LINK_MODE_40GbaseKR4_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_40000baseKR4_Full_BIT, + .speed = 40000, + .pr_mask = 1 << PRESTERA_LINK_MODE_40GbaseKR4_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_TP, + }, + [PRESTERA_LINK_MODE_40GbaseCR4_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_40000baseCR4_Full_BIT, + .speed = 40000, + .pr_mask = 1 << PRESTERA_LINK_MODE_40GbaseCR4_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_DA, + }, + [PRESTERA_LINK_MODE_40GbaseSR4_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_40000baseSR4_Full_BIT, + .speed = 40000, + .pr_mask = 1 << PRESTERA_LINK_MODE_40GbaseSR4_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_FIBRE, + }, + [PRESTERA_LINK_MODE_50GbaseCR2_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_50000baseCR2_Full_BIT, + .speed = 50000, + .pr_mask = 1 << PRESTERA_LINK_MODE_50GbaseCR2_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_DA, + }, + [PRESTERA_LINK_MODE_50GbaseKR2_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_50000baseKR2_Full_BIT, + .speed = 50000, + .pr_mask = 1 << PRESTERA_LINK_MODE_50GbaseKR2_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_TP, + }, + [PRESTERA_LINK_MODE_50GbaseSR2_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_50000baseSR2_Full_BIT, + .speed = 50000, + .pr_mask = 1 << PRESTERA_LINK_MODE_50GbaseSR2_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_FIBRE, + }, + [PRESTERA_LINK_MODE_100GbaseKR4_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_100000baseKR4_Full_BIT, + .speed = 100000, + .pr_mask = 1 << PRESTERA_LINK_MODE_100GbaseKR4_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_TP, + }, + [PRESTERA_LINK_MODE_100GbaseSR4_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_100000baseSR4_Full_BIT, + .speed = 100000, + .pr_mask = 1 << PRESTERA_LINK_MODE_100GbaseSR4_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_FIBRE, + }, + [PRESTERA_LINK_MODE_100GbaseCR4_Full] = { + .eth_mode = ETHTOOL_LINK_MODE_100000baseCR4_Full_BIT, + .speed = 100000, + .pr_mask = 1 << PRESTERA_LINK_MODE_100GbaseCR4_Full, + .duplex = PRESTERA_PORT_DUPLEX_FULL, + .port_type = PRESTERA_PORT_TYPE_DA, + } +}; + +static const struct prestera_fec { + u32 eth_fec; + enum ethtool_link_mode_bit_indices eth_mode; + u8 pr_fec; +} port_fec_caps[PRESTERA_PORT_FEC_MAX] = { + [PRESTERA_PORT_FEC_OFF] = { + .eth_fec = ETHTOOL_FEC_OFF, + .eth_mode = ETHTOOL_LINK_MODE_FEC_NONE_BIT, + .pr_fec = 1 << PRESTERA_PORT_FEC_OFF, + }, + [PRESTERA_PORT_FEC_BASER] = { + .eth_fec = ETHTOOL_FEC_BASER, + .eth_mode = ETHTOOL_LINK_MODE_FEC_BASER_BIT, + .pr_fec = 1 << PRESTERA_PORT_FEC_BASER, + }, + [PRESTERA_PORT_FEC_RS] = { + .eth_fec = ETHTOOL_FEC_RS, + .eth_mode = ETHTOOL_LINK_MODE_FEC_RS_BIT, + .pr_fec = 1 << PRESTERA_PORT_FEC_RS, + } +}; + +static const struct prestera_port_type { + enum ethtool_link_mode_bit_indices eth_mode; + u8 eth_type; +} port_types[PRESTERA_PORT_TYPE_MAX] = { + [PRESTERA_PORT_TYPE_NONE] = { + .eth_mode = __ETHTOOL_LINK_MODE_MASK_NBITS, + .eth_type = PORT_NONE, + }, + [PRESTERA_PORT_TYPE_TP] = { + .eth_mode = ETHTOOL_LINK_MODE_TP_BIT, + .eth_type = PORT_TP, + }, + [PRESTERA_PORT_TYPE_AUI] = { + .eth_mode = ETHTOOL_LINK_MODE_AUI_BIT, + .eth_type = PORT_AUI, + }, + [PRESTERA_PORT_TYPE_MII] = { + .eth_mode = ETHTOOL_LINK_MODE_MII_BIT, + .eth_type = PORT_MII, + }, + [PRESTERA_PORT_TYPE_FIBRE] = { + .eth_mode = ETHTOOL_LINK_MODE_FIBRE_BIT, + .eth_type = PORT_FIBRE, + }, + [PRESTERA_PORT_TYPE_BNC] = { + .eth_mode = ETHTOOL_LINK_MODE_BNC_BIT, + .eth_type = PORT_BNC, + }, + [PRESTERA_PORT_TYPE_DA] = { + .eth_mode = ETHTOOL_LINK_MODE_TP_BIT, + .eth_type = PORT_TP, + }, + [PRESTERA_PORT_TYPE_OTHER] = { + .eth_mode = __ETHTOOL_LINK_MODE_MASK_NBITS, + .eth_type = PORT_OTHER, + } +}; + +static const char prestera_cnt_name[PRESTERA_STATS_CNT][ETH_GSTRING_LEN] = { + PRESTERA_STATS_FIELD(good_octets_received), + PRESTERA_STATS_FIELD(bad_octets_received), + PRESTERA_STATS_FIELD(mac_trans_error), + PRESTERA_STATS_FIELD(broadcast_frames_received), + PRESTERA_STATS_FIELD(multicast_frames_received), + PRESTERA_STATS_FIELD(frames_64_octets), + PRESTERA_STATS_FIELD(frames_65_to_127_octets), + PRESTERA_STATS_FIELD(frames_128_to_255_octets), + PRESTERA_STATS_FIELD(frames_256_to_511_octets), + PRESTERA_STATS_FIELD(frames_512_to_1023_octets), + PRESTERA_STATS_FIELD(frames_1024_to_max_octets), + PRESTERA_STATS_FIELD(excessive_collision), + PRESTERA_STATS_FIELD(multicast_frames_sent), + PRESTERA_STATS_FIELD(broadcast_frames_sent), + PRESTERA_STATS_FIELD(fc_sent), + PRESTERA_STATS_FIELD(fc_received), + PRESTERA_STATS_FIELD(buffer_overrun), + PRESTERA_STATS_FIELD(undersize), + PRESTERA_STATS_FIELD(fragments), + PRESTERA_STATS_FIELD(oversize), + PRESTERA_STATS_FIELD(jabber), + PRESTERA_STATS_FIELD(rx_error_frame_received), + PRESTERA_STATS_FIELD(bad_crc), + PRESTERA_STATS_FIELD(collisions), + PRESTERA_STATS_FIELD(late_collision), + PRESTERA_STATS_FIELD(unicast_frames_received), + PRESTERA_STATS_FIELD(unicast_frames_sent), + PRESTERA_STATS_FIELD(sent_multiple), + PRESTERA_STATS_FIELD(sent_deferred), + PRESTERA_STATS_FIELD(good_octets_sent), +}; + +static void prestera_ethtool_get_drvinfo(struct net_device *dev, + struct ethtool_drvinfo *drvinfo) +{ + struct prestera_port *port = netdev_priv(dev); + struct prestera_switch *sw = port->sw; + + strlcpy(drvinfo->driver, driver_kind, sizeof(drvinfo->driver)); + strlcpy(drvinfo->bus_info, dev_name(prestera_dev(sw)), + sizeof(drvinfo->bus_info)); + snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), + "%d.%d.%d", + sw->dev->fw_rev.maj, + sw->dev->fw_rev.min, + sw->dev->fw_rev.sub); +} + +static u8 prestera_port_type_get(struct prestera_port *port) +{ + if (port->caps.type < PRESTERA_PORT_TYPE_MAX) + return port_types[port->caps.type].eth_type; + + return PORT_OTHER; +} + +static int prestera_port_type_set(const struct ethtool_link_ksettings *ecmd, + struct prestera_port *port) +{ + u32 new_mode = PRESTERA_LINK_MODE_MAX; + u32 type, mode; + int err; + + for (type = 0; type < PRESTERA_PORT_TYPE_MAX; type++) { + if (port_types[type].eth_type == ecmd->base.port && + test_bit(port_types[type].eth_mode, + ecmd->link_modes.supported)) { + break; + } + } + + if (type == port->caps.type) + return 0; + if (type != port->caps.type && ecmd->base.autoneg == AUTONEG_ENABLE) + return -EINVAL; + if (type == PRESTERA_PORT_TYPE_MAX) + return -EOPNOTSUPP; + + for (mode = 0; mode < PRESTERA_LINK_MODE_MAX; mode++) { + if ((port_link_modes[mode].pr_mask & + port->caps.supp_link_modes) && + type == port_link_modes[mode].port_type) { + new_mode = mode; + } + } + + if (new_mode < PRESTERA_LINK_MODE_MAX) + err = prestera_hw_port_link_mode_set(port, new_mode); + else + err = -EINVAL; + + if (err) + return err; + + port->caps.type = type; + port->autoneg = false; + + return 0; +} + +static void prestera_modes_to_eth(unsigned long *eth_modes, u64 link_modes, + u8 fec, u8 type) +{ + u32 mode; + + for (mode = 0; mode < PRESTERA_LINK_MODE_MAX; mode++) { + if ((port_link_modes[mode].pr_mask & link_modes) == 0) + continue; + + if (type != PRESTERA_PORT_TYPE_NONE && + port_link_modes[mode].port_type != type) + continue; + + __set_bit(port_link_modes[mode].eth_mode, eth_modes); + } + + for (mode = 0; mode < PRESTERA_PORT_FEC_MAX; mode++) { + if ((port_fec_caps[mode].pr_fec & fec) == 0) + continue; + + __set_bit(port_fec_caps[mode].eth_mode, eth_modes); + } +} + +static void prestera_modes_from_eth(const unsigned long *eth_modes, + u64 *link_modes, u8 *fec, u8 type) +{ + u64 adver_modes = 0; + u32 fec_modes = 0; + u32 mode; + + for (mode = 0; mode < PRESTERA_LINK_MODE_MAX; mode++) { + if (!test_bit(port_link_modes[mode].eth_mode, eth_modes)) + continue; + + if (port_link_modes[mode].port_type != type) + continue; + + adver_modes |= port_link_modes[mode].pr_mask; + } + + for (mode = 0; mode < PRESTERA_PORT_FEC_MAX; mode++) { + if (!test_bit(port_fec_caps[mode].eth_mode, eth_modes)) + continue; + + fec_modes |= port_fec_caps[mode].pr_fec; + } + + *link_modes = adver_modes; + *fec = fec_modes; +} + +static void prestera_port_supp_types_get(struct ethtool_link_ksettings *ecmd, + struct prestera_port *port) +{ + u32 mode; + u8 ptype; + + for (mode = 0; mode < PRESTERA_LINK_MODE_MAX; mode++) { + if ((port_link_modes[mode].pr_mask & + port->caps.supp_link_modes) == 0) + continue; + + ptype = port_link_modes[mode].port_type; + __set_bit(port_types[ptype].eth_mode, + ecmd->link_modes.supported); + } +} + +static void prestera_port_remote_cap_get(struct ethtool_link_ksettings *ecmd, + struct prestera_port *port) +{ + bool asym_pause; + bool pause; + u64 bitmap; + int err; + + err = prestera_hw_port_remote_cap_get(port, &bitmap); + if (!err) { + prestera_modes_to_eth(ecmd->link_modes.lp_advertising, + bitmap, 0, PRESTERA_PORT_TYPE_NONE); + + if (!bitmap_empty(ecmd->link_modes.lp_advertising, + __ETHTOOL_LINK_MODE_MASK_NBITS)) { + ethtool_link_ksettings_add_link_mode(ecmd, + lp_advertising, + Autoneg); + } + } + + err = prestera_hw_port_remote_fc_get(port, &pause, &asym_pause); + if (err) + return; + + if (pause) + ethtool_link_ksettings_add_link_mode(ecmd, + lp_advertising, + Pause); + if (asym_pause) + ethtool_link_ksettings_add_link_mode(ecmd, + lp_advertising, + Asym_Pause); +} + +static void prestera_port_speed_get(struct ethtool_link_ksettings *ecmd, + struct prestera_port *port) +{ + u32 speed; + int err; + + err = prestera_hw_port_speed_get(port, &speed); + ecmd->base.speed = err ? SPEED_UNKNOWN : speed; +} + +static void prestera_port_duplex_get(struct ethtool_link_ksettings *ecmd, + struct prestera_port *port) +{ + u8 duplex; + int err; + + err = prestera_hw_port_duplex_get(port, &duplex); + if (err) { + ecmd->base.duplex = DUPLEX_UNKNOWN; + return; + } + + ecmd->base.duplex = duplex == PRESTERA_PORT_DUPLEX_FULL ? + DUPLEX_FULL : DUPLEX_HALF; +} + +static int +prestera_ethtool_get_link_ksettings(struct net_device *dev, + struct ethtool_link_ksettings *ecmd) +{ + struct prestera_port *port = netdev_priv(dev); + + ethtool_link_ksettings_zero_link_mode(ecmd, supported); + ethtool_link_ksettings_zero_link_mode(ecmd, advertising); + ethtool_link_ksettings_zero_link_mode(ecmd, lp_advertising); + + ecmd->base.autoneg = port->autoneg ? AUTONEG_ENABLE : AUTONEG_DISABLE; + + if (port->caps.type == PRESTERA_PORT_TYPE_TP) { + ethtool_link_ksettings_add_link_mode(ecmd, supported, Autoneg); + + if (netif_running(dev) && + (port->autoneg || + port->caps.transceiver == PRESTERA_PORT_TCVR_COPPER)) + ethtool_link_ksettings_add_link_mode(ecmd, advertising, + Autoneg); + } + + prestera_modes_to_eth(ecmd->link_modes.supported, + port->caps.supp_link_modes, + port->caps.supp_fec, + port->caps.type); + + prestera_port_supp_types_get(ecmd, port); + + if (netif_carrier_ok(dev)) { + prestera_port_speed_get(ecmd, port); + prestera_port_duplex_get(ecmd, port); + } else { + ecmd->base.speed = SPEED_UNKNOWN; + ecmd->base.duplex = DUPLEX_UNKNOWN; + } + + ecmd->base.port = prestera_port_type_get(port); + + if (port->autoneg) { + if (netif_running(dev)) + prestera_modes_to_eth(ecmd->link_modes.advertising, + port->adver_link_modes, + port->adver_fec, + port->caps.type); + + if (netif_carrier_ok(dev) && + port->caps.transceiver == PRESTERA_PORT_TCVR_COPPER) + prestera_port_remote_cap_get(ecmd, port); + } + + if (port->caps.type == PRESTERA_PORT_TYPE_TP && + port->caps.transceiver == PRESTERA_PORT_TCVR_COPPER) + prestera_hw_port_mdix_get(port, &ecmd->base.eth_tp_mdix, + &ecmd->base.eth_tp_mdix_ctrl); + + return 0; +} + +static int prestera_port_mdix_set(const struct ethtool_link_ksettings *ecmd, + struct prestera_port *port) +{ + if (ecmd->base.eth_tp_mdix_ctrl != ETH_TP_MDI_INVALID && + port->caps.transceiver == PRESTERA_PORT_TCVR_COPPER && + port->caps.type == PRESTERA_PORT_TYPE_TP) + return prestera_hw_port_mdix_set(port, + ecmd->base.eth_tp_mdix_ctrl); + + return 0; +} + +static int prestera_port_link_mode_set(struct prestera_port *port, + u32 speed, u8 duplex, u8 type) +{ + u32 new_mode = PRESTERA_LINK_MODE_MAX; + u32 mode; + + for (mode = 0; mode < PRESTERA_LINK_MODE_MAX; mode++) { + if (speed != port_link_modes[mode].speed) + continue; + + if (duplex != port_link_modes[mode].duplex) + continue; + + if (!(port_link_modes[mode].pr_mask & + port->caps.supp_link_modes)) + continue; + + if (type != port_link_modes[mode].port_type) + continue; + + new_mode = mode; + break; + } + + if (new_mode == PRESTERA_LINK_MODE_MAX) + return -EOPNOTSUPP; + + return prestera_hw_port_link_mode_set(port, new_mode); +} + +static int +prestera_port_speed_duplex_set(const struct ethtool_link_ksettings *ecmd, + struct prestera_port *port) +{ + u32 curr_mode; + u8 duplex; + u32 speed; + int err; + + err = prestera_hw_port_link_mode_get(port, &curr_mode); + if (err) + return err; + if (curr_mode >= PRESTERA_LINK_MODE_MAX) + return -EINVAL; + + if (ecmd->base.duplex != DUPLEX_UNKNOWN) + duplex = ecmd->base.duplex == DUPLEX_FULL ? + PRESTERA_PORT_DUPLEX_FULL : PRESTERA_PORT_DUPLEX_HALF; + else + duplex = port_link_modes[curr_mode].duplex; + + if (ecmd->base.speed != SPEED_UNKNOWN) + speed = ecmd->base.speed; + else + speed = port_link_modes[curr_mode].speed; + + return prestera_port_link_mode_set(port, speed, duplex, + port->caps.type); +} + +static int +prestera_ethtool_set_link_ksettings(struct net_device *dev, + const struct ethtool_link_ksettings *ecmd) +{ + struct prestera_port *port = netdev_priv(dev); + u64 adver_modes; + u8 adver_fec; + int err; + + err = prestera_port_type_set(ecmd, port); + if (err) + return err; + + if (port->caps.transceiver == PRESTERA_PORT_TCVR_COPPER) { + err = prestera_port_mdix_set(ecmd, port); + if (err) + return err; + } + + prestera_modes_from_eth(ecmd->link_modes.advertising, &adver_modes, + &adver_fec, port->caps.type); + + err = prestera_port_autoneg_set(port, + ecmd->base.autoneg == AUTONEG_ENABLE, + adver_modes, adver_fec); + if (err) + return err; + + if (ecmd->base.autoneg == AUTONEG_DISABLE) { + err = prestera_port_speed_duplex_set(ecmd, port); + if (err) + return err; + } + + return 0; +} + +static int prestera_ethtool_get_fecparam(struct net_device *dev, + struct ethtool_fecparam *fecparam) +{ + struct prestera_port *port = netdev_priv(dev); + u8 active; + u32 mode; + int err; + + err = prestera_hw_port_fec_get(port, &active); + if (err) + return err; + + fecparam->fec = 0; + + for (mode = 0; mode < PRESTERA_PORT_FEC_MAX; mode++) { + if ((port_fec_caps[mode].pr_fec & port->caps.supp_fec) == 0) + continue; + + fecparam->fec |= port_fec_caps[mode].eth_fec; + } + + if (active < PRESTERA_PORT_FEC_MAX) + fecparam->active_fec = port_fec_caps[active].eth_fec; + else + fecparam->active_fec = ETHTOOL_FEC_AUTO; + + return 0; +} + +static int prestera_ethtool_set_fecparam(struct net_device *dev, + struct ethtool_fecparam *fecparam) +{ + struct prestera_port *port = netdev_priv(dev); + u8 fec, active; + u32 mode; + int err; + + if (port->autoneg) { + netdev_err(dev, "FEC set is not allowed while autoneg is on\n"); + return -EINVAL; + } + + err = prestera_hw_port_fec_get(port, &active); + if (err) + return err; + + fec = PRESTERA_PORT_FEC_MAX; + for (mode = 0; mode < PRESTERA_PORT_FEC_MAX; mode++) { + if ((port_fec_caps[mode].eth_fec & fecparam->fec) && + (port_fec_caps[mode].pr_fec & port->caps.supp_fec)) { + fec = mode; + break; + } + } + + if (fec == active) + return 0; + + if (fec == PRESTERA_PORT_FEC_MAX) + return -EOPNOTSUPP; + + return prestera_hw_port_fec_set(port, fec); +} + +static int prestera_ethtool_get_sset_count(struct net_device *dev, int sset) +{ + switch (sset) { + case ETH_SS_STATS: + return PRESTERA_STATS_CNT; + default: + return -EOPNOTSUPP; + } +} + +static void prestera_ethtool_get_strings(struct net_device *dev, + u32 stringset, u8 *data) +{ + if (stringset != ETH_SS_STATS) + return; + + memcpy(data, prestera_cnt_name, sizeof(prestera_cnt_name)); +} + +static void prestera_ethtool_get_stats(struct net_device *dev, + struct ethtool_stats *stats, u64 *data) +{ + struct prestera_port *port = netdev_priv(dev); + struct prestera_port_stats *port_stats; + + port_stats = &port->cached_hw_stats.stats; + + memcpy(data, port_stats, sizeof(*port_stats)); +} + +static int prestera_ethtool_nway_reset(struct net_device *dev) +{ + struct prestera_port *port = netdev_priv(dev); + + if (netif_running(dev) && + port->caps.transceiver == PRESTERA_PORT_TCVR_COPPER && + port->caps.type == PRESTERA_PORT_TYPE_TP) + return prestera_hw_port_autoneg_restart(port); + + return -EINVAL; +} + +const struct ethtool_ops prestera_ethtool_ops = { + .get_drvinfo = prestera_ethtool_get_drvinfo, + .get_link_ksettings = prestera_ethtool_get_link_ksettings, + .set_link_ksettings = prestera_ethtool_set_link_ksettings, + .get_fecparam = prestera_ethtool_get_fecparam, + .set_fecparam = prestera_ethtool_set_fecparam, + .get_sset_count = prestera_ethtool_get_sset_count, + .get_strings = prestera_ethtool_get_strings, + .get_ethtool_stats = prestera_ethtool_get_stats, + .get_link = ethtool_op_get_link, + .nway_reset = prestera_ethtool_nway_reset +}; diff --git a/drivers/net/ethernet/marvell/prestera/prestera_ethtool.h b/drivers/net/ethernet/marvell/prestera/prestera_ethtool.h new file mode 100644 index 000000000000..523ef1f592ce --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/prestera_ethtool.h @@ -0,0 +1,11 @@ +/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 */ +/* Copyright (c) 2019-2020 Marvell International Ltd. All rights reserved. */ + +#ifndef __PRESTERA_ETHTOOL_H_ +#define __PRESTERA_ETHTOOL_H_ + +#include <linux/ethtool.h> + +extern const struct ethtool_ops prestera_ethtool_ops; + +#endif /* _PRESTERA_ETHTOOL_H_ */ diff --git a/drivers/net/ethernet/marvell/prestera/prestera_hw.c b/drivers/net/ethernet/marvell/prestera/prestera_hw.c new file mode 100644 index 000000000000..0424718d5998 --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/prestera_hw.c @@ -0,0 +1,1253 @@ +// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 +/* Copyright (c) 2019-2020 Marvell International Ltd. All rights reserved */ + +#include <linux/etherdevice.h> +#include <linux/ethtool.h> +#include <linux/list.h> + +#include "prestera.h" +#include "prestera_hw.h" + +#define PRESTERA_SWITCH_INIT_TIMEOUT_MS (30 * 1000) + +#define PRESTERA_MIN_MTU 64 + +enum prestera_cmd_type_t { + PRESTERA_CMD_TYPE_SWITCH_INIT = 0x1, + PRESTERA_CMD_TYPE_SWITCH_ATTR_SET = 0x2, + + PRESTERA_CMD_TYPE_PORT_ATTR_SET = 0x100, + PRESTERA_CMD_TYPE_PORT_ATTR_GET = 0x101, + PRESTERA_CMD_TYPE_PORT_INFO_GET = 0x110, + + PRESTERA_CMD_TYPE_VLAN_CREATE = 0x200, + PRESTERA_CMD_TYPE_VLAN_DELETE = 0x201, + PRESTERA_CMD_TYPE_VLAN_PORT_SET = 0x202, + PRESTERA_CMD_TYPE_VLAN_PVID_SET = 0x203, + + PRESTERA_CMD_TYPE_FDB_ADD = 0x300, + PRESTERA_CMD_TYPE_FDB_DELETE = 0x301, + PRESTERA_CMD_TYPE_FDB_FLUSH_PORT = 0x310, + PRESTERA_CMD_TYPE_FDB_FLUSH_VLAN = 0x311, + PRESTERA_CMD_TYPE_FDB_FLUSH_PORT_VLAN = 0x312, + + PRESTERA_CMD_TYPE_BRIDGE_CREATE = 0x400, + PRESTERA_CMD_TYPE_BRIDGE_DELETE = 0x401, + PRESTERA_CMD_TYPE_BRIDGE_PORT_ADD = 0x402, + PRESTERA_CMD_TYPE_BRIDGE_PORT_DELETE = 0x403, + + PRESTERA_CMD_TYPE_RXTX_INIT = 0x800, + PRESTERA_CMD_TYPE_RXTX_PORT_INIT = 0x801, + + PRESTERA_CMD_TYPE_STP_PORT_SET = 0x1000, + + PRESTERA_CMD_TYPE_ACK = 0x10000, + PRESTERA_CMD_TYPE_MAX +}; + +enum { + PRESTERA_CMD_PORT_ATTR_ADMIN_STATE = 1, + PRESTERA_CMD_PORT_ATTR_MTU = 3, + PRESTERA_CMD_PORT_ATTR_MAC = 4, + PRESTERA_CMD_PORT_ATTR_SPEED = 5, + PRESTERA_CMD_PORT_ATTR_ACCEPT_FRAME_TYPE = 6, + PRESTERA_CMD_PORT_ATTR_LEARNING = 7, + PRESTERA_CMD_PORT_ATTR_FLOOD = 8, + PRESTERA_CMD_PORT_ATTR_CAPABILITY = 9, + PRESTERA_CMD_PORT_ATTR_REMOTE_CAPABILITY = 10, + PRESTERA_CMD_PORT_ATTR_REMOTE_FC = 11, + PRESTERA_CMD_PORT_ATTR_LINK_MODE = 12, + PRESTERA_CMD_PORT_ATTR_TYPE = 13, + PRESTERA_CMD_PORT_ATTR_FEC = 14, + PRESTERA_CMD_PORT_ATTR_AUTONEG = 15, + PRESTERA_CMD_PORT_ATTR_DUPLEX = 16, + PRESTERA_CMD_PORT_ATTR_STATS = 17, + PRESTERA_CMD_PORT_ATTR_MDIX = 18, + PRESTERA_CMD_PORT_ATTR_AUTONEG_RESTART = 19, +}; + +enum { + PRESTERA_CMD_SWITCH_ATTR_MAC = 1, + PRESTERA_CMD_SWITCH_ATTR_AGEING = 2, +}; + +enum { + PRESTERA_CMD_ACK_OK, + PRESTERA_CMD_ACK_FAILED, + + PRESTERA_CMD_ACK_MAX +}; + +enum { + PRESTERA_PORT_TP_NA, + PRESTERA_PORT_TP_MDI, + PRESTERA_PORT_TP_MDIX, + PRESTERA_PORT_TP_AUTO, +}; + +enum { + PRESTERA_PORT_GOOD_OCTETS_RCV_CNT, + PRESTERA_PORT_BAD_OCTETS_RCV_CNT, + PRESTERA_PORT_MAC_TRANSMIT_ERR_CNT, + PRESTERA_PORT_BRDC_PKTS_RCV_CNT, + PRESTERA_PORT_MC_PKTS_RCV_CNT, + PRESTERA_PORT_PKTS_64L_CNT, + PRESTERA_PORT_PKTS_65TO127L_CNT, + PRESTERA_PORT_PKTS_128TO255L_CNT, + PRESTERA_PORT_PKTS_256TO511L_CNT, + PRESTERA_PORT_PKTS_512TO1023L_CNT, + PRESTERA_PORT_PKTS_1024TOMAXL_CNT, + PRESTERA_PORT_EXCESSIVE_COLLISIONS_CNT, + PRESTERA_PORT_MC_PKTS_SENT_CNT, + PRESTERA_PORT_BRDC_PKTS_SENT_CNT, + PRESTERA_PORT_FC_SENT_CNT, + PRESTERA_PORT_GOOD_FC_RCV_CNT, + PRESTERA_PORT_DROP_EVENTS_CNT, + PRESTERA_PORT_UNDERSIZE_PKTS_CNT, + PRESTERA_PORT_FRAGMENTS_PKTS_CNT, + PRESTERA_PORT_OVERSIZE_PKTS_CNT, + PRESTERA_PORT_JABBER_PKTS_CNT, + PRESTERA_PORT_MAC_RCV_ERROR_CNT, + PRESTERA_PORT_BAD_CRC_CNT, + PRESTERA_PORT_COLLISIONS_CNT, + PRESTERA_PORT_LATE_COLLISIONS_CNT, + PRESTERA_PORT_GOOD_UC_PKTS_RCV_CNT, + PRESTERA_PORT_GOOD_UC_PKTS_SENT_CNT, + PRESTERA_PORT_MULTIPLE_PKTS_SENT_CNT, + PRESTERA_PORT_DEFERRED_PKTS_SENT_CNT, + PRESTERA_PORT_GOOD_OCTETS_SENT_CNT, + + PRESTERA_PORT_CNT_MAX +}; + +enum { + PRESTERA_FC_NONE, + PRESTERA_FC_SYMMETRIC, + PRESTERA_FC_ASYMMETRIC, + PRESTERA_FC_SYMM_ASYMM, +}; + +struct prestera_fw_event_handler { + struct list_head list; + struct rcu_head rcu; + enum prestera_event_type type; + prestera_event_cb_t func; + void *arg; +}; + +struct prestera_msg_cmd { + u32 type; +}; + +struct prestera_msg_ret { + struct prestera_msg_cmd cmd; + u32 status; +}; + +struct prestera_msg_common_req { + struct prestera_msg_cmd cmd; +}; + +struct prestera_msg_common_resp { + struct prestera_msg_ret ret; +}; + +union prestera_msg_switch_param { + u8 mac[ETH_ALEN]; + u32 ageing_timeout_ms; +}; + +struct prestera_msg_switch_attr_req { + struct prestera_msg_cmd cmd; + u32 attr; + union prestera_msg_switch_param param; +}; + +struct prestera_msg_switch_init_resp { + struct prestera_msg_ret ret; + u32 port_count; + u32 mtu_max; + u8 switch_id; +}; + +struct prestera_msg_port_autoneg_param { + u64 link_mode; + u8 enable; + u8 fec; +}; + +struct prestera_msg_port_cap_param { + u64 link_mode; + u8 type; + u8 fec; + u8 transceiver; +}; + +struct prestera_msg_port_mdix_param { + u8 status; + u8 admin_mode; +}; + +union prestera_msg_port_param { + u8 admin_state; + u8 oper_state; + u32 mtu; + u8 mac[ETH_ALEN]; + u8 accept_frm_type; + u32 speed; + u8 learning; + u8 flood; + u32 link_mode; + u8 type; + u8 duplex; + u8 fec; + u8 fc; + struct prestera_msg_port_mdix_param mdix; + struct prestera_msg_port_autoneg_param autoneg; + struct prestera_msg_port_cap_param cap; +}; + +struct prestera_msg_port_attr_req { + struct prestera_msg_cmd cmd; + u32 attr; + u32 port; + u32 dev; + union prestera_msg_port_param param; +}; + +struct prestera_msg_port_attr_resp { + struct prestera_msg_ret ret; + union prestera_msg_port_param param; +}; + +struct prestera_msg_port_stats_resp { + struct prestera_msg_ret ret; + u64 stats[PRESTERA_PORT_CNT_MAX]; +}; + +struct prestera_msg_port_info_req { + struct prestera_msg_cmd cmd; + u32 port; +}; + +struct prestera_msg_port_info_resp { + struct prestera_msg_ret ret; + u32 hw_id; + u32 dev_id; + u16 fp_id; +}; + +struct prestera_msg_vlan_req { + struct prestera_msg_cmd cmd; + u32 port; + u32 dev; + u16 vid; + u8 is_member; + u8 is_tagged; +}; + +struct prestera_msg_fdb_req { + struct prestera_msg_cmd cmd; + u8 dest_type; + u32 port; + u32 dev; + u8 mac[ETH_ALEN]; + u16 vid; + u8 dynamic; + u32 flush_mode; +}; + +struct prestera_msg_bridge_req { + struct prestera_msg_cmd cmd; + u32 port; + u32 dev; + u16 bridge; +}; + +struct prestera_msg_bridge_resp { + struct prestera_msg_ret ret; + u16 bridge; +}; + +struct prestera_msg_stp_req { + struct prestera_msg_cmd cmd; + u32 port; + u32 dev; + u16 vid; + u8 state; +}; + +struct prestera_msg_rxtx_req { + struct prestera_msg_cmd cmd; + u8 use_sdma; +}; + +struct prestera_msg_rxtx_resp { + struct prestera_msg_ret ret; + u32 map_addr; +}; + +struct prestera_msg_rxtx_port_req { + struct prestera_msg_cmd cmd; + u32 port; + u32 dev; +}; + +struct prestera_msg_event { + u16 type; + u16 id; +}; + +union prestera_msg_event_port_param { + u32 oper_state; +}; + +struct prestera_msg_event_port { + struct prestera_msg_event id; + u32 port_id; + union prestera_msg_event_port_param param; +}; + +union prestera_msg_event_fdb_param { + u8 mac[ETH_ALEN]; +}; + +struct prestera_msg_event_fdb { + struct prestera_msg_event id; + u8 dest_type; + u32 port_id; + u32 vid; + union prestera_msg_event_fdb_param param; +}; + +static int __prestera_cmd_ret(struct prestera_switch *sw, + enum prestera_cmd_type_t type, + struct prestera_msg_cmd *cmd, size_t clen, + struct prestera_msg_ret *ret, size_t rlen, + int waitms) +{ + struct prestera_device *dev = sw->dev; + int err; + + cmd->type = type; + + err = dev->send_req(dev, cmd, clen, ret, rlen, waitms); + if (err) + return err; + + if (ret->cmd.type != PRESTERA_CMD_TYPE_ACK) + return -EBADE; + if (ret->status != PRESTERA_CMD_ACK_OK) + return -EINVAL; + + return 0; +} + +static int prestera_cmd_ret(struct prestera_switch *sw, + enum prestera_cmd_type_t type, + struct prestera_msg_cmd *cmd, size_t clen, + struct prestera_msg_ret *ret, size_t rlen) +{ + return __prestera_cmd_ret(sw, type, cmd, clen, ret, rlen, 0); +} + +static int prestera_cmd_ret_wait(struct prestera_switch *sw, + enum prestera_cmd_type_t type, + struct prestera_msg_cmd *cmd, size_t clen, + struct prestera_msg_ret *ret, size_t rlen, + int waitms) +{ + return __prestera_cmd_ret(sw, type, cmd, clen, ret, rlen, waitms); +} + +static int prestera_cmd(struct prestera_switch *sw, + enum prestera_cmd_type_t type, + struct prestera_msg_cmd *cmd, size_t clen) +{ + struct prestera_msg_common_resp resp; + + return prestera_cmd_ret(sw, type, cmd, clen, &resp.ret, sizeof(resp)); +} + +static int prestera_fw_parse_port_evt(void *msg, struct prestera_event *evt) +{ + struct prestera_msg_event_port *hw_evt = msg; + + if (evt->id != PRESTERA_PORT_EVENT_STATE_CHANGED) + return -EINVAL; + + evt->port_evt.data.oper_state = hw_evt->param.oper_state; + evt->port_evt.port_id = hw_evt->port_id; + + return 0; +} + +static int prestera_fw_parse_fdb_evt(void *msg, struct prestera_event *evt) +{ + struct prestera_msg_event_fdb *hw_evt = msg; + + evt->fdb_evt.port_id = hw_evt->port_id; + evt->fdb_evt.vid = hw_evt->vid; + + ether_addr_copy(evt->fdb_evt.data.mac, hw_evt->param.mac); + + return 0; +} + +static struct prestera_fw_evt_parser { + int (*func)(void *msg, struct prestera_event *evt); +} fw_event_parsers[PRESTERA_EVENT_TYPE_MAX] = { + [PRESTERA_EVENT_TYPE_PORT] = { .func = prestera_fw_parse_port_evt }, + [PRESTERA_EVENT_TYPE_FDB] = { .func = prestera_fw_parse_fdb_evt }, +}; + +static struct prestera_fw_event_handler * +__find_event_handler(const struct prestera_switch *sw, + enum prestera_event_type type) +{ + struct prestera_fw_event_handler *eh; + + list_for_each_entry_rcu(eh, &sw->event_handlers, list) { + if (eh->type == type) + return eh; + } + + return NULL; +} + +static int prestera_find_event_handler(const struct prestera_switch *sw, + enum prestera_event_type type, + struct prestera_fw_event_handler *eh) +{ + struct prestera_fw_event_handler *tmp; + int err = 0; + + rcu_read_lock(); + tmp = __find_event_handler(sw, type); + if (tmp) + *eh = *tmp; + else + err = -ENOENT; + rcu_read_unlock(); + + return err; +} + +static int prestera_evt_recv(struct prestera_device *dev, void *buf, size_t size) +{ + struct prestera_switch *sw = dev->priv; + struct prestera_msg_event *msg = buf; + struct prestera_fw_event_handler eh; + struct prestera_event evt; + int err; + + if (msg->type >= PRESTERA_EVENT_TYPE_MAX) + return -EINVAL; + if (!fw_event_parsers[msg->type].func) + return -ENOENT; + + err = prestera_find_event_handler(sw, msg->type, &eh); + if (err) + return err; + + evt.id = msg->id; + + err = fw_event_parsers[msg->type].func(buf, &evt); + if (err) + return err; + + eh.func(sw, &evt, eh.arg); + + return 0; +} + +static void prestera_pkt_recv(struct prestera_device *dev) +{ + struct prestera_switch *sw = dev->priv; + struct prestera_fw_event_handler eh; + struct prestera_event ev; + int err; + + ev.id = PRESTERA_RXTX_EVENT_RCV_PKT; + + err = prestera_find_event_handler(sw, PRESTERA_EVENT_TYPE_RXTX, &eh); + if (err) + return; + + eh.func(sw, &ev, eh.arg); +} + +int prestera_hw_port_info_get(const struct prestera_port *port, + u32 *dev_id, u32 *hw_id, u16 *fp_id) +{ + struct prestera_msg_port_info_req req = { + .port = port->id, + }; + struct prestera_msg_port_info_resp resp; + int err; + + err = prestera_cmd_ret(port->sw, PRESTERA_CMD_TYPE_PORT_INFO_GET, + &req.cmd, sizeof(req), &resp.ret, sizeof(resp)); + if (err) + return err; + + *dev_id = resp.dev_id; + *hw_id = resp.hw_id; + *fp_id = resp.fp_id; + + return 0; +} + +int prestera_hw_switch_mac_set(struct prestera_switch *sw, const char *mac) +{ + struct prestera_msg_switch_attr_req req = { + .attr = PRESTERA_CMD_SWITCH_ATTR_MAC, + }; + + ether_addr_copy(req.param.mac, mac); + + return prestera_cmd(sw, PRESTERA_CMD_TYPE_SWITCH_ATTR_SET, + &req.cmd, sizeof(req)); +} + +int prestera_hw_switch_init(struct prestera_switch *sw) +{ + struct prestera_msg_switch_init_resp resp; + struct prestera_msg_common_req req; + int err; + + INIT_LIST_HEAD(&sw->event_handlers); + + err = prestera_cmd_ret_wait(sw, PRESTERA_CMD_TYPE_SWITCH_INIT, + &req.cmd, sizeof(req), + &resp.ret, sizeof(resp), + PRESTERA_SWITCH_INIT_TIMEOUT_MS); + if (err) + return err; + + sw->dev->recv_msg = prestera_evt_recv; + sw->dev->recv_pkt = prestera_pkt_recv; + sw->port_count = resp.port_count; + sw->mtu_min = PRESTERA_MIN_MTU; + sw->mtu_max = resp.mtu_max; + sw->id = resp.switch_id; + + return 0; +} + +void prestera_hw_switch_fini(struct prestera_switch *sw) +{ + WARN_ON(!list_empty(&sw->event_handlers)); +} + +int prestera_hw_switch_ageing_set(struct prestera_switch *sw, u32 ageing_ms) +{ + struct prestera_msg_switch_attr_req req = { + .attr = PRESTERA_CMD_SWITCH_ATTR_AGEING, + .param = { + .ageing_timeout_ms = ageing_ms, + }, + }; + + return prestera_cmd(sw, PRESTERA_CMD_TYPE_SWITCH_ATTR_SET, + &req.cmd, sizeof(req)); +} + +int prestera_hw_port_state_set(const struct prestera_port *port, + bool admin_state) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_ADMIN_STATE, + .port = port->hw_id, + .dev = port->dev_id, + .param = { + .admin_state = admin_state, + } + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_SET, + &req.cmd, sizeof(req)); +} + +int prestera_hw_port_mtu_set(const struct prestera_port *port, u32 mtu) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_MTU, + .port = port->hw_id, + .dev = port->dev_id, + .param = { + .mtu = mtu, + } + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_SET, + &req.cmd, sizeof(req)); +} + +int prestera_hw_port_mac_set(const struct prestera_port *port, const char *mac) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_MAC, + .port = port->hw_id, + .dev = port->dev_id, + }; + + ether_addr_copy(req.param.mac, mac); + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_SET, + &req.cmd, sizeof(req)); +} + +int prestera_hw_port_accept_frm_type(struct prestera_port *port, + enum prestera_accept_frm_type type) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_ACCEPT_FRAME_TYPE, + .port = port->hw_id, + .dev = port->dev_id, + .param = { + .accept_frm_type = type, + } + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_SET, + &req.cmd, sizeof(req)); +} + +int prestera_hw_port_cap_get(const struct prestera_port *port, + struct prestera_port_caps *caps) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_CAPABILITY, + .port = port->hw_id, + .dev = port->dev_id, + }; + struct prestera_msg_port_attr_resp resp; + int err; + + err = prestera_cmd_ret(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_GET, + &req.cmd, sizeof(req), &resp.ret, sizeof(resp)); + if (err) + return err; + + caps->supp_link_modes = resp.param.cap.link_mode; + caps->transceiver = resp.param.cap.transceiver; + caps->supp_fec = resp.param.cap.fec; + caps->type = resp.param.cap.type; + + return err; +} + +int prestera_hw_port_remote_cap_get(const struct prestera_port *port, + u64 *link_mode_bitmap) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_REMOTE_CAPABILITY, + .port = port->hw_id, + .dev = port->dev_id, + }; + struct prestera_msg_port_attr_resp resp; + int err; + + err = prestera_cmd_ret(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_GET, + &req.cmd, sizeof(req), &resp.ret, sizeof(resp)); + if (err) + return err; + + *link_mode_bitmap = resp.param.cap.link_mode; + + return 0; +} + +int prestera_hw_port_remote_fc_get(const struct prestera_port *port, + bool *pause, bool *asym_pause) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_REMOTE_FC, + .port = port->hw_id, + .dev = port->dev_id, + }; + struct prestera_msg_port_attr_resp resp; + int err; + + err = prestera_cmd_ret(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_GET, + &req.cmd, sizeof(req), &resp.ret, sizeof(resp)); + if (err) + return err; + + switch (resp.param.fc) { + case PRESTERA_FC_SYMMETRIC: + *pause = true; + *asym_pause = false; + break; + case PRESTERA_FC_ASYMMETRIC: + *pause = false; + *asym_pause = true; + break; + case PRESTERA_FC_SYMM_ASYMM: + *pause = true; + *asym_pause = true; + break; + default: + *pause = false; + *asym_pause = false; + } + + return 0; +} + +int prestera_hw_port_type_get(const struct prestera_port *port, u8 *type) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_TYPE, + .port = port->hw_id, + .dev = port->dev_id, + }; + struct prestera_msg_port_attr_resp resp; + int err; + + err = prestera_cmd_ret(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_GET, + &req.cmd, sizeof(req), &resp.ret, sizeof(resp)); + if (err) + return err; + + *type = resp.param.type; + + return 0; +} + +int prestera_hw_port_fec_get(const struct prestera_port *port, u8 *fec) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_FEC, + .port = port->hw_id, + .dev = port->dev_id, + }; + struct prestera_msg_port_attr_resp resp; + int err; + + err = prestera_cmd_ret(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_GET, + &req.cmd, sizeof(req), &resp.ret, sizeof(resp)); + if (err) + return err; + + *fec = resp.param.fec; + + return 0; +} + +int prestera_hw_port_fec_set(const struct prestera_port *port, u8 fec) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_FEC, + .port = port->hw_id, + .dev = port->dev_id, + .param = { + .fec = fec, + } + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_SET, + &req.cmd, sizeof(req)); +} + +static u8 prestera_hw_mdix_to_eth(u8 mode) +{ + switch (mode) { + case PRESTERA_PORT_TP_MDI: + return ETH_TP_MDI; + case PRESTERA_PORT_TP_MDIX: + return ETH_TP_MDI_X; + case PRESTERA_PORT_TP_AUTO: + return ETH_TP_MDI_AUTO; + default: + return ETH_TP_MDI_INVALID; + } +} + +static u8 prestera_hw_mdix_from_eth(u8 mode) +{ + switch (mode) { + case ETH_TP_MDI: + return PRESTERA_PORT_TP_MDI; + case ETH_TP_MDI_X: + return PRESTERA_PORT_TP_MDIX; + case ETH_TP_MDI_AUTO: + return PRESTERA_PORT_TP_AUTO; + default: + return PRESTERA_PORT_TP_NA; + } +} + +int prestera_hw_port_mdix_get(const struct prestera_port *port, u8 *status, + u8 *admin_mode) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_MDIX, + .port = port->hw_id, + .dev = port->dev_id, + }; + struct prestera_msg_port_attr_resp resp; + int err; + + err = prestera_cmd_ret(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_GET, + &req.cmd, sizeof(req), &resp.ret, sizeof(resp)); + if (err) + return err; + + *status = prestera_hw_mdix_to_eth(resp.param.mdix.status); + *admin_mode = prestera_hw_mdix_to_eth(resp.param.mdix.admin_mode); + + return 0; +} + +int prestera_hw_port_mdix_set(const struct prestera_port *port, u8 mode) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_MDIX, + .port = port->hw_id, + .dev = port->dev_id, + }; + + req.param.mdix.admin_mode = prestera_hw_mdix_from_eth(mode); + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_SET, + &req.cmd, sizeof(req)); +} + +int prestera_hw_port_link_mode_set(const struct prestera_port *port, u32 mode) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_LINK_MODE, + .port = port->hw_id, + .dev = port->dev_id, + .param = { + .link_mode = mode, + } + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_SET, + &req.cmd, sizeof(req)); +} + +int prestera_hw_port_link_mode_get(const struct prestera_port *port, u32 *mode) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_LINK_MODE, + .port = port->hw_id, + .dev = port->dev_id, + }; + struct prestera_msg_port_attr_resp resp; + int err; + + err = prestera_cmd_ret(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_GET, + &req.cmd, sizeof(req), &resp.ret, sizeof(resp)); + if (err) + return err; + + *mode = resp.param.link_mode; + + return 0; +} + +int prestera_hw_port_speed_get(const struct prestera_port *port, u32 *speed) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_SPEED, + .port = port->hw_id, + .dev = port->dev_id, + }; + struct prestera_msg_port_attr_resp resp; + int err; + + err = prestera_cmd_ret(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_GET, + &req.cmd, sizeof(req), &resp.ret, sizeof(resp)); + if (err) + return err; + + *speed = resp.param.speed; + + return 0; +} + +int prestera_hw_port_autoneg_set(const struct prestera_port *port, + bool autoneg, u64 link_modes, u8 fec) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_AUTONEG, + .port = port->hw_id, + .dev = port->dev_id, + .param = { + .autoneg = { + .link_mode = link_modes, + .enable = autoneg, + .fec = fec, + } + } + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_SET, + &req.cmd, sizeof(req)); +} + +int prestera_hw_port_autoneg_restart(struct prestera_port *port) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_AUTONEG_RESTART, + .port = port->hw_id, + .dev = port->dev_id, + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_SET, + &req.cmd, sizeof(req)); +} + +int prestera_hw_port_duplex_get(const struct prestera_port *port, u8 *duplex) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_DUPLEX, + .port = port->hw_id, + .dev = port->dev_id, + }; + struct prestera_msg_port_attr_resp resp; + int err; + + err = prestera_cmd_ret(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_GET, + &req.cmd, sizeof(req), &resp.ret, sizeof(resp)); + if (err) + return err; + + *duplex = resp.param.duplex; + + return 0; +} + +int prestera_hw_port_stats_get(const struct prestera_port *port, + struct prestera_port_stats *st) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_STATS, + .port = port->hw_id, + .dev = port->dev_id, + }; + struct prestera_msg_port_stats_resp resp; + u64 *hw = resp.stats; + int err; + + err = prestera_cmd_ret(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_GET, + &req.cmd, sizeof(req), &resp.ret, sizeof(resp)); + if (err) + return err; + + st->good_octets_received = hw[PRESTERA_PORT_GOOD_OCTETS_RCV_CNT]; + st->bad_octets_received = hw[PRESTERA_PORT_BAD_OCTETS_RCV_CNT]; + st->mac_trans_error = hw[PRESTERA_PORT_MAC_TRANSMIT_ERR_CNT]; + st->broadcast_frames_received = hw[PRESTERA_PORT_BRDC_PKTS_RCV_CNT]; + st->multicast_frames_received = hw[PRESTERA_PORT_MC_PKTS_RCV_CNT]; + st->frames_64_octets = hw[PRESTERA_PORT_PKTS_64L_CNT]; + st->frames_65_to_127_octets = hw[PRESTERA_PORT_PKTS_65TO127L_CNT]; + st->frames_128_to_255_octets = hw[PRESTERA_PORT_PKTS_128TO255L_CNT]; + st->frames_256_to_511_octets = hw[PRESTERA_PORT_PKTS_256TO511L_CNT]; + st->frames_512_to_1023_octets = hw[PRESTERA_PORT_PKTS_512TO1023L_CNT]; + st->frames_1024_to_max_octets = hw[PRESTERA_PORT_PKTS_1024TOMAXL_CNT]; + st->excessive_collision = hw[PRESTERA_PORT_EXCESSIVE_COLLISIONS_CNT]; + st->multicast_frames_sent = hw[PRESTERA_PORT_MC_PKTS_SENT_CNT]; + st->broadcast_frames_sent = hw[PRESTERA_PORT_BRDC_PKTS_SENT_CNT]; + st->fc_sent = hw[PRESTERA_PORT_FC_SENT_CNT]; + st->fc_received = hw[PRESTERA_PORT_GOOD_FC_RCV_CNT]; + st->buffer_overrun = hw[PRESTERA_PORT_DROP_EVENTS_CNT]; + st->undersize = hw[PRESTERA_PORT_UNDERSIZE_PKTS_CNT]; + st->fragments = hw[PRESTERA_PORT_FRAGMENTS_PKTS_CNT]; + st->oversize = hw[PRESTERA_PORT_OVERSIZE_PKTS_CNT]; + st->jabber = hw[PRESTERA_PORT_JABBER_PKTS_CNT]; + st->rx_error_frame_received = hw[PRESTERA_PORT_MAC_RCV_ERROR_CNT]; + st->bad_crc = hw[PRESTERA_PORT_BAD_CRC_CNT]; + st->collisions = hw[PRESTERA_PORT_COLLISIONS_CNT]; + st->late_collision = hw[PRESTERA_PORT_LATE_COLLISIONS_CNT]; + st->unicast_frames_received = hw[PRESTERA_PORT_GOOD_UC_PKTS_RCV_CNT]; + st->unicast_frames_sent = hw[PRESTERA_PORT_GOOD_UC_PKTS_SENT_CNT]; + st->sent_multiple = hw[PRESTERA_PORT_MULTIPLE_PKTS_SENT_CNT]; + st->sent_deferred = hw[PRESTERA_PORT_DEFERRED_PKTS_SENT_CNT]; + st->good_octets_sent = hw[PRESTERA_PORT_GOOD_OCTETS_SENT_CNT]; + + return 0; +} + +int prestera_hw_port_learning_set(struct prestera_port *port, bool enable) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_LEARNING, + .port = port->hw_id, + .dev = port->dev_id, + .param = { + .learning = enable, + } + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_SET, + &req.cmd, sizeof(req)); +} + +int prestera_hw_port_flood_set(struct prestera_port *port, bool flood) +{ + struct prestera_msg_port_attr_req req = { + .attr = PRESTERA_CMD_PORT_ATTR_FLOOD, + .port = port->hw_id, + .dev = port->dev_id, + .param = { + .flood = flood, + } + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_PORT_ATTR_SET, + &req.cmd, sizeof(req)); +} + +int prestera_hw_vlan_create(struct prestera_switch *sw, u16 vid) +{ + struct prestera_msg_vlan_req req = { + .vid = vid, + }; + + return prestera_cmd(sw, PRESTERA_CMD_TYPE_VLAN_CREATE, + &req.cmd, sizeof(req)); +} + +int prestera_hw_vlan_delete(struct prestera_switch *sw, u16 vid) +{ + struct prestera_msg_vlan_req req = { + .vid = vid, + }; + + return prestera_cmd(sw, PRESTERA_CMD_TYPE_VLAN_DELETE, + &req.cmd, sizeof(req)); +} + +int prestera_hw_vlan_port_set(struct prestera_port *port, u16 vid, + bool is_member, bool untagged) +{ + struct prestera_msg_vlan_req req = { + .port = port->hw_id, + .dev = port->dev_id, + .vid = vid, + .is_member = is_member, + .is_tagged = !untagged, + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_VLAN_PORT_SET, + &req.cmd, sizeof(req)); +} + +int prestera_hw_vlan_port_vid_set(struct prestera_port *port, u16 vid) +{ + struct prestera_msg_vlan_req req = { + .port = port->hw_id, + .dev = port->dev_id, + .vid = vid, + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_VLAN_PVID_SET, + &req.cmd, sizeof(req)); +} + +int prestera_hw_vlan_port_stp_set(struct prestera_port *port, u16 vid, u8 state) +{ + struct prestera_msg_stp_req req = { + .port = port->hw_id, + .dev = port->dev_id, + .vid = vid, + .state = state, + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_STP_PORT_SET, + &req.cmd, sizeof(req)); +} + +int prestera_hw_fdb_add(struct prestera_port *port, const unsigned char *mac, + u16 vid, bool dynamic) +{ + struct prestera_msg_fdb_req req = { + .port = port->hw_id, + .dev = port->dev_id, + .vid = vid, + .dynamic = dynamic, + }; + + ether_addr_copy(req.mac, mac); + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_FDB_ADD, + &req.cmd, sizeof(req)); +} + +int prestera_hw_fdb_del(struct prestera_port *port, const unsigned char *mac, + u16 vid) +{ + struct prestera_msg_fdb_req req = { + .port = port->hw_id, + .dev = port->dev_id, + .vid = vid, + }; + + ether_addr_copy(req.mac, mac); + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_FDB_DELETE, + &req.cmd, sizeof(req)); +} + +int prestera_hw_fdb_flush_port(struct prestera_port *port, u32 mode) +{ + struct prestera_msg_fdb_req req = { + .port = port->hw_id, + .dev = port->dev_id, + .flush_mode = mode, + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_FDB_FLUSH_PORT, + &req.cmd, sizeof(req)); +} + +int prestera_hw_fdb_flush_vlan(struct prestera_switch *sw, u16 vid, u32 mode) +{ + struct prestera_msg_fdb_req req = { + .vid = vid, + .flush_mode = mode, + }; + + return prestera_cmd(sw, PRESTERA_CMD_TYPE_FDB_FLUSH_VLAN, + &req.cmd, sizeof(req)); +} + +int prestera_hw_fdb_flush_port_vlan(struct prestera_port *port, u16 vid, + u32 mode) +{ + struct prestera_msg_fdb_req req = { + .port = port->hw_id, + .dev = port->dev_id, + .vid = vid, + .flush_mode = mode, + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_FDB_FLUSH_PORT_VLAN, + &req.cmd, sizeof(req)); +} + +int prestera_hw_bridge_create(struct prestera_switch *sw, u16 *bridge_id) +{ + struct prestera_msg_bridge_resp resp; + struct prestera_msg_bridge_req req; + int err; + + err = prestera_cmd_ret(sw, PRESTERA_CMD_TYPE_BRIDGE_CREATE, + &req.cmd, sizeof(req), + &resp.ret, sizeof(resp)); + if (err) + return err; + + *bridge_id = resp.bridge; + + return 0; +} + +int prestera_hw_bridge_delete(struct prestera_switch *sw, u16 bridge_id) +{ + struct prestera_msg_bridge_req req = { + .bridge = bridge_id, + }; + + return prestera_cmd(sw, PRESTERA_CMD_TYPE_BRIDGE_DELETE, + &req.cmd, sizeof(req)); +} + +int prestera_hw_bridge_port_add(struct prestera_port *port, u16 bridge_id) +{ + struct prestera_msg_bridge_req req = { + .bridge = bridge_id, + .port = port->hw_id, + .dev = port->dev_id, + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_BRIDGE_PORT_ADD, + &req.cmd, sizeof(req)); +} + +int prestera_hw_bridge_port_delete(struct prestera_port *port, u16 bridge_id) +{ + struct prestera_msg_bridge_req req = { + .bridge = bridge_id, + .port = port->hw_id, + .dev = port->dev_id, + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_BRIDGE_PORT_DELETE, + &req.cmd, sizeof(req)); +} + +int prestera_hw_rxtx_init(struct prestera_switch *sw, + struct prestera_rxtx_params *params) +{ + struct prestera_msg_rxtx_resp resp; + struct prestera_msg_rxtx_req req; + int err; + + req.use_sdma = params->use_sdma; + + err = prestera_cmd_ret(sw, PRESTERA_CMD_TYPE_RXTX_INIT, + &req.cmd, sizeof(req), &resp.ret, sizeof(resp)); + if (err) + return err; + + params->map_addr = resp.map_addr; + + return 0; +} + +int prestera_hw_rxtx_port_init(struct prestera_port *port) +{ + struct prestera_msg_rxtx_port_req req = { + .port = port->hw_id, + .dev = port->dev_id, + }; + + return prestera_cmd(port->sw, PRESTERA_CMD_TYPE_RXTX_PORT_INIT, + &req.cmd, sizeof(req)); +} + +int prestera_hw_event_handler_register(struct prestera_switch *sw, + enum prestera_event_type type, + prestera_event_cb_t fn, + void *arg) +{ + struct prestera_fw_event_handler *eh; + + eh = __find_event_handler(sw, type); + if (eh) + return -EEXIST; + + eh = kmalloc(sizeof(*eh), GFP_KERNEL); + if (!eh) + return -ENOMEM; + + eh->type = type; + eh->func = fn; + eh->arg = arg; + + INIT_LIST_HEAD(&eh->list); + + list_add_rcu(&eh->list, &sw->event_handlers); + + return 0; +} + +void prestera_hw_event_handler_unregister(struct prestera_switch *sw, + enum prestera_event_type type, + prestera_event_cb_t fn) +{ + struct prestera_fw_event_handler *eh; + + eh = __find_event_handler(sw, type); + if (!eh) + return; + + list_del_rcu(&eh->list); + kfree_rcu(eh, rcu); +} diff --git a/drivers/net/ethernet/marvell/prestera/prestera_hw.h b/drivers/net/ethernet/marvell/prestera/prestera_hw.h new file mode 100644 index 000000000000..b2b5ac95b4e3 --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/prestera_hw.h @@ -0,0 +1,182 @@ +/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 */ +/* Copyright (c) 2019-2020 Marvell International Ltd. All rights reserved. */ + +#ifndef _PRESTERA_HW_H_ +#define _PRESTERA_HW_H_ + +#include <linux/types.h> + +enum prestera_accept_frm_type { + PRESTERA_ACCEPT_FRAME_TYPE_TAGGED, + PRESTERA_ACCEPT_FRAME_TYPE_UNTAGGED, + PRESTERA_ACCEPT_FRAME_TYPE_ALL, +}; + +enum prestera_fdb_flush_mode { + PRESTERA_FDB_FLUSH_MODE_DYNAMIC = BIT(0), + PRESTERA_FDB_FLUSH_MODE_STATIC = BIT(1), + PRESTERA_FDB_FLUSH_MODE_ALL = PRESTERA_FDB_FLUSH_MODE_DYNAMIC + | PRESTERA_FDB_FLUSH_MODE_STATIC, +}; + +enum { + PRESTERA_LINK_MODE_10baseT_Half, + PRESTERA_LINK_MODE_10baseT_Full, + PRESTERA_LINK_MODE_100baseT_Half, + PRESTERA_LINK_MODE_100baseT_Full, + PRESTERA_LINK_MODE_1000baseT_Half, + PRESTERA_LINK_MODE_1000baseT_Full, + PRESTERA_LINK_MODE_1000baseX_Full, + PRESTERA_LINK_MODE_1000baseKX_Full, + PRESTERA_LINK_MODE_2500baseX_Full, + PRESTERA_LINK_MODE_10GbaseKR_Full, + PRESTERA_LINK_MODE_10GbaseSR_Full, + PRESTERA_LINK_MODE_10GbaseLR_Full, + PRESTERA_LINK_MODE_20GbaseKR2_Full, + PRESTERA_LINK_MODE_25GbaseCR_Full, + PRESTERA_LINK_MODE_25GbaseKR_Full, + PRESTERA_LINK_MODE_25GbaseSR_Full, + PRESTERA_LINK_MODE_40GbaseKR4_Full, + PRESTERA_LINK_MODE_40GbaseCR4_Full, + PRESTERA_LINK_MODE_40GbaseSR4_Full, + PRESTERA_LINK_MODE_50GbaseCR2_Full, + PRESTERA_LINK_MODE_50GbaseKR2_Full, + PRESTERA_LINK_MODE_50GbaseSR2_Full, + PRESTERA_LINK_MODE_100GbaseKR4_Full, + PRESTERA_LINK_MODE_100GbaseSR4_Full, + PRESTERA_LINK_MODE_100GbaseCR4_Full, + + PRESTERA_LINK_MODE_MAX +}; + +enum { + PRESTERA_PORT_TYPE_NONE, + PRESTERA_PORT_TYPE_TP, + PRESTERA_PORT_TYPE_AUI, + PRESTERA_PORT_TYPE_MII, + PRESTERA_PORT_TYPE_FIBRE, + PRESTERA_PORT_TYPE_BNC, + PRESTERA_PORT_TYPE_DA, + PRESTERA_PORT_TYPE_OTHER, + + PRESTERA_PORT_TYPE_MAX +}; + +enum { + PRESTERA_PORT_TCVR_COPPER, + PRESTERA_PORT_TCVR_SFP, + + PRESTERA_PORT_TCVR_MAX +}; + +enum { + PRESTERA_PORT_FEC_OFF, + PRESTERA_PORT_FEC_BASER, + PRESTERA_PORT_FEC_RS, + + PRESTERA_PORT_FEC_MAX +}; + +enum { + PRESTERA_PORT_DUPLEX_HALF, + PRESTERA_PORT_DUPLEX_FULL, +}; + +enum { + PRESTERA_STP_DISABLED, + PRESTERA_STP_BLOCK_LISTEN, + PRESTERA_STP_LEARN, + PRESTERA_STP_FORWARD, +}; + +struct prestera_switch; +struct prestera_port; +struct prestera_port_stats; +struct prestera_port_caps; +enum prestera_event_type; +struct prestera_event; + +typedef void (*prestera_event_cb_t) + (struct prestera_switch *sw, struct prestera_event *evt, void *arg); + +struct prestera_rxtx_params; + +/* Switch API */ +int prestera_hw_switch_init(struct prestera_switch *sw); +void prestera_hw_switch_fini(struct prestera_switch *sw); +int prestera_hw_switch_ageing_set(struct prestera_switch *sw, u32 ageing_ms); +int prestera_hw_switch_mac_set(struct prestera_switch *sw, const char *mac); + +/* Port API */ +int prestera_hw_port_info_get(const struct prestera_port *port, + u32 *dev_id, u32 *hw_id, u16 *fp_id); +int prestera_hw_port_state_set(const struct prestera_port *port, + bool admin_state); +int prestera_hw_port_mtu_set(const struct prestera_port *port, u32 mtu); +int prestera_hw_port_mtu_get(const struct prestera_port *port, u32 *mtu); +int prestera_hw_port_mac_set(const struct prestera_port *port, const char *mac); +int prestera_hw_port_mac_get(const struct prestera_port *port, char *mac); +int prestera_hw_port_cap_get(const struct prestera_port *port, + struct prestera_port_caps *caps); +int prestera_hw_port_remote_cap_get(const struct prestera_port *port, + u64 *link_mode_bitmap); +int prestera_hw_port_remote_fc_get(const struct prestera_port *port, + bool *pause, bool *asym_pause); +int prestera_hw_port_type_get(const struct prestera_port *port, u8 *type); +int prestera_hw_port_fec_get(const struct prestera_port *port, u8 *fec); +int prestera_hw_port_fec_set(const struct prestera_port *port, u8 fec); +int prestera_hw_port_autoneg_set(const struct prestera_port *port, + bool autoneg, u64 link_modes, u8 fec); +int prestera_hw_port_autoneg_restart(struct prestera_port *port); +int prestera_hw_port_duplex_get(const struct prestera_port *port, u8 *duplex); +int prestera_hw_port_stats_get(const struct prestera_port *port, + struct prestera_port_stats *stats); +int prestera_hw_port_link_mode_set(const struct prestera_port *port, u32 mode); +int prestera_hw_port_link_mode_get(const struct prestera_port *port, u32 *mode); +int prestera_hw_port_mdix_get(const struct prestera_port *port, u8 *status, + u8 *admin_mode); +int prestera_hw_port_mdix_set(const struct prestera_port *port, u8 mode); +int prestera_hw_port_speed_get(const struct prestera_port *port, u32 *speed); +int prestera_hw_port_learning_set(struct prestera_port *port, bool enable); +int prestera_hw_port_flood_set(struct prestera_port *port, bool flood); +int prestera_hw_port_accept_frm_type(struct prestera_port *port, + enum prestera_accept_frm_type type); +/* Vlan API */ +int prestera_hw_vlan_create(struct prestera_switch *sw, u16 vid); +int prestera_hw_vlan_delete(struct prestera_switch *sw, u16 vid); +int prestera_hw_vlan_port_set(struct prestera_port *port, u16 vid, + bool is_member, bool untagged); +int prestera_hw_vlan_port_vid_set(struct prestera_port *port, u16 vid); +int prestera_hw_vlan_port_stp_set(struct prestera_port *port, u16 vid, u8 state); + +/* FDB API */ +int prestera_hw_fdb_add(struct prestera_port *port, const unsigned char *mac, + u16 vid, bool dynamic); +int prestera_hw_fdb_del(struct prestera_port *port, const unsigned char *mac, + u16 vid); +int prestera_hw_fdb_flush_port(struct prestera_port *port, u32 mode); +int prestera_hw_fdb_flush_vlan(struct prestera_switch *sw, u16 vid, u32 mode); +int prestera_hw_fdb_flush_port_vlan(struct prestera_port *port, u16 vid, + u32 mode); + +/* Bridge API */ +int prestera_hw_bridge_create(struct prestera_switch *sw, u16 *bridge_id); +int prestera_hw_bridge_delete(struct prestera_switch *sw, u16 bridge_id); +int prestera_hw_bridge_port_add(struct prestera_port *port, u16 bridge_id); +int prestera_hw_bridge_port_delete(struct prestera_port *port, u16 bridge_id); + +/* Event handlers */ +int prestera_hw_event_handler_register(struct prestera_switch *sw, + enum prestera_event_type type, + prestera_event_cb_t fn, + void *arg); +void prestera_hw_event_handler_unregister(struct prestera_switch *sw, + enum prestera_event_type type, + prestera_event_cb_t fn); + +/* RX/TX */ +int prestera_hw_rxtx_init(struct prestera_switch *sw, + struct prestera_rxtx_params *params); +int prestera_hw_rxtx_port_init(struct prestera_port *port); + +#endif /* _PRESTERA_HW_H_ */ diff --git a/drivers/net/ethernet/marvell/prestera/prestera_main.c b/drivers/net/ethernet/marvell/prestera/prestera_main.c new file mode 100644 index 000000000000..0f20e0788cce --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/prestera_main.c @@ -0,0 +1,667 @@ +// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 +/* Copyright (c) 2019-2020 Marvell International Ltd. All rights reserved */ + +#include <linux/etherdevice.h> +#include <linux/jiffies.h> +#include <linux/list.h> +#include <linux/module.h> +#include <linux/netdev_features.h> +#include <linux/of.h> +#include <linux/of_net.h> + +#include "prestera.h" +#include "prestera_hw.h" +#include "prestera_rxtx.h" +#include "prestera_devlink.h" +#include "prestera_ethtool.h" +#include "prestera_switchdev.h" + +#define PRESTERA_MTU_DEFAULT 1536 + +#define PRESTERA_STATS_DELAY_MS 1000 + +#define PRESTERA_MAC_ADDR_NUM_MAX 255 + +static struct workqueue_struct *prestera_wq; + +int prestera_port_pvid_set(struct prestera_port *port, u16 vid) +{ + enum prestera_accept_frm_type frm_type; + int err; + + frm_type = PRESTERA_ACCEPT_FRAME_TYPE_TAGGED; + + if (vid) { + err = prestera_hw_vlan_port_vid_set(port, vid); + if (err) + return err; + + frm_type = PRESTERA_ACCEPT_FRAME_TYPE_ALL; + } + + err = prestera_hw_port_accept_frm_type(port, frm_type); + if (err && frm_type == PRESTERA_ACCEPT_FRAME_TYPE_ALL) + prestera_hw_vlan_port_vid_set(port, port->pvid); + + port->pvid = vid; + return 0; +} + +struct prestera_port *prestera_port_find_by_hwid(struct prestera_switch *sw, + u32 dev_id, u32 hw_id) +{ + struct prestera_port *port = NULL; + + read_lock(&sw->port_list_lock); + list_for_each_entry(port, &sw->port_list, list) { + if (port->dev_id == dev_id && port->hw_id == hw_id) + break; + } + read_unlock(&sw->port_list_lock); + + return port; +} + +struct prestera_port *prestera_find_port(struct prestera_switch *sw, u32 id) +{ + struct prestera_port *port = NULL; + + read_lock(&sw->port_list_lock); + list_for_each_entry(port, &sw->port_list, list) { + if (port->id == id) + break; + } + read_unlock(&sw->port_list_lock); + + return port; +} + +static int prestera_port_open(struct net_device *dev) +{ + struct prestera_port *port = netdev_priv(dev); + int err; + + err = prestera_hw_port_state_set(port, true); + if (err) + return err; + + netif_start_queue(dev); + + return 0; +} + +static int prestera_port_close(struct net_device *dev) +{ + struct prestera_port *port = netdev_priv(dev); + int err; + + netif_stop_queue(dev); + + err = prestera_hw_port_state_set(port, false); + if (err) + return err; + + return 0; +} + +static netdev_tx_t prestera_port_xmit(struct sk_buff *skb, + struct net_device *dev) +{ + return prestera_rxtx_xmit(netdev_priv(dev), skb); +} + +static int prestera_is_valid_mac_addr(struct prestera_port *port, u8 *addr) +{ + if (!is_valid_ether_addr(addr)) + return -EADDRNOTAVAIL; + + /* firmware requires that port's MAC address contains first 5 bytes + * of the base MAC address + */ + if (memcmp(port->sw->base_mac, addr, ETH_ALEN - 1)) + return -EINVAL; + + return 0; +} + +static int prestera_port_set_mac_address(struct net_device *dev, void *p) +{ + struct prestera_port *port = netdev_priv(dev); + struct sockaddr *addr = p; + int err; + + err = prestera_is_valid_mac_addr(port, addr->sa_data); + if (err) + return err; + + err = prestera_hw_port_mac_set(port, addr->sa_data); + if (err) + return err; + + ether_addr_copy(dev->dev_addr, addr->sa_data); + + return 0; +} + +static int prestera_port_change_mtu(struct net_device *dev, int mtu) +{ + struct prestera_port *port = netdev_priv(dev); + int err; + + err = prestera_hw_port_mtu_set(port, mtu); + if (err) + return err; + + dev->mtu = mtu; + + return 0; +} + +static void prestera_port_get_stats64(struct net_device *dev, + struct rtnl_link_stats64 *stats) +{ + struct prestera_port *port = netdev_priv(dev); + struct prestera_port_stats *port_stats = &port->cached_hw_stats.stats; + + stats->rx_packets = port_stats->broadcast_frames_received + + port_stats->multicast_frames_received + + port_stats->unicast_frames_received; + + stats->tx_packets = port_stats->broadcast_frames_sent + + port_stats->multicast_frames_sent + + port_stats->unicast_frames_sent; + + stats->rx_bytes = port_stats->good_octets_received; + + stats->tx_bytes = port_stats->good_octets_sent; + + stats->rx_errors = port_stats->rx_error_frame_received; + stats->tx_errors = port_stats->mac_trans_error; + + stats->rx_dropped = port_stats->buffer_overrun; + stats->tx_dropped = 0; + + stats->multicast = port_stats->multicast_frames_received; + stats->collisions = port_stats->excessive_collision; + + stats->rx_crc_errors = port_stats->bad_crc; +} + +static void prestera_port_get_hw_stats(struct prestera_port *port) +{ + prestera_hw_port_stats_get(port, &port->cached_hw_stats.stats); +} + +static void prestera_port_stats_update(struct work_struct *work) +{ + struct prestera_port *port = + container_of(work, struct prestera_port, + cached_hw_stats.caching_dw.work); + + prestera_port_get_hw_stats(port); + + queue_delayed_work(prestera_wq, &port->cached_hw_stats.caching_dw, + msecs_to_jiffies(PRESTERA_STATS_DELAY_MS)); +} + +static const struct net_device_ops prestera_netdev_ops = { + .ndo_open = prestera_port_open, + .ndo_stop = prestera_port_close, + .ndo_start_xmit = prestera_port_xmit, + .ndo_change_mtu = prestera_port_change_mtu, + .ndo_get_stats64 = prestera_port_get_stats64, + .ndo_set_mac_address = prestera_port_set_mac_address, + .ndo_get_devlink_port = prestera_devlink_get_port, +}; + +int prestera_port_autoneg_set(struct prestera_port *port, bool enable, + u64 adver_link_modes, u8 adver_fec) +{ + bool refresh = false; + u64 link_modes; + int err; + u8 fec; + + if (port->caps.type != PRESTERA_PORT_TYPE_TP) + return enable ? -EINVAL : 0; + + if (!enable) + goto set_autoneg; + + link_modes = port->caps.supp_link_modes & adver_link_modes; + fec = port->caps.supp_fec & adver_fec; + + if (!link_modes && !fec) + return -EOPNOTSUPP; + + if (link_modes && port->adver_link_modes != link_modes) { + port->adver_link_modes = link_modes; + refresh = true; + } + + if (fec && port->adver_fec != fec) { + port->adver_fec = fec; + refresh = true; + } + +set_autoneg: + if (port->autoneg == enable && !refresh) + return 0; + + err = prestera_hw_port_autoneg_set(port, enable, port->adver_link_modes, + port->adver_fec); + if (err) + return err; + + port->autoneg = enable; + + return 0; +} + +static void prestera_port_list_add(struct prestera_port *port) +{ + write_lock(&port->sw->port_list_lock); + list_add(&port->list, &port->sw->port_list); + write_unlock(&port->sw->port_list_lock); +} + +static void prestera_port_list_del(struct prestera_port *port) +{ + write_lock(&port->sw->port_list_lock); + list_del(&port->list); + write_unlock(&port->sw->port_list_lock); +} + +static int prestera_port_create(struct prestera_switch *sw, u32 id) +{ + struct prestera_port *port; + struct net_device *dev; + int err; + + dev = alloc_etherdev(sizeof(*port)); + if (!dev) + return -ENOMEM; + + port = netdev_priv(dev); + + INIT_LIST_HEAD(&port->vlans_list); + port->pvid = PRESTERA_DEFAULT_VID; + port->dev = dev; + port->id = id; + port->sw = sw; + + err = prestera_hw_port_info_get(port, &port->dev_id, &port->hw_id, + &port->fp_id); + if (err) { + dev_err(prestera_dev(sw), "Failed to get port(%u) info\n", id); + goto err_port_info_get; + } + + err = prestera_devlink_port_register(port); + if (err) + goto err_dl_port_register; + + dev->features |= NETIF_F_NETNS_LOCAL; + dev->netdev_ops = &prestera_netdev_ops; + dev->ethtool_ops = &prestera_ethtool_ops; + + netif_carrier_off(dev); + + dev->mtu = min_t(unsigned int, sw->mtu_max, PRESTERA_MTU_DEFAULT); + dev->min_mtu = sw->mtu_min; + dev->max_mtu = sw->mtu_max; + + err = prestera_hw_port_mtu_set(port, dev->mtu); + if (err) { + dev_err(prestera_dev(sw), "Failed to set port(%u) mtu(%d)\n", + id, dev->mtu); + goto err_port_init; + } + + if (port->fp_id >= PRESTERA_MAC_ADDR_NUM_MAX) + goto err_port_init; + + /* firmware requires that port's MAC address consist of the first + * 5 bytes of the base MAC address + */ + memcpy(dev->dev_addr, sw->base_mac, dev->addr_len - 1); + dev->dev_addr[dev->addr_len - 1] = port->fp_id; + + err = prestera_hw_port_mac_set(port, dev->dev_addr); + if (err) { + dev_err(prestera_dev(sw), "Failed to set port(%u) mac addr\n", id); + goto err_port_init; + } + + err = prestera_hw_port_cap_get(port, &port->caps); + if (err) { + dev_err(prestera_dev(sw), "Failed to get port(%u) caps\n", id); + goto err_port_init; + } + + port->adver_fec = BIT(PRESTERA_PORT_FEC_OFF); + prestera_port_autoneg_set(port, true, port->caps.supp_link_modes, + port->caps.supp_fec); + + err = prestera_hw_port_state_set(port, false); + if (err) { + dev_err(prestera_dev(sw), "Failed to set port(%u) down\n", id); + goto err_port_init; + } + + err = prestera_rxtx_port_init(port); + if (err) + goto err_port_init; + + INIT_DELAYED_WORK(&port->cached_hw_stats.caching_dw, + &prestera_port_stats_update); + + prestera_port_list_add(port); + + err = register_netdev(dev); + if (err) + goto err_register_netdev; + + prestera_devlink_port_set(port); + + return 0; + +err_register_netdev: + prestera_port_list_del(port); +err_port_init: + prestera_devlink_port_unregister(port); +err_dl_port_register: +err_port_info_get: + free_netdev(dev); + return err; +} + +static void prestera_port_destroy(struct prestera_port *port) +{ + struct net_device *dev = port->dev; + + cancel_delayed_work_sync(&port->cached_hw_stats.caching_dw); + prestera_devlink_port_clear(port); + unregister_netdev(dev); + prestera_port_list_del(port); + prestera_devlink_port_unregister(port); + free_netdev(dev); +} + +static void prestera_destroy_ports(struct prestera_switch *sw) +{ + struct prestera_port *port, *tmp; + + list_for_each_entry_safe(port, tmp, &sw->port_list, list) + prestera_port_destroy(port); +} + +static int prestera_create_ports(struct prestera_switch *sw) +{ + struct prestera_port *port, *tmp; + u32 port_idx; + int err; + + for (port_idx = 0; port_idx < sw->port_count; port_idx++) { + err = prestera_port_create(sw, port_idx); + if (err) + goto err_port_create; + } + + return 0; + +err_port_create: + list_for_each_entry_safe(port, tmp, &sw->port_list, list) + prestera_port_destroy(port); + + return err; +} + +static void prestera_port_handle_event(struct prestera_switch *sw, + struct prestera_event *evt, void *arg) +{ + struct delayed_work *caching_dw; + struct prestera_port *port; + + port = prestera_find_port(sw, evt->port_evt.port_id); + if (!port || !port->dev) + return; + + caching_dw = &port->cached_hw_stats.caching_dw; + + if (evt->id == PRESTERA_PORT_EVENT_STATE_CHANGED) { + if (evt->port_evt.data.oper_state) { + netif_carrier_on(port->dev); + if (!delayed_work_pending(caching_dw)) + queue_delayed_work(prestera_wq, caching_dw, 0); + } else { + netif_carrier_off(port->dev); + if (delayed_work_pending(caching_dw)) + cancel_delayed_work(caching_dw); + } + } +} + +static int prestera_event_handlers_register(struct prestera_switch *sw) +{ + return prestera_hw_event_handler_register(sw, PRESTERA_EVENT_TYPE_PORT, + prestera_port_handle_event, + NULL); +} + +static void prestera_event_handlers_unregister(struct prestera_switch *sw) +{ + prestera_hw_event_handler_unregister(sw, PRESTERA_EVENT_TYPE_PORT, + prestera_port_handle_event); +} + +static int prestera_switch_set_base_mac_addr(struct prestera_switch *sw) +{ + struct device_node *base_mac_np; + struct device_node *np; + const char *base_mac; + + np = of_find_compatible_node(NULL, NULL, "marvell,prestera"); + base_mac_np = of_parse_phandle(np, "base-mac-provider", 0); + + base_mac = of_get_mac_address(base_mac_np); + of_node_put(base_mac_np); + if (!IS_ERR(base_mac)) + ether_addr_copy(sw->base_mac, base_mac); + + if (!is_valid_ether_addr(sw->base_mac)) { + eth_random_addr(sw->base_mac); + dev_info(prestera_dev(sw), "using random base mac address\n"); + } + + return prestera_hw_switch_mac_set(sw, sw->base_mac); +} + +bool prestera_netdev_check(const struct net_device *dev) +{ + return dev->netdev_ops == &prestera_netdev_ops; +} + +static int prestera_lower_dev_walk(struct net_device *dev, + struct netdev_nested_priv *priv) +{ + struct prestera_port **pport = (struct prestera_port **)priv->data; + + if (prestera_netdev_check(dev)) { + *pport = netdev_priv(dev); + return 1; + } + + return 0; +} + +struct prestera_port *prestera_port_dev_lower_find(struct net_device *dev) +{ + struct prestera_port *port = NULL; + struct netdev_nested_priv priv = { + .data = (void *)&port, + }; + + if (prestera_netdev_check(dev)) + return netdev_priv(dev); + + netdev_walk_all_lower_dev(dev, prestera_lower_dev_walk, &priv); + + return port; +} + +static int prestera_netdev_port_event(struct net_device *dev, + unsigned long event, void *ptr) +{ + switch (event) { + case NETDEV_PRECHANGEUPPER: + case NETDEV_CHANGEUPPER: + return prestera_bridge_port_event(dev, event, ptr); + default: + return 0; + } +} + +static int prestera_netdev_event_handler(struct notifier_block *nb, + unsigned long event, void *ptr) +{ + struct net_device *dev = netdev_notifier_info_to_dev(ptr); + int err = 0; + + if (prestera_netdev_check(dev)) + err = prestera_netdev_port_event(dev, event, ptr); + + return notifier_from_errno(err); +} + +static int prestera_netdev_event_handler_register(struct prestera_switch *sw) +{ + sw->netdev_nb.notifier_call = prestera_netdev_event_handler; + + return register_netdevice_notifier(&sw->netdev_nb); +} + +static void prestera_netdev_event_handler_unregister(struct prestera_switch *sw) +{ + unregister_netdevice_notifier(&sw->netdev_nb); +} + +static int prestera_switch_init(struct prestera_switch *sw) +{ + int err; + + err = prestera_hw_switch_init(sw); + if (err) { + dev_err(prestera_dev(sw), "Failed to init Switch device\n"); + return err; + } + + rwlock_init(&sw->port_list_lock); + INIT_LIST_HEAD(&sw->port_list); + + err = prestera_switch_set_base_mac_addr(sw); + if (err) + return err; + + err = prestera_netdev_event_handler_register(sw); + if (err) + return err; + + err = prestera_switchdev_init(sw); + if (err) + goto err_swdev_register; + + err = prestera_rxtx_switch_init(sw); + if (err) + goto err_rxtx_register; + + err = prestera_event_handlers_register(sw); + if (err) + goto err_handlers_register; + + err = prestera_devlink_register(sw); + if (err) + goto err_dl_register; + + err = prestera_create_ports(sw); + if (err) + goto err_ports_create; + + return 0; + +err_ports_create: + prestera_devlink_unregister(sw); +err_dl_register: + prestera_event_handlers_unregister(sw); +err_handlers_register: + prestera_rxtx_switch_fini(sw); +err_rxtx_register: + prestera_switchdev_fini(sw); +err_swdev_register: + prestera_netdev_event_handler_unregister(sw); + prestera_hw_switch_fini(sw); + + return err; +} + +static void prestera_switch_fini(struct prestera_switch *sw) +{ + prestera_destroy_ports(sw); + prestera_devlink_unregister(sw); + prestera_event_handlers_unregister(sw); + prestera_rxtx_switch_fini(sw); + prestera_switchdev_fini(sw); + prestera_netdev_event_handler_unregister(sw); + prestera_hw_switch_fini(sw); +} + +int prestera_device_register(struct prestera_device *dev) +{ + struct prestera_switch *sw; + int err; + + sw = prestera_devlink_alloc(); + if (!sw) + return -ENOMEM; + + dev->priv = sw; + sw->dev = dev; + + err = prestera_switch_init(sw); + if (err) { + prestera_devlink_free(sw); + return err; + } + + return 0; +} +EXPORT_SYMBOL(prestera_device_register); + +void prestera_device_unregister(struct prestera_device *dev) +{ + struct prestera_switch *sw = dev->priv; + + prestera_switch_fini(sw); + prestera_devlink_free(sw); +} +EXPORT_SYMBOL(prestera_device_unregister); + +static int __init prestera_module_init(void) +{ + prestera_wq = alloc_workqueue("prestera", 0, 0); + if (!prestera_wq) + return -ENOMEM; + + return 0; +} + +static void __exit prestera_module_exit(void) +{ + destroy_workqueue(prestera_wq); +} + +module_init(prestera_module_init); +module_exit(prestera_module_exit); + +MODULE_LICENSE("Dual BSD/GPL"); +MODULE_DESCRIPTION("Marvell Prestera switch driver"); diff --git a/drivers/net/ethernet/marvell/prestera/prestera_pci.c b/drivers/net/ethernet/marvell/prestera/prestera_pci.c new file mode 100644 index 000000000000..1b97adae542e --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/prestera_pci.c @@ -0,0 +1,769 @@ +// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 +/* Copyright (c) 2019-2020 Marvell International Ltd. All rights reserved */ + +#include <linux/circ_buf.h> +#include <linux/device.h> +#include <linux/firmware.h> +#include <linux/iopoll.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/pci.h> + +#include "prestera.h" + +#define PRESTERA_MSG_MAX_SIZE 1500 + +#define PRESTERA_SUPP_FW_MAJ_VER 2 +#define PRESTERA_SUPP_FW_MIN_VER 0 + +#define PRESTERA_FW_PATH_FMT "mrvl/prestera/mvsw_prestera_fw-v%u.%u.img" + +#define PRESTERA_FW_HDR_MAGIC 0x351D9D06 +#define PRESTERA_FW_DL_TIMEOUT_MS 50000 +#define PRESTERA_FW_BLK_SZ 1024 + +#define PRESTERA_FW_VER_MAJ_MUL 1000000 +#define PRESTERA_FW_VER_MIN_MUL 1000 + +#define PRESTERA_FW_VER_MAJ(v) ((v) / PRESTERA_FW_VER_MAJ_MUL) + +#define PRESTERA_FW_VER_MIN(v) \ + (((v) - (PRESTERA_FW_VER_MAJ(v) * PRESTERA_FW_VER_MAJ_MUL)) / \ + PRESTERA_FW_VER_MIN_MUL) + +#define PRESTERA_FW_VER_PATCH(v) \ + ((v) - (PRESTERA_FW_VER_MAJ(v) * PRESTERA_FW_VER_MAJ_MUL) - \ + (PRESTERA_FW_VER_MIN(v) * PRESTERA_FW_VER_MIN_MUL)) + +enum prestera_pci_bar_t { + PRESTERA_PCI_BAR_FW = 2, + PRESTERA_PCI_BAR_PP = 4, +}; + +struct prestera_fw_header { + __be32 magic_number; + __be32 version_value; + u8 reserved[8]; +}; + +struct prestera_ldr_regs { + u32 ldr_ready; + u32 pad1; + + u32 ldr_img_size; + u32 ldr_ctl_flags; + + u32 ldr_buf_offs; + u32 ldr_buf_size; + + u32 ldr_buf_rd; + u32 pad2; + u32 ldr_buf_wr; + + u32 ldr_status; +}; + +#define PRESTERA_LDR_REG_OFFSET(f) offsetof(struct prestera_ldr_regs, f) + +#define PRESTERA_LDR_READY_MAGIC 0xf00dfeed + +#define PRESTERA_LDR_STATUS_IMG_DL BIT(0) +#define PRESTERA_LDR_STATUS_START_FW BIT(1) +#define PRESTERA_LDR_STATUS_INVALID_IMG BIT(2) +#define PRESTERA_LDR_STATUS_NOMEM BIT(3) + +#define PRESTERA_LDR_REG_BASE(fw) ((fw)->ldr_regs) +#define PRESTERA_LDR_REG_ADDR(fw, reg) (PRESTERA_LDR_REG_BASE(fw) + (reg)) + +/* fw loader registers */ +#define PRESTERA_LDR_READY_REG PRESTERA_LDR_REG_OFFSET(ldr_ready) +#define PRESTERA_LDR_IMG_SIZE_REG PRESTERA_LDR_REG_OFFSET(ldr_img_size) +#define PRESTERA_LDR_CTL_REG PRESTERA_LDR_REG_OFFSET(ldr_ctl_flags) +#define PRESTERA_LDR_BUF_SIZE_REG PRESTERA_LDR_REG_OFFSET(ldr_buf_size) +#define PRESTERA_LDR_BUF_OFFS_REG PRESTERA_LDR_REG_OFFSET(ldr_buf_offs) +#define PRESTERA_LDR_BUF_RD_REG PRESTERA_LDR_REG_OFFSET(ldr_buf_rd) +#define PRESTERA_LDR_BUF_WR_REG PRESTERA_LDR_REG_OFFSET(ldr_buf_wr) +#define PRESTERA_LDR_STATUS_REG PRESTERA_LDR_REG_OFFSET(ldr_status) + +#define PRESTERA_LDR_CTL_DL_START BIT(0) + +#define PRESTERA_EVT_QNUM_MAX 4 + +struct prestera_fw_evtq_regs { + u32 rd_idx; + u32 pad1; + u32 wr_idx; + u32 pad2; + u32 offs; + u32 len; +}; + +struct prestera_fw_regs { + u32 fw_ready; + u32 pad; + u32 cmd_offs; + u32 cmd_len; + u32 evt_offs; + u32 evt_qnum; + + u32 cmd_req_ctl; + u32 cmd_req_len; + u32 cmd_rcv_ctl; + u32 cmd_rcv_len; + + u32 fw_status; + u32 rx_status; + + struct prestera_fw_evtq_regs evtq_list[PRESTERA_EVT_QNUM_MAX]; +}; + +#define PRESTERA_FW_REG_OFFSET(f) offsetof(struct prestera_fw_regs, f) + +#define PRESTERA_FW_READY_MAGIC 0xcafebabe + +/* fw registers */ +#define PRESTERA_FW_READY_REG PRESTERA_FW_REG_OFFSET(fw_ready) + +#define PRESTERA_CMD_BUF_OFFS_REG PRESTERA_FW_REG_OFFSET(cmd_offs) +#define PRESTERA_CMD_BUF_LEN_REG PRESTERA_FW_REG_OFFSET(cmd_len) +#define PRESTERA_EVT_BUF_OFFS_REG PRESTERA_FW_REG_OFFSET(evt_offs) +#define PRESTERA_EVT_QNUM_REG PRESTERA_FW_REG_OFFSET(evt_qnum) + +#define PRESTERA_CMD_REQ_CTL_REG PRESTERA_FW_REG_OFFSET(cmd_req_ctl) +#define PRESTERA_CMD_REQ_LEN_REG PRESTERA_FW_REG_OFFSET(cmd_req_len) + +#define PRESTERA_CMD_RCV_CTL_REG PRESTERA_FW_REG_OFFSET(cmd_rcv_ctl) +#define PRESTERA_CMD_RCV_LEN_REG PRESTERA_FW_REG_OFFSET(cmd_rcv_len) +#define PRESTERA_FW_STATUS_REG PRESTERA_FW_REG_OFFSET(fw_status) +#define PRESTERA_RX_STATUS_REG PRESTERA_FW_REG_OFFSET(rx_status) + +/* PRESTERA_CMD_REQ_CTL_REG flags */ +#define PRESTERA_CMD_F_REQ_SENT BIT(0) +#define PRESTERA_CMD_F_REPL_RCVD BIT(1) + +/* PRESTERA_CMD_RCV_CTL_REG flags */ +#define PRESTERA_CMD_F_REPL_SENT BIT(0) + +#define PRESTERA_EVTQ_REG_OFFSET(q, f) \ + (PRESTERA_FW_REG_OFFSET(evtq_list) + \ + (q) * sizeof(struct prestera_fw_evtq_regs) + \ + offsetof(struct prestera_fw_evtq_regs, f)) + +#define PRESTERA_EVTQ_RD_IDX_REG(q) PRESTERA_EVTQ_REG_OFFSET(q, rd_idx) +#define PRESTERA_EVTQ_WR_IDX_REG(q) PRESTERA_EVTQ_REG_OFFSET(q, wr_idx) +#define PRESTERA_EVTQ_OFFS_REG(q) PRESTERA_EVTQ_REG_OFFSET(q, offs) +#define PRESTERA_EVTQ_LEN_REG(q) PRESTERA_EVTQ_REG_OFFSET(q, len) + +#define PRESTERA_FW_REG_BASE(fw) ((fw)->dev.ctl_regs) +#define PRESTERA_FW_REG_ADDR(fw, reg) PRESTERA_FW_REG_BASE((fw)) + (reg) + +#define PRESTERA_FW_CMD_DEFAULT_WAIT_MS 30000 +#define PRESTERA_FW_READY_WAIT_MS 20000 + +struct prestera_fw_evtq { + u8 __iomem *addr; + size_t len; +}; + +struct prestera_fw { + struct workqueue_struct *wq; + struct prestera_device dev; + u8 __iomem *ldr_regs; + u8 __iomem *ldr_ring_buf; + u32 ldr_buf_len; + u32 ldr_wr_idx; + struct mutex cmd_mtx; /* serialize access to dev->send_req */ + size_t cmd_mbox_len; + u8 __iomem *cmd_mbox; + struct prestera_fw_evtq evt_queue[PRESTERA_EVT_QNUM_MAX]; + u8 evt_qnum; + struct work_struct evt_work; + u8 __iomem *evt_buf; + u8 *evt_msg; +}; + +static int prestera_fw_load(struct prestera_fw *fw); + +static void prestera_fw_write(struct prestera_fw *fw, u32 reg, u32 val) +{ + writel(val, PRESTERA_FW_REG_ADDR(fw, reg)); +} + +static u32 prestera_fw_read(struct prestera_fw *fw, u32 reg) +{ + return readl(PRESTERA_FW_REG_ADDR(fw, reg)); +} + +static u32 prestera_fw_evtq_len(struct prestera_fw *fw, u8 qid) +{ + return fw->evt_queue[qid].len; +} + +static u32 prestera_fw_evtq_avail(struct prestera_fw *fw, u8 qid) +{ + u32 wr_idx = prestera_fw_read(fw, PRESTERA_EVTQ_WR_IDX_REG(qid)); + u32 rd_idx = prestera_fw_read(fw, PRESTERA_EVTQ_RD_IDX_REG(qid)); + + return CIRC_CNT(wr_idx, rd_idx, prestera_fw_evtq_len(fw, qid)); +} + +static void prestera_fw_evtq_rd_set(struct prestera_fw *fw, + u8 qid, u32 idx) +{ + u32 rd_idx = idx & (prestera_fw_evtq_len(fw, qid) - 1); + + prestera_fw_write(fw, PRESTERA_EVTQ_RD_IDX_REG(qid), rd_idx); +} + +static u8 __iomem *prestera_fw_evtq_buf(struct prestera_fw *fw, u8 qid) +{ + return fw->evt_queue[qid].addr; +} + +static u32 prestera_fw_evtq_read32(struct prestera_fw *fw, u8 qid) +{ + u32 rd_idx = prestera_fw_read(fw, PRESTERA_EVTQ_RD_IDX_REG(qid)); + u32 val; + + val = readl(prestera_fw_evtq_buf(fw, qid) + rd_idx); + prestera_fw_evtq_rd_set(fw, qid, rd_idx + 4); + return val; +} + +static ssize_t prestera_fw_evtq_read_buf(struct prestera_fw *fw, + u8 qid, void *buf, size_t len) +{ + u32 idx = prestera_fw_read(fw, PRESTERA_EVTQ_RD_IDX_REG(qid)); + u8 __iomem *evtq_addr = prestera_fw_evtq_buf(fw, qid); + u32 *buf32 = buf; + int i; + + for (i = 0; i < len / 4; buf32++, i++) { + *buf32 = readl_relaxed(evtq_addr + idx); + idx = (idx + 4) & (prestera_fw_evtq_len(fw, qid) - 1); + } + + prestera_fw_evtq_rd_set(fw, qid, idx); + + return i; +} + +static u8 prestera_fw_evtq_pick(struct prestera_fw *fw) +{ + int qid; + + for (qid = 0; qid < fw->evt_qnum; qid++) { + if (prestera_fw_evtq_avail(fw, qid) >= 4) + return qid; + } + + return PRESTERA_EVT_QNUM_MAX; +} + +static void prestera_fw_evt_work_fn(struct work_struct *work) +{ + struct prestera_fw *fw; + void *msg; + u8 qid; + + fw = container_of(work, struct prestera_fw, evt_work); + msg = fw->evt_msg; + + while ((qid = prestera_fw_evtq_pick(fw)) < PRESTERA_EVT_QNUM_MAX) { + u32 idx; + u32 len; + + len = prestera_fw_evtq_read32(fw, qid); + idx = prestera_fw_read(fw, PRESTERA_EVTQ_RD_IDX_REG(qid)); + + WARN_ON(prestera_fw_evtq_avail(fw, qid) < len); + + if (WARN_ON(len > PRESTERA_MSG_MAX_SIZE)) { + prestera_fw_evtq_rd_set(fw, qid, idx + len); + continue; + } + + prestera_fw_evtq_read_buf(fw, qid, msg, len); + + if (fw->dev.recv_msg) + fw->dev.recv_msg(&fw->dev, msg, len); + } +} + +static int prestera_fw_wait_reg32(struct prestera_fw *fw, u32 reg, u32 cmp, + unsigned int waitms) +{ + u8 __iomem *addr = PRESTERA_FW_REG_ADDR(fw, reg); + u32 val; + + return readl_poll_timeout(addr, val, cmp == val, + 1 * USEC_PER_MSEC, waitms * USEC_PER_MSEC); +} + +static int prestera_fw_cmd_send(struct prestera_fw *fw, + void *in_msg, size_t in_size, + void *out_msg, size_t out_size, + unsigned int waitms) +{ + u32 ret_size; + int err; + + if (!waitms) + waitms = PRESTERA_FW_CMD_DEFAULT_WAIT_MS; + + if (ALIGN(in_size, 4) > fw->cmd_mbox_len) + return -EMSGSIZE; + + /* wait for finish previous reply from FW */ + err = prestera_fw_wait_reg32(fw, PRESTERA_CMD_RCV_CTL_REG, 0, 30); + if (err) { + dev_err(fw->dev.dev, "finish reply from FW is timed out\n"); + return err; + } + + prestera_fw_write(fw, PRESTERA_CMD_REQ_LEN_REG, in_size); + memcpy_toio(fw->cmd_mbox, in_msg, in_size); + + prestera_fw_write(fw, PRESTERA_CMD_REQ_CTL_REG, PRESTERA_CMD_F_REQ_SENT); + + /* wait for reply from FW */ + err = prestera_fw_wait_reg32(fw, PRESTERA_CMD_RCV_CTL_REG, + PRESTERA_CMD_F_REPL_SENT, waitms); + if (err) { + dev_err(fw->dev.dev, "reply from FW is timed out\n"); + goto cmd_exit; + } + + ret_size = prestera_fw_read(fw, PRESTERA_CMD_RCV_LEN_REG); + if (ret_size > out_size) { + dev_err(fw->dev.dev, "ret_size (%u) > out_len(%zu)\n", + ret_size, out_size); + err = -EMSGSIZE; + goto cmd_exit; + } + + memcpy_fromio(out_msg, fw->cmd_mbox + in_size, ret_size); + +cmd_exit: + prestera_fw_write(fw, PRESTERA_CMD_REQ_CTL_REG, PRESTERA_CMD_F_REPL_RCVD); + return err; +} + +static int prestera_fw_send_req(struct prestera_device *dev, + void *in_msg, size_t in_size, void *out_msg, + size_t out_size, unsigned int waitms) +{ + struct prestera_fw *fw; + ssize_t ret; + + fw = container_of(dev, struct prestera_fw, dev); + + mutex_lock(&fw->cmd_mtx); + ret = prestera_fw_cmd_send(fw, in_msg, in_size, out_msg, out_size, waitms); + mutex_unlock(&fw->cmd_mtx); + + return ret; +} + +static int prestera_fw_init(struct prestera_fw *fw) +{ + u8 __iomem *base; + int err; + u8 qid; + + fw->dev.send_req = prestera_fw_send_req; + fw->ldr_regs = fw->dev.ctl_regs; + + err = prestera_fw_load(fw); + if (err) + return err; + + err = prestera_fw_wait_reg32(fw, PRESTERA_FW_READY_REG, + PRESTERA_FW_READY_MAGIC, + PRESTERA_FW_READY_WAIT_MS); + if (err) { + dev_err(fw->dev.dev, "FW failed to start\n"); + return err; + } + + base = fw->dev.ctl_regs; + + fw->cmd_mbox = base + prestera_fw_read(fw, PRESTERA_CMD_BUF_OFFS_REG); + fw->cmd_mbox_len = prestera_fw_read(fw, PRESTERA_CMD_BUF_LEN_REG); + mutex_init(&fw->cmd_mtx); + + fw->evt_buf = base + prestera_fw_read(fw, PRESTERA_EVT_BUF_OFFS_REG); + fw->evt_qnum = prestera_fw_read(fw, PRESTERA_EVT_QNUM_REG); + fw->evt_msg = kmalloc(PRESTERA_MSG_MAX_SIZE, GFP_KERNEL); + if (!fw->evt_msg) + return -ENOMEM; + + for (qid = 0; qid < fw->evt_qnum; qid++) { + u32 offs = prestera_fw_read(fw, PRESTERA_EVTQ_OFFS_REG(qid)); + struct prestera_fw_evtq *evtq = &fw->evt_queue[qid]; + + evtq->len = prestera_fw_read(fw, PRESTERA_EVTQ_LEN_REG(qid)); + evtq->addr = fw->evt_buf + offs; + } + + return 0; +} + +static void prestera_fw_uninit(struct prestera_fw *fw) +{ + kfree(fw->evt_msg); +} + +static irqreturn_t prestera_pci_irq_handler(int irq, void *dev_id) +{ + struct prestera_fw *fw = dev_id; + + if (prestera_fw_read(fw, PRESTERA_RX_STATUS_REG)) { + prestera_fw_write(fw, PRESTERA_RX_STATUS_REG, 0); + + if (fw->dev.recv_pkt) + fw->dev.recv_pkt(&fw->dev); + } + + queue_work(fw->wq, &fw->evt_work); + + return IRQ_HANDLED; +} + +static void prestera_ldr_write(struct prestera_fw *fw, u32 reg, u32 val) +{ + writel(val, PRESTERA_LDR_REG_ADDR(fw, reg)); +} + +static u32 prestera_ldr_read(struct prestera_fw *fw, u32 reg) +{ + return readl(PRESTERA_LDR_REG_ADDR(fw, reg)); +} + +static int prestera_ldr_wait_reg32(struct prestera_fw *fw, + u32 reg, u32 cmp, unsigned int waitms) +{ + u8 __iomem *addr = PRESTERA_LDR_REG_ADDR(fw, reg); + u32 val; + + return readl_poll_timeout(addr, val, cmp == val, + 10 * USEC_PER_MSEC, waitms * USEC_PER_MSEC); +} + +static u32 prestera_ldr_wait_buf(struct prestera_fw *fw, size_t len) +{ + u8 __iomem *addr = PRESTERA_LDR_REG_ADDR(fw, PRESTERA_LDR_BUF_RD_REG); + u32 buf_len = fw->ldr_buf_len; + u32 wr_idx = fw->ldr_wr_idx; + u32 rd_idx; + + return readl_poll_timeout(addr, rd_idx, + CIRC_SPACE(wr_idx, rd_idx, buf_len) >= len, + 1 * USEC_PER_MSEC, 100 * USEC_PER_MSEC); +} + +static int prestera_ldr_wait_dl_finish(struct prestera_fw *fw) +{ + u8 __iomem *addr = PRESTERA_LDR_REG_ADDR(fw, PRESTERA_LDR_STATUS_REG); + unsigned long mask = ~(PRESTERA_LDR_STATUS_IMG_DL); + u32 val; + int err; + + err = readl_poll_timeout(addr, val, val & mask, 10 * USEC_PER_MSEC, + PRESTERA_FW_DL_TIMEOUT_MS * USEC_PER_MSEC); + if (err) { + dev_err(fw->dev.dev, "Timeout to load FW img [state=%d]", + prestera_ldr_read(fw, PRESTERA_LDR_STATUS_REG)); + return err; + } + + return 0; +} + +static void prestera_ldr_wr_idx_move(struct prestera_fw *fw, unsigned int n) +{ + fw->ldr_wr_idx = (fw->ldr_wr_idx + (n)) & (fw->ldr_buf_len - 1); +} + +static void prestera_ldr_wr_idx_commit(struct prestera_fw *fw) +{ + prestera_ldr_write(fw, PRESTERA_LDR_BUF_WR_REG, fw->ldr_wr_idx); +} + +static u8 __iomem *prestera_ldr_wr_ptr(struct prestera_fw *fw) +{ + return fw->ldr_ring_buf + fw->ldr_wr_idx; +} + +static int prestera_ldr_send(struct prestera_fw *fw, const u8 *buf, size_t len) +{ + int err; + int i; + + err = prestera_ldr_wait_buf(fw, len); + if (err) { + dev_err(fw->dev.dev, "failed wait for sending firmware\n"); + return err; + } + + for (i = 0; i < len; i += 4) { + writel_relaxed(*(u32 *)(buf + i), prestera_ldr_wr_ptr(fw)); + prestera_ldr_wr_idx_move(fw, 4); + } + + prestera_ldr_wr_idx_commit(fw); + return 0; +} + +static int prestera_ldr_fw_send(struct prestera_fw *fw, + const char *img, u32 fw_size) +{ + u32 status; + u32 pos; + int err; + + err = prestera_ldr_wait_reg32(fw, PRESTERA_LDR_STATUS_REG, + PRESTERA_LDR_STATUS_IMG_DL, + 5 * MSEC_PER_SEC); + if (err) { + dev_err(fw->dev.dev, "Loader is not ready to load image\n"); + return err; + } + + for (pos = 0; pos < fw_size; pos += PRESTERA_FW_BLK_SZ) { + if (pos + PRESTERA_FW_BLK_SZ > fw_size) + break; + + err = prestera_ldr_send(fw, img + pos, PRESTERA_FW_BLK_SZ); + if (err) + return err; + } + + if (pos < fw_size) { + err = prestera_ldr_send(fw, img + pos, fw_size - pos); + if (err) + return err; + } + + err = prestera_ldr_wait_dl_finish(fw); + if (err) + return err; + + status = prestera_ldr_read(fw, PRESTERA_LDR_STATUS_REG); + + switch (status) { + case PRESTERA_LDR_STATUS_INVALID_IMG: + dev_err(fw->dev.dev, "FW img has bad CRC\n"); + return -EINVAL; + case PRESTERA_LDR_STATUS_NOMEM: + dev_err(fw->dev.dev, "Loader has no enough mem\n"); + return -ENOMEM; + } + + return 0; +} + +static void prestera_fw_rev_parse(const struct prestera_fw_header *hdr, + struct prestera_fw_rev *rev) +{ + u32 version = be32_to_cpu(hdr->version_value); + + rev->maj = PRESTERA_FW_VER_MAJ(version); + rev->min = PRESTERA_FW_VER_MIN(version); + rev->sub = PRESTERA_FW_VER_PATCH(version); +} + +static int prestera_fw_rev_check(struct prestera_fw *fw) +{ + struct prestera_fw_rev *rev = &fw->dev.fw_rev; + u16 maj_supp = PRESTERA_SUPP_FW_MAJ_VER; + u16 min_supp = PRESTERA_SUPP_FW_MIN_VER; + + if (rev->maj == maj_supp && rev->min >= min_supp) + return 0; + + dev_err(fw->dev.dev, "Driver supports FW version only '%u.%u.x'", + PRESTERA_SUPP_FW_MAJ_VER, PRESTERA_SUPP_FW_MIN_VER); + + return -EINVAL; +} + +static int prestera_fw_hdr_parse(struct prestera_fw *fw, + const struct firmware *img) +{ + struct prestera_fw_header *hdr = (struct prestera_fw_header *)img->data; + struct prestera_fw_rev *rev = &fw->dev.fw_rev; + u32 magic; + + magic = be32_to_cpu(hdr->magic_number); + if (magic != PRESTERA_FW_HDR_MAGIC) { + dev_err(fw->dev.dev, "FW img hdr magic is invalid"); + return -EINVAL; + } + + prestera_fw_rev_parse(hdr, rev); + + dev_info(fw->dev.dev, "FW version '%u.%u.%u'\n", + rev->maj, rev->min, rev->sub); + + return prestera_fw_rev_check(fw); +} + +static int prestera_fw_load(struct prestera_fw *fw) +{ + size_t hlen = sizeof(struct prestera_fw_header); + const struct firmware *f; + char fw_path[128]; + int err; + + err = prestera_ldr_wait_reg32(fw, PRESTERA_LDR_READY_REG, + PRESTERA_LDR_READY_MAGIC, + 5 * MSEC_PER_SEC); + if (err) { + dev_err(fw->dev.dev, "waiting for FW loader is timed out"); + return err; + } + + fw->ldr_ring_buf = fw->ldr_regs + + prestera_ldr_read(fw, PRESTERA_LDR_BUF_OFFS_REG); + + fw->ldr_buf_len = + prestera_ldr_read(fw, PRESTERA_LDR_BUF_SIZE_REG); + + fw->ldr_wr_idx = 0; + + snprintf(fw_path, sizeof(fw_path), PRESTERA_FW_PATH_FMT, + PRESTERA_SUPP_FW_MAJ_VER, PRESTERA_SUPP_FW_MIN_VER); + + err = request_firmware_direct(&f, fw_path, fw->dev.dev); + if (err) { + dev_err(fw->dev.dev, "failed to request firmware file\n"); + return err; + } + + err = prestera_fw_hdr_parse(fw, f); + if (err) { + dev_err(fw->dev.dev, "FW image header is invalid\n"); + goto out_release; + } + + prestera_ldr_write(fw, PRESTERA_LDR_IMG_SIZE_REG, f->size - hlen); + prestera_ldr_write(fw, PRESTERA_LDR_CTL_REG, PRESTERA_LDR_CTL_DL_START); + + dev_info(fw->dev.dev, "Loading %s ...", fw_path); + + err = prestera_ldr_fw_send(fw, f->data + hlen, f->size - hlen); + +out_release: + release_firmware(f); + return err; +} + +static int prestera_pci_probe(struct pci_dev *pdev, + const struct pci_device_id *id) +{ + const char *driver_name = pdev->driver->name; + struct prestera_fw *fw; + int err; + + err = pcim_enable_device(pdev); + if (err) + return err; + + err = pcim_iomap_regions(pdev, BIT(PRESTERA_PCI_BAR_FW) | + BIT(PRESTERA_PCI_BAR_PP), + pci_name(pdev)); + if (err) + return err; + + if (dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(30))) { + dev_err(&pdev->dev, "fail to set DMA mask\n"); + goto err_dma_mask; + } + + pci_set_master(pdev); + + fw = devm_kzalloc(&pdev->dev, sizeof(*fw), GFP_KERNEL); + if (!fw) { + err = -ENOMEM; + goto err_pci_dev_alloc; + } + + fw->dev.ctl_regs = pcim_iomap_table(pdev)[PRESTERA_PCI_BAR_FW]; + fw->dev.pp_regs = pcim_iomap_table(pdev)[PRESTERA_PCI_BAR_PP]; + fw->dev.dev = &pdev->dev; + + pci_set_drvdata(pdev, fw); + + err = prestera_fw_init(fw); + if (err) + goto err_prestera_fw_init; + + dev_info(fw->dev.dev, "Prestera FW is ready\n"); + + fw->wq = alloc_workqueue("prestera_fw_wq", WQ_HIGHPRI, 1); + if (!fw->wq) + goto err_wq_alloc; + + INIT_WORK(&fw->evt_work, prestera_fw_evt_work_fn); + + err = pci_alloc_irq_vectors(pdev, 1, 1, PCI_IRQ_MSI); + if (err < 0) { + dev_err(&pdev->dev, "MSI IRQ init failed\n"); + goto err_irq_alloc; + } + + err = request_irq(pci_irq_vector(pdev, 0), prestera_pci_irq_handler, + 0, driver_name, fw); + if (err) { + dev_err(&pdev->dev, "fail to request IRQ\n"); + goto err_request_irq; + } + + err = prestera_device_register(&fw->dev); + if (err) + goto err_prestera_dev_register; + + return 0; + +err_prestera_dev_register: + free_irq(pci_irq_vector(pdev, 0), fw); +err_request_irq: + pci_free_irq_vectors(pdev); +err_irq_alloc: + destroy_workqueue(fw->wq); +err_wq_alloc: + prestera_fw_uninit(fw); +err_prestera_fw_init: +err_pci_dev_alloc: +err_dma_mask: + return err; +} + +static void prestera_pci_remove(struct pci_dev *pdev) +{ + struct prestera_fw *fw = pci_get_drvdata(pdev); + + prestera_device_unregister(&fw->dev); + free_irq(pci_irq_vector(pdev, 0), fw); + pci_free_irq_vectors(pdev); + destroy_workqueue(fw->wq); + prestera_fw_uninit(fw); +} + +static const struct pci_device_id prestera_pci_devices[] = { + { PCI_DEVICE(PCI_VENDOR_ID_MARVELL, 0xC804) }, + { } +}; +MODULE_DEVICE_TABLE(pci, prestera_pci_devices); + +static struct pci_driver prestera_pci_driver = { + .name = "Prestera DX", + .id_table = prestera_pci_devices, + .probe = prestera_pci_probe, + .remove = prestera_pci_remove, +}; +module_pci_driver(prestera_pci_driver); + +MODULE_LICENSE("Dual BSD/GPL"); +MODULE_DESCRIPTION("Marvell Prestera switch PCI interface"); diff --git a/drivers/net/ethernet/marvell/prestera/prestera_rxtx.c b/drivers/net/ethernet/marvell/prestera/prestera_rxtx.c new file mode 100644 index 000000000000..2a13c318048c --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/prestera_rxtx.c @@ -0,0 +1,820 @@ +// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 +/* Copyright (c) 2019-2020 Marvell International Ltd. All rights reserved */ + +#include <linux/bitfield.h> +#include <linux/dmapool.h> +#include <linux/etherdevice.h> +#include <linux/if_vlan.h> +#include <linux/of_address.h> +#include <linux/of_device.h> +#include <linux/of.h> +#include <linux/platform_device.h> + +#include "prestera_dsa.h" +#include "prestera.h" +#include "prestera_hw.h" +#include "prestera_rxtx.h" + +#define PRESTERA_SDMA_WAIT_MUL 10 + +struct prestera_sdma_desc { + __le32 word1; + __le32 word2; + __le32 buff; + __le32 next; +} __packed __aligned(16); + +#define PRESTERA_SDMA_BUFF_SIZE_MAX 1544 + +#define PRESTERA_SDMA_RX_DESC_PKT_LEN(desc) \ + ((le32_to_cpu((desc)->word2) >> 16) & GENMASK(13, 0)) + +#define PRESTERA_SDMA_RX_DESC_OWNER(desc) \ + ((le32_to_cpu((desc)->word1) & BIT(31)) >> 31) + +#define PRESTERA_SDMA_RX_DESC_IS_RCVD(desc) \ + (PRESTERA_SDMA_RX_DESC_OWNER(desc) == PRESTERA_SDMA_RX_DESC_CPU_OWN) + +#define PRESTERA_SDMA_RX_DESC_CPU_OWN 0 +#define PRESTERA_SDMA_RX_DESC_DMA_OWN 1 + +#define PRESTERA_SDMA_RX_QUEUE_NUM 8 + +#define PRESTERA_SDMA_RX_DESC_PER_Q 1000 + +#define PRESTERA_SDMA_TX_DESC_PER_Q 1000 +#define PRESTERA_SDMA_TX_MAX_BURST 64 + +#define PRESTERA_SDMA_TX_DESC_OWNER(desc) \ + ((le32_to_cpu((desc)->word1) & BIT(31)) >> 31) + +#define PRESTERA_SDMA_TX_DESC_CPU_OWN 0 +#define PRESTERA_SDMA_TX_DESC_DMA_OWN 1U + +#define PRESTERA_SDMA_TX_DESC_IS_SENT(desc) \ + (PRESTERA_SDMA_TX_DESC_OWNER(desc) == PRESTERA_SDMA_TX_DESC_CPU_OWN) + +#define PRESTERA_SDMA_TX_DESC_LAST BIT(20) +#define PRESTERA_SDMA_TX_DESC_FIRST BIT(21) +#define PRESTERA_SDMA_TX_DESC_CALC_CRC BIT(12) + +#define PRESTERA_SDMA_TX_DESC_SINGLE \ + (PRESTERA_SDMA_TX_DESC_FIRST | PRESTERA_SDMA_TX_DESC_LAST) + +#define PRESTERA_SDMA_TX_DESC_INIT \ + (PRESTERA_SDMA_TX_DESC_SINGLE | PRESTERA_SDMA_TX_DESC_CALC_CRC) + +#define PRESTERA_SDMA_RX_INTR_MASK_REG 0x2814 +#define PRESTERA_SDMA_RX_QUEUE_STATUS_REG 0x2680 +#define PRESTERA_SDMA_RX_QUEUE_DESC_REG(n) (0x260C + (n) * 16) + +#define PRESTERA_SDMA_TX_QUEUE_DESC_REG 0x26C0 +#define PRESTERA_SDMA_TX_QUEUE_START_REG 0x2868 + +struct prestera_sdma_buf { + struct prestera_sdma_desc *desc; + dma_addr_t desc_dma; + struct sk_buff *skb; + dma_addr_t buf_dma; + bool is_used; +}; + +struct prestera_rx_ring { + struct prestera_sdma_buf *bufs; + int next_rx; +}; + +struct prestera_tx_ring { + struct prestera_sdma_buf *bufs; + int next_tx; + int max_burst; + int burst; +}; + +struct prestera_sdma { + struct prestera_rx_ring rx_ring[PRESTERA_SDMA_RX_QUEUE_NUM]; + struct prestera_tx_ring tx_ring; + struct prestera_switch *sw; + struct dma_pool *desc_pool; + struct work_struct tx_work; + struct napi_struct rx_napi; + struct net_device napi_dev; + u32 map_addr; + u64 dma_mask; + /* protect SDMA with concurrrent access from multiple CPUs */ + spinlock_t tx_lock; +}; + +struct prestera_rxtx { + struct prestera_sdma sdma; +}; + +static int prestera_sdma_buf_init(struct prestera_sdma *sdma, + struct prestera_sdma_buf *buf) +{ + struct prestera_sdma_desc *desc; + dma_addr_t dma; + + desc = dma_pool_alloc(sdma->desc_pool, GFP_DMA | GFP_KERNEL, &dma); + if (!desc) + return -ENOMEM; + + buf->buf_dma = DMA_MAPPING_ERROR; + buf->desc_dma = dma; + buf->desc = desc; + buf->skb = NULL; + + return 0; +} + +static u32 prestera_sdma_map(struct prestera_sdma *sdma, dma_addr_t pa) +{ + return sdma->map_addr + pa; +} + +static void prestera_sdma_rx_desc_init(struct prestera_sdma *sdma, + struct prestera_sdma_desc *desc, + dma_addr_t buf) +{ + u32 word = le32_to_cpu(desc->word2); + + u32p_replace_bits(&word, PRESTERA_SDMA_BUFF_SIZE_MAX, GENMASK(15, 0)); + desc->word2 = cpu_to_le32(word); + + desc->buff = cpu_to_le32(prestera_sdma_map(sdma, buf)); + + /* make sure buffer is set before reset the descriptor */ + wmb(); + + desc->word1 = cpu_to_le32(0xA0000000); +} + +static void prestera_sdma_rx_desc_set_next(struct prestera_sdma *sdma, + struct prestera_sdma_desc *desc, + dma_addr_t next) +{ + desc->next = cpu_to_le32(prestera_sdma_map(sdma, next)); +} + +static int prestera_sdma_rx_skb_alloc(struct prestera_sdma *sdma, + struct prestera_sdma_buf *buf) +{ + struct device *dev = sdma->sw->dev->dev; + struct sk_buff *skb; + dma_addr_t dma; + + skb = alloc_skb(PRESTERA_SDMA_BUFF_SIZE_MAX, GFP_DMA | GFP_ATOMIC); + if (!skb) + return -ENOMEM; + + dma = dma_map_single(dev, skb->data, skb->len, DMA_FROM_DEVICE); + if (dma_mapping_error(dev, dma)) + goto err_dma_map; + + if (buf->skb) + dma_unmap_single(dev, buf->buf_dma, buf->skb->len, + DMA_FROM_DEVICE); + + buf->buf_dma = dma; + buf->skb = skb; + + return 0; + +err_dma_map: + kfree_skb(skb); + + return -ENOMEM; +} + +static struct sk_buff *prestera_sdma_rx_skb_get(struct prestera_sdma *sdma, + struct prestera_sdma_buf *buf) +{ + dma_addr_t buf_dma = buf->buf_dma; + struct sk_buff *skb = buf->skb; + u32 len = skb->len; + int err; + + err = prestera_sdma_rx_skb_alloc(sdma, buf); + if (err) { + buf->buf_dma = buf_dma; + buf->skb = skb; + + skb = alloc_skb(skb->len, GFP_ATOMIC); + if (skb) { + skb_put(skb, len); + skb_copy_from_linear_data(buf->skb, skb->data, len); + } + } + + prestera_sdma_rx_desc_init(sdma, buf->desc, buf->buf_dma); + + return skb; +} + +static int prestera_rxtx_process_skb(struct prestera_sdma *sdma, + struct sk_buff *skb) +{ + const struct prestera_port *port; + struct prestera_dsa dsa; + u32 hw_port, dev_id; + int err; + + skb_pull(skb, ETH_HLEN); + + /* ethertype field is part of the dsa header */ + err = prestera_dsa_parse(&dsa, skb->data - ETH_TLEN); + if (err) + return err; + + dev_id = dsa.hw_dev_num; + hw_port = dsa.port_num; + + port = prestera_port_find_by_hwid(sdma->sw, dev_id, hw_port); + if (unlikely(!port)) { + dev_warn_ratelimited(prestera_dev(sdma->sw), "received pkt for non-existent port(%u, %u)\n", + dev_id, hw_port); + return -ENOENT; + } + + if (unlikely(!pskb_may_pull(skb, PRESTERA_DSA_HLEN))) + return -EINVAL; + + /* remove DSA tag and update checksum */ + skb_pull_rcsum(skb, PRESTERA_DSA_HLEN); + + memmove(skb->data - ETH_HLEN, skb->data - ETH_HLEN - PRESTERA_DSA_HLEN, + ETH_ALEN * 2); + + skb_push(skb, ETH_HLEN); + + skb->protocol = eth_type_trans(skb, port->dev); + + if (dsa.vlan.is_tagged) { + u16 tci = dsa.vlan.vid & VLAN_VID_MASK; + + tci |= dsa.vlan.vpt << VLAN_PRIO_SHIFT; + if (dsa.vlan.cfi_bit) + tci |= VLAN_CFI_MASK; + + __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), tci); + } + + return 0; +} + +static int prestera_sdma_next_rx_buf_idx(int buf_idx) +{ + return (buf_idx + 1) % PRESTERA_SDMA_RX_DESC_PER_Q; +} + +static int prestera_sdma_rx_poll(struct napi_struct *napi, int budget) +{ + int qnum = PRESTERA_SDMA_RX_QUEUE_NUM; + unsigned int rxq_done_map = 0; + struct prestera_sdma *sdma; + struct list_head rx_list; + unsigned int qmask; + int pkts_done = 0; + int q; + + qnum = PRESTERA_SDMA_RX_QUEUE_NUM; + qmask = GENMASK(qnum - 1, 0); + + INIT_LIST_HEAD(&rx_list); + + sdma = container_of(napi, struct prestera_sdma, rx_napi); + + while (pkts_done < budget && rxq_done_map != qmask) { + for (q = 0; q < qnum && pkts_done < budget; q++) { + struct prestera_rx_ring *ring = &sdma->rx_ring[q]; + struct prestera_sdma_desc *desc; + struct prestera_sdma_buf *buf; + int buf_idx = ring->next_rx; + struct sk_buff *skb; + + buf = &ring->bufs[buf_idx]; + desc = buf->desc; + + if (PRESTERA_SDMA_RX_DESC_IS_RCVD(desc)) { + rxq_done_map &= ~BIT(q); + } else { + rxq_done_map |= BIT(q); + continue; + } + + pkts_done++; + + __skb_trim(buf->skb, PRESTERA_SDMA_RX_DESC_PKT_LEN(desc)); + + skb = prestera_sdma_rx_skb_get(sdma, buf); + if (!skb) + goto rx_next_buf; + + if (unlikely(prestera_rxtx_process_skb(sdma, skb))) + goto rx_next_buf; + + list_add_tail(&skb->list, &rx_list); +rx_next_buf: + ring->next_rx = prestera_sdma_next_rx_buf_idx(buf_idx); + } + } + + if (pkts_done < budget && napi_complete_done(napi, pkts_done)) + prestera_write(sdma->sw, PRESTERA_SDMA_RX_INTR_MASK_REG, + GENMASK(9, 2)); + + netif_receive_skb_list(&rx_list); + + return pkts_done; +} + +static void prestera_sdma_rx_fini(struct prestera_sdma *sdma) +{ + int qnum = PRESTERA_SDMA_RX_QUEUE_NUM; + int q, b; + + /* disable all rx queues */ + prestera_write(sdma->sw, PRESTERA_SDMA_RX_QUEUE_STATUS_REG, + GENMASK(15, 8)); + + for (q = 0; q < qnum; q++) { + struct prestera_rx_ring *ring = &sdma->rx_ring[q]; + + if (!ring->bufs) + break; + + for (b = 0; b < PRESTERA_SDMA_RX_DESC_PER_Q; b++) { + struct prestera_sdma_buf *buf = &ring->bufs[b]; + + if (buf->desc_dma) + dma_pool_free(sdma->desc_pool, buf->desc, + buf->desc_dma); + + if (!buf->skb) + continue; + + if (buf->buf_dma != DMA_MAPPING_ERROR) + dma_unmap_single(sdma->sw->dev->dev, + buf->buf_dma, buf->skb->len, + DMA_FROM_DEVICE); + kfree_skb(buf->skb); + } + } +} + +static int prestera_sdma_rx_init(struct prestera_sdma *sdma) +{ + int bnum = PRESTERA_SDMA_RX_DESC_PER_Q; + int qnum = PRESTERA_SDMA_RX_QUEUE_NUM; + int err; + int q; + + /* disable all rx queues */ + prestera_write(sdma->sw, PRESTERA_SDMA_RX_QUEUE_STATUS_REG, + GENMASK(15, 8)); + + for (q = 0; q < qnum; q++) { + struct prestera_sdma_buf *head, *tail, *next, *prev; + struct prestera_rx_ring *ring = &sdma->rx_ring[q]; + + ring->bufs = kmalloc_array(bnum, sizeof(*head), GFP_KERNEL); + if (!ring->bufs) + return -ENOMEM; + + ring->next_rx = 0; + + tail = &ring->bufs[bnum - 1]; + head = &ring->bufs[0]; + next = head; + prev = next; + + do { + err = prestera_sdma_buf_init(sdma, next); + if (err) + return err; + + err = prestera_sdma_rx_skb_alloc(sdma, next); + if (err) + return err; + + prestera_sdma_rx_desc_init(sdma, next->desc, + next->buf_dma); + + prestera_sdma_rx_desc_set_next(sdma, prev->desc, + next->desc_dma); + + prev = next; + next++; + } while (prev != tail); + + /* join tail with head to make a circular list */ + prestera_sdma_rx_desc_set_next(sdma, tail->desc, head->desc_dma); + + prestera_write(sdma->sw, PRESTERA_SDMA_RX_QUEUE_DESC_REG(q), + prestera_sdma_map(sdma, head->desc_dma)); + } + + /* make sure all rx descs are filled before enabling all rx queues */ + wmb(); + + prestera_write(sdma->sw, PRESTERA_SDMA_RX_QUEUE_STATUS_REG, + GENMASK(7, 0)); + + return 0; +} + +static void prestera_sdma_tx_desc_init(struct prestera_sdma *sdma, + struct prestera_sdma_desc *desc) +{ + desc->word1 = cpu_to_le32(PRESTERA_SDMA_TX_DESC_INIT); + desc->word2 = 0; +} + +static void prestera_sdma_tx_desc_set_next(struct prestera_sdma *sdma, + struct prestera_sdma_desc *desc, + dma_addr_t next) +{ + desc->next = cpu_to_le32(prestera_sdma_map(sdma, next)); +} + +static void prestera_sdma_tx_desc_set_buf(struct prestera_sdma *sdma, + struct prestera_sdma_desc *desc, + dma_addr_t buf, size_t len) +{ + u32 word = le32_to_cpu(desc->word2); + + u32p_replace_bits(&word, len + ETH_FCS_LEN, GENMASK(30, 16)); + + desc->buff = cpu_to_le32(prestera_sdma_map(sdma, buf)); + desc->word2 = cpu_to_le32(word); +} + +static void prestera_sdma_tx_desc_xmit(struct prestera_sdma_desc *desc) +{ + u32 word = le32_to_cpu(desc->word1); + + word |= PRESTERA_SDMA_TX_DESC_DMA_OWN << 31; + + /* make sure everything is written before enable xmit */ + wmb(); + + desc->word1 = cpu_to_le32(word); +} + +static int prestera_sdma_tx_buf_map(struct prestera_sdma *sdma, + struct prestera_sdma_buf *buf, + struct sk_buff *skb) +{ + struct device *dma_dev = sdma->sw->dev->dev; + dma_addr_t dma; + + dma = dma_map_single(dma_dev, skb->data, skb->len, DMA_TO_DEVICE); + if (dma_mapping_error(dma_dev, dma)) + return -ENOMEM; + + buf->buf_dma = dma; + buf->skb = skb; + + return 0; +} + +static void prestera_sdma_tx_buf_unmap(struct prestera_sdma *sdma, + struct prestera_sdma_buf *buf) +{ + struct device *dma_dev = sdma->sw->dev->dev; + + dma_unmap_single(dma_dev, buf->buf_dma, buf->skb->len, DMA_TO_DEVICE); +} + +static void prestera_sdma_tx_recycle_work_fn(struct work_struct *work) +{ + int bnum = PRESTERA_SDMA_TX_DESC_PER_Q; + struct prestera_tx_ring *tx_ring; + struct prestera_sdma *sdma; + int b; + + sdma = container_of(work, struct prestera_sdma, tx_work); + + tx_ring = &sdma->tx_ring; + + for (b = 0; b < bnum; b++) { + struct prestera_sdma_buf *buf = &tx_ring->bufs[b]; + + if (!buf->is_used) + continue; + + if (!PRESTERA_SDMA_TX_DESC_IS_SENT(buf->desc)) + continue; + + prestera_sdma_tx_buf_unmap(sdma, buf); + dev_consume_skb_any(buf->skb); + buf->skb = NULL; + + /* make sure everything is cleaned up */ + wmb(); + + buf->is_used = false; + } +} + +static int prestera_sdma_tx_init(struct prestera_sdma *sdma) +{ + struct prestera_sdma_buf *head, *tail, *next, *prev; + struct prestera_tx_ring *tx_ring = &sdma->tx_ring; + int bnum = PRESTERA_SDMA_TX_DESC_PER_Q; + int err; + + INIT_WORK(&sdma->tx_work, prestera_sdma_tx_recycle_work_fn); + spin_lock_init(&sdma->tx_lock); + + tx_ring->bufs = kmalloc_array(bnum, sizeof(*head), GFP_KERNEL); + if (!tx_ring->bufs) + return -ENOMEM; + + tail = &tx_ring->bufs[bnum - 1]; + head = &tx_ring->bufs[0]; + next = head; + prev = next; + + tx_ring->max_burst = PRESTERA_SDMA_TX_MAX_BURST; + tx_ring->burst = tx_ring->max_burst; + tx_ring->next_tx = 0; + + do { + err = prestera_sdma_buf_init(sdma, next); + if (err) + return err; + + next->is_used = false; + + prestera_sdma_tx_desc_init(sdma, next->desc); + + prestera_sdma_tx_desc_set_next(sdma, prev->desc, + next->desc_dma); + + prev = next; + next++; + } while (prev != tail); + + /* join tail with head to make a circular list */ + prestera_sdma_tx_desc_set_next(sdma, tail->desc, head->desc_dma); + + /* make sure descriptors are written */ + wmb(); + + prestera_write(sdma->sw, PRESTERA_SDMA_TX_QUEUE_DESC_REG, + prestera_sdma_map(sdma, head->desc_dma)); + + return 0; +} + +static void prestera_sdma_tx_fini(struct prestera_sdma *sdma) +{ + struct prestera_tx_ring *ring = &sdma->tx_ring; + int bnum = PRESTERA_SDMA_TX_DESC_PER_Q; + int b; + + cancel_work_sync(&sdma->tx_work); + + if (!ring->bufs) + return; + + for (b = 0; b < bnum; b++) { + struct prestera_sdma_buf *buf = &ring->bufs[b]; + + if (buf->desc) + dma_pool_free(sdma->desc_pool, buf->desc, + buf->desc_dma); + + if (!buf->skb) + continue; + + dma_unmap_single(sdma->sw->dev->dev, buf->buf_dma, + buf->skb->len, DMA_TO_DEVICE); + + dev_consume_skb_any(buf->skb); + } +} + +static void prestera_rxtx_handle_event(struct prestera_switch *sw, + struct prestera_event *evt, + void *arg) +{ + struct prestera_sdma *sdma = arg; + + if (evt->id != PRESTERA_RXTX_EVENT_RCV_PKT) + return; + + prestera_write(sdma->sw, PRESTERA_SDMA_RX_INTR_MASK_REG, 0); + napi_schedule(&sdma->rx_napi); +} + +static int prestera_sdma_switch_init(struct prestera_switch *sw) +{ + struct prestera_sdma *sdma = &sw->rxtx->sdma; + struct device *dev = sw->dev->dev; + struct prestera_rxtx_params p; + int err; + + p.use_sdma = true; + + err = prestera_hw_rxtx_init(sw, &p); + if (err) { + dev_err(dev, "failed to init rxtx by hw\n"); + return err; + } + + sdma->dma_mask = dma_get_mask(dev); + sdma->map_addr = p.map_addr; + sdma->sw = sw; + + sdma->desc_pool = dma_pool_create("desc_pool", dev, + sizeof(struct prestera_sdma_desc), + 16, 0); + if (!sdma->desc_pool) + return -ENOMEM; + + err = prestera_sdma_rx_init(sdma); + if (err) { + dev_err(dev, "failed to init rx ring\n"); + goto err_rx_init; + } + + err = prestera_sdma_tx_init(sdma); + if (err) { + dev_err(dev, "failed to init tx ring\n"); + goto err_tx_init; + } + + err = prestera_hw_event_handler_register(sw, PRESTERA_EVENT_TYPE_RXTX, + prestera_rxtx_handle_event, + sdma); + if (err) + goto err_evt_register; + + init_dummy_netdev(&sdma->napi_dev); + + netif_napi_add(&sdma->napi_dev, &sdma->rx_napi, prestera_sdma_rx_poll, 64); + napi_enable(&sdma->rx_napi); + + return 0; + +err_evt_register: +err_tx_init: + prestera_sdma_tx_fini(sdma); +err_rx_init: + prestera_sdma_rx_fini(sdma); + + dma_pool_destroy(sdma->desc_pool); + return err; +} + +static void prestera_sdma_switch_fini(struct prestera_switch *sw) +{ + struct prestera_sdma *sdma = &sw->rxtx->sdma; + + napi_disable(&sdma->rx_napi); + netif_napi_del(&sdma->rx_napi); + prestera_hw_event_handler_unregister(sw, PRESTERA_EVENT_TYPE_RXTX, + prestera_rxtx_handle_event); + prestera_sdma_tx_fini(sdma); + prestera_sdma_rx_fini(sdma); + dma_pool_destroy(sdma->desc_pool); +} + +static bool prestera_sdma_is_ready(struct prestera_sdma *sdma) +{ + return !(prestera_read(sdma->sw, PRESTERA_SDMA_TX_QUEUE_START_REG) & 1); +} + +static int prestera_sdma_tx_wait(struct prestera_sdma *sdma, + struct prestera_tx_ring *tx_ring) +{ + int tx_wait_num = PRESTERA_SDMA_WAIT_MUL * tx_ring->max_burst; + + do { + if (prestera_sdma_is_ready(sdma)) + return 0; + + udelay(1); + } while (--tx_wait_num); + + return -EBUSY; +} + +static void prestera_sdma_tx_start(struct prestera_sdma *sdma) +{ + prestera_write(sdma->sw, PRESTERA_SDMA_TX_QUEUE_START_REG, 1); + schedule_work(&sdma->tx_work); +} + +static netdev_tx_t prestera_sdma_xmit(struct prestera_sdma *sdma, + struct sk_buff *skb) +{ + struct device *dma_dev = sdma->sw->dev->dev; + struct net_device *dev = skb->dev; + struct prestera_tx_ring *tx_ring; + struct prestera_sdma_buf *buf; + int err; + + spin_lock(&sdma->tx_lock); + + tx_ring = &sdma->tx_ring; + + buf = &tx_ring->bufs[tx_ring->next_tx]; + if (buf->is_used) { + schedule_work(&sdma->tx_work); + goto drop_skb; + } + + if (unlikely(eth_skb_pad(skb))) + goto drop_skb_nofree; + + err = prestera_sdma_tx_buf_map(sdma, buf, skb); + if (err) + goto drop_skb; + + prestera_sdma_tx_desc_set_buf(sdma, buf->desc, buf->buf_dma, skb->len); + + dma_sync_single_for_device(dma_dev, buf->buf_dma, skb->len, + DMA_TO_DEVICE); + + if (tx_ring->burst) { + tx_ring->burst--; + } else { + tx_ring->burst = tx_ring->max_burst; + + err = prestera_sdma_tx_wait(sdma, tx_ring); + if (err) + goto drop_skb_unmap; + } + + tx_ring->next_tx = (tx_ring->next_tx + 1) % PRESTERA_SDMA_TX_DESC_PER_Q; + prestera_sdma_tx_desc_xmit(buf->desc); + buf->is_used = true; + + prestera_sdma_tx_start(sdma); + + goto tx_done; + +drop_skb_unmap: + prestera_sdma_tx_buf_unmap(sdma, buf); +drop_skb: + dev_consume_skb_any(skb); +drop_skb_nofree: + dev->stats.tx_dropped++; +tx_done: + spin_unlock(&sdma->tx_lock); + return NETDEV_TX_OK; +} + +int prestera_rxtx_switch_init(struct prestera_switch *sw) +{ + struct prestera_rxtx *rxtx; + + rxtx = kzalloc(sizeof(*rxtx), GFP_KERNEL); + if (!rxtx) + return -ENOMEM; + + sw->rxtx = rxtx; + + return prestera_sdma_switch_init(sw); +} + +void prestera_rxtx_switch_fini(struct prestera_switch *sw) +{ + prestera_sdma_switch_fini(sw); + kfree(sw->rxtx); +} + +int prestera_rxtx_port_init(struct prestera_port *port) +{ + int err; + + err = prestera_hw_rxtx_port_init(port); + if (err) + return err; + + port->dev->needed_headroom = PRESTERA_DSA_HLEN; + + return 0; +} + +netdev_tx_t prestera_rxtx_xmit(struct prestera_port *port, struct sk_buff *skb) +{ + struct prestera_dsa dsa; + + dsa.hw_dev_num = port->dev_id; + dsa.port_num = port->hw_id; + + if (skb_cow_head(skb, PRESTERA_DSA_HLEN) < 0) + return NET_XMIT_DROP; + + skb_push(skb, PRESTERA_DSA_HLEN); + memmove(skb->data, skb->data + PRESTERA_DSA_HLEN, 2 * ETH_ALEN); + + if (prestera_dsa_build(&dsa, skb->data + 2 * ETH_ALEN) != 0) + return NET_XMIT_DROP; + + return prestera_sdma_xmit(&port->sw->rxtx->sdma, skb); +} diff --git a/drivers/net/ethernet/marvell/prestera/prestera_rxtx.h b/drivers/net/ethernet/marvell/prestera/prestera_rxtx.h new file mode 100644 index 000000000000..882a1225c323 --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/prestera_rxtx.h @@ -0,0 +1,19 @@ +/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 */ +/* Copyright (c) 2019-2020 Marvell International Ltd. All rights reserved. */ + +#ifndef _PRESTERA_RXTX_H_ +#define _PRESTERA_RXTX_H_ + +#include <linux/netdevice.h> + +struct prestera_switch; +struct prestera_port; + +int prestera_rxtx_switch_init(struct prestera_switch *sw); +void prestera_rxtx_switch_fini(struct prestera_switch *sw); + +int prestera_rxtx_port_init(struct prestera_port *port); + +netdev_tx_t prestera_rxtx_xmit(struct prestera_port *port, struct sk_buff *skb); + +#endif /* _PRESTERA_RXTX_H_ */ diff --git a/drivers/net/ethernet/marvell/prestera/prestera_switchdev.c b/drivers/net/ethernet/marvell/prestera/prestera_switchdev.c new file mode 100644 index 000000000000..7d83e1f91ef1 --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/prestera_switchdev.c @@ -0,0 +1,1277 @@ +// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 +/* Copyright (c) 2019-2020 Marvell International Ltd. All rights reserved */ + +#include <linux/if_bridge.h> +#include <linux/if_vlan.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/notifier.h> +#include <net/netevent.h> +#include <net/switchdev.h> + +#include "prestera.h" +#include "prestera_hw.h" +#include "prestera_switchdev.h" + +#define PRESTERA_VID_ALL (0xffff) + +#define PRESTERA_DEFAULT_AGEING_TIME_MS 300000 +#define PRESTERA_MAX_AGEING_TIME_MS 1000000000 +#define PRESTERA_MIN_AGEING_TIME_MS 32000 + +struct prestera_fdb_event_work { + struct work_struct work; + struct switchdev_notifier_fdb_info fdb_info; + struct net_device *dev; + unsigned long event; +}; + +struct prestera_switchdev { + struct prestera_switch *sw; + struct list_head bridge_list; + bool bridge_8021q_exists; + struct notifier_block swdev_nb_blk; + struct notifier_block swdev_nb; +}; + +struct prestera_bridge { + struct list_head head; + struct net_device *dev; + struct prestera_switchdev *swdev; + struct list_head port_list; + bool vlan_enabled; + u16 bridge_id; +}; + +struct prestera_bridge_port { + struct list_head head; + struct net_device *dev; + struct prestera_bridge *bridge; + struct list_head vlan_list; + refcount_t ref_count; + unsigned long flags; + u8 stp_state; +}; + +struct prestera_bridge_vlan { + struct list_head head; + struct list_head port_vlan_list; + u16 vid; +}; + +struct prestera_port_vlan { + struct list_head br_vlan_head; + struct list_head port_head; + struct prestera_port *port; + struct prestera_bridge_port *br_port; + u16 vid; +}; + +static struct workqueue_struct *swdev_wq; + +static void prestera_bridge_port_put(struct prestera_bridge_port *br_port); + +static int prestera_port_vid_stp_set(struct prestera_port *port, u16 vid, + u8 state); + +static struct prestera_bridge_vlan * +prestera_bridge_vlan_create(struct prestera_bridge_port *br_port, u16 vid) +{ + struct prestera_bridge_vlan *br_vlan; + + br_vlan = kzalloc(sizeof(*br_vlan), GFP_KERNEL); + if (!br_vlan) + return NULL; + + INIT_LIST_HEAD(&br_vlan->port_vlan_list); + br_vlan->vid = vid; + list_add(&br_vlan->head, &br_port->vlan_list); + + return br_vlan; +} + +static void prestera_bridge_vlan_destroy(struct prestera_bridge_vlan *br_vlan) +{ + list_del(&br_vlan->head); + WARN_ON(!list_empty(&br_vlan->port_vlan_list)); + kfree(br_vlan); +} + +static struct prestera_bridge_vlan * +prestera_bridge_vlan_by_vid(struct prestera_bridge_port *br_port, u16 vid) +{ + struct prestera_bridge_vlan *br_vlan; + + list_for_each_entry(br_vlan, &br_port->vlan_list, head) { + if (br_vlan->vid == vid) + return br_vlan; + } + + return NULL; +} + +static int prestera_bridge_vlan_port_count(struct prestera_bridge *bridge, + u16 vid) +{ + struct prestera_bridge_port *br_port; + struct prestera_bridge_vlan *br_vlan; + int count = 0; + + list_for_each_entry(br_port, &bridge->port_list, head) { + list_for_each_entry(br_vlan, &br_port->vlan_list, head) { + if (br_vlan->vid == vid) { + count += 1; + break; + } + } + } + + return count; +} + +static void prestera_bridge_vlan_put(struct prestera_bridge_vlan *br_vlan) +{ + if (list_empty(&br_vlan->port_vlan_list)) + prestera_bridge_vlan_destroy(br_vlan); +} + +static struct prestera_port_vlan * +prestera_port_vlan_by_vid(struct prestera_port *port, u16 vid) +{ + struct prestera_port_vlan *port_vlan; + + list_for_each_entry(port_vlan, &port->vlans_list, port_head) { + if (port_vlan->vid == vid) + return port_vlan; + } + + return NULL; +} + +static struct prestera_port_vlan * +prestera_port_vlan_create(struct prestera_port *port, u16 vid, bool untagged) +{ + struct prestera_port_vlan *port_vlan; + int err; + + port_vlan = prestera_port_vlan_by_vid(port, vid); + if (port_vlan) + return ERR_PTR(-EEXIST); + + err = prestera_hw_vlan_port_set(port, vid, true, untagged); + if (err) + return ERR_PTR(err); + + port_vlan = kzalloc(sizeof(*port_vlan), GFP_KERNEL); + if (!port_vlan) { + err = -ENOMEM; + goto err_port_vlan_alloc; + } + + port_vlan->port = port; + port_vlan->vid = vid; + + list_add(&port_vlan->port_head, &port->vlans_list); + + return port_vlan; + +err_port_vlan_alloc: + prestera_hw_vlan_port_set(port, vid, false, false); + return ERR_PTR(err); +} + +static void +prestera_port_vlan_bridge_leave(struct prestera_port_vlan *port_vlan) +{ + u32 fdb_flush_mode = PRESTERA_FDB_FLUSH_MODE_DYNAMIC; + struct prestera_port *port = port_vlan->port; + struct prestera_bridge_vlan *br_vlan; + struct prestera_bridge_port *br_port; + bool last_port, last_vlan; + u16 vid = port_vlan->vid; + int port_count; + + br_port = port_vlan->br_port; + port_count = prestera_bridge_vlan_port_count(br_port->bridge, vid); + br_vlan = prestera_bridge_vlan_by_vid(br_port, vid); + + last_vlan = list_is_singular(&br_port->vlan_list); + last_port = port_count == 1; + + if (last_vlan) + prestera_hw_fdb_flush_port(port, fdb_flush_mode); + else if (last_port) + prestera_hw_fdb_flush_vlan(port->sw, vid, fdb_flush_mode); + else + prestera_hw_fdb_flush_port_vlan(port, vid, fdb_flush_mode); + + list_del(&port_vlan->br_vlan_head); + prestera_bridge_vlan_put(br_vlan); + prestera_bridge_port_put(br_port); + port_vlan->br_port = NULL; +} + +static void prestera_port_vlan_destroy(struct prestera_port_vlan *port_vlan) +{ + struct prestera_port *port = port_vlan->port; + u16 vid = port_vlan->vid; + + if (port_vlan->br_port) + prestera_port_vlan_bridge_leave(port_vlan); + + prestera_hw_vlan_port_set(port, vid, false, false); + list_del(&port_vlan->port_head); + kfree(port_vlan); +} + +static struct prestera_bridge * +prestera_bridge_create(struct prestera_switchdev *swdev, struct net_device *dev) +{ + bool vlan_enabled = br_vlan_enabled(dev); + struct prestera_bridge *bridge; + u16 bridge_id; + int err; + + if (vlan_enabled && swdev->bridge_8021q_exists) { + netdev_err(dev, "Only one VLAN-aware bridge is supported\n"); + return ERR_PTR(-EINVAL); + } + + bridge = kzalloc(sizeof(*bridge), GFP_KERNEL); + if (!bridge) + return ERR_PTR(-ENOMEM); + + if (vlan_enabled) { + swdev->bridge_8021q_exists = true; + } else { + err = prestera_hw_bridge_create(swdev->sw, &bridge_id); + if (err) { + kfree(bridge); + return ERR_PTR(err); + } + + bridge->bridge_id = bridge_id; + } + + bridge->vlan_enabled = vlan_enabled; + bridge->swdev = swdev; + bridge->dev = dev; + + INIT_LIST_HEAD(&bridge->port_list); + + list_add(&bridge->head, &swdev->bridge_list); + + return bridge; +} + +static void prestera_bridge_destroy(struct prestera_bridge *bridge) +{ + struct prestera_switchdev *swdev = bridge->swdev; + + list_del(&bridge->head); + + if (bridge->vlan_enabled) + swdev->bridge_8021q_exists = false; + else + prestera_hw_bridge_delete(swdev->sw, bridge->bridge_id); + + WARN_ON(!list_empty(&bridge->port_list)); + kfree(bridge); +} + +static void prestera_bridge_put(struct prestera_bridge *bridge) +{ + if (list_empty(&bridge->port_list)) + prestera_bridge_destroy(bridge); +} + +static +struct prestera_bridge *prestera_bridge_by_dev(struct prestera_switchdev *swdev, + const struct net_device *dev) +{ + struct prestera_bridge *bridge; + + list_for_each_entry(bridge, &swdev->bridge_list, head) + if (bridge->dev == dev) + return bridge; + + return NULL; +} + +static struct prestera_bridge_port * +__prestera_bridge_port_by_dev(struct prestera_bridge *bridge, + struct net_device *dev) +{ + struct prestera_bridge_port *br_port; + + list_for_each_entry(br_port, &bridge->port_list, head) { + if (br_port->dev == dev) + return br_port; + } + + return NULL; +} + +static struct prestera_bridge_port * +prestera_bridge_port_by_dev(struct prestera_switchdev *swdev, + struct net_device *dev) +{ + struct net_device *br_dev = netdev_master_upper_dev_get(dev); + struct prestera_bridge *bridge; + + if (!br_dev) + return NULL; + + bridge = prestera_bridge_by_dev(swdev, br_dev); + if (!bridge) + return NULL; + + return __prestera_bridge_port_by_dev(bridge, dev); +} + +static struct prestera_bridge_port * +prestera_bridge_port_create(struct prestera_bridge *bridge, + struct net_device *dev) +{ + struct prestera_bridge_port *br_port; + + br_port = kzalloc(sizeof(*br_port), GFP_KERNEL); + if (!br_port) + return NULL; + + br_port->flags = BR_LEARNING | BR_FLOOD | BR_LEARNING_SYNC | + BR_MCAST_FLOOD; + br_port->stp_state = BR_STATE_DISABLED; + refcount_set(&br_port->ref_count, 1); + br_port->bridge = bridge; + br_port->dev = dev; + + INIT_LIST_HEAD(&br_port->vlan_list); + list_add(&br_port->head, &bridge->port_list); + + return br_port; +} + +static void +prestera_bridge_port_destroy(struct prestera_bridge_port *br_port) +{ + list_del(&br_port->head); + WARN_ON(!list_empty(&br_port->vlan_list)); + kfree(br_port); +} + +static void prestera_bridge_port_get(struct prestera_bridge_port *br_port) +{ + refcount_inc(&br_port->ref_count); +} + +static void prestera_bridge_port_put(struct prestera_bridge_port *br_port) +{ + struct prestera_bridge *bridge = br_port->bridge; + + if (refcount_dec_and_test(&br_port->ref_count)) { + prestera_bridge_port_destroy(br_port); + prestera_bridge_put(bridge); + } +} + +static struct prestera_bridge_port * +prestera_bridge_port_add(struct prestera_bridge *bridge, struct net_device *dev) +{ + struct prestera_bridge_port *br_port; + + br_port = __prestera_bridge_port_by_dev(bridge, dev); + if (br_port) { + prestera_bridge_port_get(br_port); + return br_port; + } + + br_port = prestera_bridge_port_create(bridge, dev); + if (!br_port) + return ERR_PTR(-ENOMEM); + + return br_port; +} + +static int +prestera_bridge_1d_port_join(struct prestera_bridge_port *br_port) +{ + struct prestera_port *port = netdev_priv(br_port->dev); + struct prestera_bridge *bridge = br_port->bridge; + int err; + + err = prestera_hw_bridge_port_add(port, bridge->bridge_id); + if (err) + return err; + + err = prestera_hw_port_flood_set(port, br_port->flags & BR_FLOOD); + if (err) + goto err_port_flood_set; + + err = prestera_hw_port_learning_set(port, br_port->flags & BR_LEARNING); + if (err) + goto err_port_learning_set; + + return 0; + +err_port_learning_set: + prestera_hw_port_flood_set(port, false); +err_port_flood_set: + prestera_hw_bridge_port_delete(port, bridge->bridge_id); + + return err; +} + +static int prestera_port_bridge_join(struct prestera_port *port, + struct net_device *upper) +{ + struct prestera_switchdev *swdev = port->sw->swdev; + struct prestera_bridge_port *br_port; + struct prestera_bridge *bridge; + int err; + + bridge = prestera_bridge_by_dev(swdev, upper); + if (!bridge) { + bridge = prestera_bridge_create(swdev, upper); + if (IS_ERR(bridge)) + return PTR_ERR(bridge); + } + + br_port = prestera_bridge_port_add(bridge, port->dev); + if (IS_ERR(br_port)) { + err = PTR_ERR(br_port); + goto err_brport_create; + } + + if (bridge->vlan_enabled) + return 0; + + err = prestera_bridge_1d_port_join(br_port); + if (err) + goto err_port_join; + + return 0; + +err_port_join: + prestera_bridge_port_put(br_port); +err_brport_create: + prestera_bridge_put(bridge); + return err; +} + +static void prestera_bridge_1q_port_leave(struct prestera_bridge_port *br_port) +{ + struct prestera_port *port = netdev_priv(br_port->dev); + + prestera_hw_fdb_flush_port(port, PRESTERA_FDB_FLUSH_MODE_ALL); + prestera_port_pvid_set(port, PRESTERA_DEFAULT_VID); +} + +static void prestera_bridge_1d_port_leave(struct prestera_bridge_port *br_port) +{ + struct prestera_port *port = netdev_priv(br_port->dev); + + prestera_hw_fdb_flush_port(port, PRESTERA_FDB_FLUSH_MODE_ALL); + prestera_hw_bridge_port_delete(port, br_port->bridge->bridge_id); +} + +static int prestera_port_vid_stp_set(struct prestera_port *port, u16 vid, + u8 state) +{ + u8 hw_state = state; + + switch (state) { + case BR_STATE_DISABLED: + hw_state = PRESTERA_STP_DISABLED; + break; + + case BR_STATE_BLOCKING: + case BR_STATE_LISTENING: + hw_state = PRESTERA_STP_BLOCK_LISTEN; + break; + + case BR_STATE_LEARNING: + hw_state = PRESTERA_STP_LEARN; + break; + + case BR_STATE_FORWARDING: + hw_state = PRESTERA_STP_FORWARD; + break; + + default: + return -EINVAL; + } + + return prestera_hw_vlan_port_stp_set(port, vid, hw_state); +} + +static void prestera_port_bridge_leave(struct prestera_port *port, + struct net_device *upper) +{ + struct prestera_switchdev *swdev = port->sw->swdev; + struct prestera_bridge_port *br_port; + struct prestera_bridge *bridge; + + bridge = prestera_bridge_by_dev(swdev, upper); + if (!bridge) + return; + + br_port = __prestera_bridge_port_by_dev(bridge, port->dev); + if (!br_port) + return; + + bridge = br_port->bridge; + + if (bridge->vlan_enabled) + prestera_bridge_1q_port_leave(br_port); + else + prestera_bridge_1d_port_leave(br_port); + + prestera_hw_port_learning_set(port, false); + prestera_hw_port_flood_set(port, false); + prestera_port_vid_stp_set(port, PRESTERA_VID_ALL, BR_STATE_FORWARDING); + prestera_bridge_port_put(br_port); +} + +int prestera_bridge_port_event(struct net_device *dev, unsigned long event, + void *ptr) +{ + struct netdev_notifier_changeupper_info *info = ptr; + struct netlink_ext_ack *extack; + struct prestera_port *port; + struct net_device *upper; + int err; + + extack = netdev_notifier_info_to_extack(&info->info); + port = netdev_priv(dev); + upper = info->upper_dev; + + switch (event) { + case NETDEV_PRECHANGEUPPER: + if (!netif_is_bridge_master(upper)) { + NL_SET_ERR_MSG_MOD(extack, "Unknown upper device type"); + return -EINVAL; + } + + if (!info->linking) + break; + + if (netdev_has_any_upper_dev(upper)) { + NL_SET_ERR_MSG_MOD(extack, "Upper device is already enslaved"); + return -EINVAL; + } + break; + + case NETDEV_CHANGEUPPER: + if (!netif_is_bridge_master(upper)) + break; + + if (info->linking) { + err = prestera_port_bridge_join(port, upper); + if (err) + return err; + } else { + prestera_port_bridge_leave(port, upper); + } + break; + } + + return 0; +} + +static int prestera_port_attr_br_flags_set(struct prestera_port *port, + struct switchdev_trans *trans, + struct net_device *dev, + unsigned long flags) +{ + struct prestera_bridge_port *br_port; + int err; + + if (switchdev_trans_ph_prepare(trans)) + return 0; + + br_port = prestera_bridge_port_by_dev(port->sw->swdev, dev); + if (!br_port) + return 0; + + err = prestera_hw_port_flood_set(port, flags & BR_FLOOD); + if (err) + return err; + + err = prestera_hw_port_learning_set(port, flags & BR_LEARNING); + if (err) + return err; + + memcpy(&br_port->flags, &flags, sizeof(flags)); + + return 0; +} + +static int prestera_port_attr_br_ageing_set(struct prestera_port *port, + struct switchdev_trans *trans, + unsigned long ageing_clock_t) +{ + unsigned long ageing_jiffies = clock_t_to_jiffies(ageing_clock_t); + u32 ageing_time_ms = jiffies_to_msecs(ageing_jiffies); + struct prestera_switch *sw = port->sw; + + if (switchdev_trans_ph_prepare(trans)) { + if (ageing_time_ms < PRESTERA_MIN_AGEING_TIME_MS || + ageing_time_ms > PRESTERA_MAX_AGEING_TIME_MS) + return -ERANGE; + else + return 0; + } + + return prestera_hw_switch_ageing_set(sw, ageing_time_ms); +} + +static int prestera_port_attr_br_vlan_set(struct prestera_port *port, + struct switchdev_trans *trans, + struct net_device *dev, + bool vlan_enabled) +{ + struct prestera_switch *sw = port->sw; + struct prestera_bridge *bridge; + + if (!switchdev_trans_ph_prepare(trans)) + return 0; + + bridge = prestera_bridge_by_dev(sw->swdev, dev); + if (WARN_ON(!bridge)) + return -EINVAL; + + if (bridge->vlan_enabled == vlan_enabled) + return 0; + + netdev_err(bridge->dev, "VLAN filtering can't be changed for existing bridge\n"); + + return -EINVAL; +} + +static int prestera_port_bridge_vlan_stp_set(struct prestera_port *port, + struct prestera_bridge_vlan *br_vlan, + u8 state) +{ + struct prestera_port_vlan *port_vlan; + + list_for_each_entry(port_vlan, &br_vlan->port_vlan_list, br_vlan_head) { + if (port_vlan->port != port) + continue; + + return prestera_port_vid_stp_set(port, br_vlan->vid, state); + } + + return 0; +} + +static int presterar_port_attr_stp_state_set(struct prestera_port *port, + struct switchdev_trans *trans, + struct net_device *dev, + u8 state) +{ + struct prestera_bridge_port *br_port; + struct prestera_bridge_vlan *br_vlan; + int err; + u16 vid; + + if (switchdev_trans_ph_prepare(trans)) + return 0; + + br_port = prestera_bridge_port_by_dev(port->sw->swdev, dev); + if (!br_port) + return 0; + + if (!br_port->bridge->vlan_enabled) { + vid = br_port->bridge->bridge_id; + err = prestera_port_vid_stp_set(port, vid, state); + if (err) + goto err_port_stp_set; + } else { + list_for_each_entry(br_vlan, &br_port->vlan_list, head) { + err = prestera_port_bridge_vlan_stp_set(port, br_vlan, + state); + if (err) + goto err_port_vlan_stp_set; + } + } + + br_port->stp_state = state; + + return 0; + +err_port_vlan_stp_set: + list_for_each_entry_continue_reverse(br_vlan, &br_port->vlan_list, head) + prestera_port_bridge_vlan_stp_set(port, br_vlan, br_port->stp_state); + return err; + +err_port_stp_set: + prestera_port_vid_stp_set(port, vid, br_port->stp_state); + + return err; +} + +static int prestera_port_obj_attr_set(struct net_device *dev, + const struct switchdev_attr *attr, + struct switchdev_trans *trans) +{ + struct prestera_port *port = netdev_priv(dev); + int err = 0; + + switch (attr->id) { + case SWITCHDEV_ATTR_ID_PORT_STP_STATE: + err = presterar_port_attr_stp_state_set(port, trans, + attr->orig_dev, + attr->u.stp_state); + break; + case SWITCHDEV_ATTR_ID_PORT_PRE_BRIDGE_FLAGS: + if (attr->u.brport_flags & + ~(BR_LEARNING | BR_FLOOD | BR_MCAST_FLOOD)) + err = -EINVAL; + break; + case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS: + err = prestera_port_attr_br_flags_set(port, trans, + attr->orig_dev, + attr->u.brport_flags); + break; + case SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME: + err = prestera_port_attr_br_ageing_set(port, trans, + attr->u.ageing_time); + break; + case SWITCHDEV_ATTR_ID_BRIDGE_VLAN_FILTERING: + err = prestera_port_attr_br_vlan_set(port, trans, + attr->orig_dev, + attr->u.vlan_filtering); + break; + default: + err = -EOPNOTSUPP; + } + + return err; +} + +static void +prestera_fdb_offload_notify(struct prestera_port *port, + struct switchdev_notifier_fdb_info *info) +{ + struct switchdev_notifier_fdb_info send_info; + + send_info.addr = info->addr; + send_info.vid = info->vid; + send_info.offloaded = true; + + call_switchdev_notifiers(SWITCHDEV_FDB_OFFLOADED, port->dev, + &send_info.info, NULL); +} + +static int prestera_port_fdb_set(struct prestera_port *port, + struct switchdev_notifier_fdb_info *fdb_info, + bool adding) +{ + struct prestera_switch *sw = port->sw; + struct prestera_bridge_port *br_port; + struct prestera_bridge *bridge; + int err; + u16 vid; + + br_port = prestera_bridge_port_by_dev(sw->swdev, port->dev); + if (!br_port) + return -EINVAL; + + bridge = br_port->bridge; + + if (bridge->vlan_enabled) + vid = fdb_info->vid; + else + vid = bridge->bridge_id; + + if (adding) + err = prestera_hw_fdb_add(port, fdb_info->addr, vid, false); + else + err = prestera_hw_fdb_del(port, fdb_info->addr, vid); + + return err; +} + +static void prestera_fdb_event_work(struct work_struct *work) +{ + struct switchdev_notifier_fdb_info *fdb_info; + struct prestera_fdb_event_work *swdev_work; + struct prestera_port *port; + struct net_device *dev; + int err; + + swdev_work = container_of(work, struct prestera_fdb_event_work, work); + dev = swdev_work->dev; + + rtnl_lock(); + + port = prestera_port_dev_lower_find(dev); + if (!port) + goto out_unlock; + + switch (swdev_work->event) { + case SWITCHDEV_FDB_ADD_TO_DEVICE: + fdb_info = &swdev_work->fdb_info; + if (!fdb_info->added_by_user) + break; + + err = prestera_port_fdb_set(port, fdb_info, true); + if (err) + break; + + prestera_fdb_offload_notify(port, fdb_info); + break; + + case SWITCHDEV_FDB_DEL_TO_DEVICE: + fdb_info = &swdev_work->fdb_info; + prestera_port_fdb_set(port, fdb_info, false); + break; + } + +out_unlock: + rtnl_unlock(); + + kfree(swdev_work->fdb_info.addr); + kfree(swdev_work); + dev_put(dev); +} + +static int prestera_switchdev_event(struct notifier_block *unused, + unsigned long event, void *ptr) +{ + struct net_device *dev = switchdev_notifier_info_to_dev(ptr); + struct switchdev_notifier_fdb_info *fdb_info; + struct switchdev_notifier_info *info = ptr; + struct prestera_fdb_event_work *swdev_work; + struct net_device *upper; + int err; + + if (event == SWITCHDEV_PORT_ATTR_SET) { + err = switchdev_handle_port_attr_set(dev, ptr, + prestera_netdev_check, + prestera_port_obj_attr_set); + return notifier_from_errno(err); + } + + if (!prestera_netdev_check(dev)) + return NOTIFY_DONE; + + upper = netdev_master_upper_dev_get_rcu(dev); + if (!upper) + return NOTIFY_DONE; + + if (!netif_is_bridge_master(upper)) + return NOTIFY_DONE; + + swdev_work = kzalloc(sizeof(*swdev_work), GFP_ATOMIC); + if (!swdev_work) + return NOTIFY_BAD; + + swdev_work->event = event; + swdev_work->dev = dev; + + switch (event) { + case SWITCHDEV_FDB_ADD_TO_DEVICE: + case SWITCHDEV_FDB_DEL_TO_DEVICE: + fdb_info = container_of(info, + struct switchdev_notifier_fdb_info, + info); + + INIT_WORK(&swdev_work->work, prestera_fdb_event_work); + memcpy(&swdev_work->fdb_info, ptr, + sizeof(swdev_work->fdb_info)); + + swdev_work->fdb_info.addr = kzalloc(ETH_ALEN, GFP_ATOMIC); + if (!swdev_work->fdb_info.addr) + goto out_bad; + + ether_addr_copy((u8 *)swdev_work->fdb_info.addr, + fdb_info->addr); + dev_hold(dev); + break; + + default: + kfree(swdev_work); + return NOTIFY_DONE; + } + + queue_work(swdev_wq, &swdev_work->work); + return NOTIFY_DONE; + +out_bad: + kfree(swdev_work); + return NOTIFY_BAD; +} + +static int +prestera_port_vlan_bridge_join(struct prestera_port_vlan *port_vlan, + struct prestera_bridge_port *br_port) +{ + struct prestera_port *port = port_vlan->port; + struct prestera_bridge_vlan *br_vlan; + u16 vid = port_vlan->vid; + int err; + + if (port_vlan->br_port) + return 0; + + err = prestera_hw_port_flood_set(port, br_port->flags & BR_FLOOD); + if (err) + return err; + + err = prestera_hw_port_learning_set(port, br_port->flags & BR_LEARNING); + if (err) + goto err_port_learning_set; + + err = prestera_port_vid_stp_set(port, vid, br_port->stp_state); + if (err) + goto err_port_vid_stp_set; + + br_vlan = prestera_bridge_vlan_by_vid(br_port, vid); + if (!br_vlan) { + br_vlan = prestera_bridge_vlan_create(br_port, vid); + if (!br_vlan) { + err = -ENOMEM; + goto err_bridge_vlan_get; + } + } + + list_add(&port_vlan->br_vlan_head, &br_vlan->port_vlan_list); + + prestera_bridge_port_get(br_port); + port_vlan->br_port = br_port; + + return 0; + +err_bridge_vlan_get: + prestera_port_vid_stp_set(port, vid, BR_STATE_FORWARDING); +err_port_vid_stp_set: + prestera_hw_port_learning_set(port, false); +err_port_learning_set: + return err; +} + +static int +prestera_bridge_port_vlan_add(struct prestera_port *port, + struct prestera_bridge_port *br_port, + u16 vid, bool is_untagged, bool is_pvid, + struct netlink_ext_ack *extack) +{ + struct prestera_port_vlan *port_vlan; + u16 old_pvid = port->pvid; + u16 pvid; + int err; + + if (is_pvid) + pvid = vid; + else + pvid = port->pvid == vid ? 0 : port->pvid; + + port_vlan = prestera_port_vlan_by_vid(port, vid); + if (port_vlan && port_vlan->br_port != br_port) + return -EEXIST; + + if (!port_vlan) { + port_vlan = prestera_port_vlan_create(port, vid, is_untagged); + if (IS_ERR(port_vlan)) + return PTR_ERR(port_vlan); + } else { + err = prestera_hw_vlan_port_set(port, vid, true, is_untagged); + if (err) + goto err_port_vlan_set; + } + + err = prestera_port_pvid_set(port, pvid); + if (err) + goto err_port_pvid_set; + + err = prestera_port_vlan_bridge_join(port_vlan, br_port); + if (err) + goto err_port_vlan_bridge_join; + + return 0; + +err_port_vlan_bridge_join: + prestera_port_pvid_set(port, old_pvid); +err_port_pvid_set: + prestera_hw_vlan_port_set(port, vid, false, false); +err_port_vlan_set: + prestera_port_vlan_destroy(port_vlan); + + return err; +} + +static void +prestera_bridge_port_vlan_del(struct prestera_port *port, + struct prestera_bridge_port *br_port, u16 vid) +{ + u16 pvid = port->pvid == vid ? 0 : port->pvid; + struct prestera_port_vlan *port_vlan; + + port_vlan = prestera_port_vlan_by_vid(port, vid); + if (WARN_ON(!port_vlan)) + return; + + prestera_port_vlan_bridge_leave(port_vlan); + prestera_port_pvid_set(port, pvid); + prestera_port_vlan_destroy(port_vlan); +} + +static int prestera_port_vlans_add(struct prestera_port *port, + const struct switchdev_obj_port_vlan *vlan, + struct switchdev_trans *trans, + struct netlink_ext_ack *extack) +{ + bool flag_untagged = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED; + bool flag_pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID; + struct net_device *dev = vlan->obj.orig_dev; + struct prestera_bridge_port *br_port; + struct prestera_switch *sw = port->sw; + struct prestera_bridge *bridge; + u16 vid; + + if (netif_is_bridge_master(dev)) + return 0; + + if (switchdev_trans_ph_commit(trans)) + return 0; + + br_port = prestera_bridge_port_by_dev(sw->swdev, dev); + if (WARN_ON(!br_port)) + return -EINVAL; + + bridge = br_port->bridge; + if (!bridge->vlan_enabled) + return 0; + + for (vid = vlan->vid_begin; vid <= vlan->vid_end; vid++) { + int err; + + err = prestera_bridge_port_vlan_add(port, br_port, + vid, flag_untagged, + flag_pvid, extack); + if (err) + return err; + } + + return 0; +} + +static int prestera_port_obj_add(struct net_device *dev, + const struct switchdev_obj *obj, + struct switchdev_trans *trans, + struct netlink_ext_ack *extack) +{ + struct prestera_port *port = netdev_priv(dev); + const struct switchdev_obj_port_vlan *vlan; + + switch (obj->id) { + case SWITCHDEV_OBJ_ID_PORT_VLAN: + vlan = SWITCHDEV_OBJ_PORT_VLAN(obj); + return prestera_port_vlans_add(port, vlan, trans, extack); + default: + return -EOPNOTSUPP; + } +} + +static int prestera_port_vlans_del(struct prestera_port *port, + const struct switchdev_obj_port_vlan *vlan) +{ + struct net_device *dev = vlan->obj.orig_dev; + struct prestera_bridge_port *br_port; + struct prestera_switch *sw = port->sw; + u16 vid; + + if (netif_is_bridge_master(dev)) + return -EOPNOTSUPP; + + br_port = prestera_bridge_port_by_dev(sw->swdev, dev); + if (WARN_ON(!br_port)) + return -EINVAL; + + if (!br_port->bridge->vlan_enabled) + return 0; + + for (vid = vlan->vid_begin; vid <= vlan->vid_end; vid++) + prestera_bridge_port_vlan_del(port, br_port, vid); + + return 0; +} + +static int prestera_port_obj_del(struct net_device *dev, + const struct switchdev_obj *obj) +{ + struct prestera_port *port = netdev_priv(dev); + + switch (obj->id) { + case SWITCHDEV_OBJ_ID_PORT_VLAN: + return prestera_port_vlans_del(port, SWITCHDEV_OBJ_PORT_VLAN(obj)); + default: + return -EOPNOTSUPP; + } +} + +static int prestera_switchdev_blk_event(struct notifier_block *unused, + unsigned long event, void *ptr) +{ + struct net_device *dev = switchdev_notifier_info_to_dev(ptr); + int err; + + switch (event) { + case SWITCHDEV_PORT_OBJ_ADD: + err = switchdev_handle_port_obj_add(dev, ptr, + prestera_netdev_check, + prestera_port_obj_add); + break; + case SWITCHDEV_PORT_OBJ_DEL: + err = switchdev_handle_port_obj_del(dev, ptr, + prestera_netdev_check, + prestera_port_obj_del); + break; + case SWITCHDEV_PORT_ATTR_SET: + err = switchdev_handle_port_attr_set(dev, ptr, + prestera_netdev_check, + prestera_port_obj_attr_set); + break; + default: + err = -EOPNOTSUPP; + } + + return notifier_from_errno(err); +} + +static void prestera_fdb_event(struct prestera_switch *sw, + struct prestera_event *evt, void *arg) +{ + struct switchdev_notifier_fdb_info info; + struct prestera_port *port; + + port = prestera_find_port(sw, evt->fdb_evt.port_id); + if (!port) + return; + + info.addr = evt->fdb_evt.data.mac; + info.vid = evt->fdb_evt.vid; + info.offloaded = true; + + rtnl_lock(); + + switch (evt->id) { + case PRESTERA_FDB_EVENT_LEARNED: + call_switchdev_notifiers(SWITCHDEV_FDB_ADD_TO_BRIDGE, + port->dev, &info.info, NULL); + break; + case PRESTERA_FDB_EVENT_AGED: + call_switchdev_notifiers(SWITCHDEV_FDB_DEL_TO_BRIDGE, + port->dev, &info.info, NULL); + break; + } + + rtnl_unlock(); +} + +static int prestera_fdb_init(struct prestera_switch *sw) +{ + int err; + + err = prestera_hw_event_handler_register(sw, PRESTERA_EVENT_TYPE_FDB, + prestera_fdb_event, NULL); + if (err) + return err; + + err = prestera_hw_switch_ageing_set(sw, PRESTERA_DEFAULT_AGEING_TIME_MS); + if (err) + goto err_ageing_set; + + return 0; + +err_ageing_set: + prestera_hw_event_handler_unregister(sw, PRESTERA_EVENT_TYPE_FDB, + prestera_fdb_event); + return err; +} + +static void prestera_fdb_fini(struct prestera_switch *sw) +{ + prestera_hw_event_handler_unregister(sw, PRESTERA_EVENT_TYPE_FDB, + prestera_fdb_event); +} + +static int prestera_switchdev_handler_init(struct prestera_switchdev *swdev) +{ + int err; + + swdev->swdev_nb.notifier_call = prestera_switchdev_event; + err = register_switchdev_notifier(&swdev->swdev_nb); + if (err) + goto err_register_swdev_notifier; + + swdev->swdev_nb_blk.notifier_call = prestera_switchdev_blk_event; + err = register_switchdev_blocking_notifier(&swdev->swdev_nb_blk); + if (err) + goto err_register_blk_swdev_notifier; + + return 0; + +err_register_blk_swdev_notifier: + unregister_switchdev_notifier(&swdev->swdev_nb); +err_register_swdev_notifier: + destroy_workqueue(swdev_wq); + return err; +} + +static void prestera_switchdev_handler_fini(struct prestera_switchdev *swdev) +{ + unregister_switchdev_blocking_notifier(&swdev->swdev_nb_blk); + unregister_switchdev_notifier(&swdev->swdev_nb); +} + +int prestera_switchdev_init(struct prestera_switch *sw) +{ + struct prestera_switchdev *swdev; + int err; + + swdev = kzalloc(sizeof(*swdev), GFP_KERNEL); + if (!swdev) + return -ENOMEM; + + sw->swdev = swdev; + swdev->sw = sw; + + INIT_LIST_HEAD(&swdev->bridge_list); + + swdev_wq = alloc_ordered_workqueue("%s_ordered", 0, "prestera_br"); + if (!swdev_wq) { + err = -ENOMEM; + goto err_alloc_wq; + } + + err = prestera_switchdev_handler_init(swdev); + if (err) + goto err_swdev_init; + + err = prestera_fdb_init(sw); + if (err) + goto err_fdb_init; + + return 0; + +err_fdb_init: +err_swdev_init: + destroy_workqueue(swdev_wq); +err_alloc_wq: + kfree(swdev); + + return err; +} + +void prestera_switchdev_fini(struct prestera_switch *sw) +{ + struct prestera_switchdev *swdev = sw->swdev; + + prestera_fdb_fini(sw); + prestera_switchdev_handler_fini(swdev); + destroy_workqueue(swdev_wq); + kfree(swdev); +} diff --git a/drivers/net/ethernet/marvell/prestera/prestera_switchdev.h b/drivers/net/ethernet/marvell/prestera/prestera_switchdev.h new file mode 100644 index 000000000000..606e21d2355b --- /dev/null +++ b/drivers/net/ethernet/marvell/prestera/prestera_switchdev.h @@ -0,0 +1,13 @@ +/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 */ +/* Copyright (c) 2019-2020 Marvell International Ltd. All rights reserved. */ + +#ifndef _PRESTERA_SWITCHDEV_H_ +#define _PRESTERA_SWITCHDEV_H_ + +int prestera_switchdev_init(struct prestera_switch *sw); +void prestera_switchdev_fini(struct prestera_switch *sw); + +int prestera_bridge_port_event(struct net_device *dev, unsigned long event, + void *ptr); + +#endif /* _PRESTERA_SWITCHDEV_H_ */ diff --git a/drivers/net/ethernet/marvell/pxa168_eth.c b/drivers/net/ethernet/marvell/pxa168_eth.c index eb8cf60ecf12..d1e4d42e497d 100644 --- a/drivers/net/ethernet/marvell/pxa168_eth.c +++ b/drivers/net/ethernet/marvell/pxa168_eth.c @@ -1187,11 +1187,10 @@ static int pxa168_eth_stop(struct net_device *dev) static int pxa168_eth_change_mtu(struct net_device *dev, int mtu) { - int retval; struct pxa168_eth_private *pep = netdev_priv(dev); dev->mtu = mtu; - retval = set_port_config_ext(pep); + set_port_config_ext(pep); if (!netif_running(dev)) return 0; @@ -1541,10 +1540,8 @@ static int pxa168_eth_remove(struct platform_device *pdev) } if (dev->phydev) phy_disconnect(dev->phydev); - if (pep->clk) { - clk_disable_unprepare(pep->clk); - } + clk_disable_unprepare(pep->clk); mdiobus_unregister(pep->smi_bus); mdiobus_free(pep->smi_bus); unregister_netdev(dev); diff --git a/drivers/net/ethernet/marvell/skge.c b/drivers/net/ethernet/marvell/skge.c index 6a930351cb23..8a9c0f490bfb 100644 --- a/drivers/net/ethernet/marvell/skge.c +++ b/drivers/net/ethernet/marvell/skge.c @@ -3338,9 +3338,9 @@ static void skge_error_irq(struct skge_hw *hw) * because accessing phy registers requires spin wait which might * cause excess interrupt latency. */ -static void skge_extirq(unsigned long arg) +static void skge_extirq(struct tasklet_struct *t) { - struct skge_hw *hw = (struct skge_hw *) arg; + struct skge_hw *hw = from_tasklet(hw, t, phy_task); int port; for (port = 0; port < hw->ports; port++) { @@ -3927,7 +3927,7 @@ static int skge_probe(struct pci_dev *pdev, const struct pci_device_id *ent) hw->pdev = pdev; spin_lock_init(&hw->hw_lock); spin_lock_init(&hw->phy_lock); - tasklet_init(&hw->phy_task, skge_extirq, (unsigned long) hw); + tasklet_setup(&hw->phy_task, skge_extirq); hw->regs = ioremap(pci_resource_start(pdev, 0), 0x4000); if (!hw->regs) { diff --git a/drivers/net/ethernet/mellanox/mlx4/cq.c b/drivers/net/ethernet/mellanox/mlx4/cq.c index 65f8a4b6ed0c..3b8576b9c2f9 100644 --- a/drivers/net/ethernet/mellanox/mlx4/cq.c +++ b/drivers/net/ethernet/mellanox/mlx4/cq.c @@ -55,11 +55,11 @@ #define TASKLET_MAX_TIME 2 #define TASKLET_MAX_TIME_JIFFIES msecs_to_jiffies(TASKLET_MAX_TIME) -void mlx4_cq_tasklet_cb(unsigned long data) +void mlx4_cq_tasklet_cb(struct tasklet_struct *t) { unsigned long flags; unsigned long end = jiffies + TASKLET_MAX_TIME_JIFFIES; - struct mlx4_eq_tasklet *ctx = (struct mlx4_eq_tasklet *)data; + struct mlx4_eq_tasklet *ctx = from_tasklet(ctx, t, task); struct mlx4_cq *mcq, *temp; spin_lock_irqsave(&ctx->lock, flags); diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c index b816154bc79a..23849f2b9c25 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c @@ -1106,6 +1106,24 @@ static int mlx4_en_set_pauseparam(struct net_device *dev, return err; } +static void mlx4_en_get_pause_stats(struct net_device *dev, + struct ethtool_pause_stats *stats) +{ + struct mlx4_en_priv *priv = netdev_priv(dev); + struct bitmap_iterator it; + + bitmap_iterator_init(&it, priv->stats_bitmap.bitmap, NUM_ALL_STATS); + + spin_lock_bh(&priv->stats_lock); + if (test_bit(FLOW_PRIORITY_STATS_IDX_TX_FRAMES, + priv->stats_bitmap.bitmap)) + stats->tx_pause_frames = priv->tx_flowstats.tx_pause; + if (test_bit(FLOW_PRIORITY_STATS_IDX_RX_FRAMES, + priv->stats_bitmap.bitmap)) + stats->rx_pause_frames = priv->rx_flowstats.rx_pause; + spin_unlock_bh(&priv->stats_lock); +} + static void mlx4_en_get_pauseparam(struct net_device *dev, struct ethtool_pauseparam *pause) { @@ -2138,6 +2156,7 @@ const struct ethtool_ops mlx4_en_ethtool_ops = { .set_msglevel = mlx4_en_set_msglevel, .get_coalesce = mlx4_en_get_coalesce, .set_coalesce = mlx4_en_set_coalesce, + .get_pause_stats = mlx4_en_get_pause_stats, .get_pauseparam = mlx4_en_get_pauseparam, .set_pauseparam = mlx4_en_set_pauseparam, .get_ringparam = mlx4_en_get_ringparam, diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c index b50c567ef508..502d1b97855c 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -705,7 +705,7 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud frags = ring->rx_info + (index << priv->log_rx_info); va = page_address(frags[0].page) + frags[0].page_offset; - prefetchw(va); + net_prefetchw(va); /* * make sure we read the CQE after we read the ownership bit */ @@ -943,6 +943,9 @@ int mlx4_en_poll_rx_cq(struct napi_struct *napi, int budget) bool clean_complete = true; int done; + if (!budget) + return 0; + if (priv->tx_ring_num[TX_XDP]) { xdp_tx_cq = priv->tx_cq[TX_XDP][cq->ring]; if (xdp_tx_cq->xdp_busy) { diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c index 9dff7b086c9f..3ddb7268e415 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c @@ -350,7 +350,7 @@ u32 mlx4_en_recycle_tx_desc(struct mlx4_en_priv *priv, .dma = tx_info->map0_dma, }; - if (!mlx4_en_rx_recycle(ring->recycle_ring, &frame)) { + if (!napi_mode || !mlx4_en_rx_recycle(ring->recycle_ring, &frame)) { dma_unmap_page(priv->ddev, tx_info->map0_dma, PAGE_SIZE, priv->dma_dir); put_page(tx_info->page); @@ -842,6 +842,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) struct mlx4_en_tx_desc *tx_desc; struct mlx4_wqe_data_seg *data; struct mlx4_en_tx_info *tx_info; + u32 __maybe_unused ring_cons; int tx_ind; int nr_txbb; int desc_size; @@ -855,7 +856,6 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) bool stop_queue; bool inline_ok; u8 data_offset; - u32 ring_cons; bool bf_ok; tx_ind = skb_get_queue_mapping(skb); @@ -1075,7 +1075,6 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) */ smp_rmb(); - ring_cons = READ_ONCE(ring->cons); if (unlikely(!mlx4_en_is_tx_ring_full(ring))) { netif_tx_wake_queue(ring->tx_queue); ring->wake_queue++; diff --git a/drivers/net/ethernet/mellanox/mlx4/eq.c b/drivers/net/ethernet/mellanox/mlx4/eq.c index ae305c2e9225..9e48509ed3b2 100644 --- a/drivers/net/ethernet/mellanox/mlx4/eq.c +++ b/drivers/net/ethernet/mellanox/mlx4/eq.c @@ -1057,8 +1057,7 @@ static int mlx4_create_eq(struct mlx4_dev *dev, int nent, INIT_LIST_HEAD(&eq->tasklet_ctx.list); INIT_LIST_HEAD(&eq->tasklet_ctx.process_list); spin_lock_init(&eq->tasklet_ctx.lock); - tasklet_init(&eq->tasklet_ctx.task, mlx4_cq_tasklet_cb, - (unsigned long)&eq->tasklet_ctx); + tasklet_setup(&eq->tasklet_ctx.task, mlx4_cq_tasklet_cb); return err; diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c index 258c7a96f269..c326b434734e 100644 --- a/drivers/net/ethernet/mellanox/mlx4/main.c +++ b/drivers/net/ethernet/mellanox/mlx4/main.c @@ -3031,6 +3031,17 @@ static int mlx4_init_port_info(struct mlx4_dev *dev, int port) if (err) return err; + /* Ethernet and IB drivers will normally set the port type, + * but if they are not built set the type now to prevent + * devlink_port_type_warn() from firing. + */ + if (!IS_ENABLED(CONFIG_MLX4_EN) && + dev->caps.port_type[port] == MLX4_PORT_TYPE_ETH) + devlink_port_type_eth_set(&info->devlink_port, NULL); + else if (!IS_ENABLED(CONFIG_MLX4_INFINIBAND) && + dev->caps.port_type[port] == MLX4_PORT_TYPE_IB) + devlink_port_type_ib_set(&info->devlink_port, NULL); + info->dev = dev; info->port = port; if (!mlx4_is_slave(dev)) { @@ -3935,6 +3946,8 @@ static int mlx4_restart_one_up(struct pci_dev *pdev, bool reload, struct devlink *devlink); static int mlx4_devlink_reload_down(struct devlink *devlink, bool netns_change, + enum devlink_reload_action action, + enum devlink_reload_limit limit, struct netlink_ext_ack *extack) { struct mlx4_priv *priv = devlink_priv(devlink); @@ -3951,7 +3964,8 @@ static int mlx4_devlink_reload_down(struct devlink *devlink, bool netns_change, return 0; } -static int mlx4_devlink_reload_up(struct devlink *devlink, +static int mlx4_devlink_reload_up(struct devlink *devlink, enum devlink_reload_action action, + enum devlink_reload_limit limit, u32 *actions_performed, struct netlink_ext_ack *extack) { struct mlx4_priv *priv = devlink_priv(devlink); @@ -3959,6 +3973,7 @@ static int mlx4_devlink_reload_up(struct devlink *devlink, struct mlx4_dev_persistent *persist = dev->persist; int err; + *actions_performed = BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT); err = mlx4_restart_one_up(persist->pdev, true, devlink); if (err) mlx4_err(persist->dev, "mlx4_restart_one_up failed, ret=%d\n", @@ -3969,6 +3984,7 @@ static int mlx4_devlink_reload_up(struct devlink *devlink, static const struct devlink_ops mlx4_devlink_ops = { .port_type_set = mlx4_devlink_port_type_set, + .reload_actions = BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT), .reload_down = mlx4_devlink_reload_down, .reload_up = mlx4_devlink_reload_up, }; diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h index 527b52e48276..64bed7ac3836 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h @@ -1217,7 +1217,7 @@ void mlx4_cmd_use_polling(struct mlx4_dev *dev); int mlx4_comm_cmd(struct mlx4_dev *dev, u8 cmd, u16 param, u16 op, unsigned long timeout); -void mlx4_cq_tasklet_cb(unsigned long data); +void mlx4_cq_tasklet_cb(struct tasklet_struct *t); void mlx4_cq_completion(struct mlx4_dev *dev, u32 cqn); void mlx4_cq_event(struct mlx4_dev *dev, u32 cqn, int event_type); diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_stats.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_stats.h index 86b6051da8ec..51d4eaab6a2f 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_stats.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_stats.h @@ -84,6 +84,11 @@ struct mlx4_en_flow_stats_rx { MLX4_NUM_PRIORITIES) }; +#define FLOW_PRIORITY_STATS_IDX_RX_FRAMES (NUM_MAIN_STATS + \ + NUM_PORT_STATS + \ + NUM_PF_STATS + \ + NUM_FLOW_PRIORITY_STATS_RX) + struct mlx4_en_flow_stats_tx { u64 tx_pause; u64 tx_pause_duration; @@ -93,6 +98,13 @@ struct mlx4_en_flow_stats_tx { MLX4_NUM_PRIORITIES) }; +#define FLOW_PRIORITY_STATS_IDX_TX_FRAMES (NUM_MAIN_STATS + \ + NUM_PORT_STATS + \ + NUM_PF_STATS + \ + NUM_FLOW_PRIORITY_STATS_RX + \ + NUM_FLOW_STATS_RX + \ + NUM_FLOW_PRIORITY_STATS_TX) + #define NUM_FLOW_STATS (NUM_FLOW_STATS_RX + NUM_FLOW_STATS_TX + \ NUM_FLOW_PRIORITY_STATS_TX + \ NUM_FLOW_PRIORITY_STATS_RX) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index 10e6886c96ba..2d477f9a8cb7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -16,7 +16,7 @@ mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ transobj.o vport.o sriov.o fs_cmd.o fs_core.o pci_irq.o \ fs_counters.o rl.o lag.o dev.o events.o wq.o lib/gid.o \ lib/devcom.o lib/pci_vsc.o lib/dm.o diag/fs_tracepoint.o \ - diag/fw_tracer.o diag/crdump.o devlink.o diag/rsc_dump.o + diag/fw_tracer.o diag/crdump.o devlink.o diag/rsc_dump.o fw_reset.o # # Netdev basic @@ -24,7 +24,7 @@ mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ mlx5_core-$(CONFIG_MLX5_CORE_EN) += en_main.o en_common.o en_fs.o en_ethtool.o \ en_tx.o en_rx.o en_dim.o en_txrx.o en/xdp.o en_stats.o \ en_selftest.o en/port.o en/monitor_stats.o en/health.o \ - en/reporter_tx.o en/reporter_rx.o en/params.o en/xsk/umem.o \ + en/reporter_tx.o en/reporter_rx.o en/params.o en/xsk/pool.o \ en/xsk/setup.o en/xsk/rx.o en/xsk/tx.o en/devlink.o # @@ -37,7 +37,7 @@ mlx5_core-$(CONFIG_PCI_HYPERV_INTERFACE) += en/hv_vhca_stats.o mlx5_core-$(CONFIG_MLX5_ESWITCH) += lag_mp.o lib/geneve.o lib/port_tun.o \ en_rep.o en/rep/bond.o en/mod_hdr.o mlx5_core-$(CONFIG_MLX5_CLS_ACT) += en_tc.o en/rep/tc.o en/rep/neigh.o \ - en/mapping.o esw/chains.o en/tc_tun.o \ + en/mapping.o lib/fs_chains.o en/tc_tun.o \ en/tc_tun_vxlan.o en/tc_tun_gre.o en/tc_tun_geneve.o \ en/tc_tun_mplsoudp.o diag/en_tc_tracepoint.o mlx5_core-$(CONFIG_MLX5_TC_CT) += en/tc_ct.o @@ -49,7 +49,8 @@ mlx5_core-$(CONFIG_MLX5_ESWITCH) += eswitch.o eswitch_offloads.o eswitch_offlo ecpf.o rdma.o mlx5_core-$(CONFIG_MLX5_ESWITCH) += esw/acl/helper.o \ esw/acl/egress_lgcy.o esw/acl/egress_ofld.o \ - esw/acl/ingress_lgcy.o esw/acl/ingress_ofld.o + esw/acl/ingress_lgcy.o esw/acl/ingress_ofld.o \ + esw/devlink_port.o mlx5_core-$(CONFIG_MLX5_MPFS) += lib/mpfs.o mlx5_core-$(CONFIG_VXLAN) += lib/vxlan.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/alloc.c b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c index 8db4b5f0f963..291e427e9e4f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/alloc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c @@ -56,8 +56,8 @@ static void *mlx5_dma_zalloc_coherent_node(struct mlx5_core_dev *dev, size_t size, dma_addr_t *dma_handle, int node) { + struct device *device = mlx5_core_dma_dev(dev); struct mlx5_priv *priv = &dev->priv; - struct device *device = dev->device; int original_node; void *cpu_handle; @@ -111,7 +111,7 @@ EXPORT_SYMBOL(mlx5_buf_alloc); void mlx5_buf_free(struct mlx5_core_dev *dev, struct mlx5_frag_buf *buf) { - dma_free_coherent(dev->device, buf->size, buf->frags->buf, + dma_free_coherent(mlx5_core_dma_dev(dev), buf->size, buf->frags->buf, buf->frags->map); kfree(buf->frags); @@ -140,7 +140,7 @@ int mlx5_frag_buf_alloc_node(struct mlx5_core_dev *dev, int size, if (!frag->buf) goto err_free_buf; if (frag->map & ((1 << buf->page_shift) - 1)) { - dma_free_coherent(dev->device, frag_sz, + dma_free_coherent(mlx5_core_dma_dev(dev), frag_sz, buf->frags[i].buf, buf->frags[i].map); mlx5_core_warn(dev, "unexpected map alignment: %pad, page_shift=%d\n", &frag->map, buf->page_shift); @@ -153,7 +153,7 @@ int mlx5_frag_buf_alloc_node(struct mlx5_core_dev *dev, int size, err_free_buf: while (i--) - dma_free_coherent(dev->device, PAGE_SIZE, buf->frags[i].buf, + dma_free_coherent(mlx5_core_dma_dev(dev), PAGE_SIZE, buf->frags[i].buf, buf->frags[i].map); kfree(buf->frags); err_out: @@ -169,7 +169,7 @@ void mlx5_frag_buf_free(struct mlx5_core_dev *dev, struct mlx5_frag_buf *buf) for (i = 0; i < buf->npages; i++) { int frag_sz = min_t(int, size, PAGE_SIZE); - dma_free_coherent(dev->device, frag_sz, buf->frags[i].buf, + dma_free_coherent(mlx5_core_dma_dev(dev), frag_sz, buf->frags[i].buf, buf->frags[i].map); size -= frag_sz; } @@ -275,7 +275,7 @@ void mlx5_db_free(struct mlx5_core_dev *dev, struct mlx5_db *db) __set_bit(db->index, db->u.pgdir->bitmap); if (bitmap_full(db->u.pgdir->bitmap, db_per_page)) { - dma_free_coherent(dev->device, PAGE_SIZE, + dma_free_coherent(mlx5_core_dma_dev(dev), PAGE_SIZE, db->u.pgdir->db_page, db->u.pgdir->db_dma); list_del(&db->u.pgdir->list); bitmap_free(db->u.pgdir->bitmap); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index 2d1f4b3be9bf..e49387dbef98 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -1989,9 +1989,7 @@ static void create_msg_cache(struct mlx5_core_dev *dev) static int alloc_cmd_page(struct mlx5_core_dev *dev, struct mlx5_cmd *cmd) { - struct device *ddev = dev->device; - - cmd->cmd_alloc_buf = dma_alloc_coherent(ddev, MLX5_ADAPTER_PAGE_SIZE, + cmd->cmd_alloc_buf = dma_alloc_coherent(mlx5_core_dma_dev(dev), MLX5_ADAPTER_PAGE_SIZE, &cmd->alloc_dma, GFP_KERNEL); if (!cmd->cmd_alloc_buf) return -ENOMEM; @@ -2004,9 +2002,9 @@ static int alloc_cmd_page(struct mlx5_core_dev *dev, struct mlx5_cmd *cmd) return 0; } - dma_free_coherent(ddev, MLX5_ADAPTER_PAGE_SIZE, cmd->cmd_alloc_buf, + dma_free_coherent(mlx5_core_dma_dev(dev), MLX5_ADAPTER_PAGE_SIZE, cmd->cmd_alloc_buf, cmd->alloc_dma); - cmd->cmd_alloc_buf = dma_alloc_coherent(ddev, + cmd->cmd_alloc_buf = dma_alloc_coherent(mlx5_core_dma_dev(dev), 2 * MLX5_ADAPTER_PAGE_SIZE - 1, &cmd->alloc_dma, GFP_KERNEL); if (!cmd->cmd_alloc_buf) @@ -2020,9 +2018,7 @@ static int alloc_cmd_page(struct mlx5_core_dev *dev, struct mlx5_cmd *cmd) static void free_cmd_page(struct mlx5_core_dev *dev, struct mlx5_cmd *cmd) { - struct device *ddev = dev->device; - - dma_free_coherent(ddev, cmd->alloc_size, cmd->cmd_alloc_buf, + dma_free_coherent(mlx5_core_dma_dev(dev), cmd->alloc_size, cmd->cmd_alloc_buf, cmd->alloc_dma); } @@ -2054,7 +2050,7 @@ int mlx5_cmd_init(struct mlx5_core_dev *dev) if (!cmd->stats) return -ENOMEM; - cmd->pool = dma_pool_create("mlx5_cmd", dev->device, size, align, 0); + cmd->pool = dma_pool_create("mlx5_cmd", mlx5_core_dma_dev(dev), size, align, 0); if (!cmd->pool) { err = -ENOMEM; goto dma_pool_err; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cq.c b/drivers/net/ethernet/mellanox/mlx5/core/cq.c index 8379b24cb838..df3e4938ecdd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cq.c @@ -42,11 +42,11 @@ #define TASKLET_MAX_TIME 2 #define TASKLET_MAX_TIME_JIFFIES msecs_to_jiffies(TASKLET_MAX_TIME) -void mlx5_cq_tasklet_cb(unsigned long data) +void mlx5_cq_tasklet_cb(struct tasklet_struct *t) { unsigned long flags; unsigned long end = jiffies + TASKLET_MAX_TIME_JIFFIES; - struct mlx5_eq_tasklet *ctx = (struct mlx5_eq_tasklet *)data; + struct mlx5_eq_tasklet *ctx = from_tasklet(ctx, t, task); struct mlx5_core_cq *mcq; struct mlx5_core_cq *temp; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c index c709e9a385f6..a28f95df2901 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c @@ -4,22 +4,19 @@ #include <devlink.h> #include "mlx5_core.h" +#include "fw_reset.h" #include "fs_core.h" #include "eswitch.h" static int mlx5_devlink_flash_update(struct devlink *devlink, - const char *file_name, - const char *component, + struct devlink_flash_update_params *params, struct netlink_ext_ack *extack) { struct mlx5_core_dev *dev = devlink_priv(devlink); const struct firmware *fw; int err; - if (component) - return -EOPNOTSUPP; - - err = request_firmware_direct(&fw, file_name, &dev->pdev->dev); + err = request_firmware_direct(&fw, params->file_name, &dev->pdev->dev); if (err) return err; @@ -88,21 +85,96 @@ mlx5_devlink_info_get(struct devlink *devlink, struct devlink_info_req *req, return 0; } +static int mlx5_devlink_reload_fw_activate(struct devlink *devlink, struct netlink_ext_ack *extack) +{ + struct mlx5_core_dev *dev = devlink_priv(devlink); + u8 reset_level, reset_type, net_port_alive; + int err; + + err = mlx5_fw_reset_query(dev, &reset_level, &reset_type); + if (err) + return err; + if (!(reset_level & MLX5_MFRL_REG_RESET_LEVEL3)) { + NL_SET_ERR_MSG_MOD(extack, "FW activate requires reboot"); + return -EINVAL; + } + + net_port_alive = !!(reset_type & MLX5_MFRL_REG_RESET_TYPE_NET_PORT_ALIVE); + err = mlx5_fw_reset_set_reset_sync(dev, net_port_alive); + if (err) + goto out; + + err = mlx5_fw_reset_wait_reset_done(dev); +out: + if (err) + NL_SET_ERR_MSG_MOD(extack, "FW activate command failed"); + return err; +} + +static int mlx5_devlink_trigger_fw_live_patch(struct devlink *devlink, + struct netlink_ext_ack *extack) +{ + struct mlx5_core_dev *dev = devlink_priv(devlink); + u8 reset_level; + int err; + + err = mlx5_fw_reset_query(dev, &reset_level, NULL); + if (err) + return err; + if (!(reset_level & MLX5_MFRL_REG_RESET_LEVEL0)) { + NL_SET_ERR_MSG_MOD(extack, + "FW upgrade to the stored FW can't be done by FW live patching"); + return -EINVAL; + } + + return mlx5_fw_reset_set_live_patch(dev); +} + static int mlx5_devlink_reload_down(struct devlink *devlink, bool netns_change, + enum devlink_reload_action action, + enum devlink_reload_limit limit, struct netlink_ext_ack *extack) { struct mlx5_core_dev *dev = devlink_priv(devlink); - mlx5_unload_one(dev, false); - return 0; + switch (action) { + case DEVLINK_RELOAD_ACTION_DRIVER_REINIT: + mlx5_unload_one(dev, false); + return 0; + case DEVLINK_RELOAD_ACTION_FW_ACTIVATE: + if (limit == DEVLINK_RELOAD_LIMIT_NO_RESET) + return mlx5_devlink_trigger_fw_live_patch(devlink, extack); + return mlx5_devlink_reload_fw_activate(devlink, extack); + default: + /* Unsupported action should not get to this function */ + WARN_ON(1); + return -EOPNOTSUPP; + } } -static int mlx5_devlink_reload_up(struct devlink *devlink, +static int mlx5_devlink_reload_up(struct devlink *devlink, enum devlink_reload_action action, + enum devlink_reload_limit limit, u32 *actions_performed, struct netlink_ext_ack *extack) { struct mlx5_core_dev *dev = devlink_priv(devlink); - return mlx5_load_one(dev, false); + *actions_performed = BIT(action); + switch (action) { + case DEVLINK_RELOAD_ACTION_DRIVER_REINIT: + return mlx5_load_one(dev, false); + case DEVLINK_RELOAD_ACTION_FW_ACTIVATE: + if (limit == DEVLINK_RELOAD_LIMIT_NO_RESET) + break; + /* On fw_activate action, also driver is reloaded and reinit performed */ + *actions_performed |= BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT); + return mlx5_load_one(dev, false); + default: + /* Unsupported action should not get to this function */ + WARN_ON(1); + return -EOPNOTSUPP; + } + + return 0; } static const struct devlink_ops mlx5_devlink_ops = { @@ -118,6 +190,9 @@ static const struct devlink_ops mlx5_devlink_ops = { #endif .flash_update = mlx5_devlink_flash_update, .info_get = mlx5_devlink_info_get, + .reload_actions = BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT) | + BIT(DEVLINK_RELOAD_ACTION_FW_ACTIVATE), + .reload_limits = BIT(DEVLINK_RELOAD_LIMIT_NO_RESET), .reload_down = mlx5_devlink_reload_down, .reload_up = mlx5_devlink_reload_up, }; @@ -228,6 +303,24 @@ static int mlx5_devlink_large_group_num_validate(struct devlink *devlink, u32 id } #endif +static int mlx5_devlink_enable_remote_dev_reset_set(struct devlink *devlink, u32 id, + struct devlink_param_gset_ctx *ctx) +{ + struct mlx5_core_dev *dev = devlink_priv(devlink); + + mlx5_fw_reset_enable_remote_dev_reset_set(dev, ctx->val.vbool); + return 0; +} + +static int mlx5_devlink_enable_remote_dev_reset_get(struct devlink *devlink, u32 id, + struct devlink_param_gset_ctx *ctx) +{ + struct mlx5_core_dev *dev = devlink_priv(devlink); + + ctx->val.vbool = mlx5_fw_reset_enable_remote_dev_reset_get(dev); + return 0; +} + static const struct devlink_param mlx5_devlink_params[] = { DEVLINK_PARAM_DRIVER(MLX5_DEVLINK_PARAM_ID_FLOW_STEERING_MODE, "flow_steering_mode", DEVLINK_PARAM_TYPE_STRING, @@ -243,6 +336,9 @@ static const struct devlink_param mlx5_devlink_params[] = { NULL, NULL, mlx5_devlink_large_group_num_validate), #endif + DEVLINK_PARAM_GENERIC(ENABLE_REMOTE_DEV_RESET, BIT(DEVLINK_PARAM_CMODE_RUNTIME), + mlx5_devlink_enable_remote_dev_reset_get, + mlx5_devlink_enable_remote_dev_reset_set, NULL), }; static void mlx5_devlink_set_params_init_values(struct devlink *devlink) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c index ad3594c4afcb..2eb022ad7fd0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c @@ -124,7 +124,7 @@ static void mlx5_fw_tracer_ownership_release(struct mlx5_fw_tracer *tracer) static int mlx5_fw_tracer_create_log_buf(struct mlx5_fw_tracer *tracer) { struct mlx5_core_dev *dev = tracer->dev; - struct device *ddev = &dev->pdev->dev; + struct device *ddev; dma_addr_t dma; void *buff; gfp_t gfp; @@ -142,6 +142,7 @@ static int mlx5_fw_tracer_create_log_buf(struct mlx5_fw_tracer *tracer) } tracer->buff.log_buf = buff; + ddev = mlx5_core_dma_dev(dev); dma = dma_map_single(ddev, buff, tracer->buff.size, DMA_FROM_DEVICE); if (dma_mapping_error(ddev, dma)) { mlx5_core_warn(dev, "FWTracer: Unable to map DMA: %d\n", @@ -162,11 +163,12 @@ free_pages: static void mlx5_fw_tracer_destroy_log_buf(struct mlx5_fw_tracer *tracer) { struct mlx5_core_dev *dev = tracer->dev; - struct device *ddev = &dev->pdev->dev; + struct device *ddev; if (!tracer->buff.log_buf) return; + ddev = mlx5_core_dma_dev(dev); dma_unmap_single(ddev, tracer->buff.dma, tracer->buff.size, DMA_FROM_DEVICE); free_pages((unsigned long)tracer->buff.log_buf, get_order(tracer->buff.size)); } @@ -1064,6 +1066,58 @@ void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer) kvfree(tracer); } +static int mlx5_fw_tracer_recreate_strings_db(struct mlx5_fw_tracer *tracer) +{ + struct mlx5_core_dev *dev; + int err; + + cancel_work_sync(&tracer->read_fw_strings_work); + mlx5_fw_tracer_clean_ready_list(tracer); + mlx5_fw_tracer_clean_print_hash(tracer); + mlx5_fw_tracer_clean_saved_traces_array(tracer); + mlx5_fw_tracer_free_strings_db(tracer); + + dev = tracer->dev; + err = mlx5_query_mtrc_caps(tracer); + if (err) { + mlx5_core_dbg(dev, "FWTracer: Failed to query capabilities %d\n", err); + return err; + } + + err = mlx5_fw_tracer_allocate_strings_db(tracer); + if (err) { + mlx5_core_warn(dev, "FWTracer: Allocate strings DB failed %d\n", err); + return err; + } + mlx5_fw_tracer_init_saved_traces_array(tracer); + + return 0; +} + +int mlx5_fw_tracer_reload(struct mlx5_fw_tracer *tracer) +{ + struct mlx5_core_dev *dev; + int err; + + if (IS_ERR_OR_NULL(tracer)) + return -EINVAL; + + dev = tracer->dev; + mlx5_fw_tracer_cleanup(tracer); + err = mlx5_fw_tracer_recreate_strings_db(tracer); + if (err) { + mlx5_core_warn(dev, "Failed to recreate FW tracer strings DB\n"); + return err; + } + err = mlx5_fw_tracer_init(tracer); + if (err) { + mlx5_core_warn(dev, "Failed to re-initialize FW tracer\n"); + return err; + } + + return 0; +} + static int fw_tracer_event(struct notifier_block *nb, unsigned long action, void *data) { struct mlx5_fw_tracer *tracer = mlx5_nb_cof(nb, struct mlx5_fw_tracer, nb); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h index 40601fba80ba..97252a85d65e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h @@ -191,5 +191,6 @@ void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer); int mlx5_fw_tracer_trigger_core_dump_general(struct mlx5_core_dev *dev); int mlx5_fw_tracer_get_saved_traces_objects(struct mlx5_fw_tracer *tracer, struct devlink_fmsg *fmsg); +int mlx5_fw_tracer_reload(struct mlx5_fw_tracer *tracer); #endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/rsc_dump.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/rsc_dump.c index 4924a5658853..ed4fb79b4db7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/rsc_dump.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/rsc_dump.c @@ -78,7 +78,7 @@ static int mlx5_rsc_dump_trigger(struct mlx5_core_dev *dev, struct mlx5_rsc_dump struct page *page) { struct mlx5_rsc_dump *rsc_dump = dev->rsc_dump; - struct device *ddev = &dev->pdev->dev; + struct device *ddev = mlx5_core_dma_dev(dev); u32 out_seq_num; u32 in_seq_num; dma_addr_t dma; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ecpf.c b/drivers/net/ethernet/mellanox/mlx5/core/ecpf.c index a894ea98c95a..3dc9dd3f24dc 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ecpf.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ecpf.c @@ -43,19 +43,13 @@ static void mlx5_peer_pf_cleanup(struct mlx5_core_dev *dev) int mlx5_ec_init(struct mlx5_core_dev *dev) { - int err = 0; - if (!mlx5_core_is_ecpf(dev)) return 0; /* ECPF shall enable HCA for peer PF in the same way a PF * does this for its VFs. */ - err = mlx5_peer_pf_init(dev); - if (err) - return err; - - return 0; + return mlx5_peer_pf_init(dev); } void mlx5_ec_cleanup(struct mlx5_core_dev *dev) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 356f5852955f..2f05b0f9de01 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -226,6 +226,7 @@ enum mlx5e_priv_flag { MLX5E_PFLAG_RX_STRIDING_RQ, MLX5E_PFLAG_RX_NO_CSUM_COMPLETE, MLX5E_PFLAG_XDP_TX_MPWQE, + MLX5E_PFLAG_SKB_TX_MPWQE, MLX5E_NUM_PFLAGS, /* Keep last */ }; @@ -270,6 +271,7 @@ enum { MLX5E_RQ_STATE_NO_CSUM_COMPLETE, MLX5E_RQ_STATE_CSUM_FULL, /* cqe_csum_full hw bit is set */ MLX5E_RQ_STATE_FPGA_TLS, /* FPGA TLS enabled */ + MLX5E_RQ_STATE_MINI_CQE_HW_STRIDX /* set when mini_cqe_resp_stride_index cap is used */ }; struct mlx5e_cq { @@ -309,6 +311,7 @@ struct mlx5e_sq_dma { enum { MLX5E_SQ_STATE_ENABLED, + MLX5E_SQ_STATE_MPWQE, MLX5E_SQ_STATE_RECOVERING, MLX5E_SQ_STATE_IPSEC, MLX5E_SQ_STATE_AM, @@ -317,26 +320,40 @@ enum { MLX5E_SQ_STATE_PENDING_XSK_TX, }; +struct mlx5e_tx_mpwqe { + /* Current MPWQE session */ + struct mlx5e_tx_wqe *wqe; + u32 bytes_count; + u8 ds_count; + u8 pkt_count; + u8 inline_on; +}; + struct mlx5e_txqsq { /* data path */ /* dirtied @completion */ u16 cc; + u16 skb_fifo_cc; u32 dma_fifo_cc; struct dim dim; /* Adaptive Moderation */ /* dirtied @xmit */ u16 pc ____cacheline_aligned_in_smp; + u16 skb_fifo_pc; u32 dma_fifo_pc; + struct mlx5e_tx_mpwqe mpwqe; struct mlx5e_cq cq; /* read only */ struct mlx5_wq_cyc wq; u32 dma_fifo_mask; + u16 skb_fifo_mask; struct mlx5e_sq_stats *stats; struct { struct mlx5e_sq_dma *dma_fifo; + struct sk_buff **skb_fifo; struct mlx5e_tx_wqe_info *wqe_info; } db; void __iomem *uar_map; @@ -403,7 +420,7 @@ struct mlx5e_xdp_info { }; }; -struct mlx5e_xdp_xmit_data { +struct mlx5e_xmit_data { dma_addr_t dma_addr; void *data; u32 len; @@ -416,18 +433,10 @@ struct mlx5e_xdp_info_fifo { u32 mask; }; -struct mlx5e_xdp_mpwqe { - /* Current MPWQE session */ - struct mlx5e_tx_wqe *wqe; - u8 ds_count; - u8 pkt_count; - u8 inline_on; -}; - struct mlx5e_xdpsq; typedef int (*mlx5e_fp_xmit_xdp_frame_check)(struct mlx5e_xdpsq *); typedef bool (*mlx5e_fp_xmit_xdp_frame)(struct mlx5e_xdpsq *, - struct mlx5e_xdp_xmit_data *, + struct mlx5e_xmit_data *, struct mlx5e_xdp_info *, int); @@ -442,12 +451,12 @@ struct mlx5e_xdpsq { u32 xdpi_fifo_pc ____cacheline_aligned_in_smp; u16 pc; struct mlx5_wqe_ctrl_seg *doorbell_cseg; - struct mlx5e_xdp_mpwqe mpwqe; + struct mlx5e_tx_mpwqe mpwqe; struct mlx5e_cq cq; /* read only */ - struct xdp_umem *umem; + struct xsk_buff_pool *xsk_pool; struct mlx5_wq_cyc wq; struct mlx5e_xdpsq_stats *stats; mlx5e_fp_xmit_xdp_frame_check xmit_xdp_frame_check; @@ -611,7 +620,7 @@ struct mlx5e_rq { struct page_pool *page_pool; /* AF_XDP zero-copy */ - struct xdp_umem *umem; + struct xsk_buff_pool *xsk_pool; struct work_struct recover_work; @@ -735,12 +744,13 @@ struct mlx5e_hv_vhca_stats_agent { #endif struct mlx5e_xsk { - /* UMEMs are stored separately from channels, because we don't want to - * lose them when channels are recreated. The kernel also stores UMEMs, - * but it doesn't distinguish between zero-copy and non-zero-copy UMEMs, - * so rely on our mechanism. + /* XSK buffer pools are stored separately from channels, + * because we don't want to lose them when channels are + * recreated. The kernel also stores buffer pool, but it doesn't + * distinguish between zero-copy and non-zero-copy UMEMs, so + * rely on our mechanism. */ - struct xdp_umem **umems; + struct xsk_buff_pool **pools; u16 refcnt; bool ever_used; }; @@ -899,7 +909,7 @@ struct mlx5e_xsk_param; struct mlx5e_rq_param; int mlx5e_open_rq(struct mlx5e_channel *c, struct mlx5e_params *params, struct mlx5e_rq_param *param, struct mlx5e_xsk_param *xsk, - struct xdp_umem *umem, struct mlx5e_rq *rq); + struct xsk_buff_pool *xsk_pool, struct mlx5e_rq *rq); int mlx5e_wait_for_min_rx_wqes(struct mlx5e_rq *rq, int wait_time); void mlx5e_deactivate_rq(struct mlx5e_rq *rq); void mlx5e_close_rq(struct mlx5e_rq *rq); @@ -909,7 +919,7 @@ int mlx5e_open_icosq(struct mlx5e_channel *c, struct mlx5e_params *params, struct mlx5e_sq_param *param, struct mlx5e_icosq *sq); void mlx5e_close_icosq(struct mlx5e_icosq *sq); int mlx5e_open_xdpsq(struct mlx5e_channel *c, struct mlx5e_params *params, - struct mlx5e_sq_param *param, struct xdp_umem *umem, + struct mlx5e_sq_param *param, struct xsk_buff_pool *xsk_pool, struct mlx5e_xdpsq *sq, bool is_redirect); void mlx5e_close_xdpsq(struct mlx5e_xdpsq *sq); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h b/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h index 6fdcd5e69476..dc744702aee4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h @@ -12,9 +12,12 @@ enum { }; struct mlx5e_tc_table { - /* protects flow table */ + /* Protects the dynamic assignment of the t parameter + * which is the nic tc root table. + */ struct mutex t_lock; struct mlx5_flow_table *t; + struct mlx5_fs_chains *chains; struct rhashtable ht; @@ -24,6 +27,8 @@ struct mlx5e_tc_table { struct notifier_block netdevice_nb; struct netdev_net_notifier netdevice_nn; + + struct mlx5_tc_ct_priv *ct; }; struct mlx5e_flow_table { @@ -231,6 +236,7 @@ struct mlx5e_accel_fs_tcp; struct mlx5e_flow_steering { struct mlx5_flow_namespace *ns; + struct mlx5_flow_namespace *egress_ns; #ifdef CONFIG_MLX5_EN_RXNFC struct mlx5e_ethtool_steering ethtool; #endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/health.c b/drivers/net/ethernet/mellanox/mlx5/core/en/health.c index 3dc200bcfabd..69a05da0e3e3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/health.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/health.c @@ -242,8 +242,8 @@ static int mlx5e_health_rsc_fmsg_binary(struct devlink_fmsg *fmsg, { u32 data_size; + int err = 0; u32 offset; - int err; for (offset = 0; offset < value_len; offset += data_size) { data_size = value_len - offset; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c index 79cc42d88eec..e36e505d38ad 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c @@ -12,7 +12,7 @@ #include "neigh.h" #include "en_rep.h" #include "eswitch.h" -#include "esw/chains.h" +#include "lib/fs_chains.h" #include "en/tc_ct.h" #include "en/mapping.h" #include "en/tc_tun.h" @@ -191,7 +191,7 @@ static int mlx5e_rep_setup_ft_cb(enum tc_setup_type type, void *type_data, case TC_SETUP_CLSFLOWER: memcpy(&tmp, f, sizeof(*f)); - if (!mlx5_esw_chains_prios_supported(esw)) + if (!mlx5_chains_prios_supported(esw_chains(esw))) return -EOPNOTSUPP; /* Re-use tc offload path by moving the ft flow to the @@ -203,12 +203,12 @@ static int mlx5e_rep_setup_ft_cb(enum tc_setup_type type, void *type_data, * * We only support chain 0 of FT offload. */ - if (tmp.common.prio >= mlx5_esw_chains_get_prio_range(esw)) + if (tmp.common.prio >= mlx5_chains_get_prio_range(esw_chains(esw))) return -EOPNOTSUPP; if (tmp.common.chain_index != 0) return -EOPNOTSUPP; - tmp.common.chain_index = mlx5_esw_chains_get_ft_chain(esw); + tmp.common.chain_index = mlx5_chains_get_nf_ft_chain(esw_chains(esw)); tmp.common.prio++; err = mlx5e_rep_setup_tc_cls_flower(priv, &tmp, flags); memcpy(&f->stats, &tmp.stats, sizeof(f->stats)); @@ -378,12 +378,12 @@ static int mlx5e_rep_indr_setup_ft_cb(enum tc_setup_type type, * * We only support chain 0 of FT offload. */ - if (!mlx5_esw_chains_prios_supported(esw) || - tmp.common.prio >= mlx5_esw_chains_get_prio_range(esw) || + if (!mlx5_chains_prios_supported(esw_chains(esw)) || + tmp.common.prio >= mlx5_chains_get_prio_range(esw_chains(esw)) || tmp.common.chain_index) return -EOPNOTSUPP; - tmp.common.chain_index = mlx5_esw_chains_get_ft_chain(esw); + tmp.common.chain_index = mlx5_chains_get_nf_ft_chain(esw_chains(esw)); tmp.common.prio++; err = mlx5e_rep_indr_offload(priv->netdev, &tmp, priv, flags); memcpy(&f->stats, &tmp.stats, sizeof(f->stats)); @@ -612,7 +612,6 @@ bool mlx5e_rep_tc_update_skb(struct mlx5_cqe64 *cqe, struct tc_skb_ext *tc_skb_ext; struct mlx5_eswitch *esw; struct mlx5e_priv *priv; - int tunnel_moffset; int err; reg_c0 = (be32_to_cpu(cqe->sop_drop_qpn) & MLX5E_TC_FLOW_ID_MASK); @@ -626,7 +625,7 @@ bool mlx5e_rep_tc_update_skb(struct mlx5_cqe64 *cqe, priv = netdev_priv(skb->dev); esw = priv->mdev->priv.eswitch; - err = mlx5_eswitch_get_chain_for_tag(esw, reg_c0, &chain); + err = mlx5_get_chain_for_tag(esw_chains(esw), reg_c0, &chain); if (err) { netdev_dbg(priv->netdev, "Couldn't find chain for chain tag: %d, err: %d\n", @@ -647,13 +646,12 @@ bool mlx5e_rep_tc_update_skb(struct mlx5_cqe64 *cqe, uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH); uplink_priv = &uplink_rpriv->uplink_priv; - if (!mlx5e_tc_ct_restore_flow(uplink_priv, skb, + if (!mlx5e_tc_ct_restore_flow(uplink_priv->ct_priv, skb, zone_restore_id)) return false; } - tunnel_moffset = mlx5e_tc_attr_to_reg_mappings[TUNNEL_TO_REG].moffset; - tunnel_id = reg_c1 >> (8 * tunnel_moffset); + tunnel_id = reg_c1 >> REG_MAPPING_SHIFT(TUNNEL_TO_REG); return mlx5e_restore_tunnel(priv, skb, tc_priv, tunnel_id); #endif /* CONFIG_NET_TC_SKB_EXT */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c index a8be40cbe325..e521254d886e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c @@ -14,7 +14,7 @@ #include <linux/workqueue.h> #include <linux/xarray.h> -#include "esw/chains.h" +#include "lib/fs_chains.h" #include "en/tc_ct.h" #include "en/mod_hdr.h" #include "en/mapping.h" @@ -39,8 +39,9 @@ netdev_dbg(ct_priv->netdev, "ct_debug: " fmt "\n", ##args) struct mlx5_tc_ct_priv { - struct mlx5_eswitch *esw; + struct mlx5_core_dev *dev; const struct net_device *netdev; + struct mod_hdr_tbl *mod_hdr_tbl; struct idr fte_ids; struct xarray tuple_ids; struct rhashtable zone_ht; @@ -50,13 +51,16 @@ struct mlx5_tc_ct_priv { struct mlx5_flow_table *ct_nat; struct mlx5_flow_table *post_ct; struct mutex control_lock; /* guards parallel adds/dels */ + struct mutex shared_counter_lock; struct mapping_ctx *zone_mapping; struct mapping_ctx *labels_mapping; + enum mlx5_flow_namespace_type ns_type; + struct mlx5_fs_chains *chains; }; struct mlx5_ct_flow { - struct mlx5_esw_flow_attr pre_ct_attr; - struct mlx5_esw_flow_attr post_ct_attr; + struct mlx5_flow_attr *pre_ct_attr; + struct mlx5_flow_attr *post_ct_attr; struct mlx5_flow_handle *pre_ct_rule; struct mlx5_flow_handle *post_ct_rule; struct mlx5_ct_ft *ft; @@ -67,12 +71,12 @@ struct mlx5_ct_flow { struct mlx5_ct_zone_rule { struct mlx5_flow_handle *rule; struct mlx5e_mod_hdr_handle *mh; - struct mlx5_esw_flow_attr attr; + struct mlx5_flow_attr *attr; bool nat; }; struct mlx5_tc_ct_pre { - struct mlx5_flow_table *fdb; + struct mlx5_flow_table *ft; struct mlx5_flow_group *flow_grp; struct mlx5_flow_group *miss_grp; struct mlx5_flow_handle *flow_rule; @@ -114,11 +118,16 @@ struct mlx5_ct_tuple { u16 zone; }; +struct mlx5_ct_shared_counter { + struct mlx5_fc *counter; + refcount_t refcount; +}; + struct mlx5_ct_entry { struct rhash_head node; struct rhash_head tuple_node; struct rhash_head tuple_nat_node; - struct mlx5_fc *counter; + struct mlx5_ct_shared_counter *shared_counter; unsigned long cookie; unsigned long restore_cookie; struct mlx5_ct_tuple tuple; @@ -157,18 +166,6 @@ static const struct rhashtable_params tuples_nat_ht_params = { .min_size = 16 * 1024, }; -static struct mlx5_tc_ct_priv * -mlx5_tc_ct_get_ct_priv(struct mlx5e_priv *priv) -{ - struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; - struct mlx5_rep_uplink_priv *uplink_priv; - struct mlx5e_rep_priv *uplink_rpriv; - - uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH); - uplink_priv = &uplink_rpriv->uplink_priv; - return uplink_priv->ct_priv; -} - static int mlx5_tc_ct_rule_to_tuple(struct mlx5_ct_tuple *tuple, struct flow_rule *rule) { @@ -397,20 +394,30 @@ mlx5_tc_ct_set_tuple_match(struct mlx5e_priv *priv, struct mlx5_flow_spec *spec, } static void +mlx5_tc_ct_shared_counter_put(struct mlx5_tc_ct_priv *ct_priv, struct mlx5_ct_entry *entry) +{ + if (!refcount_dec_and_test(&entry->shared_counter->refcount)) + return; + + mlx5_fc_destroy(ct_priv->dev, entry->shared_counter->counter); + kfree(entry->shared_counter); +} + +static void mlx5_tc_ct_entry_del_rule(struct mlx5_tc_ct_priv *ct_priv, struct mlx5_ct_entry *entry, bool nat) { struct mlx5_ct_zone_rule *zone_rule = &entry->zone_rules[nat]; - struct mlx5_esw_flow_attr *attr = &zone_rule->attr; - struct mlx5_eswitch *esw = ct_priv->esw; + struct mlx5_flow_attr *attr = zone_rule->attr; ct_dbg("Deleting ct entry rule in zone %d", entry->tuple.zone); - mlx5_eswitch_del_offloaded_rule(esw, zone_rule->rule, attr); - mlx5e_mod_hdr_detach(ct_priv->esw->dev, - &esw->offloads.mod_hdr, zone_rule->mh); + mlx5_tc_rule_delete(netdev_priv(ct_priv->netdev), zone_rule->rule, attr); + mlx5e_mod_hdr_detach(ct_priv->dev, + ct_priv->mod_hdr_tbl, zone_rule->mh); mapping_remove(ct_priv->labels_mapping, attr->ct_attr.ct_labels_id); + kfree(attr); } static void @@ -419,8 +426,6 @@ mlx5_tc_ct_entry_del_rules(struct mlx5_tc_ct_priv *ct_priv, { mlx5_tc_ct_entry_del_rule(ct_priv, entry, true); mlx5_tc_ct_entry_del_rule(ct_priv, entry, false); - - mlx5_fc_destroy(ct_priv->esw->dev, entry->counter); } static struct flow_action_entry * @@ -446,29 +451,40 @@ mlx5_tc_ct_entry_set_registers(struct mlx5_tc_ct_priv *ct_priv, u32 labels_id, u8 zone_restore_id) { - struct mlx5_eswitch *esw = ct_priv->esw; + enum mlx5_flow_namespace_type ns = ct_priv->ns_type; + struct mlx5_core_dev *dev = ct_priv->dev; int err; - err = mlx5e_tc_match_to_reg_set(esw->dev, mod_acts, + err = mlx5e_tc_match_to_reg_set(dev, mod_acts, ns, CTSTATE_TO_REG, ct_state); if (err) return err; - err = mlx5e_tc_match_to_reg_set(esw->dev, mod_acts, + err = mlx5e_tc_match_to_reg_set(dev, mod_acts, ns, MARK_TO_REG, mark); if (err) return err; - err = mlx5e_tc_match_to_reg_set(esw->dev, mod_acts, + err = mlx5e_tc_match_to_reg_set(dev, mod_acts, ns, LABELS_TO_REG, labels_id); if (err) return err; - err = mlx5e_tc_match_to_reg_set(esw->dev, mod_acts, + err = mlx5e_tc_match_to_reg_set(dev, mod_acts, ns, ZONE_RESTORE_TO_REG, zone_restore_id); if (err) return err; + /* Make another copy of zone id in reg_b for + * NIC rx flows since we don't copy reg_c1 to + * reg_b upon miss. + */ + if (ns != MLX5_FLOW_NAMESPACE_FDB) { + err = mlx5e_tc_match_to_reg_set(dev, mod_acts, ns, + NIC_ZONE_RESTORE_TO_REG, zone_restore_id); + if (err) + return err; + } return 0; } @@ -549,7 +565,7 @@ mlx5_tc_ct_entry_create_nat(struct mlx5_tc_ct_priv *ct_priv, struct mlx5e_tc_mod_hdr_acts *mod_acts) { struct flow_action *flow_action = &flow_rule->action; - struct mlx5_core_dev *mdev = ct_priv->esw->dev; + struct mlx5_core_dev *mdev = ct_priv->dev; struct flow_action_entry *act; size_t action_size; char *modact; @@ -560,8 +576,7 @@ mlx5_tc_ct_entry_create_nat(struct mlx5_tc_ct_priv *ct_priv, flow_action_for_each(i, act, flow_action) { switch (act->id) { case FLOW_ACTION_MANGLE: { - err = alloc_mod_hdr_actions(mdev, - MLX5_FLOW_NAMESPACE_FDB, + err = alloc_mod_hdr_actions(mdev, ct_priv->ns_type, mod_acts); if (err) return err; @@ -590,7 +605,7 @@ mlx5_tc_ct_entry_create_nat(struct mlx5_tc_ct_priv *ct_priv, static int mlx5_tc_ct_entry_create_mod_hdr(struct mlx5_tc_ct_priv *ct_priv, - struct mlx5_esw_flow_attr *attr, + struct mlx5_flow_attr *attr, struct flow_rule *flow_rule, struct mlx5e_mod_hdr_handle **mh, u8 zone_restore_id, bool nat) @@ -626,9 +641,9 @@ mlx5_tc_ct_entry_create_mod_hdr(struct mlx5_tc_ct_priv *ct_priv, if (err) goto err_mapping; - *mh = mlx5e_mod_hdr_attach(ct_priv->esw->dev, - &ct_priv->esw->offloads.mod_hdr, - MLX5_FLOW_NAMESPACE_FDB, + *mh = mlx5e_mod_hdr_attach(ct_priv->dev, + ct_priv->mod_hdr_tbl, + ct_priv->ns_type, &mod_acts); if (IS_ERR(*mh)) { err = PTR_ERR(*mh); @@ -652,9 +667,9 @@ mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv, bool nat, u8 zone_restore_id) { struct mlx5_ct_zone_rule *zone_rule = &entry->zone_rules[nat]; - struct mlx5_esw_flow_attr *attr = &zone_rule->attr; - struct mlx5_eswitch *esw = ct_priv->esw; + struct mlx5e_priv *priv = netdev_priv(ct_priv->netdev); struct mlx5_flow_spec *spec = NULL; + struct mlx5_flow_attr *attr; int err; zone_rule->nat = nat; @@ -663,6 +678,12 @@ mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv, if (!spec) return -ENOMEM; + attr = mlx5_alloc_flow_attr(ct_priv->ns_type); + if (!attr) { + err = -ENOMEM; + goto err_attr; + } + err = mlx5_tc_ct_entry_create_mod_hdr(ct_priv, attr, flow_rule, &zone_rule->mh, zone_restore_id, nat); @@ -676,9 +697,9 @@ mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv, MLX5_FLOW_CONTEXT_ACTION_COUNT; attr->dest_chain = 0; attr->dest_ft = ct_priv->post_ct; - attr->fdb = nat ? ct_priv->ct_nat : ct_priv->ct; + attr->ft = nat ? ct_priv->ct_nat : ct_priv->ct; attr->outer_match_level = MLX5_MATCH_L4; - attr->counter = entry->counter; + attr->counter = entry->shared_counter->counter; attr->flags |= MLX5_ESW_ATTR_FLAG_NO_IN_PORT; mlx5_tc_ct_set_tuple_match(netdev_priv(ct_priv->netdev), spec, flow_rule); @@ -686,39 +707,100 @@ mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv, entry->tuple.zone & MLX5_CT_ZONE_MASK, MLX5_CT_ZONE_MASK); - zone_rule->rule = mlx5_eswitch_add_offloaded_rule(esw, spec, attr); + zone_rule->rule = mlx5_tc_rule_insert(priv, spec, attr); if (IS_ERR(zone_rule->rule)) { err = PTR_ERR(zone_rule->rule); ct_dbg("Failed to add ct entry rule, nat: %d", nat); goto err_rule; } + zone_rule->attr = attr; + kfree(spec); ct_dbg("Offloaded ct entry rule in zone %d", entry->tuple.zone); return 0; err_rule: - mlx5e_mod_hdr_detach(ct_priv->esw->dev, - &esw->offloads.mod_hdr, zone_rule->mh); + mlx5e_mod_hdr_detach(ct_priv->dev, + ct_priv->mod_hdr_tbl, zone_rule->mh); mapping_remove(ct_priv->labels_mapping, attr->ct_attr.ct_labels_id); err_mod_hdr: + kfree(attr); +err_attr: kfree(spec); return err; } +static struct mlx5_ct_shared_counter * +mlx5_tc_ct_shared_counter_get(struct mlx5_tc_ct_priv *ct_priv, + struct mlx5_ct_entry *entry) +{ + struct mlx5_ct_tuple rev_tuple = entry->tuple; + struct mlx5_ct_shared_counter *shared_counter; + struct mlx5_core_dev *dev = ct_priv->dev; + struct mlx5_ct_entry *rev_entry; + __be16 tmp_port; + int ret; + + /* get the reversed tuple */ + tmp_port = rev_tuple.port.src; + rev_tuple.port.src = rev_tuple.port.dst; + rev_tuple.port.dst = tmp_port; + + if (rev_tuple.addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) { + __be32 tmp_addr = rev_tuple.ip.src_v4; + + rev_tuple.ip.src_v4 = rev_tuple.ip.dst_v4; + rev_tuple.ip.dst_v4 = tmp_addr; + } else if (rev_tuple.addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS) { + struct in6_addr tmp_addr = rev_tuple.ip.src_v6; + + rev_tuple.ip.src_v6 = rev_tuple.ip.dst_v6; + rev_tuple.ip.dst_v6 = tmp_addr; + } else { + return ERR_PTR(-EOPNOTSUPP); + } + + /* Use the same counter as the reverse direction */ + mutex_lock(&ct_priv->shared_counter_lock); + rev_entry = rhashtable_lookup_fast(&ct_priv->ct_tuples_ht, &rev_tuple, + tuples_ht_params); + if (rev_entry) { + if (refcount_inc_not_zero(&rev_entry->shared_counter->refcount)) { + mutex_unlock(&ct_priv->shared_counter_lock); + return rev_entry->shared_counter; + } + } + mutex_unlock(&ct_priv->shared_counter_lock); + + shared_counter = kzalloc(sizeof(*shared_counter), GFP_KERNEL); + if (!shared_counter) + return ERR_PTR(-ENOMEM); + + shared_counter->counter = mlx5_fc_create(dev, true); + if (IS_ERR(shared_counter->counter)) { + ct_dbg("Failed to create counter for ct entry"); + ret = PTR_ERR(shared_counter->counter); + kfree(shared_counter); + return ERR_PTR(ret); + } + + refcount_set(&shared_counter->refcount, 1); + return shared_counter; +} + static int mlx5_tc_ct_entry_add_rules(struct mlx5_tc_ct_priv *ct_priv, struct flow_rule *flow_rule, struct mlx5_ct_entry *entry, u8 zone_restore_id) { - struct mlx5_eswitch *esw = ct_priv->esw; int err; - entry->counter = mlx5_fc_create(esw->dev, true); - if (IS_ERR(entry->counter)) { - err = PTR_ERR(entry->counter); + entry->shared_counter = mlx5_tc_ct_shared_counter_get(ct_priv, entry); + if (IS_ERR(entry->shared_counter)) { + err = PTR_ERR(entry->shared_counter); ct_dbg("Failed to create counter for ct entry"); return err; } @@ -738,7 +820,7 @@ mlx5_tc_ct_entry_add_rules(struct mlx5_tc_ct_priv *ct_priv, err_nat: mlx5_tc_ct_entry_del_rule(ct_priv, entry, false); err_orig: - mlx5_fc_destroy(esw->dev, entry->counter); + mlx5_tc_ct_shared_counter_put(ct_priv, entry); return err; } @@ -828,12 +910,16 @@ mlx5_tc_ct_del_ft_entry(struct mlx5_tc_ct_priv *ct_priv, struct mlx5_ct_entry *entry) { mlx5_tc_ct_entry_del_rules(ct_priv, entry); + mutex_lock(&ct_priv->shared_counter_lock); if (entry->tuple_node.next) rhashtable_remove_fast(&ct_priv->ct_tuples_nat_ht, &entry->tuple_nat_node, tuples_nat_ht_params); rhashtable_remove_fast(&ct_priv->ct_tuples_ht, &entry->tuple_node, tuples_ht_params); + mutex_unlock(&ct_priv->shared_counter_lock); + mlx5_tc_ct_shared_counter_put(ct_priv, entry); + } static int @@ -870,7 +956,7 @@ mlx5_tc_ct_block_flow_offload_stats(struct mlx5_ct_ft *ft, if (!entry) return -ENOENT; - mlx5_fc_query_cached(entry->counter, &bytes, &packets, &lastuse); + mlx5_fc_query_cached(entry->shared_counter->counter, &bytes, &packets, &lastuse); flow_stats_update(&f->stats, bytes, packets, 0, lastuse, FLOW_ACTION_HW_STATS_DELAYED); @@ -943,9 +1029,7 @@ out: return false; } -int -mlx5_tc_ct_add_no_trk_match(struct mlx5e_priv *priv, - struct mlx5_flow_spec *spec) +int mlx5_tc_ct_add_no_trk_match(struct mlx5_flow_spec *spec) { u32 ctstate = 0, ctstate_mask = 0; @@ -961,24 +1045,21 @@ mlx5_tc_ct_add_no_trk_match(struct mlx5e_priv *priv, return 0; } -void mlx5_tc_ct_match_del(struct mlx5e_priv *priv, struct mlx5_ct_attr *ct_attr) +void mlx5_tc_ct_match_del(struct mlx5_tc_ct_priv *priv, struct mlx5_ct_attr *ct_attr) { - struct mlx5_tc_ct_priv *ct_priv = mlx5_tc_ct_get_ct_priv(priv); - - if (!ct_priv || !ct_attr->ct_labels_id) + if (!priv || !ct_attr->ct_labels_id) return; - mapping_remove(ct_priv->labels_mapping, ct_attr->ct_labels_id); + mapping_remove(priv->labels_mapping, ct_attr->ct_labels_id); } int -mlx5_tc_ct_match_add(struct mlx5e_priv *priv, +mlx5_tc_ct_match_add(struct mlx5_tc_ct_priv *priv, struct mlx5_flow_spec *spec, struct flow_cls_offload *f, struct mlx5_ct_attr *ct_attr, struct netlink_ext_ack *extack) { - struct mlx5_tc_ct_priv *ct_priv = mlx5_tc_ct_get_ct_priv(priv); struct flow_rule *rule = flow_cls_offload_flow_rule(f); struct flow_dissector_key_ct *mask, *key; bool trk, est, untrk, unest, new; @@ -991,7 +1072,7 @@ mlx5_tc_ct_match_add(struct mlx5e_priv *priv, if (!flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_CT)) return 0; - if (!ct_priv) { + if (!priv) { NL_SET_ERR_MSG_MOD(extack, "offload of ct matching isn't available"); return -EOPNOTSUPP; @@ -1047,7 +1128,7 @@ mlx5_tc_ct_match_add(struct mlx5e_priv *priv, ct_labels[1] = key->ct_labels[1] & mask->ct_labels[1]; ct_labels[2] = key->ct_labels[2] & mask->ct_labels[2]; ct_labels[3] = key->ct_labels[3] & mask->ct_labels[3]; - if (mapping_add(ct_priv->labels_mapping, ct_labels, &ct_attr->ct_labels_id)) + if (mapping_add(priv->labels_mapping, ct_labels, &ct_attr->ct_labels_id)) return -EOPNOTSUPP; mlx5e_tc_match_to_reg_match(spec, LABELS_TO_REG, ct_attr->ct_labels_id, MLX5_CT_LABELS_MASK); @@ -1057,14 +1138,12 @@ mlx5_tc_ct_match_add(struct mlx5e_priv *priv, } int -mlx5_tc_ct_parse_action(struct mlx5e_priv *priv, - struct mlx5_esw_flow_attr *attr, +mlx5_tc_ct_parse_action(struct mlx5_tc_ct_priv *priv, + struct mlx5_flow_attr *attr, const struct flow_action_entry *act, struct netlink_ext_ack *extack) { - struct mlx5_tc_ct_priv *ct_priv = mlx5_tc_ct_get_ct_priv(priv); - - if (!ct_priv) { + if (!priv) { NL_SET_ERR_MSG_MOD(extack, "offload of ct action isn't available"); return -EOPNOTSUPP; @@ -1083,8 +1162,8 @@ static int tc_ct_pre_ct_add_rules(struct mlx5_ct_ft *ct_ft, { struct mlx5_tc_ct_priv *ct_priv = ct_ft->ct_priv; struct mlx5e_tc_mod_hdr_acts pre_mod_acts = {}; - struct mlx5_core_dev *dev = ct_priv->esw->dev; - struct mlx5_flow_table *fdb = pre_ct->fdb; + struct mlx5_core_dev *dev = ct_priv->dev; + struct mlx5_flow_table *ft = pre_ct->ft; struct mlx5_flow_destination dest = {}; struct mlx5_flow_act flow_act = {}; struct mlx5_modify_hdr *mod_hdr; @@ -1099,14 +1178,14 @@ static int tc_ct_pre_ct_add_rules(struct mlx5_ct_ft *ct_ft, return -ENOMEM; zone = ct_ft->zone & MLX5_CT_ZONE_MASK; - err = mlx5e_tc_match_to_reg_set(dev, &pre_mod_acts, ZONE_TO_REG, zone); + err = mlx5e_tc_match_to_reg_set(dev, &pre_mod_acts, ct_priv->ns_type, + ZONE_TO_REG, zone); if (err) { ct_dbg("Failed to set zone register mapping"); goto err_mapping; } - mod_hdr = mlx5_modify_header_alloc(dev, - MLX5_FLOW_NAMESPACE_FDB, + mod_hdr = mlx5_modify_header_alloc(dev, ct_priv->ns_type, pre_mod_acts.num_actions, pre_mod_acts.actions); @@ -1132,7 +1211,7 @@ static int tc_ct_pre_ct_add_rules(struct mlx5_ct_ft *ct_ft, mlx5e_tc_match_to_reg_match(spec, CTSTATE_TO_REG, ctstate, ctstate); dest.ft = ct_priv->post_ct; - rule = mlx5_add_flow_rules(fdb, spec, &flow_act, &dest, 1); + rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1); if (IS_ERR(rule)) { err = PTR_ERR(rule); ct_dbg("Failed to add pre ct flow rule zone %d", zone); @@ -1143,7 +1222,7 @@ static int tc_ct_pre_ct_add_rules(struct mlx5_ct_ft *ct_ft, /* add miss rule */ memset(spec, 0, sizeof(*spec)); dest.ft = nat ? ct_priv->ct_nat : ct_priv->ct; - rule = mlx5_add_flow_rules(fdb, spec, &flow_act, &dest, 1); + rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1); if (IS_ERR(rule)) { err = PTR_ERR(rule); ct_dbg("Failed to add pre ct miss rule zone %d", zone); @@ -1170,7 +1249,7 @@ tc_ct_pre_ct_del_rules(struct mlx5_ct_ft *ct_ft, struct mlx5_tc_ct_pre *pre_ct) { struct mlx5_tc_ct_priv *ct_priv = ct_ft->ct_priv; - struct mlx5_core_dev *dev = ct_priv->esw->dev; + struct mlx5_core_dev *dev = ct_priv->dev; mlx5_del_flow_rules(pre_ct->flow_rule); mlx5_del_flow_rules(pre_ct->miss_rule); @@ -1184,7 +1263,7 @@ mlx5_tc_ct_alloc_pre_ct(struct mlx5_ct_ft *ct_ft, { int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); struct mlx5_tc_ct_priv *ct_priv = ct_ft->ct_priv; - struct mlx5_core_dev *dev = ct_priv->esw->dev; + struct mlx5_core_dev *dev = ct_priv->dev; struct mlx5_flow_table_attr ft_attr = {}; struct mlx5_flow_namespace *ns; struct mlx5_flow_table *ft; @@ -1194,10 +1273,10 @@ mlx5_tc_ct_alloc_pre_ct(struct mlx5_ct_ft *ct_ft, void *misc; int err; - ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_FDB); + ns = mlx5_get_flow_namespace(dev, ct_priv->ns_type); if (!ns) { err = -EOPNOTSUPP; - ct_dbg("Failed to get FDB flow namespace"); + ct_dbg("Failed to get flow namespace"); return err; } @@ -1206,7 +1285,8 @@ mlx5_tc_ct_alloc_pre_ct(struct mlx5_ct_ft *ct_ft, return -ENOMEM; ft_attr.flags = MLX5_FLOW_TABLE_UNMANAGED; - ft_attr.prio = FDB_TC_OFFLOAD; + ft_attr.prio = ct_priv->ns_type == MLX5_FLOW_NAMESPACE_FDB ? + FDB_TC_OFFLOAD : MLX5E_TC_PRIO; ft_attr.max_fte = 2; ft_attr.level = 1; ft = mlx5_create_flow_table(ns, &ft_attr); @@ -1215,7 +1295,7 @@ mlx5_tc_ct_alloc_pre_ct(struct mlx5_ct_ft *ct_ft, ct_dbg("Failed to create pre ct table"); goto out_free; } - pre_ct->fdb = ft; + pre_ct->ft = ft; /* create flow group */ MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0); @@ -1279,7 +1359,7 @@ mlx5_tc_ct_free_pre_ct(struct mlx5_ct_ft *ct_ft, tc_ct_pre_ct_del_rules(ct_ft, pre_ct); mlx5_destroy_flow_group(pre_ct->miss_grp); mlx5_destroy_flow_group(pre_ct->flow_grp); - mlx5_destroy_flow_table(pre_ct->fdb); + mlx5_destroy_flow_table(pre_ct->ft); } static int @@ -1398,7 +1478,7 @@ mlx5_tc_ct_del_ft_cb(struct mlx5_tc_ct_priv *ct_priv, struct mlx5_ct_ft *ft) /* We translate the tc filter with CT action to the following HW model: * * +---------------------+ - * + fdb prio (tc chain) + + * + ft prio (tc chain) + * + original match + * +---------------------+ * | set chain miss mapping @@ -1428,17 +1508,17 @@ mlx5_tc_ct_del_ft_cb(struct mlx5_tc_ct_priv *ct_priv, struct mlx5_ct_ft *ft) * +--------------+ */ static struct mlx5_flow_handle * -__mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, +__mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *ct_priv, struct mlx5e_tc_flow *flow, struct mlx5_flow_spec *orig_spec, - struct mlx5_esw_flow_attr *attr) + struct mlx5_flow_attr *attr) { - struct mlx5_tc_ct_priv *ct_priv = mlx5_tc_ct_get_ct_priv(priv); bool nat = attr->ct_attr.ct_action & TCA_CT_ACT_NAT; + struct mlx5e_priv *priv = netdev_priv(ct_priv->netdev); struct mlx5e_tc_mod_hdr_acts pre_mod_acts = {}; + u32 attr_sz = ns_to_attr_sz(ct_priv->ns_type); struct mlx5_flow_spec *post_ct_spec = NULL; - struct mlx5_eswitch *esw = ct_priv->esw; - struct mlx5_esw_flow_attr *pre_ct_attr; + struct mlx5_flow_attr *pre_ct_attr; struct mlx5_modify_hdr *mod_hdr; struct mlx5_flow_handle *rule; struct mlx5_ct_flow *ct_flow; @@ -1473,10 +1553,22 @@ __mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, } ct_flow->fte_id = fte_id; - /* Base esw attributes of both rules on original rule attribute */ - pre_ct_attr = &ct_flow->pre_ct_attr; - memcpy(pre_ct_attr, attr, sizeof(*attr)); - memcpy(&ct_flow->post_ct_attr, attr, sizeof(*attr)); + /* Base flow attributes of both rules on original rule attribute */ + ct_flow->pre_ct_attr = mlx5_alloc_flow_attr(ct_priv->ns_type); + if (!ct_flow->pre_ct_attr) { + err = -ENOMEM; + goto err_alloc_pre; + } + + ct_flow->post_ct_attr = mlx5_alloc_flow_attr(ct_priv->ns_type); + if (!ct_flow->post_ct_attr) { + err = -ENOMEM; + goto err_alloc_post; + } + + pre_ct_attr = ct_flow->pre_ct_attr; + memcpy(pre_ct_attr, attr, attr_sz); + memcpy(ct_flow->post_ct_attr, attr, attr_sz); /* Modify the original rule's action to fwd and modify, leave decap */ pre_ct_attr->action = attr->action & MLX5_FLOW_CONTEXT_ACTION_DECAP; @@ -1487,22 +1579,22 @@ __mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, * don't go though all prios of this chain as normal tc rules * miss. */ - err = mlx5_esw_chains_get_chain_mapping(esw, attr->chain, - &chain_mapping); + err = mlx5_chains_get_chain_mapping(ct_priv->chains, attr->chain, + &chain_mapping); if (err) { ct_dbg("Failed to get chain register mapping for chain"); goto err_get_chain; } ct_flow->chain_mapping = chain_mapping; - err = mlx5e_tc_match_to_reg_set(esw->dev, &pre_mod_acts, + err = mlx5e_tc_match_to_reg_set(priv->mdev, &pre_mod_acts, ct_priv->ns_type, CHAIN_TO_REG, chain_mapping); if (err) { ct_dbg("Failed to set chain register mapping"); goto err_mapping; } - err = mlx5e_tc_match_to_reg_set(esw->dev, &pre_mod_acts, + err = mlx5e_tc_match_to_reg_set(priv->mdev, &pre_mod_acts, ct_priv->ns_type, FTEID_TO_REG, fte_id); if (err) { ct_dbg("Failed to set fte_id register mapping"); @@ -1516,7 +1608,8 @@ __mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, attr->chain == 0) { u32 tun_id = mlx5e_tc_get_flow_tun_id(flow); - err = mlx5e_tc_match_to_reg_set(esw->dev, &pre_mod_acts, + err = mlx5e_tc_match_to_reg_set(priv->mdev, &pre_mod_acts, + ct_priv->ns_type, TUNNEL_TO_REG, tun_id); if (err) { @@ -1525,8 +1618,7 @@ __mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, } } - mod_hdr = mlx5_modify_header_alloc(esw->dev, - MLX5_FLOW_NAMESPACE_FDB, + mod_hdr = mlx5_modify_header_alloc(priv->mdev, ct_priv->ns_type, pre_mod_acts.num_actions, pre_mod_acts.actions); if (IS_ERR(mod_hdr)) { @@ -1542,16 +1634,16 @@ __mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, mlx5e_tc_match_to_reg_match(post_ct_spec, FTEID_TO_REG, fte_id, MLX5_FTE_ID_MASK); - /* Put post_ct rule on post_ct fdb */ - ct_flow->post_ct_attr.chain = 0; - ct_flow->post_ct_attr.prio = 0; - ct_flow->post_ct_attr.fdb = ct_priv->post_ct; + /* Put post_ct rule on post_ct flow table */ + ct_flow->post_ct_attr->chain = 0; + ct_flow->post_ct_attr->prio = 0; + ct_flow->post_ct_attr->ft = ct_priv->post_ct; - ct_flow->post_ct_attr.inner_match_level = MLX5_MATCH_NONE; - ct_flow->post_ct_attr.outer_match_level = MLX5_MATCH_NONE; - ct_flow->post_ct_attr.action &= ~(MLX5_FLOW_CONTEXT_ACTION_DECAP); - rule = mlx5_eswitch_add_offloaded_rule(esw, post_ct_spec, - &ct_flow->post_ct_attr); + ct_flow->post_ct_attr->inner_match_level = MLX5_MATCH_NONE; + ct_flow->post_ct_attr->outer_match_level = MLX5_MATCH_NONE; + ct_flow->post_ct_attr->action &= ~(MLX5_FLOW_CONTEXT_ACTION_DECAP); + rule = mlx5_tc_rule_insert(priv, post_ct_spec, + ct_flow->post_ct_attr); ct_flow->post_ct_rule = rule; if (IS_ERR(ct_flow->post_ct_rule)) { err = PTR_ERR(ct_flow->post_ct_rule); @@ -1561,10 +1653,9 @@ __mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, /* Change original rule point to ct table */ pre_ct_attr->dest_chain = 0; - pre_ct_attr->dest_ft = nat ? ft->pre_ct_nat.fdb : ft->pre_ct.fdb; - ct_flow->pre_ct_rule = mlx5_eswitch_add_offloaded_rule(esw, - orig_spec, - pre_ct_attr); + pre_ct_attr->dest_ft = nat ? ft->pre_ct_nat.ft : ft->pre_ct.ft; + ct_flow->pre_ct_rule = mlx5_tc_rule_insert(priv, orig_spec, + pre_ct_attr); if (IS_ERR(ct_flow->pre_ct_rule)) { err = PTR_ERR(ct_flow->pre_ct_rule); ct_dbg("Failed to add pre ct rule"); @@ -1578,14 +1669,18 @@ __mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, return rule; err_insert_orig: - mlx5_eswitch_del_offloaded_rule(ct_priv->esw, ct_flow->post_ct_rule, - &ct_flow->post_ct_attr); + mlx5_tc_rule_delete(priv, ct_flow->post_ct_rule, + ct_flow->post_ct_attr); err_insert_post_ct: mlx5_modify_header_dealloc(priv->mdev, pre_ct_attr->modify_hdr); err_mapping: dealloc_mod_hdr_actions(&pre_mod_acts); - mlx5_esw_chains_put_chain_mapping(esw, ct_flow->chain_mapping); + mlx5_chains_put_chain_mapping(ct_priv->chains, ct_flow->chain_mapping); err_get_chain: + kfree(ct_flow->post_ct_attr); +err_alloc_post: + kfree(ct_flow->pre_ct_attr); +err_alloc_pre: idr_remove(&ct_priv->fte_ids, fte_id); err_idr: mlx5_tc_ct_del_ft_cb(ct_priv, ft); @@ -1597,14 +1692,14 @@ err_ft: } static struct mlx5_flow_handle * -__mlx5_tc_ct_flow_offload_clear(struct mlx5e_priv *priv, +__mlx5_tc_ct_flow_offload_clear(struct mlx5_tc_ct_priv *ct_priv, struct mlx5_flow_spec *orig_spec, - struct mlx5_esw_flow_attr *attr, + struct mlx5_flow_attr *attr, struct mlx5e_tc_mod_hdr_acts *mod_acts) { - struct mlx5_tc_ct_priv *ct_priv = mlx5_tc_ct_get_ct_priv(priv); - struct mlx5_eswitch *esw = ct_priv->esw; - struct mlx5_esw_flow_attr *pre_ct_attr; + struct mlx5e_priv *priv = netdev_priv(ct_priv->netdev); + u32 attr_sz = ns_to_attr_sz(ct_priv->ns_type); + struct mlx5_flow_attr *pre_ct_attr; struct mlx5_modify_hdr *mod_hdr; struct mlx5_flow_handle *rule; struct mlx5_ct_flow *ct_flow; @@ -1615,8 +1710,13 @@ __mlx5_tc_ct_flow_offload_clear(struct mlx5e_priv *priv, return ERR_PTR(-ENOMEM); /* Base esw attributes on original rule attribute */ - pre_ct_attr = &ct_flow->pre_ct_attr; - memcpy(pre_ct_attr, attr, sizeof(*attr)); + pre_ct_attr = mlx5_alloc_flow_attr(ct_priv->ns_type); + if (!pre_ct_attr) { + err = -ENOMEM; + goto err_attr; + } + + memcpy(pre_ct_attr, attr, attr_sz); err = mlx5_tc_ct_entry_set_registers(ct_priv, mod_acts, 0, 0, 0, 0); if (err) { @@ -1624,8 +1724,7 @@ __mlx5_tc_ct_flow_offload_clear(struct mlx5e_priv *priv, goto err_set_registers; } - mod_hdr = mlx5_modify_header_alloc(esw->dev, - MLX5_FLOW_NAMESPACE_FDB, + mod_hdr = mlx5_modify_header_alloc(priv->mdev, ct_priv->ns_type, mod_acts->num_actions, mod_acts->actions); if (IS_ERR(mod_hdr)) { @@ -1638,7 +1737,7 @@ __mlx5_tc_ct_flow_offload_clear(struct mlx5e_priv *priv, pre_ct_attr->modify_hdr = mod_hdr; pre_ct_attr->action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; - rule = mlx5_eswitch_add_offloaded_rule(esw, orig_spec, pre_ct_attr); + rule = mlx5_tc_rule_insert(priv, orig_spec, pre_ct_attr); if (IS_ERR(rule)) { err = PTR_ERR(rule); ct_dbg("Failed to add ct clear rule"); @@ -1646,6 +1745,7 @@ __mlx5_tc_ct_flow_offload_clear(struct mlx5e_priv *priv, } attr->ct_attr.ct_flow = ct_flow; + ct_flow->pre_ct_attr = pre_ct_attr; ct_flow->pre_ct_rule = rule; return rule; @@ -1654,61 +1754,67 @@ err_insert: err_set_registers: netdev_warn(priv->netdev, "Failed to offload ct clear flow, err %d\n", err); + kfree(pre_ct_attr); +err_attr: + kfree(ct_flow); + return ERR_PTR(err); } struct mlx5_flow_handle * -mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, +mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *priv, struct mlx5e_tc_flow *flow, struct mlx5_flow_spec *spec, - struct mlx5_esw_flow_attr *attr, + struct mlx5_flow_attr *attr, struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts) { bool clear_action = attr->ct_attr.ct_action & TCA_CT_ACT_CLEAR; - struct mlx5_tc_ct_priv *ct_priv = mlx5_tc_ct_get_ct_priv(priv); struct mlx5_flow_handle *rule; - if (!ct_priv) + if (!priv) return ERR_PTR(-EOPNOTSUPP); - mutex_lock(&ct_priv->control_lock); + mutex_lock(&priv->control_lock); if (clear_action) rule = __mlx5_tc_ct_flow_offload_clear(priv, spec, attr, mod_hdr_acts); else rule = __mlx5_tc_ct_flow_offload(priv, flow, spec, attr); - mutex_unlock(&ct_priv->control_lock); + mutex_unlock(&priv->control_lock); return rule; } static void __mlx5_tc_ct_delete_flow(struct mlx5_tc_ct_priv *ct_priv, + struct mlx5e_tc_flow *flow, struct mlx5_ct_flow *ct_flow) { - struct mlx5_esw_flow_attr *pre_ct_attr = &ct_flow->pre_ct_attr; - struct mlx5_eswitch *esw = ct_priv->esw; + struct mlx5_flow_attr *pre_ct_attr = ct_flow->pre_ct_attr; + struct mlx5e_priv *priv = netdev_priv(ct_priv->netdev); - mlx5_eswitch_del_offloaded_rule(esw, ct_flow->pre_ct_rule, - pre_ct_attr); - mlx5_modify_header_dealloc(esw->dev, pre_ct_attr->modify_hdr); + mlx5_tc_rule_delete(priv, ct_flow->pre_ct_rule, + pre_ct_attr); + mlx5_modify_header_dealloc(priv->mdev, pre_ct_attr->modify_hdr); if (ct_flow->post_ct_rule) { - mlx5_eswitch_del_offloaded_rule(esw, ct_flow->post_ct_rule, - &ct_flow->post_ct_attr); - mlx5_esw_chains_put_chain_mapping(esw, ct_flow->chain_mapping); + mlx5_tc_rule_delete(priv, ct_flow->post_ct_rule, + ct_flow->post_ct_attr); + mlx5_chains_put_chain_mapping(ct_priv->chains, ct_flow->chain_mapping); idr_remove(&ct_priv->fte_ids, ct_flow->fte_id); mlx5_tc_ct_del_ft_cb(ct_priv, ct_flow->ft); } + kfree(ct_flow->pre_ct_attr); + kfree(ct_flow->post_ct_attr); kfree(ct_flow); } void -mlx5_tc_ct_delete_flow(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow, - struct mlx5_esw_flow_attr *attr) +mlx5_tc_ct_delete_flow(struct mlx5_tc_ct_priv *priv, + struct mlx5e_tc_flow *flow, + struct mlx5_flow_attr *attr) { - struct mlx5_tc_ct_priv *ct_priv = mlx5_tc_ct_get_ct_priv(priv); struct mlx5_ct_flow *ct_flow = attr->ct_attr.ct_flow; /* We are called on error to clean up stuff from parsing @@ -1717,22 +1823,15 @@ mlx5_tc_ct_delete_flow(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow, if (!ct_flow) return; - mutex_lock(&ct_priv->control_lock); - __mlx5_tc_ct_delete_flow(ct_priv, ct_flow); - mutex_unlock(&ct_priv->control_lock); + mutex_lock(&priv->control_lock); + __mlx5_tc_ct_delete_flow(priv, flow, ct_flow); + mutex_unlock(&priv->control_lock); } static int -mlx5_tc_ct_init_check_support(struct mlx5_eswitch *esw, - const char **err_msg) +mlx5_tc_ct_init_check_esw_support(struct mlx5_eswitch *esw, + const char **err_msg) { -#if !IS_ENABLED(CONFIG_NET_TC_SKB_EXT) - /* cannot restore chain ID on HW miss */ - - *err_msg = "tc skb extension missing"; - return -EOPNOTSUPP; -#endif - if (!MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, ignore_flow_level)) { *err_msg = "firmware level support is missing"; return -EOPNOTSUPP; @@ -1766,44 +1865,61 @@ mlx5_tc_ct_init_check_support(struct mlx5_eswitch *esw, return 0; } -static void -mlx5_tc_ct_init_err(struct mlx5e_rep_priv *rpriv, const char *msg, int err) +static int +mlx5_tc_ct_init_check_nic_support(struct mlx5e_priv *priv, + const char **err_msg) +{ + if (!MLX5_CAP_FLOWTABLE_NIC_RX(priv->mdev, ignore_flow_level)) { + *err_msg = "firmware level support is missing"; + return -EOPNOTSUPP; + } + + return 0; +} + +static int +mlx5_tc_ct_init_check_support(struct mlx5e_priv *priv, + enum mlx5_flow_namespace_type ns_type, + const char **err_msg) { - if (msg) - netdev_warn(rpriv->netdev, - "tc ct offload not supported, %s, err: %d\n", - msg, err); + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + +#if !IS_ENABLED(CONFIG_NET_TC_SKB_EXT) + /* cannot restore chain ID on HW miss */ + + *err_msg = "tc skb extension missing"; + return -EOPNOTSUPP; +#endif + if (ns_type == MLX5_FLOW_NAMESPACE_FDB) + return mlx5_tc_ct_init_check_esw_support(esw, err_msg); else - netdev_warn(rpriv->netdev, - "tc ct offload not supported, err: %d\n", - err); + return mlx5_tc_ct_init_check_nic_support(priv, err_msg); } -int -mlx5_tc_ct_init(struct mlx5_rep_uplink_priv *uplink_priv) +#define INIT_ERR_PREFIX "tc ct offload init failed" + +struct mlx5_tc_ct_priv * +mlx5_tc_ct_init(struct mlx5e_priv *priv, struct mlx5_fs_chains *chains, + struct mod_hdr_tbl *mod_hdr, + enum mlx5_flow_namespace_type ns_type) { struct mlx5_tc_ct_priv *ct_priv; - struct mlx5e_rep_priv *rpriv; - struct mlx5_eswitch *esw; - struct mlx5e_priv *priv; + struct mlx5_core_dev *dev; const char *msg; int err; - rpriv = container_of(uplink_priv, struct mlx5e_rep_priv, uplink_priv); - priv = netdev_priv(rpriv->netdev); - esw = priv->mdev->priv.eswitch; - - err = mlx5_tc_ct_init_check_support(esw, &msg); + dev = priv->mdev; + err = mlx5_tc_ct_init_check_support(priv, ns_type, &msg); if (err) { - mlx5_tc_ct_init_err(rpriv, msg, err); + mlx5_core_warn(dev, + "tc ct offload not supported, %s\n", + msg); goto err_support; } ct_priv = kzalloc(sizeof(*ct_priv), GFP_KERNEL); - if (!ct_priv) { - mlx5_tc_ct_init_err(rpriv, NULL, -ENOMEM); + if (!ct_priv) goto err_alloc; - } ct_priv->zone_mapping = mapping_create(sizeof(u16), 0, true); if (IS_ERR(ct_priv->zone_mapping)) { @@ -1817,46 +1933,51 @@ mlx5_tc_ct_init(struct mlx5_rep_uplink_priv *uplink_priv) goto err_mapping_labels; } - ct_priv->esw = esw; - ct_priv->netdev = rpriv->netdev; - ct_priv->ct = mlx5_esw_chains_create_global_table(esw); + ct_priv->ns_type = ns_type; + ct_priv->chains = chains; + ct_priv->netdev = priv->netdev; + ct_priv->dev = priv->mdev; + ct_priv->mod_hdr_tbl = mod_hdr; + ct_priv->ct = mlx5_chains_create_global_table(chains); if (IS_ERR(ct_priv->ct)) { err = PTR_ERR(ct_priv->ct); - mlx5_tc_ct_init_err(rpriv, "failed to create ct table", err); + mlx5_core_warn(dev, + "%s, failed to create ct table err: %d\n", + INIT_ERR_PREFIX, err); goto err_ct_tbl; } - ct_priv->ct_nat = mlx5_esw_chains_create_global_table(esw); + ct_priv->ct_nat = mlx5_chains_create_global_table(chains); if (IS_ERR(ct_priv->ct_nat)) { err = PTR_ERR(ct_priv->ct_nat); - mlx5_tc_ct_init_err(rpriv, "failed to create ct nat table", - err); + mlx5_core_warn(dev, + "%s, failed to create ct nat table err: %d\n", + INIT_ERR_PREFIX, err); goto err_ct_nat_tbl; } - ct_priv->post_ct = mlx5_esw_chains_create_global_table(esw); + ct_priv->post_ct = mlx5_chains_create_global_table(chains); if (IS_ERR(ct_priv->post_ct)) { err = PTR_ERR(ct_priv->post_ct); - mlx5_tc_ct_init_err(rpriv, "failed to create post ct table", - err); + mlx5_core_warn(dev, + "%s, failed to create post ct table err: %d\n", + INIT_ERR_PREFIX, err); goto err_post_ct_tbl; } idr_init(&ct_priv->fte_ids); mutex_init(&ct_priv->control_lock); + mutex_init(&ct_priv->shared_counter_lock); rhashtable_init(&ct_priv->zone_ht, &zone_params); rhashtable_init(&ct_priv->ct_tuples_ht, &tuples_ht_params); rhashtable_init(&ct_priv->ct_tuples_nat_ht, &tuples_nat_ht_params); - /* Done, set ct_priv to know it initializted */ - uplink_priv->ct_priv = ct_priv; - - return 0; + return ct_priv; err_post_ct_tbl: - mlx5_esw_chains_destroy_global_table(esw, ct_priv->ct_nat); + mlx5_chains_destroy_global_table(chains, ct_priv->ct_nat); err_ct_nat_tbl: - mlx5_esw_chains_destroy_global_table(esw, ct_priv->ct); + mlx5_chains_destroy_global_table(chains, ct_priv->ct); err_ct_tbl: mapping_destroy(ct_priv->labels_mapping); err_mapping_labels: @@ -1866,20 +1987,22 @@ err_mapping_zone: err_alloc: err_support: - return 0; + return NULL; } void -mlx5_tc_ct_clean(struct mlx5_rep_uplink_priv *uplink_priv) +mlx5_tc_ct_clean(struct mlx5_tc_ct_priv *ct_priv) { - struct mlx5_tc_ct_priv *ct_priv = uplink_priv->ct_priv; + struct mlx5_fs_chains *chains; if (!ct_priv) return; - mlx5_esw_chains_destroy_global_table(ct_priv->esw, ct_priv->post_ct); - mlx5_esw_chains_destroy_global_table(ct_priv->esw, ct_priv->ct_nat); - mlx5_esw_chains_destroy_global_table(ct_priv->esw, ct_priv->ct); + chains = ct_priv->chains; + + mlx5_chains_destroy_global_table(chains, ct_priv->post_ct); + mlx5_chains_destroy_global_table(chains, ct_priv->ct_nat); + mlx5_chains_destroy_global_table(chains, ct_priv->ct); mapping_destroy(ct_priv->zone_mapping); mapping_destroy(ct_priv->labels_mapping); @@ -1887,17 +2010,15 @@ mlx5_tc_ct_clean(struct mlx5_rep_uplink_priv *uplink_priv) rhashtable_destroy(&ct_priv->ct_tuples_nat_ht); rhashtable_destroy(&ct_priv->zone_ht); mutex_destroy(&ct_priv->control_lock); + mutex_destroy(&ct_priv->shared_counter_lock); idr_destroy(&ct_priv->fte_ids); kfree(ct_priv); - - uplink_priv->ct_priv = NULL; } bool -mlx5e_tc_ct_restore_flow(struct mlx5_rep_uplink_priv *uplink_priv, +mlx5e_tc_ct_restore_flow(struct mlx5_tc_ct_priv *ct_priv, struct sk_buff *skb, u8 zone_restore_id) { - struct mlx5_tc_ct_priv *ct_priv = uplink_priv->ct_priv; struct mlx5_ct_tuple tuple = {}; struct mlx5_ct_entry *entry; u16 zone; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h index 708c216325d3..6503b614337c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h @@ -10,12 +10,14 @@ #include "en.h" -struct mlx5_esw_flow_attr; +struct mlx5_flow_attr; struct mlx5e_tc_mod_hdr_acts; struct mlx5_rep_uplink_priv; struct mlx5e_tc_flow; struct mlx5e_priv; +struct mlx5_fs_chains; +struct mlx5_tc_ct_priv; struct mlx5_ct_flow; struct nf_flowtable; @@ -76,68 +78,78 @@ struct mlx5_ct_attr { misc_parameters_2.metadata_reg_c_1) + 3,\ } +#define nic_zone_restore_to_reg_ct {\ + .mfield = MLX5_ACTION_IN_FIELD_METADATA_REG_B,\ + .moffset = 2,\ + .mlen = 1,\ +} + #define REG_MAPPING_MLEN(reg) (mlx5e_tc_attr_to_reg_mappings[reg].mlen) +#define REG_MAPPING_MOFFSET(reg) (mlx5e_tc_attr_to_reg_mappings[reg].moffset) +#define REG_MAPPING_SHIFT(reg) (REG_MAPPING_MOFFSET(reg) * 8) #define ZONE_RESTORE_BITS (REG_MAPPING_MLEN(ZONE_RESTORE_TO_REG) * 8) #define ZONE_RESTORE_MAX GENMASK(ZONE_RESTORE_BITS - 1, 0) #if IS_ENABLED(CONFIG_MLX5_TC_CT) -int -mlx5_tc_ct_init(struct mlx5_rep_uplink_priv *uplink_priv); +struct mlx5_tc_ct_priv * +mlx5_tc_ct_init(struct mlx5e_priv *priv, struct mlx5_fs_chains *chains, + struct mod_hdr_tbl *mod_hdr, + enum mlx5_flow_namespace_type ns_type); void -mlx5_tc_ct_clean(struct mlx5_rep_uplink_priv *uplink_priv); +mlx5_tc_ct_clean(struct mlx5_tc_ct_priv *ct_priv); void -mlx5_tc_ct_match_del(struct mlx5e_priv *priv, struct mlx5_ct_attr *ct_attr); +mlx5_tc_ct_match_del(struct mlx5_tc_ct_priv *priv, struct mlx5_ct_attr *ct_attr); int -mlx5_tc_ct_match_add(struct mlx5e_priv *priv, +mlx5_tc_ct_match_add(struct mlx5_tc_ct_priv *priv, struct mlx5_flow_spec *spec, struct flow_cls_offload *f, struct mlx5_ct_attr *ct_attr, struct netlink_ext_ack *extack); +int mlx5_tc_ct_add_no_trk_match(struct mlx5_flow_spec *spec); int -mlx5_tc_ct_add_no_trk_match(struct mlx5e_priv *priv, - struct mlx5_flow_spec *spec); -int -mlx5_tc_ct_parse_action(struct mlx5e_priv *priv, - struct mlx5_esw_flow_attr *attr, +mlx5_tc_ct_parse_action(struct mlx5_tc_ct_priv *priv, + struct mlx5_flow_attr *attr, const struct flow_action_entry *act, struct netlink_ext_ack *extack); struct mlx5_flow_handle * -mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, +mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *priv, struct mlx5e_tc_flow *flow, struct mlx5_flow_spec *spec, - struct mlx5_esw_flow_attr *attr, + struct mlx5_flow_attr *attr, struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts); void -mlx5_tc_ct_delete_flow(struct mlx5e_priv *priv, +mlx5_tc_ct_delete_flow(struct mlx5_tc_ct_priv *priv, struct mlx5e_tc_flow *flow, - struct mlx5_esw_flow_attr *attr); + struct mlx5_flow_attr *attr); bool -mlx5e_tc_ct_restore_flow(struct mlx5_rep_uplink_priv *uplink_priv, +mlx5e_tc_ct_restore_flow(struct mlx5_tc_ct_priv *ct_priv, struct sk_buff *skb, u8 zone_restore_id); #else /* CONFIG_MLX5_TC_CT */ -static inline int -mlx5_tc_ct_init(struct mlx5_rep_uplink_priv *uplink_priv) +static inline struct mlx5_tc_ct_priv * +mlx5_tc_ct_init(struct mlx5e_priv *priv, struct mlx5_fs_chains *chains, + struct mod_hdr_tbl *mod_hdr, + enum mlx5_flow_namespace_type ns_type) { - return 0; + return NULL; } static inline void -mlx5_tc_ct_clean(struct mlx5_rep_uplink_priv *uplink_priv) +mlx5_tc_ct_clean(struct mlx5_tc_ct_priv *ct_priv) { } static inline void -mlx5_tc_ct_match_del(struct mlx5e_priv *priv, struct mlx5_ct_attr *ct_attr) {} +mlx5_tc_ct_match_del(struct mlx5_tc_ct_priv *priv, struct mlx5_ct_attr *ct_attr) {} static inline int -mlx5_tc_ct_match_add(struct mlx5e_priv *priv, +mlx5_tc_ct_match_add(struct mlx5_tc_ct_priv *priv, struct mlx5_flow_spec *spec, struct flow_cls_offload *f, struct mlx5_ct_attr *ct_attr, @@ -149,47 +161,44 @@ mlx5_tc_ct_match_add(struct mlx5e_priv *priv, return 0; NL_SET_ERR_MSG_MOD(extack, "mlx5 tc ct offload isn't enabled."); - netdev_warn(priv->netdev, "mlx5 tc ct offload isn't enabled.\n"); return -EOPNOTSUPP; } static inline int -mlx5_tc_ct_add_no_trk_match(struct mlx5e_priv *priv, - struct mlx5_flow_spec *spec) +mlx5_tc_ct_add_no_trk_match(struct mlx5_flow_spec *spec) { return 0; } static inline int -mlx5_tc_ct_parse_action(struct mlx5e_priv *priv, - struct mlx5_esw_flow_attr *attr, +mlx5_tc_ct_parse_action(struct mlx5_tc_ct_priv *priv, + struct mlx5_flow_attr *attr, const struct flow_action_entry *act, struct netlink_ext_ack *extack) { NL_SET_ERR_MSG_MOD(extack, "mlx5 tc ct offload isn't enabled."); - netdev_warn(priv->netdev, "mlx5 tc ct offload isn't enabled.\n"); return -EOPNOTSUPP; } static inline struct mlx5_flow_handle * -mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, +mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *priv, struct mlx5e_tc_flow *flow, struct mlx5_flow_spec *spec, - struct mlx5_esw_flow_attr *attr, + struct mlx5_flow_attr *attr, struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts) { return ERR_PTR(-EOPNOTSUPP); } static inline void -mlx5_tc_ct_delete_flow(struct mlx5e_priv *priv, +mlx5_tc_ct_delete_flow(struct mlx5_tc_ct_priv *priv, struct mlx5e_tc_flow *flow, - struct mlx5_esw_flow_attr *attr) + struct mlx5_flow_attr *attr) { } static inline bool -mlx5e_tc_ct_restore_flow(struct mlx5_rep_uplink_priv *uplink_priv, +mlx5e_tc_ct_restore_flow(struct mlx5_tc_ct_priv *ct_priv, struct sk_buff *skb, u8 zone_restore_id) { if (!zone_restore_id) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h index 24336c60123a..07ee1d236ab3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h @@ -7,6 +7,21 @@ #include "en.h" #include <linux/indirect_call_wrapper.h> +#define MLX5E_TX_WQE_EMPTY_DS_COUNT (sizeof(struct mlx5e_tx_wqe) / MLX5_SEND_WQE_DS) + +/* The mult of MLX5_SEND_WQE_MAX_WQEBBS * MLX5_SEND_WQEBB_NUM_DS + * (16 * 4 == 64) does not fit in the 6-bit DS field of Ctrl Segment. + * We use a bound lower that MLX5_SEND_WQE_MAX_WQEBBS to let a + * full-session WQE be cache-aligned. + */ +#if L1_CACHE_BYTES < 128 +#define MLX5E_TX_MPW_MAX_WQEBBS (MLX5_SEND_WQE_MAX_WQEBBS - 1) +#else +#define MLX5E_TX_MPW_MAX_WQEBBS (MLX5_SEND_WQE_MAX_WQEBBS - 2) +#endif + +#define MLX5E_TX_MPW_MAX_NUM_DS (MLX5E_TX_MPW_MAX_WQEBBS * MLX5_SEND_WQEBB_NUM_DS) + #define INL_HDR_START_SZ (sizeof(((struct mlx5_wqe_eth_seg *)NULL)->inline_hdr.start)) enum mlx5e_icosq_wqe_type { @@ -46,8 +61,6 @@ void mlx5e_free_rx_in_progress_descs(struct mlx5e_rq *rq); u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb, struct net_device *sb_dev); netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev); -void mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, - struct mlx5e_tx_wqe *wqe, u16 pi, bool xmit_more); bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget); void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq); @@ -110,6 +123,7 @@ struct mlx5e_tx_wqe_info { u32 num_bytes; u8 num_wqebbs; u8 num_dma; + u8 num_fifo_pkts; #ifdef CONFIG_MLX5_EN_TLS struct page *resync_dump_frag_page; #endif @@ -194,23 +208,6 @@ static inline u16 mlx5e_icosq_get_next_pi(struct mlx5e_icosq *sq, u16 size) } static inline void -mlx5e_fill_sq_frag_edge(struct mlx5e_txqsq *sq, struct mlx5_wq_cyc *wq, - u16 pi, u16 nnops) -{ - struct mlx5e_tx_wqe_info *edge_wi, *wi = &sq->db.wqe_info[pi]; - - edge_wi = wi + nnops; - - /* fill sq frag edge with nops to avoid wqe wrapping two pages */ - for (; wi < edge_wi; wi++) { - memset(wi, 0, sizeof(*wi)); - wi->num_wqebbs = 1; - mlx5e_post_nop(wq, sq->sqn, &sq->pc); - } - sq->stats->nop += nnops; -} - -static inline void mlx5e_notify_hw(struct mlx5_wq_cyc *wq, u16 pc, void __iomem *uar_map, struct mlx5_wqe_ctrl_seg *ctrl) { @@ -228,29 +225,6 @@ mlx5e_notify_hw(struct mlx5_wq_cyc *wq, u16 pc, void __iomem *uar_map, mlx5_write64((__be32 *)ctrl, uar_map); } -static inline bool mlx5e_transport_inline_tx_wqe(struct mlx5_wqe_ctrl_seg *cseg) -{ - return cseg && !!cseg->tis_tir_num; -} - -static inline u8 -mlx5e_tx_wqe_inline_mode(struct mlx5e_txqsq *sq, struct mlx5_wqe_ctrl_seg *cseg, - struct sk_buff *skb) -{ - u8 mode; - - if (mlx5e_transport_inline_tx_wqe(cseg)) - return MLX5_INLINE_MODE_TCP_UDP; - - mode = sq->min_inline_mode; - - if (skb_vlan_tag_present(skb) && - test_bit(MLX5E_SQ_STATE_VLAN_NEED_L2_INLINE, &sq->state)) - mode = max_t(u8, MLX5_INLINE_MODE_L2, mode); - - return mode; -} - static inline void mlx5e_cq_arm(struct mlx5e_cq *cq) { struct mlx5_core_cq *mcq; @@ -276,6 +250,23 @@ mlx5e_dma_push(struct mlx5e_txqsq *sq, dma_addr_t addr, u32 size, dma->type = map_type; } +static inline struct sk_buff **mlx5e_skb_fifo_get(struct mlx5e_txqsq *sq, u16 i) +{ + return &sq->db.skb_fifo[i & sq->skb_fifo_mask]; +} + +static inline void mlx5e_skb_fifo_push(struct mlx5e_txqsq *sq, struct sk_buff *skb) +{ + struct sk_buff **skb_item = mlx5e_skb_fifo_get(sq, sq->skb_fifo_pc++); + + *skb_item = skb; +} + +static inline struct sk_buff *mlx5e_skb_fifo_pop(struct mlx5e_txqsq *sq) +{ + return *mlx5e_skb_fifo_get(sq, sq->skb_fifo_cc++); +} + static inline void mlx5e_tx_dma_unmap(struct device *pdev, struct mlx5e_sq_dma *dma) { @@ -291,6 +282,14 @@ mlx5e_tx_dma_unmap(struct device *pdev, struct mlx5e_sq_dma *dma) } } +void mlx5e_sq_xmit_simple(struct mlx5e_txqsq *sq, struct sk_buff *skb, bool xmit_more); +void mlx5e_tx_mpwqe_ensure_complete(struct mlx5e_txqsq *sq); + +static inline bool mlx5e_tx_mpwqe_is_full(struct mlx5e_tx_mpwqe *session) +{ + return session->ds_count == MLX5E_TX_MPW_MAX_NUM_DS; +} + static inline void mlx5e_rqwq_reset(struct mlx5e_rq *rq) { if (rq->wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c index b28df21981a1..ae90d533a350 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c @@ -59,7 +59,7 @@ static inline bool mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq, struct mlx5e_dma_info *di, struct xdp_buff *xdp) { - struct mlx5e_xdp_xmit_data xdptxd; + struct mlx5e_xmit_data xdptxd; struct mlx5e_xdp_info xdpi; struct xdp_frame *xdpf; dma_addr_t dma_addr; @@ -194,18 +194,22 @@ static u16 mlx5e_xdpsq_get_next_pi(struct mlx5e_xdpsq *sq, u16 size) static void mlx5e_xdp_mpwqe_session_start(struct mlx5e_xdpsq *sq) { - struct mlx5e_xdp_mpwqe *session = &sq->mpwqe; + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; struct mlx5e_xdpsq_stats *stats = sq->stats; + struct mlx5e_tx_wqe *wqe; u16 pi; - pi = mlx5e_xdpsq_get_next_pi(sq, MLX5_SEND_WQE_MAX_WQEBBS); - session->wqe = MLX5E_TX_FETCH_WQE(sq, pi); + pi = mlx5e_xdpsq_get_next_pi(sq, MLX5E_TX_MPW_MAX_WQEBBS); + wqe = MLX5E_TX_FETCH_WQE(sq, pi); + net_prefetchw(wqe->data); - prefetchw(session->wqe->data); - session->ds_count = MLX5E_XDP_TX_EMPTY_DS_COUNT; - session->pkt_count = 0; - - mlx5e_xdp_update_inline_state(sq); + *session = (struct mlx5e_tx_mpwqe) { + .wqe = wqe, + .bytes_count = 0, + .ds_count = MLX5E_TX_WQE_EMPTY_DS_COUNT, + .pkt_count = 0, + .inline_on = mlx5e_xdp_get_inline_state(sq, session->inline_on), + }; stats->mpwqe++; } @@ -213,7 +217,7 @@ static void mlx5e_xdp_mpwqe_session_start(struct mlx5e_xdpsq *sq) void mlx5e_xdp_mpwqe_complete(struct mlx5e_xdpsq *sq) { struct mlx5_wq_cyc *wq = &sq->wq; - struct mlx5e_xdp_mpwqe *session = &sq->mpwqe; + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; struct mlx5_wqe_ctrl_seg *cseg = &session->wqe->ctrl; u16 ds_count = session->ds_count; u16 pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); @@ -258,10 +262,10 @@ INDIRECT_CALLABLE_SCOPE int mlx5e_xmit_xdp_frame_check_mpwqe(struct mlx5e_xdpsq } INDIRECT_CALLABLE_SCOPE bool -mlx5e_xmit_xdp_frame_mpwqe(struct mlx5e_xdpsq *sq, struct mlx5e_xdp_xmit_data *xdptxd, +mlx5e_xmit_xdp_frame_mpwqe(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd, struct mlx5e_xdp_info *xdpi, int check_result) { - struct mlx5e_xdp_mpwqe *session = &sq->mpwqe; + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; struct mlx5e_xdpsq_stats *stats = sq->stats; if (unlikely(xdptxd->len > sq->hw_mtu)) { @@ -284,8 +288,7 @@ mlx5e_xmit_xdp_frame_mpwqe(struct mlx5e_xdpsq *sq, struct mlx5e_xdp_xmit_data *x mlx5e_xdp_mpwqe_add_dseg(sq, xdptxd, stats); - if (unlikely(mlx5e_xdp_no_room_for_inline_pkt(session) || - session->ds_count == MLX5E_XDP_MPW_MAX_NUM_DS)) + if (unlikely(mlx5e_xdp_mpqwe_is_full(session))) mlx5e_xdp_mpwqe_complete(sq); mlx5e_xdpi_fifo_push(&sq->db.xdpi_fifo, xdpi); @@ -306,7 +309,7 @@ INDIRECT_CALLABLE_SCOPE int mlx5e_xmit_xdp_frame_check(struct mlx5e_xdpsq *sq) } INDIRECT_CALLABLE_SCOPE bool -mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xdp_xmit_data *xdptxd, +mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd, struct mlx5e_xdp_info *xdpi, int check_result) { struct mlx5_wq_cyc *wq = &sq->wq; @@ -322,7 +325,7 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xdp_xmit_data *xdptxd, struct mlx5e_xdpsq_stats *stats = sq->stats; - prefetchw(wqe); + net_prefetchw(wqe); if (unlikely(dma_len < MLX5E_XDP_MIN_INLINE || sq->hw_mtu < dma_len)) { stats->err++; @@ -445,7 +448,7 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq) } while ((++i < MLX5E_TX_CQ_POLL_BUDGET) && (cqe = mlx5_cqwq_get_cqe(&cq->wq))); if (xsk_frames) - xsk_umem_complete_tx(sq->umem, xsk_frames); + xsk_tx_completed(sq->xsk_pool, xsk_frames); sq->stats->cqes += i; @@ -475,7 +478,7 @@ void mlx5e_free_xdpsq_descs(struct mlx5e_xdpsq *sq) } if (xsk_frames) - xsk_umem_complete_tx(sq->umem, xsk_frames); + xsk_tx_completed(sq->xsk_pool, xsk_frames); } int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames, @@ -503,7 +506,7 @@ int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames, for (i = 0; i < n; i++) { struct xdp_frame *xdpf = frames[i]; - struct mlx5e_xdp_xmit_data xdptxd; + struct mlx5e_xmit_data xdptxd; struct mlx5e_xdp_info xdpi; bool ret; @@ -563,4 +566,3 @@ void mlx5e_set_xmit_fp(struct mlx5e_xdpsq *sq, bool is_mpw) sq->xmit_xdp_frame = is_mpw ? mlx5e_xmit_xdp_frame_mpwqe : mlx5e_xmit_xdp_frame; } - diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h index e806c13d491f..d487e5e37162 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h @@ -38,27 +38,12 @@ #include "en/txrx.h" #define MLX5E_XDP_MIN_INLINE (ETH_HLEN + VLAN_HLEN) -#define MLX5E_XDP_TX_EMPTY_DS_COUNT \ - (sizeof(struct mlx5e_tx_wqe) / MLX5_SEND_WQE_DS) -#define MLX5E_XDP_TX_DS_COUNT (MLX5E_XDP_TX_EMPTY_DS_COUNT + 1 /* SG DS */) - -#define MLX5E_XDP_INLINE_WQE_SZ_THRSD (256 - sizeof(struct mlx5_wqe_inline_seg)) -#define MLX5E_XDP_INLINE_WQE_MAX_DS_CNT \ - DIV_ROUND_UP(MLX5E_XDP_INLINE_WQE_SZ_THRSD, MLX5_SEND_WQE_DS) - -/* The mult of MLX5_SEND_WQE_MAX_WQEBBS * MLX5_SEND_WQEBB_NUM_DS - * (16 * 4 == 64) does not fit in the 6-bit DS field of Ctrl Segment. - * We use a bound lower that MLX5_SEND_WQE_MAX_WQEBBS to let a - * full-session WQE be cache-aligned. - */ -#if L1_CACHE_BYTES < 128 -#define MLX5E_XDP_MPW_MAX_WQEBBS (MLX5_SEND_WQE_MAX_WQEBBS - 1) -#else -#define MLX5E_XDP_MPW_MAX_WQEBBS (MLX5_SEND_WQE_MAX_WQEBBS - 2) -#endif +#define MLX5E_XDP_TX_DS_COUNT (MLX5E_TX_WQE_EMPTY_DS_COUNT + 1 /* SG DS */) -#define MLX5E_XDP_MPW_MAX_NUM_DS \ - (MLX5E_XDP_MPW_MAX_WQEBBS * MLX5_SEND_WQEBB_NUM_DS) +#define MLX5E_XDP_INLINE_WQE_MAX_DS_CNT 16 +#define MLX5E_XDP_INLINE_WQE_SZ_THRSD \ + (MLX5E_XDP_INLINE_WQE_MAX_DS_CNT * MLX5_SEND_WQE_DS - \ + sizeof(struct mlx5_wqe_inline_seg)) struct mlx5e_xsk_param; int mlx5e_xdp_max_mtu(struct mlx5e_params *params, struct mlx5e_xsk_param *xsk); @@ -73,11 +58,11 @@ int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames, u32 flags); INDIRECT_CALLABLE_DECLARE(bool mlx5e_xmit_xdp_frame_mpwqe(struct mlx5e_xdpsq *sq, - struct mlx5e_xdp_xmit_data *xdptxd, + struct mlx5e_xmit_data *xdptxd, struct mlx5e_xdp_info *xdpi, int check_result)); INDIRECT_CALLABLE_DECLARE(bool mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, - struct mlx5e_xdp_xmit_data *xdptxd, + struct mlx5e_xmit_data *xdptxd, struct mlx5e_xdp_info *xdpi, int check_result)); INDIRECT_CALLABLE_DECLARE(int mlx5e_xmit_xdp_frame_check_mpwqe(struct mlx5e_xdpsq *sq)); @@ -122,30 +107,28 @@ static inline void mlx5e_xmit_xdp_doorbell(struct mlx5e_xdpsq *sq) /* Enable inline WQEs to shift some load from a congested HCA (HW) to * a less congested cpu (SW). */ -static inline void mlx5e_xdp_update_inline_state(struct mlx5e_xdpsq *sq) +static inline bool mlx5e_xdp_get_inline_state(struct mlx5e_xdpsq *sq, bool cur) { u16 outstanding = sq->xdpi_fifo_pc - sq->xdpi_fifo_cc; - struct mlx5e_xdp_mpwqe *session = &sq->mpwqe; #define MLX5E_XDP_INLINE_WATERMARK_LOW 10 #define MLX5E_XDP_INLINE_WATERMARK_HIGH 128 - if (session->inline_on) { - if (outstanding <= MLX5E_XDP_INLINE_WATERMARK_LOW) - session->inline_on = 0; - return; - } + if (cur && outstanding <= MLX5E_XDP_INLINE_WATERMARK_LOW) + return false; + + if (!cur && outstanding >= MLX5E_XDP_INLINE_WATERMARK_HIGH) + return true; - /* inline is false */ - if (outstanding >= MLX5E_XDP_INLINE_WATERMARK_HIGH) - session->inline_on = 1; + return cur; } -static inline bool -mlx5e_xdp_no_room_for_inline_pkt(struct mlx5e_xdp_mpwqe *session) +static inline bool mlx5e_xdp_mpqwe_is_full(struct mlx5e_tx_mpwqe *session) { - return session->inline_on && - session->ds_count + MLX5E_XDP_INLINE_WQE_MAX_DS_CNT > MLX5E_XDP_MPW_MAX_NUM_DS; + if (session->inline_on) + return session->ds_count + MLX5E_XDP_INLINE_WQE_MAX_DS_CNT > + MLX5E_TX_MPW_MAX_NUM_DS; + return mlx5e_tx_mpwqe_is_full(session); } struct mlx5e_xdp_wqe_info { @@ -155,15 +138,16 @@ struct mlx5e_xdp_wqe_info { static inline void mlx5e_xdp_mpwqe_add_dseg(struct mlx5e_xdpsq *sq, - struct mlx5e_xdp_xmit_data *xdptxd, + struct mlx5e_xmit_data *xdptxd, struct mlx5e_xdpsq_stats *stats) { - struct mlx5e_xdp_mpwqe *session = &sq->mpwqe; + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; struct mlx5_wqe_data_seg *dseg = (struct mlx5_wqe_data_seg *)session->wqe + session->ds_count; u32 dma_len = xdptxd->len; session->pkt_count++; + session->bytes_count += dma_len; if (session->inline_on && dma_len <= MLX5E_XDP_INLINE_WQE_SZ_THRSD) { struct mlx5_wqe_inline_seg *inline_dseg = diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/umem.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/pool.c index 331ca2b0f8a4..71e8d66fa150 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/umem.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/pool.c @@ -1,31 +1,31 @@ // SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB -/* Copyright (c) 2019 Mellanox Technologies. */ +/* Copyright (c) 2019-2020, Mellanox Technologies inc. All rights reserved. */ #include <net/xdp_sock_drv.h> -#include "umem.h" +#include "pool.h" #include "setup.h" #include "en/params.h" -static int mlx5e_xsk_map_umem(struct mlx5e_priv *priv, - struct xdp_umem *umem) +static int mlx5e_xsk_map_pool(struct mlx5e_priv *priv, + struct xsk_buff_pool *pool) { - struct device *dev = priv->mdev->device; + struct device *dev = mlx5_core_dma_dev(priv->mdev); - return xsk_buff_dma_map(umem, dev, 0); + return xsk_pool_dma_map(pool, dev, 0); } -static void mlx5e_xsk_unmap_umem(struct mlx5e_priv *priv, - struct xdp_umem *umem) +static void mlx5e_xsk_unmap_pool(struct mlx5e_priv *priv, + struct xsk_buff_pool *pool) { - return xsk_buff_dma_unmap(umem, 0); + return xsk_pool_dma_unmap(pool, 0); } -static int mlx5e_xsk_get_umems(struct mlx5e_xsk *xsk) +static int mlx5e_xsk_get_pools(struct mlx5e_xsk *xsk) { - if (!xsk->umems) { - xsk->umems = kcalloc(MLX5E_MAX_NUM_CHANNELS, - sizeof(*xsk->umems), GFP_KERNEL); - if (unlikely(!xsk->umems)) + if (!xsk->pools) { + xsk->pools = kcalloc(MLX5E_MAX_NUM_CHANNELS, + sizeof(*xsk->pools), GFP_KERNEL); + if (unlikely(!xsk->pools)) return -ENOMEM; } @@ -35,68 +35,68 @@ static int mlx5e_xsk_get_umems(struct mlx5e_xsk *xsk) return 0; } -static void mlx5e_xsk_put_umems(struct mlx5e_xsk *xsk) +static void mlx5e_xsk_put_pools(struct mlx5e_xsk *xsk) { if (!--xsk->refcnt) { - kfree(xsk->umems); - xsk->umems = NULL; + kfree(xsk->pools); + xsk->pools = NULL; } } -static int mlx5e_xsk_add_umem(struct mlx5e_xsk *xsk, struct xdp_umem *umem, u16 ix) +static int mlx5e_xsk_add_pool(struct mlx5e_xsk *xsk, struct xsk_buff_pool *pool, u16 ix) { int err; - err = mlx5e_xsk_get_umems(xsk); + err = mlx5e_xsk_get_pools(xsk); if (unlikely(err)) return err; - xsk->umems[ix] = umem; + xsk->pools[ix] = pool; return 0; } -static void mlx5e_xsk_remove_umem(struct mlx5e_xsk *xsk, u16 ix) +static void mlx5e_xsk_remove_pool(struct mlx5e_xsk *xsk, u16 ix) { - xsk->umems[ix] = NULL; + xsk->pools[ix] = NULL; - mlx5e_xsk_put_umems(xsk); + mlx5e_xsk_put_pools(xsk); } -static bool mlx5e_xsk_is_umem_sane(struct xdp_umem *umem) +static bool mlx5e_xsk_is_pool_sane(struct xsk_buff_pool *pool) { - return xsk_umem_get_headroom(umem) <= 0xffff && - xsk_umem_get_chunk_size(umem) <= 0xffff; + return xsk_pool_get_headroom(pool) <= 0xffff && + xsk_pool_get_chunk_size(pool) <= 0xffff; } -void mlx5e_build_xsk_param(struct xdp_umem *umem, struct mlx5e_xsk_param *xsk) +void mlx5e_build_xsk_param(struct xsk_buff_pool *pool, struct mlx5e_xsk_param *xsk) { - xsk->headroom = xsk_umem_get_headroom(umem); - xsk->chunk_size = xsk_umem_get_chunk_size(umem); + xsk->headroom = xsk_pool_get_headroom(pool); + xsk->chunk_size = xsk_pool_get_chunk_size(pool); } static int mlx5e_xsk_enable_locked(struct mlx5e_priv *priv, - struct xdp_umem *umem, u16 ix) + struct xsk_buff_pool *pool, u16 ix) { struct mlx5e_params *params = &priv->channels.params; struct mlx5e_xsk_param xsk; struct mlx5e_channel *c; int err; - if (unlikely(mlx5e_xsk_get_umem(&priv->channels.params, &priv->xsk, ix))) + if (unlikely(mlx5e_xsk_get_pool(&priv->channels.params, &priv->xsk, ix))) return -EBUSY; - if (unlikely(!mlx5e_xsk_is_umem_sane(umem))) + if (unlikely(!mlx5e_xsk_is_pool_sane(pool))) return -EINVAL; - err = mlx5e_xsk_map_umem(priv, umem); + err = mlx5e_xsk_map_pool(priv, pool); if (unlikely(err)) return err; - err = mlx5e_xsk_add_umem(&priv->xsk, umem, ix); + err = mlx5e_xsk_add_pool(&priv->xsk, pool, ix); if (unlikely(err)) - goto err_unmap_umem; + goto err_unmap_pool; - mlx5e_build_xsk_param(umem, &xsk); + mlx5e_build_xsk_param(pool, &xsk); if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) { /* XSK objects will be created on open. */ @@ -112,9 +112,9 @@ static int mlx5e_xsk_enable_locked(struct mlx5e_priv *priv, c = priv->channels.c[ix]; - err = mlx5e_open_xsk(priv, params, &xsk, umem, c); + err = mlx5e_open_xsk(priv, params, &xsk, pool, c); if (unlikely(err)) - goto err_remove_umem; + goto err_remove_pool; mlx5e_activate_xsk(c); @@ -132,11 +132,11 @@ err_deactivate: mlx5e_deactivate_xsk(c); mlx5e_close_xsk(c); -err_remove_umem: - mlx5e_xsk_remove_umem(&priv->xsk, ix); +err_remove_pool: + mlx5e_xsk_remove_pool(&priv->xsk, ix); -err_unmap_umem: - mlx5e_xsk_unmap_umem(priv, umem); +err_unmap_pool: + mlx5e_xsk_unmap_pool(priv, pool); return err; @@ -146,7 +146,7 @@ validate_closed: */ if (!mlx5e_validate_xsk_param(params, &xsk, priv->mdev)) { err = -EINVAL; - goto err_remove_umem; + goto err_remove_pool; } return 0; @@ -154,45 +154,45 @@ validate_closed: static int mlx5e_xsk_disable_locked(struct mlx5e_priv *priv, u16 ix) { - struct xdp_umem *umem = mlx5e_xsk_get_umem(&priv->channels.params, + struct xsk_buff_pool *pool = mlx5e_xsk_get_pool(&priv->channels.params, &priv->xsk, ix); struct mlx5e_channel *c; - if (unlikely(!umem)) + if (unlikely(!pool)) return -EINVAL; if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) - goto remove_umem; + goto remove_pool; /* XSK RQ and SQ are only created if XDP program is set. */ if (!priv->channels.params.xdp_prog) - goto remove_umem; + goto remove_pool; c = priv->channels.c[ix]; mlx5e_xsk_redirect_rqt_to_drop(priv, ix); mlx5e_deactivate_xsk(c); mlx5e_close_xsk(c); -remove_umem: - mlx5e_xsk_remove_umem(&priv->xsk, ix); - mlx5e_xsk_unmap_umem(priv, umem); +remove_pool: + mlx5e_xsk_remove_pool(&priv->xsk, ix); + mlx5e_xsk_unmap_pool(priv, pool); return 0; } -static int mlx5e_xsk_enable_umem(struct mlx5e_priv *priv, struct xdp_umem *umem, +static int mlx5e_xsk_enable_pool(struct mlx5e_priv *priv, struct xsk_buff_pool *pool, u16 ix) { int err; mutex_lock(&priv->state_lock); - err = mlx5e_xsk_enable_locked(priv, umem, ix); + err = mlx5e_xsk_enable_locked(priv, pool, ix); mutex_unlock(&priv->state_lock); return err; } -static int mlx5e_xsk_disable_umem(struct mlx5e_priv *priv, u16 ix) +static int mlx5e_xsk_disable_pool(struct mlx5e_priv *priv, u16 ix) { int err; @@ -203,7 +203,7 @@ static int mlx5e_xsk_disable_umem(struct mlx5e_priv *priv, u16 ix) return err; } -int mlx5e_xsk_setup_umem(struct net_device *dev, struct xdp_umem *umem, u16 qid) +int mlx5e_xsk_setup_pool(struct net_device *dev, struct xsk_buff_pool *pool, u16 qid) { struct mlx5e_priv *priv = netdev_priv(dev); struct mlx5e_params *params = &priv->channels.params; @@ -212,6 +212,6 @@ int mlx5e_xsk_setup_umem(struct net_device *dev, struct xdp_umem *umem, u16 qid) if (unlikely(!mlx5e_qid_get_ch_if_in_group(params, qid, MLX5E_RQ_GROUP_XSK, &ix))) return -EINVAL; - return umem ? mlx5e_xsk_enable_umem(priv, umem, ix) : - mlx5e_xsk_disable_umem(priv, ix); + return pool ? mlx5e_xsk_enable_pool(priv, pool, ix) : + mlx5e_xsk_disable_pool(priv, ix); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/pool.h b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/pool.h new file mode 100644 index 000000000000..dca0010a0866 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/pool.h @@ -0,0 +1,27 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ +/* Copyright (c) 2019-2020, Mellanox Technologies inc. All rights reserved. */ + +#ifndef __MLX5_EN_XSK_POOL_H__ +#define __MLX5_EN_XSK_POOL_H__ + +#include "en.h" + +static inline struct xsk_buff_pool *mlx5e_xsk_get_pool(struct mlx5e_params *params, + struct mlx5e_xsk *xsk, u16 ix) +{ + if (!xsk || !xsk->pools) + return NULL; + + if (unlikely(ix >= params->num_channels)) + return NULL; + + return xsk->pools[ix]; +} + +struct mlx5e_xsk_param; +void mlx5e_build_xsk_param(struct xsk_buff_pool *pool, struct mlx5e_xsk_param *xsk); + +/* .ndo_bpf callback. */ +int mlx5e_xsk_setup_pool(struct net_device *dev, struct xsk_buff_pool *pool, u16 qid); + +#endif /* __MLX5_EN_XSK_POOL_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c index 40db27bf790b..8e7b877d8a12 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c @@ -47,8 +47,8 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, xdp->data_end = xdp->data + cqe_bcnt32; xdp_set_data_meta_invalid(xdp); - xsk_buff_dma_sync_for_cpu(xdp); - prefetch(xdp->data); + xsk_buff_dma_sync_for_cpu(xdp, rq->xsk_pool); + net_prefetch(xdp->data); /* Possible flows: * - XDP_REDIRECT to XSKMAP: @@ -93,8 +93,8 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_linear(struct mlx5e_rq *rq, xdp->data_end = xdp->data + cqe_bcnt; xdp_set_data_meta_invalid(xdp); - xsk_buff_dma_sync_for_cpu(xdp); - prefetch(xdp->data); + xsk_buff_dma_sync_for_cpu(xdp, rq->xsk_pool); + net_prefetch(xdp->data); if (unlikely(get_cqe_opcode(cqe) != MLX5_CQE_RESP_SEND)) { rq->stats->wqe_err++; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.h index d147b2f13b54..7f88ccf67fdd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.h @@ -19,10 +19,10 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_linear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi, u32 cqe_bcnt); -static inline int mlx5e_xsk_page_alloc_umem(struct mlx5e_rq *rq, +static inline int mlx5e_xsk_page_alloc_pool(struct mlx5e_rq *rq, struct mlx5e_dma_info *dma_info) { - dma_info->xsk = xsk_buff_alloc(rq->umem); + dma_info->xsk = xsk_buff_alloc(rq->xsk_pool); if (!dma_info->xsk) return -ENOMEM; @@ -38,13 +38,13 @@ static inline int mlx5e_xsk_page_alloc_umem(struct mlx5e_rq *rq, static inline bool mlx5e_xsk_update_rx_wakeup(struct mlx5e_rq *rq, bool alloc_err) { - if (!xsk_umem_uses_need_wakeup(rq->umem)) + if (!xsk_uses_need_wakeup(rq->xsk_pool)) return alloc_err; if (unlikely(alloc_err)) - xsk_set_rx_need_wakeup(rq->umem); + xsk_set_rx_need_wakeup(rq->xsk_pool); else - xsk_clear_rx_need_wakeup(rq->umem); + xsk_clear_rx_need_wakeup(rq->xsk_pool); return false; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c index 55e65a438de7..4e574ac73019 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c @@ -45,7 +45,7 @@ static void mlx5e_build_xsk_cparam(struct mlx5e_priv *priv, } int mlx5e_open_xsk(struct mlx5e_priv *priv, struct mlx5e_params *params, - struct mlx5e_xsk_param *xsk, struct xdp_umem *umem, + struct mlx5e_xsk_param *xsk, struct xsk_buff_pool *pool, struct mlx5e_channel *c) { struct mlx5e_channel_param *cparam; @@ -64,7 +64,7 @@ int mlx5e_open_xsk(struct mlx5e_priv *priv, struct mlx5e_params *params, if (unlikely(err)) goto err_free_cparam; - err = mlx5e_open_rq(c, params, &cparam->rq, xsk, umem, &c->xskrq); + err = mlx5e_open_rq(c, params, &cparam->rq, xsk, pool, &c->xskrq); if (unlikely(err)) goto err_close_rx_cq; @@ -72,13 +72,13 @@ int mlx5e_open_xsk(struct mlx5e_priv *priv, struct mlx5e_params *params, if (unlikely(err)) goto err_close_rq; - /* Create a separate SQ, so that when the UMEM is disabled, we could + /* Create a separate SQ, so that when the buff pool is disabled, we could * close this SQ safely and stop receiving CQEs. In other case, e.g., if - * the XDPSQ was used instead, we might run into trouble when the UMEM + * the XDPSQ was used instead, we might run into trouble when the buff pool * is disabled and then reenabled, but the SQ continues receiving CQEs - * from the old UMEM. + * from the old buff pool. */ - err = mlx5e_open_xdpsq(c, params, &cparam->xdp_sq, umem, &c->xsksq, true); + err = mlx5e_open_xdpsq(c, params, &cparam->xdp_sq, pool, &c->xsksq, true); if (unlikely(err)) goto err_close_tx_cq; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.h b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.h index 0dd11b81c046..ca20f1ff5e39 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.h @@ -12,7 +12,7 @@ bool mlx5e_validate_xsk_param(struct mlx5e_params *params, struct mlx5e_xsk_param *xsk, struct mlx5_core_dev *mdev); int mlx5e_open_xsk(struct mlx5e_priv *priv, struct mlx5e_params *params, - struct mlx5e_xsk_param *xsk, struct xdp_umem *umem, + struct mlx5e_xsk_param *xsk, struct xsk_buff_pool *pool, struct mlx5e_channel *c); void mlx5e_close_xsk(struct mlx5e_channel *c); void mlx5e_activate_xsk(struct mlx5e_channel *c); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c index 4d892f6cecb3..fb671a457129 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c @@ -2,7 +2,7 @@ /* Copyright (c) 2019 Mellanox Technologies. */ #include "tx.h" -#include "umem.h" +#include "pool.h" #include "en/xdp.h" #include "en/params.h" #include <net/xdp_sock_drv.h> @@ -66,9 +66,9 @@ static void mlx5e_xsk_tx_post_err(struct mlx5e_xdpsq *sq, bool mlx5e_xsk_tx(struct mlx5e_xdpsq *sq, unsigned int budget) { - struct xdp_umem *umem = sq->umem; + struct xsk_buff_pool *pool = sq->xsk_pool; + struct mlx5e_xmit_data xdptxd; struct mlx5e_xdp_info xdpi; - struct mlx5e_xdp_xmit_data xdptxd; bool work_done = true; bool flush = false; @@ -87,7 +87,7 @@ bool mlx5e_xsk_tx(struct mlx5e_xdpsq *sq, unsigned int budget) break; } - if (!xsk_umem_consume_tx(umem, &desc)) { + if (!xsk_tx_peek_desc(pool, &desc)) { /* TX will get stuck until something wakes it up by * triggering NAPI. Currently it's expected that the * application calls sendto() if there are consumed, but @@ -96,11 +96,11 @@ bool mlx5e_xsk_tx(struct mlx5e_xdpsq *sq, unsigned int budget) break; } - xdptxd.dma_addr = xsk_buff_raw_get_dma(umem, desc.addr); - xdptxd.data = xsk_buff_raw_get_data(umem, desc.addr); + xdptxd.dma_addr = xsk_buff_raw_get_dma(pool, desc.addr); + xdptxd.data = xsk_buff_raw_get_data(pool, desc.addr); xdptxd.len = desc.len; - xsk_buff_raw_dma_sync_for_device(umem, xdptxd.dma_addr, xdptxd.len); + xsk_buff_raw_dma_sync_for_device(pool, xdptxd.dma_addr, xdptxd.len); ret = INDIRECT_CALL_2(sq->xmit_xdp_frame, mlx5e_xmit_xdp_frame_mpwqe, mlx5e_xmit_xdp_frame, sq, &xdptxd, &xdpi, check_result); @@ -119,7 +119,7 @@ bool mlx5e_xsk_tx(struct mlx5e_xdpsq *sq, unsigned int budget) mlx5e_xdp_mpwqe_complete(sq); mlx5e_xmit_xdp_doorbell(sq); - xsk_umem_consume_tx_done(umem); + xsk_tx_release(pool); } return !(budget && work_done); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.h index 39fa0a705856..a05085035f23 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.h @@ -15,13 +15,13 @@ bool mlx5e_xsk_tx(struct mlx5e_xdpsq *sq, unsigned int budget); static inline void mlx5e_xsk_update_tx_wakeup(struct mlx5e_xdpsq *sq) { - if (!xsk_umem_uses_need_wakeup(sq->umem)) + if (!xsk_uses_need_wakeup(sq->xsk_pool)) return; if (sq->pc != sq->cc) - xsk_clear_tx_need_wakeup(sq->umem); + xsk_clear_tx_need_wakeup(sq->xsk_pool); else - xsk_set_tx_need_wakeup(sq->umem); + xsk_set_tx_need_wakeup(sq->xsk_pool); } #endif /* __MLX5_EN_XSK_TX_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/umem.h b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/umem.h deleted file mode 100644 index bada94973586..000000000000 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/umem.h +++ /dev/null @@ -1,29 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ -/* Copyright (c) 2019 Mellanox Technologies. */ - -#ifndef __MLX5_EN_XSK_UMEM_H__ -#define __MLX5_EN_XSK_UMEM_H__ - -#include "en.h" - -static inline struct xdp_umem *mlx5e_xsk_get_umem(struct mlx5e_params *params, - struct mlx5e_xsk *xsk, u16 ix) -{ - if (!xsk || !xsk->umems) - return NULL; - - if (unlikely(ix >= params->num_channels)) - return NULL; - - return xsk->umems[ix]; -} - -struct mlx5e_xsk_param; -void mlx5e_build_xsk_param(struct xdp_umem *umem, struct mlx5e_xsk_param *xsk); - -/* .ndo_bpf callback. */ -int mlx5e_xsk_setup_umem(struct net_device *dev, struct xdp_umem *umem, u16 qid); - -int mlx5e_xsk_resize_reuseq(struct xdp_umem *umem, u32 nentries); - -#endif /* __MLX5_EN_XSK_UMEM_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h index 110476bdeffb..899b98aca0d3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h @@ -107,6 +107,9 @@ struct mlx5e_accel_tx_state { #ifdef CONFIG_MLX5_EN_TLS struct mlx5e_accel_tx_tls_state tls; #endif +#ifdef CONFIG_MLX5_EN_IPSEC + struct mlx5e_accel_tx_ipsec_state ipsec; +#endif }; static inline bool mlx5e_accel_tx_begin(struct net_device *dev, @@ -125,27 +128,70 @@ static inline bool mlx5e_accel_tx_begin(struct net_device *dev, } #endif +#ifdef CONFIG_MLX5_EN_IPSEC + if (test_bit(MLX5E_SQ_STATE_IPSEC, &sq->state) && xfrm_offload(skb)) { + if (unlikely(!mlx5e_ipsec_handle_tx_skb(dev, skb, &state->ipsec))) + return false; + } +#endif + return true; } -static inline bool mlx5e_accel_tx_finish(struct mlx5e_priv *priv, - struct mlx5e_txqsq *sq, - struct sk_buff *skb, +static inline bool mlx5e_accel_tx_is_ipsec_flow(struct mlx5e_accel_tx_state *state) +{ +#ifdef CONFIG_MLX5_EN_IPSEC + return mlx5e_ipsec_is_tx_flow(&state->ipsec); +#endif + + return false; +} + +static inline unsigned int mlx5e_accel_tx_ids_len(struct mlx5e_txqsq *sq, + struct mlx5e_accel_tx_state *state) +{ +#ifdef CONFIG_MLX5_EN_IPSEC + if (test_bit(MLX5E_SQ_STATE_IPSEC, &sq->state)) + return mlx5e_ipsec_tx_ids_len(&state->ipsec); +#endif + + return 0; +} + +/* Part of the eseg touched by TX offloads */ +#define MLX5E_ACCEL_ESEG_LEN offsetof(struct mlx5_wqe_eth_seg, mss) + +static inline bool mlx5e_accel_tx_eseg(struct mlx5e_priv *priv, + struct sk_buff *skb, + struct mlx5_wqe_eth_seg *eseg) +{ +#ifdef CONFIG_MLX5_EN_IPSEC + if (xfrm_offload(skb)) + mlx5e_ipsec_tx_build_eseg(priv, skb, eseg); +#endif + +#if IS_ENABLED(CONFIG_GENEVE) + if (skb->encapsulation) + mlx5e_tx_tunnel_accel(skb, eseg); +#endif + + return true; +} + +static inline void mlx5e_accel_tx_finish(struct mlx5e_txqsq *sq, struct mlx5e_tx_wqe *wqe, - struct mlx5e_accel_tx_state *state) + struct mlx5e_accel_tx_state *state, + struct mlx5_wqe_inline_seg *inlseg) { #ifdef CONFIG_MLX5_EN_TLS mlx5e_tls_handle_tx_wqe(sq, &wqe->ctrl, &state->tls); #endif #ifdef CONFIG_MLX5_EN_IPSEC - if (test_bit(MLX5E_SQ_STATE_IPSEC, &sq->state)) { - if (unlikely(!mlx5e_ipsec_handle_tx_skb(priv, &wqe->eth, skb))) - return false; - } + if (test_bit(MLX5E_SQ_STATE_IPSEC, &sq->state) && + state->ipsec.xo && state->ipsec.tailen) + mlx5e_ipsec_handle_tx_wqe(wqe, &state->ipsec, inlseg); #endif - - return true; } static inline int mlx5e_accel_init_rx(struct mlx5e_priv *priv) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/fs_tcp.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/fs_tcp.c index 4cdd9eac647d..97f1594cee11 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/fs_tcp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/fs_tcp.c @@ -191,7 +191,7 @@ static int accel_fs_tcp_create_groups(struct mlx5e_flow_table *ft, ft->g = kcalloc(MLX5E_ACCEL_FS_TCP_NUM_GROUPS, sizeof(*ft->g), GFP_KERNEL); in = kvzalloc(inlen, GFP_KERNEL); if (!in || !ft->g) { - kvfree(ft->g); + kfree(ft->g); kvfree(in); return -ENOMEM; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c index d39989cddd90..3d45341e2216 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c @@ -560,6 +560,9 @@ void mlx5e_ipsec_build_netdev(struct mlx5e_priv *priv) return; } + if (mlx5_is_ipsec_device(mdev)) + netdev->gso_partial_features |= NETIF_F_GSO_ESP; + mlx5_core_dbg(mdev, "mlx5e: ESP GSO capability turned on\n"); netdev->features |= NETIF_F_GSO_ESP; netdev->hw_features |= NETIF_F_GSO_ESP; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h index 0fc8b4d4f4a3..6164c7f59efb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h @@ -76,6 +76,7 @@ struct mlx5e_ipsec_stats { }; struct mlx5e_accel_fs_esp; +struct mlx5e_ipsec_tx; struct mlx5e_ipsec { struct mlx5e_priv *en_priv; @@ -87,6 +88,7 @@ struct mlx5e_ipsec { struct mlx5e_ipsec_stats stats; struct workqueue_struct *wq; struct mlx5e_accel_fs_esp *rx_fs; + struct mlx5e_ipsec_tx *tx_fs; }; struct mlx5e_ipsec_esn_state { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c index 429428bbc903..0e45590662a8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c @@ -34,6 +34,12 @@ struct mlx5e_accel_fs_esp { struct mlx5e_accel_fs_esp_prot fs_prot[ACCEL_FS_ESP_NUM_TYPES]; }; +struct mlx5e_ipsec_tx { + struct mlx5_flow_table *ft; + struct mutex mutex; /* Protect IPsec TX steering */ + u32 refcnt; +}; + /* IPsec RX flow steering */ static enum mlx5e_traffic_types fs_esp2tt(enum accel_fs_esp_type i) { @@ -228,8 +234,8 @@ static int rx_fs_create(struct mlx5e_priv *priv, fs_prot->miss_rule = miss_rule; out: - kfree(flow_group_in); - kfree(spec); + kvfree(flow_group_in); + kvfree(spec); return err; } @@ -323,6 +329,77 @@ out: mutex_unlock(&fs_prot->prot_mutex); } +/* IPsec TX flow steering */ +static int tx_create(struct mlx5e_priv *priv) +{ + struct mlx5_flow_table_attr ft_attr = {}; + struct mlx5e_ipsec *ipsec = priv->ipsec; + struct mlx5_flow_table *ft; + int err; + + priv->fs.egress_ns = + mlx5_get_flow_namespace(priv->mdev, + MLX5_FLOW_NAMESPACE_EGRESS_KERNEL); + if (!priv->fs.egress_ns) + return -EOPNOTSUPP; + + ft_attr.max_fte = NUM_IPSEC_FTE; + ft_attr.autogroup.max_num_groups = 1; + ft = mlx5_create_auto_grouped_flow_table(priv->fs.egress_ns, &ft_attr); + if (IS_ERR(ft)) { + err = PTR_ERR(ft); + netdev_err(priv->netdev, "fail to create ipsec tx ft err=%d\n", err); + return err; + } + ipsec->tx_fs->ft = ft; + return 0; +} + +static void tx_destroy(struct mlx5e_priv *priv) +{ + struct mlx5e_ipsec *ipsec = priv->ipsec; + + if (IS_ERR_OR_NULL(ipsec->tx_fs->ft)) + return; + + mlx5_destroy_flow_table(ipsec->tx_fs->ft); + ipsec->tx_fs->ft = NULL; +} + +static int tx_ft_get(struct mlx5e_priv *priv) +{ + struct mlx5e_ipsec_tx *tx_fs = priv->ipsec->tx_fs; + int err = 0; + + mutex_lock(&tx_fs->mutex); + if (tx_fs->refcnt++) + goto out; + + err = tx_create(priv); + if (err) { + tx_fs->refcnt--; + goto out; + } + +out: + mutex_unlock(&tx_fs->mutex); + return err; +} + +static void tx_ft_put(struct mlx5e_priv *priv) +{ + struct mlx5e_ipsec_tx *tx_fs = priv->ipsec->tx_fs; + + mutex_lock(&tx_fs->mutex); + if (--tx_fs->refcnt) + goto out; + + tx_destroy(priv); + +out: + mutex_unlock(&tx_fs->mutex); +} + static void setup_fte_common(struct mlx5_accel_esp_xfrm_attrs *attrs, u32 ipsec_obj_id, struct mlx5_flow_spec *spec, @@ -457,6 +534,54 @@ out: return err; } +static int tx_add_rule(struct mlx5e_priv *priv, + struct mlx5_accel_esp_xfrm_attrs *attrs, + u32 ipsec_obj_id, + struct mlx5e_ipsec_rule *ipsec_rule) +{ + struct mlx5_flow_act flow_act = {}; + struct mlx5_flow_handle *rule; + struct mlx5_flow_spec *spec; + int err = 0; + + err = tx_ft_get(priv); + if (err) + return err; + + spec = kvzalloc(sizeof(*spec), GFP_KERNEL); + if (!spec) { + err = -ENOMEM; + goto out; + } + + setup_fte_common(attrs, ipsec_obj_id, spec, &flow_act); + + /* Add IPsec indicator in metadata_reg_a */ + spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS_2; + MLX5_SET(fte_match_param, spec->match_criteria, misc_parameters_2.metadata_reg_a, + MLX5_ETH_WQE_FT_META_IPSEC); + MLX5_SET(fte_match_param, spec->match_value, misc_parameters_2.metadata_reg_a, + MLX5_ETH_WQE_FT_META_IPSEC); + + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_ALLOW | + MLX5_FLOW_CONTEXT_ACTION_IPSEC_ENCRYPT; + rule = mlx5_add_flow_rules(priv->ipsec->tx_fs->ft, spec, &flow_act, NULL, 0); + if (IS_ERR(rule)) { + err = PTR_ERR(rule); + netdev_err(priv->netdev, "fail to add ipsec rule attrs->action=0x%x, err=%d\n", + attrs->action, err); + goto out; + } + + ipsec_rule->rule = rule; + +out: + kvfree(spec); + if (err) + tx_ft_put(priv); + return err; +} + static void rx_del_rule(struct mlx5e_priv *priv, struct mlx5_accel_esp_xfrm_attrs *attrs, struct mlx5e_ipsec_rule *ipsec_rule) @@ -470,15 +595,27 @@ static void rx_del_rule(struct mlx5e_priv *priv, rx_ft_put(priv, attrs->is_ipv6 ? ACCEL_FS_ESP6 : ACCEL_FS_ESP4); } +static void tx_del_rule(struct mlx5e_priv *priv, + struct mlx5e_ipsec_rule *ipsec_rule) +{ + mlx5_del_flow_rules(ipsec_rule->rule); + ipsec_rule->rule = NULL; + + tx_ft_put(priv); +} + int mlx5e_accel_ipsec_fs_add_rule(struct mlx5e_priv *priv, struct mlx5_accel_esp_xfrm_attrs *attrs, u32 ipsec_obj_id, struct mlx5e_ipsec_rule *ipsec_rule) { - if (!priv->ipsec->rx_fs || attrs->action != MLX5_ACCEL_ESP_ACTION_DECRYPT) + if (!priv->ipsec->rx_fs) return -EOPNOTSUPP; - return rx_add_rule(priv, attrs, ipsec_obj_id, ipsec_rule); + if (attrs->action == MLX5_ACCEL_ESP_ACTION_DECRYPT) + return rx_add_rule(priv, attrs, ipsec_obj_id, ipsec_rule); + else + return tx_add_rule(priv, attrs, ipsec_obj_id, ipsec_rule); } void mlx5e_accel_ipsec_fs_del_rule(struct mlx5e_priv *priv, @@ -488,7 +625,18 @@ void mlx5e_accel_ipsec_fs_del_rule(struct mlx5e_priv *priv, if (!priv->ipsec->rx_fs) return; - rx_del_rule(priv, attrs, ipsec_rule); + if (attrs->action == MLX5_ACCEL_ESP_ACTION_DECRYPT) + rx_del_rule(priv, attrs, ipsec_rule); + else + tx_del_rule(priv, ipsec_rule); +} + +static void fs_cleanup_tx(struct mlx5e_priv *priv) +{ + mutex_destroy(&priv->ipsec->tx_fs->mutex); + WARN_ON(priv->ipsec->tx_fs->refcnt); + kfree(priv->ipsec->tx_fs); + priv->ipsec->tx_fs = NULL; } static void fs_cleanup_rx(struct mlx5e_priv *priv) @@ -507,6 +655,17 @@ static void fs_cleanup_rx(struct mlx5e_priv *priv) priv->ipsec->rx_fs = NULL; } +static int fs_init_tx(struct mlx5e_priv *priv) +{ + priv->ipsec->tx_fs = + kzalloc(sizeof(struct mlx5e_ipsec_tx), GFP_KERNEL); + if (!priv->ipsec->tx_fs) + return -ENOMEM; + + mutex_init(&priv->ipsec->tx_fs->mutex); + return 0; +} + static int fs_init_rx(struct mlx5e_priv *priv) { struct mlx5e_accel_fs_esp_prot *fs_prot; @@ -532,13 +691,24 @@ void mlx5e_accel_ipsec_fs_cleanup(struct mlx5e_priv *priv) if (!priv->ipsec->rx_fs) return; + fs_cleanup_tx(priv); fs_cleanup_rx(priv); } int mlx5e_accel_ipsec_fs_init(struct mlx5e_priv *priv) { + int err; + if (!mlx5_is_ipsec_device(priv->mdev) || !priv->ipsec) return -EOPNOTSUPP; - return fs_init_rx(priv); + err = fs_init_tx(priv); + if (err) + return err; + + err = fs_init_rx(priv); + if (err) + fs_cleanup_tx(priv); + + return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c index 93a8d68815ad..11e31a3db2be 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c @@ -34,7 +34,7 @@ #include <crypto/aead.h> #include <net/xfrm.h> #include <net/esp.h> - +#include "accel/ipsec_offload.h" #include "en_accel/ipsec_rxtx.h" #include "en_accel/ipsec.h" #include "accel/accel.h" @@ -233,18 +233,94 @@ static void mlx5e_ipsec_set_metadata(struct sk_buff *skb, ntohs(mdata->content.tx.seq)); } -bool mlx5e_ipsec_handle_tx_skb(struct mlx5e_priv *priv, - struct mlx5_wqe_eth_seg *eseg, - struct sk_buff *skb) +void mlx5e_ipsec_handle_tx_wqe(struct mlx5e_tx_wqe *wqe, + struct mlx5e_accel_tx_ipsec_state *ipsec_st, + struct mlx5_wqe_inline_seg *inlseg) +{ + inlseg->byte_count = cpu_to_be32(ipsec_st->tailen | MLX5_INLINE_SEG); + esp_output_fill_trailer((u8 *)inlseg->data, 0, ipsec_st->plen, ipsec_st->xo->proto); +} + +static int mlx5e_ipsec_set_state(struct mlx5e_priv *priv, + struct sk_buff *skb, + struct xfrm_state *x, + struct xfrm_offload *xo, + struct mlx5e_accel_tx_ipsec_state *ipsec_st) +{ + unsigned int blksize, clen, alen, plen; + struct crypto_aead *aead; + unsigned int tailen; + + ipsec_st->x = x; + ipsec_st->xo = xo; + if (mlx5_is_ipsec_device(priv->mdev)) { + aead = x->data; + alen = crypto_aead_authsize(aead); + blksize = ALIGN(crypto_aead_blocksize(aead), 4); + clen = ALIGN(skb->len + 2, blksize); + plen = max_t(u32, clen - skb->len, 4); + tailen = plen + alen; + ipsec_st->plen = plen; + ipsec_st->tailen = tailen; + } + + return 0; +} + +void mlx5e_ipsec_tx_build_eseg(struct mlx5e_priv *priv, struct sk_buff *skb, + struct mlx5_wqe_eth_seg *eseg) { struct xfrm_offload *xo = xfrm_offload(skb); - struct mlx5e_ipsec_metadata *mdata; - struct mlx5e_ipsec_sa_entry *sa_entry; + struct xfrm_encap_tmpl *encap; struct xfrm_state *x; struct sec_path *sp; + u8 l3_proto; + + sp = skb_sec_path(skb); + if (unlikely(sp->len != 1)) + return; + + x = xfrm_input_state(skb); + if (unlikely(!x)) + return; + + if (unlikely(!x->xso.offload_handle || + (skb->protocol != htons(ETH_P_IP) && + skb->protocol != htons(ETH_P_IPV6)))) + return; + + mlx5e_ipsec_set_swp(skb, eseg, x->props.mode, xo); - if (!xo) - return true; + l3_proto = (x->props.family == AF_INET) ? + ((struct iphdr *)skb_network_header(skb))->protocol : + ((struct ipv6hdr *)skb_network_header(skb))->nexthdr; + + if (mlx5_is_ipsec_device(priv->mdev)) { + eseg->flow_table_metadata |= cpu_to_be32(MLX5_ETH_WQE_FT_META_IPSEC); + eseg->trailer |= cpu_to_be32(MLX5_ETH_WQE_INSERT_TRAILER); + encap = x->encap; + if (!encap) { + eseg->trailer |= (l3_proto == IPPROTO_ESP) ? + cpu_to_be32(MLX5_ETH_WQE_TRAILER_HDR_OUTER_IP_ASSOC) : + cpu_to_be32(MLX5_ETH_WQE_TRAILER_HDR_OUTER_L4_ASSOC); + } else if (encap->encap_type == UDP_ENCAP_ESPINUDP) { + eseg->trailer |= (l3_proto == IPPROTO_ESP) ? + cpu_to_be32(MLX5_ETH_WQE_TRAILER_HDR_INNER_IP_ASSOC) : + cpu_to_be32(MLX5_ETH_WQE_TRAILER_HDR_INNER_L4_ASSOC); + } + } +} + +bool mlx5e_ipsec_handle_tx_skb(struct net_device *netdev, + struct sk_buff *skb, + struct mlx5e_accel_tx_ipsec_state *ipsec_st) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct xfrm_offload *xo = xfrm_offload(skb); + struct mlx5e_ipsec_sa_entry *sa_entry; + struct mlx5e_ipsec_metadata *mdata; + struct xfrm_state *x; + struct sec_path *sp; sp = skb_sec_path(skb); if (unlikely(sp->len != 1)) { @@ -270,15 +346,21 @@ bool mlx5e_ipsec_handle_tx_skb(struct mlx5e_priv *priv, atomic64_inc(&priv->ipsec->sw_stats.ipsec_tx_drop_trailer); goto drop; } - mdata = mlx5e_ipsec_add_metadata(skb); - if (IS_ERR(mdata)) { - atomic64_inc(&priv->ipsec->sw_stats.ipsec_tx_drop_metadata); - goto drop; + + if (MLX5_CAP_GEN(priv->mdev, fpga)) { + mdata = mlx5e_ipsec_add_metadata(skb); + if (IS_ERR(mdata)) { + atomic64_inc(&priv->ipsec->sw_stats.ipsec_tx_drop_metadata); + goto drop; + } } - mlx5e_ipsec_set_swp(skb, eseg, x->props.mode, xo); + sa_entry = (struct mlx5e_ipsec_sa_entry *)x->xso.offload_handle; sa_entry->set_iv_op(skb, x, xo); - mlx5e_ipsec_set_metadata(skb, mdata, xo); + if (MLX5_CAP_GEN(priv->mdev, fpga)) + mlx5e_ipsec_set_metadata(skb, mdata, xo); + + mlx5e_ipsec_set_state(priv, skb, x, xo, ipsec_st); return true; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h index f96e786db158..056dacb612b0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h @@ -43,6 +43,13 @@ #define MLX5_IPSEC_METADATA_SYNDROM_MASK (0x7F) #define MLX5_IPSEC_METADATA_HANDLE(metadata) (((metadata) >> 8) & 0xFF) +struct mlx5e_accel_tx_ipsec_state { + struct xfrm_offload *xo; + struct xfrm_state *x; + u32 tailen; + u32 plen; +}; + #ifdef CONFIG_MLX5_EN_IPSEC struct sk_buff *mlx5e_ipsec_handle_rx_skb(struct net_device *netdev, @@ -55,16 +62,32 @@ void mlx5e_ipsec_set_iv_esn(struct sk_buff *skb, struct xfrm_state *x, struct xfrm_offload *xo); void mlx5e_ipsec_set_iv(struct sk_buff *skb, struct xfrm_state *x, struct xfrm_offload *xo); -bool mlx5e_ipsec_handle_tx_skb(struct mlx5e_priv *priv, - struct mlx5_wqe_eth_seg *eseg, - struct sk_buff *skb); +bool mlx5e_ipsec_handle_tx_skb(struct net_device *netdev, + struct sk_buff *skb, + struct mlx5e_accel_tx_ipsec_state *ipsec_st); +void mlx5e_ipsec_handle_tx_wqe(struct mlx5e_tx_wqe *wqe, + struct mlx5e_accel_tx_ipsec_state *ipsec_st, + struct mlx5_wqe_inline_seg *inlseg); void mlx5e_ipsec_offload_handle_rx_skb(struct net_device *netdev, struct sk_buff *skb, struct mlx5_cqe64 *cqe); +static inline unsigned int mlx5e_ipsec_tx_ids_len(struct mlx5e_accel_tx_ipsec_state *ipsec_st) +{ + return ipsec_st->tailen; +} + static inline bool mlx5_ipsec_is_rx_flow(struct mlx5_cqe64 *cqe) { return !!(MLX5_IPSEC_METADATA_MARKER_MASK & be32_to_cpu(cqe->ft_metadata)); } + +static inline bool mlx5e_ipsec_is_tx_flow(struct mlx5e_accel_tx_ipsec_state *ipsec_st) +{ + return ipsec_st->x; +} + +void mlx5e_ipsec_tx_build_eseg(struct mlx5e_priv *priv, struct sk_buff *skb, + struct mlx5_wqe_eth_seg *eseg); #else static inline void mlx5e_ipsec_offload_handle_rx_skb(struct net_device *netdev, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c index 6bbfcf18107d..ccaccb9fc2f7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c @@ -253,7 +253,7 @@ resync_post_get_progress_params(struct mlx5e_icosq *sq, goto err_out; } - pdev = sq->channel->priv->mdev->device; + pdev = mlx5_core_dma_dev(sq->channel->priv->mdev); buf->dma_addr = dma_map_single(pdev, &buf->progress, PROGRESS_PARAMS_PADDED_SIZE, DMA_FROM_DEVICE); if (unlikely(dma_mapping_error(pdev, buf->dma_addr))) { @@ -390,7 +390,7 @@ void mlx5e_ktls_handle_get_psv_completion(struct mlx5e_icosq_wqe_info *wi, priv_rx = buf->priv_rx; resync = &priv_rx->resync; - dev = resync->priv->mdev->device; + dev = mlx5_core_dma_dev(resync->priv->mdev); if (unlikely(test_bit(MLX5E_PRIV_RX_FLAG_DELETING, priv_rx->flags))) goto out; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c index f4861545b236..b140e13fdcc8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c @@ -345,9 +345,6 @@ void mlx5e_ktls_tx_handle_resync_dump_comp(struct mlx5e_txqsq *sq, struct mlx5e_sq_stats *stats; struct mlx5e_sq_dma *dma; - if (!wi->resync_dump_frag_page) - return; - dma = mlx5e_dma_get(sq, (*dma_fifo_cc)++); stats = sq->stats; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_txrx.h index ff4c740af10b..7521c9be735b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_txrx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_txrx.h @@ -29,12 +29,24 @@ void mlx5e_ktls_handle_get_psv_completion(struct mlx5e_icosq_wqe_info *wi, void mlx5e_ktls_tx_handle_resync_dump_comp(struct mlx5e_txqsq *sq, struct mlx5e_tx_wqe_info *wi, u32 *dma_fifo_cc); +static inline bool +mlx5e_ktls_tx_try_handle_resync_dump_comp(struct mlx5e_txqsq *sq, + struct mlx5e_tx_wqe_info *wi, + u32 *dma_fifo_cc) +{ + if (unlikely(wi->resync_dump_frag_page)) { + mlx5e_ktls_tx_handle_resync_dump_comp(sq, wi, dma_fifo_cc); + return true; + } + return false; +} #else -static inline void -mlx5e_ktls_tx_handle_resync_dump_comp(struct mlx5e_txqsq *sq, - struct mlx5e_tx_wqe_info *wi, - u32 *dma_fifo_cc) +static inline bool +mlx5e_ktls_tx_try_handle_resync_dump_comp(struct mlx5e_txqsq *sq, + struct mlx5e_tx_wqe_info *wi, + u32 *dma_fifo_cc) { + return false; } #endif /* CONFIG_MLX5_EN_TLS */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c index b0c31d49ff8d..6982b193ee8a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c @@ -189,12 +189,10 @@ static bool mlx5e_tls_handle_ooo(struct mlx5e_tls_offload_context_tx *context, struct mlx5e_tls *tls) { u32 tcp_seq = ntohl(tcp_hdr(skb)->seq); - struct mlx5e_tx_wqe *wqe; struct sync_info info; struct sk_buff *nskb; int linear_len = 0; int headln; - u16 pi; int i; sq->stats->tls_ooo++; @@ -246,9 +244,7 @@ static bool mlx5e_tls_handle_ooo(struct mlx5e_tls_offload_context_tx *context, sq->stats->tls_resync_bytes += nskb->len; mlx5e_tls_complete_sync_skb(skb, nskb, tcp_seq, headln, cpu_to_be64(info.rcd_sn)); - pi = mlx5_wq_cyc_ctr2ix(&sq->wq, sq->pc); - wqe = MLX5E_TX_FETCH_WQE(sq, pi); - mlx5e_sq_xmit(sq, nskb, wqe, pi, true); + mlx5e_sq_xmit_simple(sq, nskb, true); return true; @@ -274,6 +270,8 @@ bool mlx5e_tls_handle_tx_skb(struct net_device *netdev, struct mlx5e_txqsq *sq, if (!datalen) return true; + mlx5e_tx_mpwqe_ensure_complete(sq); + tls_ctx = tls_get_ctx(skb->sk); if (WARN_ON_ONCE(tls_ctx->netdev != netdev)) goto err_out; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 08270987c506..d25a56ec6876 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -32,7 +32,7 @@ #include "en.h" #include "en/port.h" -#include "en/xsk/umem.h" +#include "en/xsk/pool.h" #include "lib/clock.h" void mlx5e_ethtool_get_drvinfo(struct mlx5e_priv *priv, @@ -243,7 +243,6 @@ int mlx5e_ethtool_get_sset_count(struct mlx5e_priv *priv, int sset) return MLX5E_NUM_PFLAGS; case ETH_SS_TEST: return mlx5e_self_test_num(priv); - fallthrough; default: return -EOPNOTSUPP; } @@ -1341,6 +1340,14 @@ static int mlx5e_set_tunable(struct net_device *dev, return err; } +static void mlx5e_get_pause_stats(struct net_device *netdev, + struct ethtool_pause_stats *pause_stats) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + + mlx5e_stats_pause_get(priv, pause_stats); +} + void mlx5e_ethtool_get_pauseparam(struct mlx5e_priv *priv, struct ethtool_pauseparam *pauseparam) { @@ -1901,7 +1908,7 @@ static int set_pflag_rx_no_csum_complete(struct net_device *netdev, bool enable) return 0; } -static int set_pflag_xdp_tx_mpwqe(struct net_device *netdev, bool enable) +static int set_pflag_tx_mpwqe_common(struct net_device *netdev, u32 flag, bool enable) { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; @@ -1913,7 +1920,7 @@ static int set_pflag_xdp_tx_mpwqe(struct net_device *netdev, bool enable) new_channels.params = priv->channels.params; - MLX5E_SET_PFLAG(&new_channels.params, MLX5E_PFLAG_XDP_TX_MPWQE, enable); + MLX5E_SET_PFLAG(&new_channels.params, flag, enable); if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) { priv->channels.params = new_channels.params; @@ -1924,6 +1931,16 @@ static int set_pflag_xdp_tx_mpwqe(struct net_device *netdev, bool enable) return err; } +static int set_pflag_xdp_tx_mpwqe(struct net_device *netdev, bool enable) +{ + return set_pflag_tx_mpwqe_common(netdev, MLX5E_PFLAG_XDP_TX_MPWQE, enable); +} + +static int set_pflag_skb_tx_mpwqe(struct net_device *netdev, bool enable) +{ + return set_pflag_tx_mpwqe_common(netdev, MLX5E_PFLAG_SKB_TX_MPWQE, enable); +} + static const struct pflag_desc mlx5e_priv_flags[MLX5E_NUM_PFLAGS] = { { "rx_cqe_moder", set_pflag_rx_cqe_based_moder }, { "tx_cqe_moder", set_pflag_tx_cqe_based_moder }, @@ -1931,6 +1948,7 @@ static const struct pflag_desc mlx5e_priv_flags[MLX5E_NUM_PFLAGS] = { { "rx_striding_rq", set_pflag_rx_striding_rq }, { "rx_no_csum_complete", set_pflag_rx_no_csum_complete }, { "xdp_tx_mpwqe", set_pflag_xdp_tx_mpwqe }, + { "skb_tx_mpwqe", set_pflag_skb_tx_mpwqe }, }; static int mlx5e_handle_pflag(struct net_device *netdev, @@ -2033,6 +2051,7 @@ const struct ethtool_ops mlx5e_ethtool_ops = { .set_rxnfc = mlx5e_set_rxnfc, .get_tunable = mlx5e_get_tunable, .set_tunable = mlx5e_set_tunable, + .get_pause_stats = mlx5e_get_pause_stats, .get_pauseparam = mlx5e_get_pauseparam, .set_pauseparam = mlx5e_set_pauseparam, .get_ts_info = mlx5e_get_ts_info, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c index 83c9b2bbc4af..b416a8ee2eed 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c @@ -33,7 +33,7 @@ #include <linux/mlx5/fs.h> #include "en.h" #include "en/params.h" -#include "en/xsk/umem.h" +#include "en/xsk/pool.h" struct mlx5e_ethtool_rule { struct list_head list; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 42ec28e29834..b3f02aac7f26 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -57,7 +57,7 @@ #include "en/monitor_stats.h" #include "en/health.h" #include "en/params.h" -#include "en/xsk/umem.h" +#include "en/xsk/pool.h" #include "en/xsk/setup.h" #include "en/xsk/rx.h" #include "en/xsk/tx.h" @@ -393,7 +393,7 @@ static void mlx5e_free_mpwqe_rq_drop_page(struct mlx5e_rq *rq) static int mlx5e_alloc_rq(struct mlx5e_channel *c, struct mlx5e_params *params, struct mlx5e_xsk_param *xsk, - struct xdp_umem *umem, + struct xsk_buff_pool *xsk_pool, struct mlx5e_rq_param *rqp, struct mlx5e_rq *rq) { @@ -419,9 +419,9 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, rq->mdev = mdev; rq->hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu); rq->xdpsq = &c->rq_xdpsq; - rq->umem = umem; + rq->xsk_pool = xsk_pool; - if (rq->umem) + if (rq->xsk_pool) rq->stats = &c->priv->channel_stats[c->ix].xskrq; else rq->stats = &c->priv->channel_stats[c->ix].rq; @@ -511,7 +511,7 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, if (xsk) { err = xdp_rxq_info_reg_mem_model(&rq->xdp_rxq, MEM_TYPE_XSK_BUFF_POOL, NULL); - xsk_buff_set_rxq_info(rq->umem, &rq->xdp_rxq); + xsk_pool_set_rxq_info(rq->xsk_pool, &rq->xdp_rxq); } else { /* Create a page_pool and register it with rxq */ pp_params.order = 0; @@ -861,11 +861,11 @@ void mlx5e_free_rx_descs(struct mlx5e_rq *rq) int mlx5e_open_rq(struct mlx5e_channel *c, struct mlx5e_params *params, struct mlx5e_rq_param *param, struct mlx5e_xsk_param *xsk, - struct xdp_umem *umem, struct mlx5e_rq *rq) + struct xsk_buff_pool *xsk_pool, struct mlx5e_rq *rq) { int err; - err = mlx5e_alloc_rq(c, params, xsk, umem, param, rq); + err = mlx5e_alloc_rq(c, params, xsk, xsk_pool, param, rq); if (err) return err; @@ -893,6 +893,13 @@ int mlx5e_open_rq(struct mlx5e_channel *c, struct mlx5e_params *params, if (MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_NO_CSUM_COMPLETE) || c->xdp) __set_bit(MLX5E_RQ_STATE_NO_CSUM_COMPLETE, &c->rq.state); + /* For CQE compression on striding RQ, use stride index provided by + * HW if capability is supported. + */ + if (MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_STRIDING_RQ) && + MLX5_CAP_GEN(c->mdev, mini_cqe_resp_stride_index)) + __set_bit(MLX5E_RQ_STATE_MINI_CQE_HW_STRIDX, &c->rq.state); + return 0; err_destroy_rq: @@ -970,7 +977,7 @@ static int mlx5e_alloc_xdpsq_db(struct mlx5e_xdpsq *sq, int numa) static int mlx5e_alloc_xdpsq(struct mlx5e_channel *c, struct mlx5e_params *params, - struct xdp_umem *umem, + struct xsk_buff_pool *xsk_pool, struct mlx5e_sq_param *param, struct mlx5e_xdpsq *sq, bool is_redirect) @@ -986,9 +993,9 @@ static int mlx5e_alloc_xdpsq(struct mlx5e_channel *c, sq->uar_map = mdev->mlx5e_res.bfreg.map; sq->min_inline_mode = params->tx_min_inline_mode; sq->hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu); - sq->umem = umem; + sq->xsk_pool = xsk_pool; - sq->stats = sq->umem ? + sq->stats = sq->xsk_pool ? &c->priv->channel_stats[c->ix].xsksq : is_redirect ? &c->priv->channel_stats[c->ix].xdpsq : @@ -1085,6 +1092,7 @@ static void mlx5e_free_icosq(struct mlx5e_icosq *sq) static void mlx5e_free_txqsq_db(struct mlx5e_txqsq *sq) { kvfree(sq->db.wqe_info); + kvfree(sq->db.skb_fifo); kvfree(sq->db.dma_fifo); } @@ -1096,15 +1104,19 @@ static int mlx5e_alloc_txqsq_db(struct mlx5e_txqsq *sq, int numa) sq->db.dma_fifo = kvzalloc_node(array_size(df_sz, sizeof(*sq->db.dma_fifo)), GFP_KERNEL, numa); + sq->db.skb_fifo = kvzalloc_node(array_size(df_sz, + sizeof(*sq->db.skb_fifo)), + GFP_KERNEL, numa); sq->db.wqe_info = kvzalloc_node(array_size(wq_sz, sizeof(*sq->db.wqe_info)), GFP_KERNEL, numa); - if (!sq->db.dma_fifo || !sq->db.wqe_info) { + if (!sq->db.dma_fifo || !sq->db.skb_fifo || !sq->db.wqe_info) { mlx5e_free_txqsq_db(sq); return -ENOMEM; } sq->dma_fifo_mask = df_sz - 1; + sq->skb_fifo_mask = df_sz - 1; return 0; } @@ -1115,6 +1127,12 @@ static int mlx5e_calc_sq_stop_room(struct mlx5e_txqsq *sq, u8 log_sq_size) sq->stop_room = mlx5e_tls_get_stop_room(sq); sq->stop_room += mlx5e_stop_room_for_wqe(MLX5_SEND_WQE_MAX_WQEBBS); + if (test_bit(MLX5E_SQ_STATE_MPWQE, &sq->state)) + /* A MPWQE can take up to the maximum-sized WQE + all the normal + * stop room can be taken if a new packet breaks the active + * MPWQE session and allocates its WQEs right away. + */ + sq->stop_room += mlx5e_stop_room_for_wqe(MLX5_SEND_WQE_MAX_WQEBBS); if (WARN_ON(sq->stop_room >= sq_size)) { netdev_err(sq->channel->netdev, "Stop room %hu is bigger than the SQ size %d\n", @@ -1156,6 +1174,8 @@ static int mlx5e_alloc_txqsq(struct mlx5e_channel *c, set_bit(MLX5E_SQ_STATE_IPSEC, &sq->state); if (mlx5_accel_is_tls_device(c->priv->mdev)) set_bit(MLX5E_SQ_STATE_TLS, &sq->state); + if (param->is_mpw) + set_bit(MLX5E_SQ_STATE_MPWQE, &sq->state); err = mlx5e_calc_sq_stop_room(sq, params->log_sq_size); if (err) return err; @@ -1449,13 +1469,13 @@ void mlx5e_close_icosq(struct mlx5e_icosq *sq) } int mlx5e_open_xdpsq(struct mlx5e_channel *c, struct mlx5e_params *params, - struct mlx5e_sq_param *param, struct xdp_umem *umem, + struct mlx5e_sq_param *param, struct xsk_buff_pool *xsk_pool, struct mlx5e_xdpsq *sq, bool is_redirect) { struct mlx5e_create_sq_param csp = {}; int err; - err = mlx5e_alloc_xdpsq(c, params, umem, param, sq, is_redirect); + err = mlx5e_alloc_xdpsq(c, params, xsk_pool, param, sq, is_redirect); if (err) return err; @@ -1948,7 +1968,7 @@ static u8 mlx5e_enumerate_lag_port(struct mlx5_core_dev *mdev, int ix) static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, struct mlx5e_params *params, struct mlx5e_channel_param *cparam, - struct xdp_umem *umem, + struct xsk_buff_pool *xsk_pool, struct mlx5e_channel **cp) { int cpu = cpumask_first(mlx5_comp_irq_get_affinity_mask(priv->mdev, ix)); @@ -1972,7 +1992,7 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, c->tstamp = &priv->tstamp; c->ix = ix; c->cpu = cpu; - c->pdev = priv->mdev->device; + c->pdev = mlx5_core_dma_dev(priv->mdev); c->netdev = priv->netdev; c->mkey_be = cpu_to_be32(priv->mdev->mlx5e_res.mkey.key); c->num_tc = params->num_tc; @@ -1987,9 +2007,9 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, if (unlikely(err)) goto err_napi_del; - if (umem) { - mlx5e_build_xsk_param(umem, &xsk); - err = mlx5e_open_xsk(priv, params, &xsk, umem, c); + if (xsk_pool) { + mlx5e_build_xsk_param(xsk_pool, &xsk); + err = mlx5e_open_xsk(priv, params, &xsk, xsk_pool, c); if (unlikely(err)) goto err_close_queues; } @@ -2160,7 +2180,7 @@ void mlx5e_build_rq_param(struct mlx5e_priv *priv, MLX5_SET(rqc, rqc, vsd, params->vlan_strip_disable); MLX5_SET(rqc, rqc, scatter_fcs, params->scatter_fcs_en); - param->wq.buf_numa_node = dev_to_node(mdev->device); + param->wq.buf_numa_node = dev_to_node(mlx5_core_dma_dev(mdev)); mlx5e_build_rx_cq_param(priv, params, xsk, ¶m->cqp); } @@ -2176,7 +2196,7 @@ static void mlx5e_build_drop_rq_param(struct mlx5e_priv *priv, mlx5e_get_rqwq_log_stride(MLX5_WQ_TYPE_CYCLIC, 1)); MLX5_SET(rqc, rqc, counter_set_id, priv->drop_rq_q_counter); - param->wq.buf_numa_node = dev_to_node(mdev->device); + param->wq.buf_numa_node = dev_to_node(mlx5_core_dma_dev(mdev)); } void mlx5e_build_sq_param_common(struct mlx5e_priv *priv, @@ -2188,7 +2208,7 @@ void mlx5e_build_sq_param_common(struct mlx5e_priv *priv, MLX5_SET(wq, wq, log_wq_stride, ilog2(MLX5_SEND_WQE_BB)); MLX5_SET(wq, wq, pd, priv->mdev->mlx5e_res.pdn); - param->wq.buf_numa_node = dev_to_node(priv->mdev->device); + param->wq.buf_numa_node = dev_to_node(mlx5_core_dma_dev(priv->mdev)); } static void mlx5e_build_sq_param(struct mlx5e_priv *priv, @@ -2204,6 +2224,7 @@ static void mlx5e_build_sq_param(struct mlx5e_priv *priv, mlx5e_build_sq_param_common(priv, param); MLX5_SET(wq, wq, log_wq_sz, params->log_sq_size); MLX5_SET(sqc, sqc, allow_swp, allow_swp); + param->is_mpw = MLX5E_GET_PFLAG(params, MLX5E_PFLAG_SKB_TX_MPWQE); mlx5e_build_tx_cq_param(priv, params, ¶m->cqp); } @@ -2223,6 +2244,7 @@ void mlx5e_build_rx_cq_param(struct mlx5e_priv *priv, struct mlx5e_cq_param *param) { struct mlx5_core_dev *mdev = priv->mdev; + bool hw_stridx = false; void *cqc = param->cqc; u8 log_cq_size; @@ -2230,6 +2252,7 @@ void mlx5e_build_rx_cq_param(struct mlx5e_priv *priv, case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ: log_cq_size = mlx5e_mpwqe_get_log_rq_size(params, xsk) + mlx5e_mpwqe_get_log_num_strides(mdev, params, xsk); + hw_stridx = MLX5_CAP_GEN(mdev, mini_cqe_resp_stride_index); break; default: /* MLX5_WQ_TYPE_CYCLIC */ log_cq_size = params->log_rq_mtu_frames; @@ -2237,7 +2260,8 @@ void mlx5e_build_rx_cq_param(struct mlx5e_priv *priv, MLX5_SET(cqc, cqc, log_cq_size, log_cq_size); if (MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_CQE_COMPRESS)) { - MLX5_SET(cqc, cqc, mini_cqe_res_format, MLX5_CQE_FORMAT_CSUM); + MLX5_SET(cqc, cqc, mini_cqe_res_format, hw_stridx ? + MLX5_CQE_FORMAT_CSUM_STRIDX : MLX5_CQE_FORMAT_CSUM); MLX5_SET(cqc, cqc, cqe_comp_en, 1); } @@ -2350,12 +2374,12 @@ int mlx5e_open_channels(struct mlx5e_priv *priv, mlx5e_build_channel_param(priv, &chs->params, cparam); for (i = 0; i < chs->num; i++) { - struct xdp_umem *umem = NULL; + struct xsk_buff_pool *xsk_pool = NULL; if (chs->params.xdp_prog) - umem = mlx5e_xsk_get_umem(&chs->params, chs->params.xsk, i); + xsk_pool = mlx5e_xsk_get_pool(&chs->params, chs->params.xsk, i); - err = mlx5e_open_channel(priv, i, &chs->params, cparam, umem, &chs->c[i]); + err = mlx5e_open_channel(priv, i, &chs->params, cparam, xsk_pool, &chs->c[i]); if (err) goto err_close_channels; } @@ -3222,8 +3246,8 @@ static int mlx5e_alloc_drop_cq(struct mlx5_core_dev *mdev, struct mlx5e_cq *cq, struct mlx5e_cq_param *param) { - param->wq.buf_numa_node = dev_to_node(mdev->device); - param->wq.db_numa_node = dev_to_node(mdev->device); + param->wq.buf_numa_node = dev_to_node(mlx5_core_dma_dev(mdev)); + param->wq.db_numa_node = dev_to_node(mlx5_core_dma_dev(mdev)); return mlx5e_alloc_cq_common(mdev, param, cq); } @@ -3927,13 +3951,14 @@ static bool mlx5e_xsk_validate_mtu(struct net_device *netdev, u16 ix; for (ix = 0; ix < chs->params.num_channels; ix++) { - struct xdp_umem *umem = mlx5e_xsk_get_umem(&chs->params, chs->params.xsk, ix); + struct xsk_buff_pool *xsk_pool = + mlx5e_xsk_get_pool(&chs->params, chs->params.xsk, ix); struct mlx5e_xsk_param xsk; - if (!umem) + if (!xsk_pool) continue; - mlx5e_build_xsk_param(umem, &xsk); + mlx5e_build_xsk_param(xsk_pool, &xsk); if (!mlx5e_validate_xsk_param(new_params, &xsk, mdev)) { u32 hr = mlx5e_get_linear_rq_headroom(new_params, &xsk); @@ -4466,8 +4491,8 @@ static int mlx5e_xdp(struct net_device *dev, struct netdev_bpf *xdp) switch (xdp->command) { case XDP_SETUP_PROG: return mlx5e_xdp_set(dev, xdp->prog); - case XDP_SETUP_XSK_UMEM: - return mlx5e_xsk_setup_umem(dev, xdp->xsk.umem, + case XDP_SETUP_XSK_POOL: + return mlx5e_xsk_setup_pool(dev, xdp->xsk.pool, xdp->xsk.queue_id); default: return -EINVAL; @@ -4758,6 +4783,8 @@ void mlx5e_build_nic_params(struct mlx5e_priv *priv, params->log_sq_size = is_kdump_kernel() ? MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE : MLX5E_PARAMS_DEFAULT_LOG_SQ_SIZE; + MLX5E_SET_PFLAG(params, MLX5E_PFLAG_SKB_TX_MPWQE, + MLX5_CAP_ETH(mdev, enhanced_multi_pkt_send_wqe)); /* XDP SQ */ MLX5E_SET_PFLAG(params, MLX5E_PFLAG_XDP_TX_MPWQE, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c index e979bff64c49..67247c33b9fd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -39,7 +39,6 @@ #include <net/ipv6_stubs.h> #include "eswitch.h" -#include "esw/chains.h" #include "en.h" #include "en_rep.h" #include "en/txrx.h" @@ -288,6 +287,14 @@ static u32 mlx5e_rep_get_rxfh_indir_size(struct net_device *netdev) return mlx5e_ethtool_get_rxfh_indir_size(priv); } +static void mlx5e_uplink_rep_get_pause_stats(struct net_device *netdev, + struct ethtool_pause_stats *stats) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + + mlx5e_stats_pause_get(priv, stats); +} + static void mlx5e_uplink_rep_get_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pauseparam) { @@ -362,23 +369,11 @@ static const struct ethtool_ops mlx5e_uplink_rep_ethtool_ops = { .set_rxfh = mlx5e_set_rxfh, .get_rxnfc = mlx5e_get_rxnfc, .set_rxnfc = mlx5e_set_rxnfc, + .get_pause_stats = mlx5e_uplink_rep_get_pause_stats, .get_pauseparam = mlx5e_uplink_rep_get_pauseparam, .set_pauseparam = mlx5e_uplink_rep_set_pauseparam, }; -static void mlx5e_rep_get_port_parent_id(struct net_device *dev, - struct netdev_phys_item_id *ppid) -{ - struct mlx5e_priv *priv; - u64 parent_id; - - priv = netdev_priv(dev); - - parent_id = mlx5_query_nic_system_image_guid(priv->mdev); - ppid->id_len = sizeof(parent_id); - memcpy(ppid->id, &parent_id, sizeof(parent_id)); -} - static void mlx5e_sqs2vport_stop(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep) { @@ -603,12 +598,13 @@ static int mlx5e_uplink_rep_set_vf_vlan(struct net_device *dev, int vf, u16 vlan return 0; } -static struct devlink_port *mlx5e_rep_get_devlink_port(struct net_device *dev) +static struct devlink_port *mlx5e_rep_get_devlink_port(struct net_device *netdev) { - struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5e_rep_priv *rpriv = priv->ppriv; + struct mlx5_core_dev *dev = priv->mdev; - return &rpriv->dl_port; + return mlx5_esw_offloads_devlink_port(dev->priv.eswitch, rpriv->rep->vport); } static int mlx5e_rep_change_carrier(struct net_device *dev, bool new_carrier) @@ -1198,63 +1194,13 @@ static const struct mlx5e_profile mlx5e_uplink_rep_profile = { .stats_grps_num = mlx5e_ul_rep_stats_grps_num, }; -static bool -is_devlink_port_supported(const struct mlx5_core_dev *dev, - const struct mlx5e_rep_priv *rpriv) -{ - return rpriv->rep->vport == MLX5_VPORT_UPLINK || - rpriv->rep->vport == MLX5_VPORT_PF || - mlx5_eswitch_is_vf_vport(dev->priv.eswitch, rpriv->rep->vport); -} - -static int register_devlink_port(struct mlx5_core_dev *dev, - struct mlx5e_rep_priv *rpriv) -{ - struct devlink *devlink = priv_to_devlink(dev); - struct mlx5_eswitch_rep *rep = rpriv->rep; - struct devlink_port_attrs attrs = {}; - struct netdev_phys_item_id ppid = {}; - unsigned int dl_port_index = 0; - u16 pfnum; - - if (!is_devlink_port_supported(dev, rpriv)) - return 0; - - mlx5e_rep_get_port_parent_id(rpriv->netdev, &ppid); - dl_port_index = mlx5_esw_vport_to_devlink_port_index(dev, rep->vport); - pfnum = PCI_FUNC(dev->pdev->devfn); - if (rep->vport == MLX5_VPORT_UPLINK) { - attrs.flavour = DEVLINK_PORT_FLAVOUR_PHYSICAL; - attrs.phys.port_number = pfnum; - memcpy(attrs.switch_id.id, &ppid.id[0], ppid.id_len); - attrs.switch_id.id_len = ppid.id_len; - devlink_port_attrs_set(&rpriv->dl_port, &attrs); - } else if (rep->vport == MLX5_VPORT_PF) { - memcpy(rpriv->dl_port.attrs.switch_id.id, &ppid.id[0], ppid.id_len); - rpriv->dl_port.attrs.switch_id.id_len = ppid.id_len; - devlink_port_attrs_pci_pf_set(&rpriv->dl_port, pfnum); - } else if (mlx5_eswitch_is_vf_vport(dev->priv.eswitch, rpriv->rep->vport)) { - memcpy(rpriv->dl_port.attrs.switch_id.id, &ppid.id[0], ppid.id_len); - rpriv->dl_port.attrs.switch_id.id_len = ppid.id_len; - devlink_port_attrs_pci_vf_set(&rpriv->dl_port, - pfnum, rep->vport - 1); - } - return devlink_port_register(devlink, &rpriv->dl_port, dl_port_index); -} - -static void unregister_devlink_port(struct mlx5_core_dev *dev, - struct mlx5e_rep_priv *rpriv) -{ - if (is_devlink_port_supported(dev, rpriv)) - devlink_port_unregister(&rpriv->dl_port); -} - /* e-Switch vport representors */ static int mlx5e_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep) { const struct mlx5e_profile *profile; struct mlx5e_rep_priv *rpriv; + struct devlink_port *dl_port; struct net_device *netdev; int nch, err; @@ -1304,28 +1250,19 @@ mlx5e_vport_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep) goto err_detach_netdev; } - err = register_devlink_port(dev, rpriv); - if (err) { - netdev_warn(netdev, "Failed to register devlink port %d\n", - rep->vport); - goto err_neigh_cleanup; - } - err = register_netdev(netdev); if (err) { netdev_warn(netdev, "Failed to register representor netdev for vport %d\n", rep->vport); - goto err_devlink_cleanup; + goto err_neigh_cleanup; } - if (is_devlink_port_supported(dev, rpriv)) - devlink_port_type_eth_set(&rpriv->dl_port, netdev); + dl_port = mlx5_esw_offloads_devlink_port(dev->priv.eswitch, rpriv->rep->vport); + if (dl_port) + devlink_port_type_eth_set(dl_port, netdev); return 0; -err_devlink_cleanup: - unregister_devlink_port(dev, rpriv); - err_neigh_cleanup: mlx5e_rep_neigh_cleanup(rpriv); @@ -1349,12 +1286,13 @@ mlx5e_vport_rep_unload(struct mlx5_eswitch_rep *rep) struct net_device *netdev = rpriv->netdev; struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *dev = priv->mdev; + struct devlink_port *dl_port; void *ppriv = priv->ppriv; - if (is_devlink_port_supported(dev, rpriv)) - devlink_port_type_clear(&rpriv->dl_port); + dl_port = mlx5_esw_offloads_devlink_port(dev->priv.eswitch, rpriv->rep->vport); + if (dl_port) + devlink_port_type_clear(dl_port); unregister_netdev(netdev); - unregister_devlink_port(dev, rpriv); mlx5e_rep_neigh_cleanup(rpriv); mlx5e_detach_netdev(priv); if (rep->vport == MLX5_VPORT_UPLINK) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h index 0d1562e20118..9020d1419bcf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h @@ -101,7 +101,6 @@ struct mlx5e_rep_priv { struct list_head vport_sqs_list; struct mlx5_rep_uplink_priv uplink_priv; /* valid for uplink rep */ struct rtnl_link_stats64 prev_vf_vport_stats; - struct devlink_port dl_port; }; static inline diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index 64c8ac5eabf6..599f5b5ebc97 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -30,7 +30,6 @@ * SOFTWARE. */ -#include <linux/prefetch.h> #include <linux/ip.h> #include <linux/ipv6.h> #include <linux/tcp.h> @@ -139,8 +138,17 @@ static inline void mlx5e_decompress_cqe(struct mlx5e_rq *rq, title->check_sum = mini_cqe->checksum; title->op_own &= 0xf0; title->op_own |= 0x01 & (cqcc >> wq->fbc.log_sz); - title->wqe_counter = cpu_to_be16(cqd->wqe_counter); + /* state bit set implies linked-list striding RQ wq type and + * HW stride index capability supported + */ + if (test_bit(MLX5E_RQ_STATE_MINI_CQE_HW_STRIDX, &rq->state)) { + title->wqe_counter = mini_cqe->stridx; + return; + } + + /* HW stride index capability not supported */ + title->wqe_counter = cpu_to_be16(cqd->wqe_counter); if (rq->wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) cqd->wqe_counter += mpwrq_get_cqe_consumed_strides(title); else @@ -282,8 +290,8 @@ static inline int mlx5e_page_alloc_pool(struct mlx5e_rq *rq, static inline int mlx5e_page_alloc(struct mlx5e_rq *rq, struct mlx5e_dma_info *dma_info) { - if (rq->umem) - return mlx5e_xsk_page_alloc_umem(rq, dma_info); + if (rq->xsk_pool) + return mlx5e_xsk_page_alloc_pool(rq, dma_info); else return mlx5e_page_alloc_pool(rq, dma_info); } @@ -314,7 +322,7 @@ static inline void mlx5e_page_release(struct mlx5e_rq *rq, struct mlx5e_dma_info *dma_info, bool recycle) { - if (rq->umem) + if (rq->xsk_pool) /* The `recycle` parameter is ignored, and the page is always * put into the Reuse Ring, because there is no way to return * the page to the userspace when the interface goes down. @@ -401,14 +409,14 @@ static int mlx5e_alloc_rx_wqes(struct mlx5e_rq *rq, u16 ix, u8 wqe_bulk) int err; int i; - if (rq->umem) { + if (rq->xsk_pool) { int pages_desired = wqe_bulk << rq->wqe.info.log_num_frags; /* Check in advance that we have enough frames, instead of * allocating one-by-one, failing and moving frames to the * Reuse Ring. */ - if (unlikely(!xsk_buff_can_alloc(rq->umem, pages_desired))) + if (unlikely(!xsk_buff_can_alloc(rq->xsk_pool, pages_desired))) return -ENOMEM; } @@ -506,8 +514,8 @@ static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix) /* Check in advance that we have enough frames, instead of allocating * one-by-one, failing and moving frames to the Reuse Ring. */ - if (rq->umem && - unlikely(!xsk_buff_can_alloc(rq->umem, MLX5_MPWRQ_PAGES_PER_WQE))) { + if (rq->xsk_pool && + unlikely(!xsk_buff_can_alloc(rq->xsk_pool, MLX5_MPWRQ_PAGES_PER_WQE))) { err = -ENOMEM; goto err; } @@ -755,7 +763,7 @@ INDIRECT_CALLABLE_SCOPE bool mlx5e_post_rx_mpwqes(struct mlx5e_rq *rq) * the driver when it refills the Fill Ring. * 2. Otherwise, busy poll by rescheduling the NAPI poll. */ - if (unlikely(alloc_err == -ENOMEM && rq->umem)) + if (unlikely(alloc_err == -ENOMEM && rq->xsk_pool)) return true; return false; @@ -1144,8 +1152,8 @@ mlx5e_skb_from_cqe_linear(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe, dma_sync_single_range_for_cpu(rq->pdev, di->addr, wi->offset, frag_size, DMA_FROM_DEVICE); - prefetchw(va); /* xdp_frame data area */ - prefetch(data); + net_prefetchw(va); /* xdp_frame data area */ + net_prefetch(data); mlx5e_fill_xdp_buff(rq, va, rx_headroom, cqe_bcnt, &xdp); if (mlx5e_xdp_handle(rq, di, &cqe_bcnt, &xdp)) @@ -1184,7 +1192,7 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe, return NULL; } - prefetchw(skb->data); + net_prefetchw(skb->data); while (byte_cnt) { u16 frag_consumed_bytes = @@ -1252,6 +1260,11 @@ static void mlx5e_handle_rx_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe) } mlx5e_complete_rx_cqe(rq, cqe, cqe_bcnt, skb); + + if (mlx5e_cqe_regb_chain(cqe)) + if (!mlx5e_tc_update_skb(cqe, skb)) + goto free_wqe; + napi_gro_receive(rq->cq.napi, skb); free_wqe: @@ -1399,7 +1412,7 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w return NULL; } - prefetchw(skb->data); + net_prefetchw(skb->data); if (unlikely(frag_offset >= PAGE_SIZE)) { di++; @@ -1451,8 +1464,8 @@ mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, dma_sync_single_range_for_cpu(rq->pdev, di->addr, head_offset, frag_size, DMA_FROM_DEVICE); - prefetchw(va); /* xdp_frame data area */ - prefetch(data); + net_prefetchw(va); /* xdp_frame data area */ + net_prefetch(data); mlx5e_fill_xdp_buff(rq, va, rx_headroom, cqe_bcnt32, &xdp); if (mlx5e_xdp_handle(rq, di, &cqe_bcnt32, &xdp)) { @@ -1513,6 +1526,11 @@ static void mlx5e_handle_rx_cqe_mpwrq(struct mlx5e_rq *rq, struct mlx5_cqe64 *cq goto mpwrq_cqe_out; mlx5e_complete_rx_cqe(rq, cqe, cqe_bcnt, skb); + + if (mlx5e_cqe_regb_chain(cqe)) + if (!mlx5e_tc_update_skb(cqe, skb)) + goto mpwrq_cqe_out; + napi_gro_receive(rq->cq.napi, skb); mpwrq_cqe_out: diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c b/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c index 46790216ce86..ce8ab1f01876 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c @@ -30,7 +30,6 @@ * SOFTWARE. */ -#include <linux/prefetch.h> #include <linux/ip.h> #include <linux/udp.h> #include <net/udp.h> @@ -115,7 +114,7 @@ static struct sk_buff *mlx5e_test_get_udp_skb(struct mlx5e_priv *priv) return NULL; } - prefetchw(skb->data); + net_prefetchw(skb->data); skb_reserve(skb, NET_IP_ALIGN); /* Reserve for ethernet and IP header */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c index f6383bc2bc3f..78f6a6f0a7e0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c @@ -110,6 +110,8 @@ static const struct counter_desc sw_stats_desc[] = { { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tso_inner_bytes) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_added_vlan_packets) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_nop) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_mpwqe_blks) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_mpwqe_pkts) }, #ifdef CONFIG_MLX5_EN_TLS { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tls_encrypted_packets) }, @@ -365,6 +367,8 @@ static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(sw) s->tx_tso_inner_bytes += sq_stats->tso_inner_bytes; s->tx_added_vlan_packets += sq_stats->added_vlan_packets; s->tx_nop += sq_stats->nop; + s->tx_mpwqe_blks += sq_stats->mpwqe_blks; + s->tx_mpwqe_pkts += sq_stats->mpwqe_pkts; s->tx_queue_stopped += sq_stats->stopped; s->tx_queue_wake += sq_stats->wake; s->tx_queue_dropped += sq_stats->dropped; @@ -689,6 +693,35 @@ static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(802_3) mlx5_core_access_reg(mdev, in, sz, out, sz, MLX5_REG_PPCNT, 0, 0); } +#define MLX5E_READ_CTR64_BE_F(ptr, c) \ + be64_to_cpu(*(__be64 *)((char *)ptr + \ + MLX5_BYTE_OFF(ppcnt_reg, \ + counter_set.eth_802_3_cntrs_grp_data_layout.c##_high))) + +void mlx5e_stats_pause_get(struct mlx5e_priv *priv, + struct ethtool_pause_stats *pause_stats) +{ + u32 ppcnt_ieee_802_3[MLX5_ST_SZ_DW(ppcnt_reg)]; + struct mlx5_core_dev *mdev = priv->mdev; + u32 in[MLX5_ST_SZ_DW(ppcnt_reg)] = {}; + int sz = MLX5_ST_SZ_BYTES(ppcnt_reg); + + if (!MLX5_BASIC_PPCNT_SUPPORTED(mdev)) + return; + + MLX5_SET(ppcnt_reg, in, local_port, 1); + MLX5_SET(ppcnt_reg, in, grp, MLX5_IEEE_802_3_COUNTERS_GROUP); + mlx5_core_access_reg(mdev, in, sz, ppcnt_ieee_802_3, + sz, MLX5_REG_PPCNT, 0, 0); + + pause_stats->tx_pause_frames = + MLX5E_READ_CTR64_BE_F(ppcnt_ieee_802_3, + a_pause_mac_ctrl_frames_transmitted); + pause_stats->rx_pause_frames = + MLX5E_READ_CTR64_BE_F(ppcnt_ieee_802_3, + a_pause_mac_ctrl_frames_received); +} + #define PPORT_2863_OFF(c) \ MLX5_BYTE_OFF(ppcnt_reg, \ counter_set.eth_2863_cntrs_grp_data_layout.c##_high) @@ -1539,6 +1572,8 @@ static const struct counter_desc sq_stats_desc[] = { { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, csum_partial_inner) }, { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, added_vlan_packets) }, { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, nop) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, mpwqe_blks) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, mpwqe_pkts) }, #ifdef CONFIG_MLX5_EN_TLS { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, tls_encrypted_packets) }, { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, tls_encrypted_bytes) }, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h index 562263d62141..162daaadb0d8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h @@ -105,6 +105,9 @@ void mlx5e_stats_fill(struct mlx5e_priv *priv, u64 *data, int idx); void mlx5e_stats_fill_strings(struct mlx5e_priv *priv, u8 *data); void mlx5e_stats_update_ndo_stats(struct mlx5e_priv *priv); +void mlx5e_stats_pause_get(struct mlx5e_priv *priv, + struct ethtool_pause_stats *pause_stats); + /* Concrete NIC Stats */ struct mlx5e_sw_stats { @@ -118,6 +121,8 @@ struct mlx5e_sw_stats { u64 tx_tso_inner_bytes; u64 tx_added_vlan_packets; u64 tx_nop; + u64 tx_mpwqe_blks; + u64 tx_mpwqe_pkts; u64 rx_lro_packets; u64 rx_lro_bytes; u64 rx_mcast_packets; @@ -348,6 +353,8 @@ struct mlx5e_sq_stats { u64 csum_partial_inner; u64 added_vlan_packets; u64 nop; + u64 mpwqe_blks; + u64 mpwqe_pkts; #ifdef CONFIG_MLX5_EN_TLS u64 tls_encrypted_packets; u64 tls_encrypted_bytes; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 1c93f92d9210..e3a968e9e2a0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -57,7 +57,6 @@ #include "en/rep/neigh.h" #include "en_tc.h" #include "eswitch.h" -#include "esw/chains.h" #include "fs_core.h" #include "en/port.h" #include "en/tc_tun.h" @@ -66,20 +65,11 @@ #include "en/mod_hdr.h" #include "lib/devcom.h" #include "lib/geneve.h" +#include "lib/fs_chains.h" #include "diag/en_tc_tracepoint.h" +#define nic_chains(priv) ((priv)->fs.tc.chains) #define MLX5_MH_ACT_SZ MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto) - -struct mlx5_nic_flow_attr { - u32 action; - u32 flow_tag; - struct mlx5_modify_hdr *modify_hdr; - u32 hairpin_tirn; - u8 match_level; - struct mlx5_flow_table *hairpin_ft; - struct mlx5_fc *counter; -}; - #define MLX5E_TC_FLOW_BASE (MLX5E_TC_FLAG_LAST_EXPORTED_BIT + 1) enum { @@ -153,11 +143,7 @@ struct mlx5e_tc_flow { struct rcu_head rcu_head; struct completion init_done; int tunnel_id; /* the mapped tunnel id of this flow */ - - union { - struct mlx5_esw_flow_attr esw_attr[0]; - struct mlx5_nic_flow_attr nic_attr[0]; - }; + struct mlx5_flow_attr *attr; }; struct mlx5e_tc_flow_parse_attr { @@ -170,7 +156,7 @@ struct mlx5e_tc_flow_parse_attr { }; #define MLX5E_TC_TABLE_NUM_GROUPS 4 -#define MLX5E_TC_TABLE_MAX_GROUP_SIZE BIT(16) +#define MLX5E_TC_TABLE_MAX_GROUP_SIZE BIT(18) struct mlx5e_tc_attr_to_reg_mapping mlx5e_tc_attr_to_reg_mappings[] = { [CHAIN_TO_REG] = { @@ -191,6 +177,16 @@ struct mlx5e_tc_attr_to_reg_mapping mlx5e_tc_attr_to_reg_mappings[] = { [MARK_TO_REG] = mark_to_reg_ct, [LABELS_TO_REG] = labels_to_reg_ct, [FTEID_TO_REG] = fteid_to_reg_ct, + /* For NIC rules we store the retore metadata directly + * into reg_b that is passed to SW since we don't + * jump between steering domains. + */ + [NIC_CHAIN_TO_REG] = { + .mfield = MLX5_ACTION_IN_FIELD_METADATA_REG_B, + .moffset = 0, + .mlen = 2, + }, + [NIC_ZONE_RESTORE_TO_REG] = nic_zone_restore_to_reg_ct, }; static void mlx5e_put_flow_tunnel_id(struct mlx5e_tc_flow *flow); @@ -244,6 +240,7 @@ mlx5e_tc_match_to_reg_get_match(struct mlx5_flow_spec *spec, int mlx5e_tc_match_to_reg_set(struct mlx5_core_dev *mdev, struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts, + enum mlx5_flow_namespace_type ns, enum mlx5e_tc_attr_to_reg type, u32 data) { @@ -253,8 +250,7 @@ mlx5e_tc_match_to_reg_set(struct mlx5_core_dev *mdev, char *modact; int err; - err = alloc_mod_hdr_actions(mdev, MLX5_FLOW_NAMESPACE_FDB, - mod_hdr_acts); + err = alloc_mod_hdr_actions(mdev, ns, mod_hdr_acts); if (err) return err; @@ -275,6 +271,54 @@ mlx5e_tc_match_to_reg_set(struct mlx5_core_dev *mdev, return 0; } +#define esw_offloads_mode(esw) (mlx5_eswitch_mode(esw) == MLX5_ESWITCH_OFFLOADS) + +static struct mlx5_tc_ct_priv * +get_ct_priv(struct mlx5e_priv *priv) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5_rep_uplink_priv *uplink_priv; + struct mlx5e_rep_priv *uplink_rpriv; + + if (esw_offloads_mode(esw)) { + uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH); + uplink_priv = &uplink_rpriv->uplink_priv; + + return uplink_priv->ct_priv; + } + + return priv->fs.tc.ct; +} + +struct mlx5_flow_handle * +mlx5_tc_rule_insert(struct mlx5e_priv *priv, + struct mlx5_flow_spec *spec, + struct mlx5_flow_attr *attr) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + + if (esw_offloads_mode(esw)) + return mlx5_eswitch_add_offloaded_rule(esw, spec, attr); + + return mlx5e_add_offloaded_nic_rule(priv, spec, attr); +} + +void +mlx5_tc_rule_delete(struct mlx5e_priv *priv, + struct mlx5_flow_handle *rule, + struct mlx5_flow_attr *attr) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + + if (esw_offloads_mode(esw)) { + mlx5_eswitch_del_offloaded_rule(esw, rule, attr); + + return; + } + + mlx5e_del_offloaded_nic_rule(priv, rule, attr); +} + struct mlx5e_hairpin { struct mlx5_hairpin *pair; @@ -370,7 +414,7 @@ static bool __flow_flag_test(struct mlx5e_tc_flow *flow, unsigned long flag) #define flow_flag_test(flow, flag) __flow_flag_test(flow, \ MLX5E_TC_FLOW_FLAG_##flag) -static bool mlx5e_is_eswitch_flow(struct mlx5e_tc_flow *flow) +bool mlx5e_is_eswitch_flow(struct mlx5e_tc_flow *flow) { return flow_flag_test(flow, ESWITCH); } @@ -415,10 +459,7 @@ static int mlx5e_attach_mod_hdr(struct mlx5e_priv *priv, return PTR_ERR(mh); modify_hdr = mlx5e_mod_hdr_get(mh); - if (mlx5e_is_eswitch_flow(flow)) - flow->esw_attr->modify_hdr = modify_hdr; - else - flow->nic_attr->modify_hdr = modify_hdr; + flow->attr->modify_hdr = modify_hdr; flow->mh = mh; return 0; @@ -858,9 +899,9 @@ static int mlx5e_hairpin_flow_add(struct mlx5e_priv *priv, attach_flow: if (hpe->hp->num_channels > 1) { flow_flag_set(flow, HAIRPIN_RSS); - flow->nic_attr->hairpin_ft = hpe->hp->ttc.ft.t; + flow->attr->nic_attr->hairpin_ft = hpe->hp->ttc.ft.t; } else { - flow->nic_attr->hairpin_tirn = hpe->hp->tirn; + flow->attr->nic_attr->hairpin_tirn = hpe->hp->tirn; } flow->hpe = hpe; @@ -890,129 +931,212 @@ static void mlx5e_hairpin_flow_del(struct mlx5e_priv *priv, flow->hpe = NULL; } -static int -mlx5e_tc_add_nic_flow(struct mlx5e_priv *priv, - struct mlx5e_tc_flow_parse_attr *parse_attr, - struct mlx5e_tc_flow *flow, - struct netlink_ext_ack *extack) +struct mlx5_flow_handle * +mlx5e_add_offloaded_nic_rule(struct mlx5e_priv *priv, + struct mlx5_flow_spec *spec, + struct mlx5_flow_attr *attr) { - struct mlx5_flow_context *flow_context = &parse_attr->spec.flow_context; - struct mlx5_nic_flow_attr *attr = flow->nic_attr; - struct mlx5_core_dev *dev = priv->mdev; + struct mlx5_flow_context *flow_context = &spec->flow_context; + struct mlx5_fs_chains *nic_chains = nic_chains(priv); + struct mlx5_nic_flow_attr *nic_attr = attr->nic_attr; + struct mlx5e_tc_table *tc = &priv->fs.tc; struct mlx5_flow_destination dest[2] = {}; struct mlx5_flow_act flow_act = { .action = attr->action, .flags = FLOW_ACT_NO_APPEND, }; - struct mlx5_fc *counter = NULL; - int err, dest_ix = 0; + struct mlx5_flow_handle *rule; + struct mlx5_flow_table *ft; + int dest_ix = 0; flow_context->flags |= FLOW_CONTEXT_HAS_TAG; - flow_context->flow_tag = attr->flow_tag; - - if (flow_flag_test(flow, HAIRPIN)) { - err = mlx5e_hairpin_flow_add(priv, flow, parse_attr, extack); - if (err) - return err; + flow_context->flow_tag = nic_attr->flow_tag; - if (flow_flag_test(flow, HAIRPIN_RSS)) { - dest[dest_ix].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; - dest[dest_ix].ft = attr->hairpin_ft; - } else { - dest[dest_ix].type = MLX5_FLOW_DESTINATION_TYPE_TIR; - dest[dest_ix].tir_num = attr->hairpin_tirn; - } + if (attr->dest_ft) { + dest[dest_ix].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; + dest[dest_ix].ft = attr->dest_ft; + dest_ix++; + } else if (nic_attr->hairpin_ft) { + dest[dest_ix].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; + dest[dest_ix].ft = nic_attr->hairpin_ft; + dest_ix++; + } else if (nic_attr->hairpin_tirn) { + dest[dest_ix].type = MLX5_FLOW_DESTINATION_TYPE_TIR; + dest[dest_ix].tir_num = nic_attr->hairpin_tirn; dest_ix++; } else if (attr->action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) { dest[dest_ix].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; - dest[dest_ix].ft = priv->fs.vlan.ft.t; + if (attr->dest_chain) { + dest[dest_ix].ft = mlx5_chains_get_table(nic_chains, + attr->dest_chain, 1, + MLX5E_TC_FT_LEVEL); + if (IS_ERR(dest[dest_ix].ft)) + return ERR_CAST(dest[dest_ix].ft); + } else { + dest[dest_ix].ft = priv->fs.vlan.ft.t; + } + dest_ix++; + } + + if (dest[0].type == MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE && + MLX5_CAP_FLOWTABLE_NIC_RX(priv->mdev, ignore_flow_level)) + flow_act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL; + + if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_COUNT) { + dest[dest_ix].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; + dest[dest_ix].counter_id = mlx5_fc_id(attr->counter); dest_ix++; } + if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) + flow_act.modify_hdr = attr->modify_hdr; + + mutex_lock(&tc->t_lock); + if (IS_ERR_OR_NULL(tc->t)) { + /* Create the root table here if doesn't exist yet */ + tc->t = + mlx5_chains_get_table(nic_chains, 0, 1, MLX5E_TC_FT_LEVEL); + + if (IS_ERR(tc->t)) { + mutex_unlock(&tc->t_lock); + netdev_err(priv->netdev, + "Failed to create tc offload table\n"); + rule = ERR_CAST(priv->fs.tc.t); + goto err_ft_get; + } + } + mutex_unlock(&tc->t_lock); + + if (attr->chain || attr->prio) + ft = mlx5_chains_get_table(nic_chains, + attr->chain, attr->prio, + MLX5E_TC_FT_LEVEL); + else + ft = attr->ft; + + if (IS_ERR(ft)) { + rule = ERR_CAST(ft); + goto err_ft_get; + } + + if (attr->outer_match_level != MLX5_MATCH_NONE) + spec->match_criteria_enable |= MLX5_MATCH_OUTER_HEADERS; + + rule = mlx5_add_flow_rules(ft, spec, + &flow_act, dest, dest_ix); + if (IS_ERR(rule)) + goto err_rule; + + return rule; + +err_rule: + if (attr->chain || attr->prio) + mlx5_chains_put_table(nic_chains, + attr->chain, attr->prio, + MLX5E_TC_FT_LEVEL); +err_ft_get: + if (attr->dest_chain) + mlx5_chains_put_table(nic_chains, + attr->dest_chain, 1, + MLX5E_TC_FT_LEVEL); + + return ERR_CAST(rule); +} + +static int +mlx5e_tc_add_nic_flow(struct mlx5e_priv *priv, + struct mlx5e_tc_flow_parse_attr *parse_attr, + struct mlx5e_tc_flow *flow, + struct netlink_ext_ack *extack) +{ + struct mlx5_flow_attr *attr = flow->attr; + struct mlx5_core_dev *dev = priv->mdev; + struct mlx5_fc *counter = NULL; + int err; + + if (flow_flag_test(flow, HAIRPIN)) { + err = mlx5e_hairpin_flow_add(priv, flow, parse_attr, extack); + if (err) + return err; + } + if (attr->action & MLX5_FLOW_CONTEXT_ACTION_COUNT) { counter = mlx5_fc_create(dev, true); if (IS_ERR(counter)) return PTR_ERR(counter); - dest[dest_ix].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; - dest[dest_ix].counter_id = mlx5_fc_id(counter); - dest_ix++; attr->counter = counter; } if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) { err = mlx5e_attach_mod_hdr(priv, flow, parse_attr); - flow_act.modify_hdr = attr->modify_hdr; dealloc_mod_hdr_actions(&parse_attr->mod_hdr_acts); if (err) return err; } - mutex_lock(&priv->fs.tc.t_lock); - if (IS_ERR_OR_NULL(priv->fs.tc.t)) { - struct mlx5_flow_table_attr ft_attr = {}; - int tc_grp_size, tc_tbl_size, tc_num_grps; - u32 max_flow_counter; - - max_flow_counter = (MLX5_CAP_GEN(dev, max_flow_counter_31_16) << 16) | - MLX5_CAP_GEN(dev, max_flow_counter_15_0); - - tc_grp_size = min_t(int, max_flow_counter, MLX5E_TC_TABLE_MAX_GROUP_SIZE); - - tc_tbl_size = min_t(int, tc_grp_size * MLX5E_TC_TABLE_NUM_GROUPS, - BIT(MLX5_CAP_FLOWTABLE_NIC_RX(dev, log_max_ft_size))); - tc_num_grps = MLX5E_TC_TABLE_NUM_GROUPS; - - ft_attr.prio = MLX5E_TC_PRIO; - ft_attr.max_fte = tc_tbl_size; - ft_attr.level = MLX5E_TC_FT_LEVEL; - ft_attr.autogroup.max_num_groups = tc_num_grps; - priv->fs.tc.t = - mlx5_create_auto_grouped_flow_table(priv->fs.ns, - &ft_attr); - if (IS_ERR(priv->fs.tc.t)) { - mutex_unlock(&priv->fs.tc.t_lock); - NL_SET_ERR_MSG_MOD(extack, - "Failed to create tc offload table"); - netdev_err(priv->netdev, - "Failed to create tc offload table\n"); - return PTR_ERR(priv->fs.tc.t); - } - } + if (flow_flag_test(flow, CT)) + flow->rule[0] = mlx5_tc_ct_flow_offload(get_ct_priv(priv), flow, &parse_attr->spec, + attr, &parse_attr->mod_hdr_acts); + else + flow->rule[0] = mlx5e_add_offloaded_nic_rule(priv, &parse_attr->spec, + attr); - if (attr->match_level != MLX5_MATCH_NONE) - parse_attr->spec.match_criteria_enable |= MLX5_MATCH_OUTER_HEADERS; + return PTR_ERR_OR_ZERO(flow->rule[0]); +} - flow->rule[0] = mlx5_add_flow_rules(priv->fs.tc.t, &parse_attr->spec, - &flow_act, dest, dest_ix); - mutex_unlock(&priv->fs.tc.t_lock); +void mlx5e_del_offloaded_nic_rule(struct mlx5e_priv *priv, + struct mlx5_flow_handle *rule, + struct mlx5_flow_attr *attr) +{ + struct mlx5_fs_chains *nic_chains = nic_chains(priv); - return PTR_ERR_OR_ZERO(flow->rule[0]); + mlx5_del_flow_rules(rule); + + if (attr->chain || attr->prio) + mlx5_chains_put_table(nic_chains, attr->chain, attr->prio, + MLX5E_TC_FT_LEVEL); + + if (attr->dest_chain) + mlx5_chains_put_table(nic_chains, attr->dest_chain, 1, + MLX5E_TC_FT_LEVEL); } static void mlx5e_tc_del_nic_flow(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow) { - struct mlx5_nic_flow_attr *attr = flow->nic_attr; - struct mlx5_fc *counter = NULL; + struct mlx5_flow_attr *attr = flow->attr; + struct mlx5e_tc_table *tc = &priv->fs.tc; + + flow_flag_clear(flow, OFFLOADED); - counter = attr->counter; - if (!IS_ERR_OR_NULL(flow->rule[0])) - mlx5_del_flow_rules(flow->rule[0]); - mlx5_fc_destroy(priv->mdev, counter); + if (flow_flag_test(flow, CT)) + mlx5_tc_ct_delete_flow(get_ct_priv(flow->priv), flow, attr); + else if (!IS_ERR_OR_NULL(flow->rule[0])) + mlx5e_del_offloaded_nic_rule(priv, flow->rule[0], attr); + /* Remove root table if no rules are left to avoid + * extra steering hops. + */ mutex_lock(&priv->fs.tc.t_lock); - if (!mlx5e_tc_num_filters(priv, MLX5_TC_FLAG(NIC_OFFLOAD)) && priv->fs.tc.t) { - mlx5_destroy_flow_table(priv->fs.tc.t); + if (!mlx5e_tc_num_filters(priv, MLX5_TC_FLAG(NIC_OFFLOAD)) && + !IS_ERR_OR_NULL(tc->t)) { + mlx5_chains_put_table(nic_chains(priv), 0, 1, MLX5E_TC_FT_LEVEL); priv->fs.tc.t = NULL; } mutex_unlock(&priv->fs.tc.t_lock); + kvfree(attr->parse_attr); + if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) mlx5e_detach_mod_hdr(priv, flow); + mlx5_fc_destroy(priv->mdev, attr->counter); + if (flow_flag_test(flow, HAIRPIN)) mlx5e_hairpin_flow_del(priv, flow); + + kfree(flow->attr); } static void mlx5e_detach_encap(struct mlx5e_priv *priv, @@ -1035,7 +1159,7 @@ static struct mlx5_flow_handle * mlx5e_tc_offload_fdb_rules(struct mlx5_eswitch *esw, struct mlx5e_tc_flow *flow, struct mlx5_flow_spec *spec, - struct mlx5_esw_flow_attr *attr) + struct mlx5_flow_attr *attr) { struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts; struct mlx5_flow_handle *rule; @@ -1043,7 +1167,8 @@ mlx5e_tc_offload_fdb_rules(struct mlx5_eswitch *esw, if (flow_flag_test(flow, CT)) { mod_hdr_acts = &attr->parse_attr->mod_hdr_acts; - return mlx5_tc_ct_flow_offload(flow->priv, flow, spec, attr, + return mlx5_tc_ct_flow_offload(get_ct_priv(flow->priv), + flow, spec, attr, mod_hdr_acts); } @@ -1051,7 +1176,7 @@ mlx5e_tc_offload_fdb_rules(struct mlx5_eswitch *esw, if (IS_ERR(rule)) return rule; - if (attr->split_count) { + if (attr->esw_attr->split_count) { flow->rule[1] = mlx5_eswitch_add_fwd_rule(esw, spec, attr); if (IS_ERR(flow->rule[1])) { mlx5_eswitch_del_offloaded_rule(esw, rule, attr); @@ -1065,16 +1190,16 @@ mlx5e_tc_offload_fdb_rules(struct mlx5_eswitch *esw, static void mlx5e_tc_unoffload_fdb_rules(struct mlx5_eswitch *esw, struct mlx5e_tc_flow *flow, - struct mlx5_esw_flow_attr *attr) + struct mlx5_flow_attr *attr) { flow_flag_clear(flow, OFFLOADED); if (flow_flag_test(flow, CT)) { - mlx5_tc_ct_delete_flow(flow->priv, flow, attr); + mlx5_tc_ct_delete_flow(get_ct_priv(flow->priv), flow, attr); return; } - if (attr->split_count) + if (attr->esw_attr->split_count) mlx5_eswitch_del_fwd_rule(esw, flow->rule[1], attr); mlx5_eswitch_del_offloaded_rule(esw, flow->rule[0], attr); @@ -1085,18 +1210,24 @@ mlx5e_tc_offload_to_slow_path(struct mlx5_eswitch *esw, struct mlx5e_tc_flow *flow, struct mlx5_flow_spec *spec) { - struct mlx5_esw_flow_attr slow_attr; + struct mlx5_flow_attr *slow_attr; struct mlx5_flow_handle *rule; - memcpy(&slow_attr, flow->esw_attr, sizeof(slow_attr)); - slow_attr.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; - slow_attr.split_count = 0; - slow_attr.flags |= MLX5_ESW_ATTR_FLAG_SLOW_PATH; + slow_attr = mlx5_alloc_flow_attr(MLX5_FLOW_NAMESPACE_FDB); + if (!slow_attr) + return ERR_PTR(-ENOMEM); + + memcpy(slow_attr, flow->attr, ESW_FLOW_ATTR_SZ); + slow_attr->action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + slow_attr->esw_attr->split_count = 0; + slow_attr->flags |= MLX5_ESW_ATTR_FLAG_SLOW_PATH; - rule = mlx5e_tc_offload_fdb_rules(esw, flow, spec, &slow_attr); + rule = mlx5e_tc_offload_fdb_rules(esw, flow, spec, slow_attr); if (!IS_ERR(rule)) flow_flag_set(flow, SLOW); + kfree(slow_attr); + return rule; } @@ -1104,14 +1235,21 @@ static void mlx5e_tc_unoffload_from_slow_path(struct mlx5_eswitch *esw, struct mlx5e_tc_flow *flow) { - struct mlx5_esw_flow_attr slow_attr; + struct mlx5_flow_attr *slow_attr; - memcpy(&slow_attr, flow->esw_attr, sizeof(slow_attr)); - slow_attr.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; - slow_attr.split_count = 0; - slow_attr.flags |= MLX5_ESW_ATTR_FLAG_SLOW_PATH; - mlx5e_tc_unoffload_fdb_rules(esw, flow, &slow_attr); + slow_attr = mlx5_alloc_flow_attr(MLX5_FLOW_NAMESPACE_FDB); + if (!slow_attr) { + mlx5_core_warn(flow->priv->mdev, "Unable to alloc attr to unoffload slow path rule\n"); + return; + } + + memcpy(slow_attr, flow->attr, ESW_FLOW_ATTR_SZ); + slow_attr->action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + slow_attr->esw_attr->split_count = 0; + slow_attr->flags |= MLX5_ESW_ATTR_FLAG_SLOW_PATH; + mlx5e_tc_unoffload_fdb_rules(esw, flow, slow_attr); flow_flag_clear(flow, SLOW); + kfree(slow_attr); } /* Caller must obtain uplink_priv->unready_flows_lock mutex before calling this @@ -1169,9 +1307,10 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, struct netlink_ext_ack *extack) { struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; - struct mlx5_esw_flow_attr *attr = flow->esw_attr; - struct mlx5e_tc_flow_parse_attr *parse_attr = attr->parse_attr; struct net_device *out_dev, *encap_dev = NULL; + struct mlx5e_tc_flow_parse_attr *parse_attr; + struct mlx5_flow_attr *attr = flow->attr; + struct mlx5_esw_flow_attr *esw_attr; struct mlx5_fc *counter = NULL; struct mlx5e_rep_priv *rpriv; struct mlx5e_priv *out_priv; @@ -1180,7 +1319,7 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, int err = 0; int out_index; - if (!mlx5_esw_chains_prios_supported(esw) && attr->prio != 1) { + if (!mlx5_chains_prios_supported(esw_chains(esw)) && attr->prio != 1) { NL_SET_ERR_MSG_MOD(extack, "E-switch priorities unsupported, upgrade FW"); return -EOPNOTSUPP; @@ -1191,14 +1330,14 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, * FDB_FT_CHAIN which is outside tc range. * See mlx5e_rep_setup_ft_cb(). */ - max_chain = mlx5_esw_chains_get_chain_range(esw); + max_chain = mlx5_chains_get_chain_range(esw_chains(esw)); if (!mlx5e_is_ft_flow(flow) && attr->chain > max_chain) { NL_SET_ERR_MSG_MOD(extack, "Requested chain is out of supported range"); return -EOPNOTSUPP; } - max_prio = mlx5_esw_chains_get_prio_range(esw); + max_prio = mlx5_chains_get_prio_range(esw_chains(esw)); if (attr->prio > max_prio) { NL_SET_ERR_MSG_MOD(extack, "Requested priority is out of supported range"); @@ -1211,10 +1350,13 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, return err; } + parse_attr = attr->parse_attr; + esw_attr = attr->esw_attr; + for (out_index = 0; out_index < MLX5_MAX_FLOW_FWD_VPORTS; out_index++) { int mirred_ifindex; - if (!(attr->dests[out_index].flags & MLX5_ESW_DEST_ENCAP)) + if (!(esw_attr->dests[out_index].flags & MLX5_ESW_DEST_ENCAP)) continue; mirred_ifindex = parse_attr->mirred_ifindex[out_index]; @@ -1227,8 +1369,8 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, out_priv = netdev_priv(encap_dev); rpriv = out_priv->ppriv; - attr->dests[out_index].rep = rpriv->rep; - attr->dests[out_index].mdev = out_priv->mdev; + esw_attr->dests[out_index].rep = rpriv->rep; + esw_attr->dests[out_index].mdev = out_priv->mdev; } err = mlx5_eswitch_add_vlan_action(esw, attr); @@ -1244,7 +1386,7 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, } if (attr->action & MLX5_FLOW_CONTEXT_ACTION_COUNT) { - counter = mlx5_fc_create(attr->counter_dev, true); + counter = mlx5_fc_create(esw_attr->counter_dev, true); if (IS_ERR(counter)) return PTR_ERR(counter); @@ -1270,7 +1412,7 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, static bool mlx5_flow_has_geneve_opt(struct mlx5e_tc_flow *flow) { - struct mlx5_flow_spec *spec = &flow->esw_attr->parse_attr->spec; + struct mlx5_flow_spec *spec = &flow->attr->parse_attr->spec; void *headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters_3); @@ -1285,7 +1427,7 @@ static void mlx5e_tc_del_fdb_flow(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow) { struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; - struct mlx5_esw_flow_attr *attr = flow->esw_attr; + struct mlx5_flow_attr *attr = flow->attr; int out_index; mlx5e_put_flow_tunnel_id(flow); @@ -1306,22 +1448,24 @@ static void mlx5e_tc_del_fdb_flow(struct mlx5e_priv *priv, mlx5_eswitch_del_vlan_action(esw, attr); for (out_index = 0; out_index < MLX5_MAX_FLOW_FWD_VPORTS; out_index++) - if (attr->dests[out_index].flags & MLX5_ESW_DEST_ENCAP) { + if (attr->esw_attr->dests[out_index].flags & MLX5_ESW_DEST_ENCAP) { mlx5e_detach_encap(priv, flow, out_index); kfree(attr->parse_attr->tun_info[out_index]); } kvfree(attr->parse_attr); - mlx5_tc_ct_match_del(priv, &flow->esw_attr->ct_attr); + mlx5_tc_ct_match_del(get_ct_priv(priv), &flow->attr->ct_attr); if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) mlx5e_detach_mod_hdr(priv, flow); if (attr->action & MLX5_FLOW_CONTEXT_ACTION_COUNT) - mlx5_fc_destroy(attr->counter_dev, attr->counter); + mlx5_fc_destroy(attr->esw_attr->counter_dev, attr->counter); if (flow_flag_test(flow, L3_TO_L2_DECAP)) mlx5e_detach_decap(priv, flow); + + kfree(flow->attr); } void mlx5e_tc_encap_flows_add(struct mlx5e_priv *priv, @@ -1331,6 +1475,7 @@ void mlx5e_tc_encap_flows_add(struct mlx5e_priv *priv, struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; struct mlx5_esw_flow_attr *esw_attr; struct mlx5_flow_handle *rule; + struct mlx5_flow_attr *attr; struct mlx5_flow_spec *spec; struct mlx5e_tc_flow *flow; int err; @@ -1353,8 +1498,9 @@ void mlx5e_tc_encap_flows_add(struct mlx5e_priv *priv, if (!mlx5e_is_offloaded_flow(flow)) continue; - esw_attr = flow->esw_attr; - spec = &esw_attr->parse_attr->spec; + attr = flow->attr; + esw_attr = attr->esw_attr; + spec = &attr->parse_attr->spec; esw_attr->dests[flow->tmp_efi_index].pkt_reformat = e->pkt_reformat; esw_attr->dests[flow->tmp_efi_index].flags |= MLX5_ESW_DEST_ENCAP_VALID; @@ -1374,7 +1520,7 @@ void mlx5e_tc_encap_flows_add(struct mlx5e_priv *priv, if (!all_flow_encaps_valid) continue; /* update from slow path rule to encap rule */ - rule = mlx5e_tc_offload_fdb_rules(esw, flow, spec, esw_attr); + rule = mlx5e_tc_offload_fdb_rules(esw, flow, spec, attr); if (IS_ERR(rule)) { err = PTR_ERR(rule); mlx5_core_warn(priv->mdev, "Failed to update cached encapsulation flow, %d\n", @@ -1394,7 +1540,9 @@ void mlx5e_tc_encap_flows_del(struct mlx5e_priv *priv, struct list_head *flow_list) { struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5_esw_flow_attr *esw_attr; struct mlx5_flow_handle *rule; + struct mlx5_flow_attr *attr; struct mlx5_flow_spec *spec; struct mlx5e_tc_flow *flow; int err; @@ -1402,12 +1550,14 @@ void mlx5e_tc_encap_flows_del(struct mlx5e_priv *priv, list_for_each_entry(flow, flow_list, tmp_list) { if (!mlx5e_is_offloaded_flow(flow)) continue; - spec = &flow->esw_attr->parse_attr->spec; + attr = flow->attr; + esw_attr = attr->esw_attr; + spec = &attr->parse_attr->spec; /* update from encap rule to slow path rule */ rule = mlx5e_tc_offload_to_slow_path(esw, flow, spec); /* mark the flow's encap dest as non-valid */ - flow->esw_attr->dests[flow->tmp_efi_index].flags &= ~MLX5_ESW_DEST_ENCAP_VALID; + esw_attr->dests[flow->tmp_efi_index].flags &= ~MLX5_ESW_DEST_ENCAP_VALID; if (IS_ERR(rule)) { err = PTR_ERR(rule); @@ -1416,7 +1566,7 @@ void mlx5e_tc_encap_flows_del(struct mlx5e_priv *priv, continue; } - mlx5e_tc_unoffload_fdb_rules(esw, flow, flow->esw_attr); + mlx5e_tc_unoffload_fdb_rules(esw, flow, attr); flow->rule[0] = rule; /* was unset when fast path rule removed */ flow_flag_set(flow, OFFLOADED); @@ -1429,10 +1579,7 @@ void mlx5e_tc_encap_flows_del(struct mlx5e_priv *priv, static struct mlx5_fc *mlx5e_tc_get_counter(struct mlx5e_tc_flow *flow) { - if (mlx5e_is_eswitch_flow(flow)) - return flow->esw_attr->counter; - else - return flow->nic_attr->counter; + return flow->attr->counter; } /* Takes reference to all flows attached to encap and adds the flows to @@ -1798,11 +1945,11 @@ static int mlx5e_get_flow_tunnel_id(struct mlx5e_priv *priv, { struct flow_rule *rule = flow_cls_offload_flow_rule(f); struct netlink_ext_ack *extack = f->common.extack; - struct mlx5_esw_flow_attr *attr = flow->esw_attr; struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts; struct flow_match_enc_opts enc_opts_match; struct tunnel_match_enc_opts tun_enc_opts; struct mlx5_rep_uplink_priv *uplink_priv; + struct mlx5_flow_attr *attr = flow->attr; struct mlx5e_rep_priv *uplink_rpriv; struct tunnel_match_key tunnel_key; bool enc_opts_is_dont_care = true; @@ -1866,7 +2013,7 @@ static int mlx5e_get_flow_tunnel_id(struct mlx5e_priv *priv, } else { mod_hdr_acts = &attr->parse_attr->mod_hdr_acts; err = mlx5e_tc_match_to_reg_set(priv->mdev, - mod_hdr_acts, + mod_hdr_acts, MLX5_FLOW_NAMESPACE_FDB, TUNNEL_TO_REG, value); if (err) goto err_set; @@ -1952,8 +2099,8 @@ static int parse_tunnel_attr(struct mlx5e_priv *priv, if (!mlx5e_is_eswitch_flow(flow)) return -EOPNOTSUPP; - needs_mapping = !!flow->esw_attr->chain; - sets_mapping = !flow->esw_attr->chain && flow_has_tc_fwd_action(f); + needs_mapping = !!flow->attr->chain; + sets_mapping = !flow->attr->chain && flow_has_tc_fwd_action(f); *match_inner = !needs_mapping; if ((needs_mapping || sets_mapping) && @@ -1965,7 +2112,7 @@ static int parse_tunnel_attr(struct mlx5e_priv *priv, return -EOPNOTSUPP; } - if (!flow->esw_attr->chain) { + if (!flow->attr->chain) { err = mlx5e_tc_tun_parse(filter_dev, priv, spec, f, match_level); if (err) { @@ -1980,7 +2127,7 @@ static int parse_tunnel_attr(struct mlx5e_priv *priv, * object */ if (!netif_is_bareudp(filter_dev)) - flow->esw_attr->action |= MLX5_FLOW_CONTEXT_ACTION_DECAP; + flow->attr->action |= MLX5_FLOW_CONTEXT_ACTION_DECAP; } if (!needs_mapping && !sets_mapping) @@ -2483,12 +2630,9 @@ static int parse_cls_flower(struct mlx5e_priv *priv, } } - if (is_eswitch_flow) { - flow->esw_attr->inner_match_level = inner_match_level; - flow->esw_attr->outer_match_level = outer_match_level; - } else { - flow->nic_attr->match_level = non_tunnel_match_level; - } + flow->attr->inner_match_level = inner_match_level; + flow->attr->outer_match_level = outer_match_level; + return err; } @@ -2614,6 +2758,7 @@ static struct mlx5_fields fields[] = { OFFLOAD(DIPV6_31_0, 32, U32_MAX, ip6.daddr.s6_addr32[3], 0, dst_ipv4_dst_ipv6.ipv6_layout.ipv6[12]), OFFLOAD(IPV6_HOPLIMIT, 8, U8_MAX, ip6.hop_limit, 0, ttl_hoplimit), + OFFLOAD(IP_DSCP, 16, 0xc00f, ip6, 0, ip_dscp), OFFLOAD(TCP_SPORT, 16, U16_MAX, tcp.source, 0, tcp_sport), OFFLOAD(TCP_DPORT, 16, U16_MAX, tcp.dest, 0, tcp_dport), @@ -3090,7 +3235,7 @@ static bool modify_header_match_supported(struct mlx5e_priv *priv, * we can't restore ct state */ if (!ct_clear && modify_tuple && - mlx5_tc_ct_add_no_trk_match(priv, spec)) { + mlx5_tc_ct_add_no_trk_match(spec)) { NL_SET_ERR_MSG_MOD(extack, "can't offload tuple modify header with ct matches"); netdev_info(priv->netdev, @@ -3121,12 +3266,13 @@ static bool actions_match_supported(struct mlx5e_priv *priv, bool ct_flow = false, ct_clear = false; u32 actions; + ct_clear = flow->attr->ct_attr.ct_action & + TCA_CT_ACT_CLEAR; + ct_flow = flow_flag_test(flow, CT) && !ct_clear; + actions = flow->attr->action; + if (mlx5e_is_eswitch_flow(flow)) { - actions = flow->esw_attr->action; - ct_clear = flow->esw_attr->ct_attr.ct_action & - TCA_CT_ACT_CLEAR; - ct_flow = flow_flag_test(flow, CT) && !ct_clear; - if (flow->esw_attr->split_count && ct_flow) { + if (flow->attr->esw_attr->split_count && ct_flow) { /* All registers used by ct are cleared when using * split rules. */ @@ -3134,8 +3280,6 @@ static bool actions_match_supported(struct mlx5e_priv *priv, "Can't offload mirroring with action ct"); return false; } - } else { - actions = flow->nic_attr->action; } if (actions & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) @@ -3233,15 +3377,67 @@ add_vlan_prio_tag_rewrite_action(struct mlx5e_priv *priv, extack); } +static int validate_goto_chain(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + const struct flow_action_entry *act, + u32 actions, + struct netlink_ext_ack *extack) +{ + bool is_esw = mlx5e_is_eswitch_flow(flow); + struct mlx5_flow_attr *attr = flow->attr; + bool ft_flow = mlx5e_is_ft_flow(flow); + u32 dest_chain = act->chain_index; + struct mlx5_fs_chains *chains; + struct mlx5_eswitch *esw; + u32 reformat_and_fwd; + u32 max_chain; + + esw = priv->mdev->priv.eswitch; + chains = is_esw ? esw_chains(esw) : nic_chains(priv); + max_chain = mlx5_chains_get_chain_range(chains); + reformat_and_fwd = is_esw ? + MLX5_CAP_ESW_FLOWTABLE_FDB(priv->mdev, reformat_and_fwd_to_table) : + MLX5_CAP_FLOWTABLE_NIC_RX(priv->mdev, reformat_and_fwd_to_table); + + if (ft_flow) { + NL_SET_ERR_MSG_MOD(extack, "Goto action is not supported"); + return -EOPNOTSUPP; + } + + if (!mlx5_chains_backwards_supported(chains) && + dest_chain <= attr->chain) { + NL_SET_ERR_MSG_MOD(extack, + "Goto lower numbered chain isn't supported"); + return -EOPNOTSUPP; + } + + if (dest_chain > max_chain) { + NL_SET_ERR_MSG_MOD(extack, + "Requested destination chain is out of supported range"); + return -EOPNOTSUPP; + } + + if (actions & (MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT | + MLX5_FLOW_CONTEXT_ACTION_DECAP) && + !reformat_and_fwd) { + NL_SET_ERR_MSG_MOD(extack, + "Goto chain is not allowed if action has reformat or decap"); + return -EOPNOTSUPP; + } + + return 0; +} + static int parse_tc_nic_actions(struct mlx5e_priv *priv, struct flow_action *flow_action, struct mlx5e_tc_flow_parse_attr *parse_attr, struct mlx5e_tc_flow *flow, struct netlink_ext_ack *extack) { - struct mlx5_nic_flow_attr *attr = flow->nic_attr; + struct mlx5_flow_attr *attr = flow->attr; struct pedit_headers_action hdrs[2] = {}; const struct flow_action_entry *act; + struct mlx5_nic_flow_attr *nic_attr; u32 action = 0; int err, i; @@ -3252,7 +3448,9 @@ static int parse_tc_nic_actions(struct mlx5e_priv *priv, FLOW_ACTION_HW_STATS_DELAYED_BIT)) return -EOPNOTSUPP; - attr->flow_tag = MLX5_FS_DEFAULT_FLOW_TAG; + nic_attr = attr->nic_attr; + + nic_attr->flow_tag = MLX5_FS_DEFAULT_FLOW_TAG; flow_action_for_each(i, act, flow_action) { switch (act->id) { @@ -3273,8 +3471,7 @@ static int parse_tc_nic_actions(struct mlx5e_priv *priv, if (err) return err; - action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR | - MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; break; case FLOW_ACTION_VLAN_MANGLE: err = add_vlan_rewrite_action(priv, @@ -3319,10 +3516,26 @@ static int parse_tc_nic_actions(struct mlx5e_priv *priv, return -EINVAL; } - attr->flow_tag = mark; + nic_attr->flow_tag = mark; action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; } break; + case FLOW_ACTION_GOTO: + err = validate_goto_chain(priv, flow, act, action, + extack); + if (err) + return err; + + action |= MLX5_FLOW_CONTEXT_ACTION_COUNT; + attr->dest_chain = act->chain_index; + break; + case FLOW_ACTION_CT: + err = mlx5_tc_ct_parse_action(get_ct_priv(priv), attr, act, extack); + if (err) + return err; + + flow_flag_set(flow, CT); + break; default: NL_SET_ERR_MSG_MOD(extack, "The offload action is not supported"); return -EOPNOTSUPP; @@ -3345,6 +3558,18 @@ static int parse_tc_nic_actions(struct mlx5e_priv *priv, } attr->action = action; + + if (attr->dest_chain) { + if (attr->action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) { + NL_SET_ERR_MSG(extack, "Mirroring goto chain rules isn't supported"); + return -EOPNOTSUPP; + } + attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + } + + if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) + attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + if (!actions_match_supported(priv, flow_action, parse_attr, flow, extack)) return -EOPNOTSUPP; @@ -3476,8 +3701,8 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv, bool *encap_valid) { struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; - struct mlx5_esw_flow_attr *attr = flow->esw_attr; struct mlx5e_tc_flow_parse_attr *parse_attr; + struct mlx5_flow_attr *attr = flow->attr; const struct ip_tunnel_info *tun_info; struct encap_key key; struct mlx5e_encap_entry *e; @@ -3563,8 +3788,8 @@ attach_flow: flow->encaps[out_index].index = out_index; *encap_dev = e->out_dev; if (e->flags & MLX5_ENCAP_ENTRY_VALID) { - attr->dests[out_index].pkt_reformat = e->pkt_reformat; - attr->dests[out_index].flags |= MLX5_ESW_DEST_ENCAP_VALID; + attr->esw_attr->dests[out_index].pkt_reformat = e->pkt_reformat; + attr->esw_attr->dests[out_index].flags |= MLX5_ESW_DEST_ENCAP_VALID; *encap_valid = true; } else { *encap_valid = false; @@ -3591,14 +3816,14 @@ static int mlx5e_attach_decap(struct mlx5e_priv *priv, struct netlink_ext_ack *extack) { struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; - struct mlx5_esw_flow_attr *attr = flow->esw_attr; + struct mlx5_esw_flow_attr *attr = flow->attr->esw_attr; struct mlx5e_tc_flow_parse_attr *parse_attr; struct mlx5e_decap_entry *d; struct mlx5e_decap_key key; uintptr_t hash_key; int err = 0; - parse_attr = attr->parse_attr; + parse_attr = flow->attr->parse_attr; if (sizeof(parse_attr->eth) > MLX5_CAP_ESW(priv->mdev, max_encap_header_size)) { NL_SET_ERR_MSG_MOD(extack, "encap header larger than max supported"); @@ -3740,7 +3965,7 @@ static struct net_device *get_fdb_out_dev(struct net_device *uplink_dev, } static int add_vlan_push_action(struct mlx5e_priv *priv, - struct mlx5_esw_flow_attr *attr, + struct mlx5_flow_attr *attr, struct net_device **out_dev, u32 *action) { @@ -3753,7 +3978,7 @@ static int add_vlan_push_action(struct mlx5e_priv *priv, }; int err; - err = parse_tc_vlan_action(priv, &vlan_act, attr, action); + err = parse_tc_vlan_action(priv, &vlan_act, attr->esw_attr, action); if (err) return err; @@ -3766,7 +3991,7 @@ static int add_vlan_push_action(struct mlx5e_priv *priv, } static int add_vlan_pop_action(struct mlx5e_priv *priv, - struct mlx5_esw_flow_attr *attr, + struct mlx5_flow_attr *attr, u32 *action) { struct flow_action_entry vlan_act = { @@ -3777,7 +4002,7 @@ static int add_vlan_pop_action(struct mlx5e_priv *priv, nest_level = attr->parse_attr->filter_dev->lower_level - priv->netdev->lower_level; while (nest_level--) { - err = parse_tc_vlan_action(priv, &vlan_act, attr, action); + err = parse_tc_vlan_action(priv, &vlan_act, attr->esw_attr, action); if (err) return err; } @@ -3838,59 +4063,20 @@ static bool is_duplicated_output_device(struct net_device *dev, return false; } -static int mlx5_validate_goto_chain(struct mlx5_eswitch *esw, - struct mlx5e_tc_flow *flow, - const struct flow_action_entry *act, - u32 actions, - struct netlink_ext_ack *extack) -{ - u32 max_chain = mlx5_esw_chains_get_chain_range(esw); - struct mlx5_esw_flow_attr *attr = flow->esw_attr; - bool ft_flow = mlx5e_is_ft_flow(flow); - u32 dest_chain = act->chain_index; - - if (ft_flow) { - NL_SET_ERR_MSG_MOD(extack, "Goto action is not supported"); - return -EOPNOTSUPP; - } - - if (!mlx5_esw_chains_backwards_supported(esw) && - dest_chain <= attr->chain) { - NL_SET_ERR_MSG_MOD(extack, - "Goto lower numbered chain isn't supported"); - return -EOPNOTSUPP; - } - if (dest_chain > max_chain) { - NL_SET_ERR_MSG_MOD(extack, - "Requested destination chain is out of supported range"); - return -EOPNOTSUPP; - } - - if (actions & (MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT | - MLX5_FLOW_CONTEXT_ACTION_DECAP) && - !MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, reformat_and_fwd_to_table)) { - NL_SET_ERR_MSG_MOD(extack, - "Goto chain is not allowed if action has reformat or decap"); - return -EOPNOTSUPP; - } - - return 0; -} - static int verify_uplink_forwarding(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow, struct net_device *out_dev, struct netlink_ext_ack *extack) { + struct mlx5_esw_flow_attr *attr = flow->attr->esw_attr; struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; - struct mlx5_esw_flow_attr *attr = flow->esw_attr; struct mlx5e_rep_priv *rep_priv; /* Forwarding non encapsulated traffic between * uplink ports is allowed only if * termination_table_raw_traffic cap is set. * - * Input vport was stored esw_attr->in_rep. + * Input vport was stored attr->in_rep. * In LAG case, *priv* is the private data of * uplink which may be not the input vport. */ @@ -3925,13 +4111,14 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, { struct pedit_headers_action hdrs[2] = {}; struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; - struct mlx5_esw_flow_attr *attr = flow->esw_attr; - struct mlx5e_tc_flow_parse_attr *parse_attr = attr->parse_attr; + struct mlx5e_tc_flow_parse_attr *parse_attr; struct mlx5e_rep_priv *rpriv = priv->ppriv; const struct ip_tunnel_info *info = NULL; + struct mlx5_flow_attr *attr = flow->attr; int ifindexes[MLX5_MAX_FLOW_FWD_VPORTS]; bool ft_flow = mlx5e_is_ft_flow(flow); const struct flow_action_entry *act; + struct mlx5_esw_flow_attr *esw_attr; bool encap = false, decap = false; u32 action = attr->action; int err, i, if_count = 0; @@ -3944,12 +4131,25 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, FLOW_ACTION_HW_STATS_DELAYED_BIT)) return -EOPNOTSUPP; + esw_attr = attr->esw_attr; + parse_attr = attr->parse_attr; + flow_action_for_each(i, act, flow_action) { switch (act->id) { case FLOW_ACTION_DROP: action |= MLX5_FLOW_CONTEXT_ACTION_DROP | MLX5_FLOW_CONTEXT_ACTION_COUNT; break; + case FLOW_ACTION_TRAP: + if (!flow_offload_has_one_action(flow_action)) { + NL_SET_ERR_MSG_MOD(extack, + "action trap is supported as a sole action only"); + return -EOPNOTSUPP; + } + action |= (MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | + MLX5_FLOW_CONTEXT_ACTION_COUNT); + attr->flags |= MLX5_ESW_ATTR_FLAG_SLOW_PATH; + break; case FLOW_ACTION_MPLS_PUSH: if (!MLX5_CAP_ESW_FLOWTABLE_FDB(priv->mdev, reformat_l2_to_l3_tunnel) || @@ -3990,7 +4190,7 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, if (!flow_flag_test(flow, L3_TO_L2_DECAP)) { action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; - attr->split_count = attr->out_count; + esw_attr->split_count = esw_attr->out_count; } break; case FLOW_ACTION_CSUM: @@ -4027,27 +4227,27 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, return -EOPNOTSUPP; } - if (attr->out_count >= MLX5_MAX_FLOW_FWD_VPORTS) { + if (esw_attr->out_count >= MLX5_MAX_FLOW_FWD_VPORTS) { NL_SET_ERR_MSG_MOD(extack, "can't support more output ports, can't offload forwarding"); netdev_warn(priv->netdev, "can't support more than %d output ports, can't offload forwarding\n", - attr->out_count); + esw_attr->out_count); return -EOPNOTSUPP; } action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | MLX5_FLOW_CONTEXT_ACTION_COUNT; if (encap) { - parse_attr->mirred_ifindex[attr->out_count] = + parse_attr->mirred_ifindex[esw_attr->out_count] = out_dev->ifindex; - parse_attr->tun_info[attr->out_count] = dup_tun_info(info); - if (!parse_attr->tun_info[attr->out_count]) + parse_attr->tun_info[esw_attr->out_count] = dup_tun_info(info); + if (!parse_attr->tun_info[esw_attr->out_count]) return -ENOMEM; encap = false; - attr->dests[attr->out_count].flags |= + esw_attr->dests[esw_attr->out_count].flags |= MLX5_ESW_DEST_ENCAP; - attr->out_count++; + esw_attr->out_count++; /* attr->dests[].rep is resolved when we * handle encap */ @@ -4096,9 +4296,9 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, out_priv = netdev_priv(out_dev); rpriv = out_priv->ppriv; - attr->dests[attr->out_count].rep = rpriv->rep; - attr->dests[attr->out_count].mdev = out_priv->mdev; - attr->out_count++; + esw_attr->dests[esw_attr->out_count].rep = rpriv->rep; + esw_attr->dests[esw_attr->out_count].mdev = out_priv->mdev; + esw_attr->out_count++; } else if (parse_attr->filter_dev != priv->netdev) { /* All mlx5 devices are called to configure * high level device filters. Therefore, the @@ -4136,12 +4336,12 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, act, parse_attr, hdrs, &action, extack); } else { - err = parse_tc_vlan_action(priv, act, attr, &action); + err = parse_tc_vlan_action(priv, act, esw_attr, &action); } if (err) return err; - attr->split_count = attr->out_count; + esw_attr->split_count = esw_attr->out_count; break; case FLOW_ACTION_VLAN_MANGLE: err = add_vlan_rewrite_action(priv, @@ -4151,14 +4351,14 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, if (err) return err; - attr->split_count = attr->out_count; + esw_attr->split_count = esw_attr->out_count; break; case FLOW_ACTION_TUNNEL_DECAP: decap = true; break; case FLOW_ACTION_GOTO: - err = mlx5_validate_goto_chain(esw, flow, act, action, - extack); + err = validate_goto_chain(priv, flow, act, action, + extack); if (err) return err; @@ -4166,7 +4366,7 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, attr->dest_chain = act->chain_index; break; case FLOW_ACTION_CT: - err = mlx5_tc_ct_parse_action(priv, attr, act, extack); + err = mlx5_tc_ct_parse_action(get_ct_priv(priv), attr, act, extack); if (err) return err; @@ -4205,7 +4405,7 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, dealloc_mod_hdr_actions(&parse_attr->mod_hdr_acts); if (!((action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP) || (action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH))) - attr->split_count = 0; + esw_attr->split_count = 0; } } @@ -4245,7 +4445,7 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, return -EOPNOTSUPP; } - if (attr->split_count > 0 && !mlx5_esw_has_fwd_fdb(priv->mdev)) { + if (esw_attr->split_count > 0 && !mlx5_esw_has_fwd_fdb(priv->mdev)) { NL_SET_ERR_MSG_MOD(extack, "current firmware doesn't support split rule for port mirroring"); netdev_warn_once(priv->netdev, "current firmware doesn't support split rule for port mirroring\n"); @@ -4296,25 +4496,37 @@ static struct rhashtable *get_tc_ht(struct mlx5e_priv *priv, static bool is_peer_flow_needed(struct mlx5e_tc_flow *flow) { - struct mlx5_esw_flow_attr *attr = flow->esw_attr; - bool is_rep_ingress = attr->in_rep->vport != MLX5_VPORT_UPLINK && + struct mlx5_esw_flow_attr *esw_attr = flow->attr->esw_attr; + struct mlx5_flow_attr *attr = flow->attr; + bool is_rep_ingress = esw_attr->in_rep->vport != MLX5_VPORT_UPLINK && flow_flag_test(flow, INGRESS); bool act_is_encap = !!(attr->action & MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT); - bool esw_paired = mlx5_devcom_is_paired(attr->in_mdev->priv.devcom, + bool esw_paired = mlx5_devcom_is_paired(esw_attr->in_mdev->priv.devcom, MLX5_DEVCOM_ESW_OFFLOADS); if (!esw_paired) return false; - if ((mlx5_lag_is_sriov(attr->in_mdev) || - mlx5_lag_is_multipath(attr->in_mdev)) && + if ((mlx5_lag_is_sriov(esw_attr->in_mdev) || + mlx5_lag_is_multipath(esw_attr->in_mdev)) && (is_rep_ingress || act_is_encap)) return true; return false; } +struct mlx5_flow_attr * +mlx5_alloc_flow_attr(enum mlx5_flow_namespace_type type) +{ + u32 ex_attr_size = (type == MLX5_FLOW_NAMESPACE_FDB) ? + sizeof(struct mlx5_esw_flow_attr) : + sizeof(struct mlx5_nic_flow_attr); + struct mlx5_flow_attr *attr; + + return kzalloc(sizeof(*attr) + ex_attr_size, GFP_KERNEL); +} + static int mlx5e_alloc_flow(struct mlx5e_priv *priv, int attr_size, struct flow_cls_offload *f, unsigned long flow_flags, @@ -4322,19 +4534,26 @@ mlx5e_alloc_flow(struct mlx5e_priv *priv, int attr_size, struct mlx5e_tc_flow **__flow) { struct mlx5e_tc_flow_parse_attr *parse_attr; + struct mlx5_flow_attr *attr; struct mlx5e_tc_flow *flow; - int out_index, err; + int err = -ENOMEM; + int out_index; - flow = kzalloc(sizeof(*flow) + attr_size, GFP_KERNEL); + flow = kzalloc(sizeof(*flow), GFP_KERNEL); parse_attr = kvzalloc(sizeof(*parse_attr), GFP_KERNEL); - if (!parse_attr || !flow) { - err = -ENOMEM; + if (!parse_attr || !flow) goto err_free; - } - flow->cookie = f->cookie; flow->flags = flow_flags; + flow->cookie = f->cookie; flow->priv = priv; + + attr = mlx5_alloc_flow_attr(get_flow_name_space(flow)); + if (!attr) + goto err_free; + + flow->attr = attr; + for (out_index = 0; out_index < MLX5_MAX_FLOW_FWD_VPORTS; out_index++) INIT_LIST_HEAD(&flow->encaps[out_index].list); INIT_LIST_HEAD(&flow->hairpin); @@ -4354,7 +4573,17 @@ err_free: } static void -mlx5e_flow_esw_attr_init(struct mlx5_esw_flow_attr *esw_attr, +mlx5e_flow_attr_init(struct mlx5_flow_attr *attr, + struct mlx5e_tc_flow_parse_attr *parse_attr, + struct flow_cls_offload *f) +{ + attr->parse_attr = parse_attr; + attr->chain = f->common.chain_index; + attr->prio = f->common.prio; +} + +static void +mlx5e_flow_esw_attr_init(struct mlx5_flow_attr *attr, struct mlx5e_priv *priv, struct mlx5e_tc_flow_parse_attr *parse_attr, struct flow_cls_offload *f, @@ -4362,10 +4591,9 @@ mlx5e_flow_esw_attr_init(struct mlx5_esw_flow_attr *esw_attr, struct mlx5_core_dev *in_mdev) { struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr; - esw_attr->parse_attr = parse_attr; - esw_attr->chain = f->common.chain_index; - esw_attr->prio = f->common.prio; + mlx5e_flow_attr_init(attr, parse_attr, f); esw_attr->in_rep = in_rep; esw_attr->in_mdev = in_mdev; @@ -4399,7 +4627,7 @@ __mlx5e_add_fdb_flow(struct mlx5e_priv *priv, goto out; parse_attr->filter_dev = filter_dev; - mlx5e_flow_esw_attr_init(flow->esw_attr, + mlx5e_flow_esw_attr_init(flow->attr, priv, parse_attr, f, in_rep, in_mdev); @@ -4409,8 +4637,8 @@ __mlx5e_add_fdb_flow(struct mlx5e_priv *priv, goto err_free; /* actions validation depends on parsing the ct matches first */ - err = mlx5_tc_ct_match_add(priv, &parse_attr->spec, f, - &flow->esw_attr->ct_attr, extack); + err = mlx5_tc_ct_match_add(get_ct_priv(priv), &parse_attr->spec, f, + &flow->attr->ct_attr, extack); if (err) goto err_free; @@ -4441,6 +4669,7 @@ static int mlx5e_tc_add_fdb_peer_flow(struct flow_cls_offload *f, { struct mlx5e_priv *priv = flow->priv, *peer_priv; struct mlx5_eswitch *esw = priv->mdev->priv.eswitch, *peer_esw; + struct mlx5_esw_flow_attr *attr = flow->attr->esw_attr; struct mlx5_devcom *devcom = priv->mdev->priv.devcom; struct mlx5e_tc_flow_parse_attr *parse_attr; struct mlx5e_rep_priv *peer_urpriv; @@ -4460,15 +4689,15 @@ static int mlx5e_tc_add_fdb_peer_flow(struct flow_cls_offload *f, * original flow and packets redirected from uplink use the * peer mdev. */ - if (flow->esw_attr->in_rep->vport == MLX5_VPORT_UPLINK) + if (attr->in_rep->vport == MLX5_VPORT_UPLINK) in_mdev = peer_priv->mdev; else in_mdev = priv->mdev; - parse_attr = flow->esw_attr->parse_attr; + parse_attr = flow->attr->parse_attr; peer_flow = __mlx5e_add_fdb_flow(peer_priv, f, flow_flags, parse_attr->filter_dev, - flow->esw_attr->in_rep, in_mdev); + attr->in_rep, in_mdev); if (IS_ERR(peer_flow)) { err = PTR_ERR(peer_flow); goto out; @@ -4532,9 +4761,12 @@ mlx5e_add_nic_flow(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow; int attr_size, err; - /* multi-chain not supported for NIC rules */ - if (!tc_cls_can_offload_and_chain0(priv->netdev, &f->common)) + if (!MLX5_CAP_FLOWTABLE_NIC_RX(priv->mdev, ignore_flow_level)) { + if (!tc_cls_can_offload_and_chain0(priv->netdev, &f->common)) + return -EOPNOTSUPP; + } else if (!tc_can_offload_extack(priv->netdev, f->common.extack)) { return -EOPNOTSUPP; + } flow_flags |= BIT(MLX5E_TC_FLOW_FLAG_NIC); attr_size = sizeof(struct mlx5_nic_flow_attr); @@ -4544,11 +4776,18 @@ mlx5e_add_nic_flow(struct mlx5e_priv *priv, goto out; parse_attr->filter_dev = filter_dev; + mlx5e_flow_attr_init(flow->attr, parse_attr, f); + err = parse_cls_flower(flow->priv, flow, &parse_attr->spec, f, filter_dev); if (err) goto err_free; + err = mlx5_tc_ct_match_add(get_ct_priv(priv), &parse_attr->spec, f, + &flow->attr->ct_attr, extack); + if (err) + goto err_free; + err = parse_tc_nic_actions(priv, &rule->action, parse_attr, flow, extack); if (err) goto err_free; @@ -4558,14 +4797,12 @@ mlx5e_add_nic_flow(struct mlx5e_priv *priv, goto err_free; flow_flag_set(flow, OFFLOADED); - kvfree(parse_attr); *__flow = flow; return 0; err_free: mlx5e_flow_put(priv, flow); - kvfree(parse_attr); out: return err; } @@ -4940,9 +5177,27 @@ static int mlx5e_tc_netdev_event(struct notifier_block *this, return NOTIFY_DONE; } +static int mlx5e_tc_nic_get_ft_size(struct mlx5_core_dev *dev) +{ + int tc_grp_size, tc_tbl_size; + u32 max_flow_counter; + + max_flow_counter = (MLX5_CAP_GEN(dev, max_flow_counter_31_16) << 16) | + MLX5_CAP_GEN(dev, max_flow_counter_15_0); + + tc_grp_size = min_t(int, max_flow_counter, MLX5E_TC_TABLE_MAX_GROUP_SIZE); + + tc_tbl_size = min_t(int, tc_grp_size * MLX5E_TC_TABLE_NUM_GROUPS, + BIT(MLX5_CAP_FLOWTABLE_NIC_RX(dev, log_max_ft_size))); + + return tc_tbl_size; +} + int mlx5e_tc_nic_init(struct mlx5e_priv *priv) { struct mlx5e_tc_table *tc = &priv->fs.tc; + struct mlx5_core_dev *dev = priv->mdev; + struct mlx5_chains_attr attr = {}; int err; mlx5e_mod_hdr_tbl_init(&tc->mod_hdr); @@ -4954,6 +5209,27 @@ int mlx5e_tc_nic_init(struct mlx5e_priv *priv) if (err) return err; + if (MLX5_CAP_FLOWTABLE_NIC_RX(priv->mdev, ignore_flow_level)) { + attr.flags = MLX5_CHAINS_AND_PRIOS_SUPPORTED | + MLX5_CHAINS_IGNORE_FLOW_LEVEL_SUPPORTED; + attr.max_restore_tag = MLX5E_TC_TABLE_CHAIN_TAG_MASK; + } + attr.ns = MLX5_FLOW_NAMESPACE_KERNEL; + attr.max_ft_sz = mlx5e_tc_nic_get_ft_size(dev); + attr.max_grp_num = MLX5E_TC_TABLE_NUM_GROUPS; + attr.default_ft = priv->fs.vlan.ft.t; + + tc->chains = mlx5_chains_create(dev, &attr); + if (IS_ERR(tc->chains)) { + err = PTR_ERR(tc->chains); + goto err_chains; + } + + tc->ct = mlx5_tc_ct_init(priv, tc->chains, &priv->fs.tc.mod_hdr, + MLX5_FLOW_NAMESPACE_KERNEL); + if (IS_ERR(tc->ct)) + goto err_ct; + tc->netdevice_nb.notifier_call = mlx5e_tc_netdev_event; err = register_netdevice_notifier_dev_net(priv->netdev, &tc->netdevice_nb, @@ -4961,8 +5237,17 @@ int mlx5e_tc_nic_init(struct mlx5e_priv *priv) if (err) { tc->netdevice_nb.notifier_call = NULL; mlx5_core_warn(priv->mdev, "Failed to register netdev notifier\n"); + goto err_reg; } + return 0; + +err_reg: + mlx5_tc_ct_clean(tc->ct); +err_ct: + mlx5_chains_destroy(tc->chains); +err_chains: + rhashtable_destroy(&tc->ht); return err; } @@ -4987,28 +5272,38 @@ void mlx5e_tc_nic_cleanup(struct mlx5e_priv *priv) mlx5e_mod_hdr_tbl_destroy(&tc->mod_hdr); mutex_destroy(&tc->hairpin_tbl_lock); - rhashtable_destroy(&tc->ht); + rhashtable_free_and_destroy(&tc->ht, _mlx5e_tc_del_flow, NULL); if (!IS_ERR_OR_NULL(tc->t)) { - mlx5_destroy_flow_table(tc->t); + mlx5_chains_put_table(tc->chains, 0, 1, MLX5E_TC_FT_LEVEL); tc->t = NULL; } mutex_destroy(&tc->t_lock); + + mlx5_tc_ct_clean(tc->ct); + mlx5_chains_destroy(tc->chains); } int mlx5e_tc_esw_init(struct rhashtable *tc_ht) { const size_t sz_enc_opts = sizeof(struct tunnel_match_enc_opts); struct mlx5_rep_uplink_priv *uplink_priv; - struct mlx5e_rep_priv *priv; + struct mlx5e_rep_priv *rpriv; struct mapping_ctx *mapping; - int err; + struct mlx5_eswitch *esw; + struct mlx5e_priv *priv; + int err = 0; uplink_priv = container_of(tc_ht, struct mlx5_rep_uplink_priv, tc_ht); - priv = container_of(uplink_priv, struct mlx5e_rep_priv, uplink_priv); + rpriv = container_of(uplink_priv, struct mlx5e_rep_priv, uplink_priv); + priv = netdev_priv(rpriv->netdev); + esw = priv->mdev->priv.eswitch; - err = mlx5_tc_ct_init(uplink_priv); - if (err) + uplink_priv->ct_priv = mlx5_tc_ct_init(netdev_priv(priv->netdev), + esw_chains(esw), + &esw->offloads.mod_hdr, + MLX5_FLOW_NAMESPACE_FDB); + if (IS_ERR(uplink_priv->ct_priv)) goto err_ct; mapping = mapping_create(sizeof(struct tunnel_match_key), @@ -5037,7 +5332,7 @@ err_ht_init: err_enc_opts_mapping: mapping_destroy(uplink_priv->tunnel_mapping); err_tun_mapping: - mlx5_tc_ct_clean(uplink_priv); + mlx5_tc_ct_clean(uplink_priv->ct_priv); err_ct: netdev_warn(priv->netdev, "Failed to initialize tc (eswitch), err: %d", err); @@ -5051,10 +5346,11 @@ void mlx5e_tc_esw_cleanup(struct rhashtable *tc_ht) rhashtable_free_and_destroy(tc_ht, _mlx5e_tc_del_flow, NULL); uplink_priv = container_of(tc_ht, struct mlx5_rep_uplink_priv, tc_ht); + mapping_destroy(uplink_priv->tunnel_enc_opts_mapping); mapping_destroy(uplink_priv->tunnel_mapping); - mlx5_tc_ct_clean(uplink_priv); + mlx5_tc_ct_clean(uplink_priv->ct_priv); } int mlx5e_tc_num_filters(struct mlx5e_priv *priv, unsigned long flags) @@ -5119,3 +5415,44 @@ int mlx5e_setup_tc_block_cb(enum tc_setup_type type, void *type_data, return -EOPNOTSUPP; } } + +bool mlx5e_tc_update_skb(struct mlx5_cqe64 *cqe, + struct sk_buff *skb) +{ +#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT) + u32 chain = 0, chain_tag, reg_b, zone_restore_id; + struct mlx5e_priv *priv = netdev_priv(skb->dev); + struct mlx5e_tc_table *tc = &priv->fs.tc; + struct tc_skb_ext *tc_skb_ext; + int err; + + reg_b = be32_to_cpu(cqe->ft_metadata); + + chain_tag = reg_b & MLX5E_TC_TABLE_CHAIN_TAG_MASK; + + err = mlx5_get_chain_for_tag(nic_chains(priv), chain_tag, &chain); + if (err) { + netdev_dbg(priv->netdev, + "Couldn't find chain for chain tag: %d, err: %d\n", + chain_tag, err); + return false; + } + + if (chain) { + tc_skb_ext = skb_ext_add(skb, TC_SKB_EXT); + if (WARN_ON(!tc_skb_ext)) + return false; + + tc_skb_ext->chain = chain; + + zone_restore_id = (reg_b >> REG_MAPPING_SHIFT(NIC_ZONE_RESTORE_TO_REG)) & + ZONE_RESTORE_MAX; + + if (!mlx5e_tc_ct_restore_flow(tc->ct, skb, + zone_restore_id)) + return false; + } +#endif /* CONFIG_NET_TC_SKB_EXT */ + + return true; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h index 437f680728fd..3b979008143d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h @@ -35,17 +35,57 @@ #include <net/pkt_cls.h> #include "en.h" +#include "eswitch.h" +#include "en/tc_ct.h" #define MLX5E_TC_FLOW_ID_MASK 0x0000ffff #ifdef CONFIG_MLX5_ESWITCH +#define NIC_FLOW_ATTR_SZ (sizeof(struct mlx5_flow_attr) +\ + sizeof(struct mlx5_nic_flow_attr)) +#define ESW_FLOW_ATTR_SZ (sizeof(struct mlx5_flow_attr) +\ + sizeof(struct mlx5_esw_flow_attr)) +#define ns_to_attr_sz(ns) (((ns) == MLX5_FLOW_NAMESPACE_FDB) ?\ + ESW_FLOW_ATTR_SZ :\ + NIC_FLOW_ATTR_SZ) + + int mlx5e_tc_num_filters(struct mlx5e_priv *priv, unsigned long flags); struct mlx5e_tc_update_priv { struct net_device *tun_dev; }; +struct mlx5_nic_flow_attr { + u32 flow_tag; + u32 hairpin_tirn; + struct mlx5_flow_table *hairpin_ft; +}; + +struct mlx5_flow_attr { + u32 action; + struct mlx5_fc *counter; + struct mlx5_modify_hdr *modify_hdr; + struct mlx5_ct_attr ct_attr; + struct mlx5e_tc_flow_parse_attr *parse_attr; + u32 chain; + u16 prio; + u32 dest_chain; + struct mlx5_flow_table *ft; + struct mlx5_flow_table *dest_ft; + u8 inner_match_level; + u8 outer_match_level; + u32 flags; + union { + struct mlx5_esw_flow_attr esw_attr[0]; + struct mlx5_nic_flow_attr nic_attr[0]; + }; +}; + +#define MLX5E_TC_TABLE_CHAIN_TAG_BITS 16 +#define MLX5E_TC_TABLE_CHAIN_TAG_MASK GENMASK(MLX5E_TC_TABLE_CHAIN_TAG_BITS - 1, 0) + #if IS_ENABLED(CONFIG_MLX5_CLS_ACT) struct tunnel_match_key { @@ -90,6 +130,7 @@ enum { int mlx5e_tc_esw_init(struct rhashtable *tc_ht); void mlx5e_tc_esw_cleanup(struct rhashtable *tc_ht); +bool mlx5e_is_eswitch_flow(struct mlx5e_tc_flow *flow); int mlx5e_configure_flower(struct net_device *dev, struct mlx5e_priv *priv, struct flow_cls_offload *f, unsigned long flags); @@ -133,6 +174,8 @@ enum mlx5e_tc_attr_to_reg { MARK_TO_REG, LABELS_TO_REG, FTEID_TO_REG, + NIC_CHAIN_TO_REG, + NIC_ZONE_RESTORE_TO_REG, }; struct mlx5e_tc_attr_to_reg_mapping { @@ -150,6 +193,7 @@ bool mlx5e_is_valid_eswitch_fwd_dev(struct mlx5e_priv *priv, int mlx5e_tc_match_to_reg_set(struct mlx5_core_dev *mdev, struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts, + enum mlx5_flow_namespace_type ns, enum mlx5e_tc_attr_to_reg type, u32 data); @@ -181,14 +225,42 @@ void mlx5e_tc_nic_cleanup(struct mlx5e_priv *priv); int mlx5e_setup_tc_block_cb(enum tc_setup_type type, void *type_data, void *cb_priv); +struct mlx5_flow_handle * +mlx5e_add_offloaded_nic_rule(struct mlx5e_priv *priv, + struct mlx5_flow_spec *spec, + struct mlx5_flow_attr *attr); +void mlx5e_del_offloaded_nic_rule(struct mlx5e_priv *priv, + struct mlx5_flow_handle *rule, + struct mlx5_flow_attr *attr); + +struct mlx5_flow_handle * +mlx5_tc_rule_insert(struct mlx5e_priv *priv, + struct mlx5_flow_spec *spec, + struct mlx5_flow_attr *attr); +void +mlx5_tc_rule_delete(struct mlx5e_priv *priv, + struct mlx5_flow_handle *rule, + struct mlx5_flow_attr *attr); + #else /* CONFIG_MLX5_CLS_ACT */ static inline int mlx5e_tc_nic_init(struct mlx5e_priv *priv) { return 0; } static inline void mlx5e_tc_nic_cleanup(struct mlx5e_priv *priv) {} static inline int mlx5e_setup_tc_block_cb(enum tc_setup_type type, void *type_data, void *cb_priv) { return -EOPNOTSUPP; } + #endif /* CONFIG_MLX5_CLS_ACT */ +struct mlx5_flow_attr *mlx5_alloc_flow_attr(enum mlx5_flow_namespace_type type); + +struct mlx5_flow_handle * +mlx5e_add_offloaded_nic_rule(struct mlx5e_priv *priv, + struct mlx5_flow_spec *spec, + struct mlx5_flow_attr *attr); +void mlx5e_del_offloaded_nic_rule(struct mlx5e_priv *priv, + struct mlx5_flow_handle *rule, + struct mlx5_flow_attr *attr); + #else /* CONFIG_MLX5_ESWITCH */ static inline int mlx5e_tc_nic_init(struct mlx5e_priv *priv) { return 0; } static inline void mlx5e_tc_nic_cleanup(struct mlx5e_priv *priv) {} @@ -203,4 +275,29 @@ mlx5e_setup_tc_block_cb(enum tc_setup_type type, void *type_data, void *cb_priv) { return -EOPNOTSUPP; } #endif +#if IS_ENABLED(CONFIG_MLX5_CLS_ACT) +static inline bool mlx5e_cqe_regb_chain(struct mlx5_cqe64 *cqe) +{ +#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT) + u32 chain, reg_b; + + reg_b = be32_to_cpu(cqe->ft_metadata); + + chain = reg_b & MLX5E_TC_TABLE_CHAIN_TAG_MASK; + if (chain) + return true; +#endif + + return false; +} + +bool mlx5e_tc_update_skb(struct mlx5_cqe64 *cqe, struct sk_buff *skb); +#else /* CONFIG_MLX5_CLS_ACT */ +static inline bool mlx5e_cqe_regb_chain(struct mlx5_cqe64 *cqe) +{ return false; } +static inline bool +mlx5e_tc_update_skb(struct mlx5_cqe64 *cqe, struct sk_buff *skb) +{ return true; } +#endif + #endif /* __MLX5_EN_TC_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index da596de3abba..82b4419af9d4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -144,9 +144,29 @@ static inline void mlx5e_insert_vlan(void *start, struct sk_buff *skb, u16 ihs) memcpy(&vhdr->h_vlan_encapsulated_proto, skb->data + cpy1_sz, cpy2_sz); } +/* RM 2311217: no L4 inner checksum for IPsec tunnel type packet */ +static void +ipsec_txwqe_build_eseg_csum(struct mlx5e_txqsq *sq, struct sk_buff *skb, + struct mlx5_wqe_eth_seg *eseg) +{ + eseg->cs_flags = MLX5_ETH_WQE_L3_CSUM; + if (skb->encapsulation) { + eseg->cs_flags |= MLX5_ETH_WQE_L3_INNER_CSUM; + sq->stats->csum_partial_inner++; + } else { + eseg->cs_flags |= MLX5_ETH_WQE_L4_CSUM; + sq->stats->csum_partial++; + } +} + static inline void mlx5e_txwqe_build_eseg_csum(struct mlx5e_txqsq *sq, struct sk_buff *skb, struct mlx5_wqe_eth_seg *eseg) { + if (unlikely(eseg->flow_table_metadata & cpu_to_be32(MLX5_ETH_WQE_FT_META_IPSEC))) { + ipsec_txwqe_build_eseg_csum(sq, skb, eseg); + return; + } + if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) { eseg->cs_flags = MLX5_ETH_WQE_L3_CSUM; if (skb->encapsulation) { @@ -232,131 +252,188 @@ dma_unmap_wqe_err: return -ENOMEM; } +struct mlx5e_tx_attr { + u32 num_bytes; + u16 headlen; + u16 ihs; + __be16 mss; + u16 insz; + u8 opcode; +}; + +struct mlx5e_tx_wqe_attr { + u16 ds_cnt; + u16 ds_cnt_inl; + u16 ds_cnt_ids; + u8 num_wqebbs; +}; + +static u8 +mlx5e_tx_wqe_inline_mode(struct mlx5e_txqsq *sq, struct sk_buff *skb, + struct mlx5e_accel_tx_state *accel) +{ + u8 mode; + +#ifdef CONFIG_MLX5_EN_TLS + if (accel && accel->tls.tls_tisn) + return MLX5_INLINE_MODE_TCP_UDP; +#endif + + mode = sq->min_inline_mode; + + if (skb_vlan_tag_present(skb) && + test_bit(MLX5E_SQ_STATE_VLAN_NEED_L2_INLINE, &sq->state)) + mode = max_t(u8, MLX5_INLINE_MODE_L2, mode); + + return mode; +} + +static void mlx5e_sq_xmit_prepare(struct mlx5e_txqsq *sq, struct sk_buff *skb, + struct mlx5e_accel_tx_state *accel, + struct mlx5e_tx_attr *attr) +{ + struct mlx5e_sq_stats *stats = sq->stats; + + if (skb_is_gso(skb)) { + u16 ihs = mlx5e_tx_get_gso_ihs(sq, skb); + + *attr = (struct mlx5e_tx_attr) { + .opcode = MLX5_OPCODE_LSO, + .mss = cpu_to_be16(skb_shinfo(skb)->gso_size), + .ihs = ihs, + .num_bytes = skb->len + (skb_shinfo(skb)->gso_segs - 1) * ihs, + .headlen = skb_headlen(skb) - ihs, + }; + + stats->packets += skb_shinfo(skb)->gso_segs; + } else { + u8 mode = mlx5e_tx_wqe_inline_mode(sq, skb, accel); + u16 ihs = mlx5e_calc_min_inline(mode, skb); + + *attr = (struct mlx5e_tx_attr) { + .opcode = MLX5_OPCODE_SEND, + .mss = cpu_to_be16(0), + .ihs = ihs, + .num_bytes = max_t(unsigned int, skb->len, ETH_ZLEN), + .headlen = skb_headlen(skb) - ihs, + }; + + stats->packets++; + } + + attr->insz = mlx5e_accel_tx_ids_len(sq, accel); + stats->bytes += attr->num_bytes; +} + +static void mlx5e_sq_calc_wqe_attr(struct sk_buff *skb, const struct mlx5e_tx_attr *attr, + struct mlx5e_tx_wqe_attr *wqe_attr) +{ + u16 ds_cnt = MLX5E_TX_WQE_EMPTY_DS_COUNT; + u16 ds_cnt_inl = 0; + u16 ds_cnt_ids = 0; + + if (attr->insz) + ds_cnt_ids = DIV_ROUND_UP(sizeof(struct mlx5_wqe_inline_seg) + attr->insz, + MLX5_SEND_WQE_DS); + + ds_cnt += !!attr->headlen + skb_shinfo(skb)->nr_frags + ds_cnt_ids; + if (attr->ihs) { + u16 inl = attr->ihs - INL_HDR_START_SZ; + + if (skb_vlan_tag_present(skb)) + inl += VLAN_HLEN; + + ds_cnt_inl = DIV_ROUND_UP(inl, MLX5_SEND_WQE_DS); + ds_cnt += ds_cnt_inl; + } + + *wqe_attr = (struct mlx5e_tx_wqe_attr) { + .ds_cnt = ds_cnt, + .ds_cnt_inl = ds_cnt_inl, + .ds_cnt_ids = ds_cnt_ids, + .num_wqebbs = DIV_ROUND_UP(ds_cnt, MLX5_SEND_WQEBB_NUM_DS), + }; +} + +static void mlx5e_tx_skb_update_hwts_flags(struct sk_buff *skb) +{ + if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) + skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; +} + +static void mlx5e_tx_check_stop(struct mlx5e_txqsq *sq) +{ + if (unlikely(!mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc, sq->stop_room))) { + netif_tx_stop_queue(sq->txq); + sq->stats->stopped++; + } +} + static inline void mlx5e_txwqe_complete(struct mlx5e_txqsq *sq, struct sk_buff *skb, - u8 opcode, u16 ds_cnt, u8 num_wqebbs, u32 num_bytes, u8 num_dma, + const struct mlx5e_tx_attr *attr, + const struct mlx5e_tx_wqe_attr *wqe_attr, u8 num_dma, struct mlx5e_tx_wqe_info *wi, struct mlx5_wqe_ctrl_seg *cseg, bool xmit_more) { struct mlx5_wq_cyc *wq = &sq->wq; bool send_doorbell; - wi->num_bytes = num_bytes; - wi->num_dma = num_dma; - wi->num_wqebbs = num_wqebbs; - wi->skb = skb; + *wi = (struct mlx5e_tx_wqe_info) { + .skb = skb, + .num_bytes = attr->num_bytes, + .num_dma = num_dma, + .num_wqebbs = wqe_attr->num_wqebbs, + .num_fifo_pkts = 0, + }; - cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | opcode); - cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | ds_cnt); + cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | attr->opcode); + cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | wqe_attr->ds_cnt); - if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) - skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; + mlx5e_tx_skb_update_hwts_flags(skb); sq->pc += wi->num_wqebbs; - if (unlikely(!mlx5e_wqc_has_room_for(wq, sq->cc, sq->pc, sq->stop_room))) { - netif_tx_stop_queue(sq->txq); - sq->stats->stopped++; - } - send_doorbell = __netdev_tx_sent_queue(sq->txq, num_bytes, - xmit_more); + mlx5e_tx_check_stop(sq); + + send_doorbell = __netdev_tx_sent_queue(sq->txq, attr->num_bytes, xmit_more); if (send_doorbell) mlx5e_notify_hw(wq, sq->pc, sq->uar_map, cseg); } -void mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, - struct mlx5e_tx_wqe *wqe, u16 pi, bool xmit_more) +static void +mlx5e_sq_xmit_wqe(struct mlx5e_txqsq *sq, struct sk_buff *skb, + const struct mlx5e_tx_attr *attr, const struct mlx5e_tx_wqe_attr *wqe_attr, + struct mlx5e_tx_wqe *wqe, u16 pi, bool xmit_more) { - struct mlx5_wq_cyc *wq = &sq->wq; struct mlx5_wqe_ctrl_seg *cseg; struct mlx5_wqe_eth_seg *eseg; struct mlx5_wqe_data_seg *dseg; struct mlx5e_tx_wqe_info *wi; struct mlx5e_sq_stats *stats = sq->stats; - u16 headlen, ihs, contig_wqebbs_room; - u16 ds_cnt, ds_cnt_inl = 0; - u8 num_wqebbs, opcode; - u32 num_bytes; int num_dma; - __be16 mss; - - /* Calc ihs and ds cnt, no writes to wqe yet */ - ds_cnt = sizeof(*wqe) / MLX5_SEND_WQE_DS; - if (skb_is_gso(skb)) { - opcode = MLX5_OPCODE_LSO; - mss = cpu_to_be16(skb_shinfo(skb)->gso_size); - ihs = mlx5e_tx_get_gso_ihs(sq, skb); - num_bytes = skb->len + (skb_shinfo(skb)->gso_segs - 1) * ihs; - stats->packets += skb_shinfo(skb)->gso_segs; - } else { - u8 mode = mlx5e_tx_wqe_inline_mode(sq, &wqe->ctrl, skb); - opcode = MLX5_OPCODE_SEND; - mss = 0; - ihs = mlx5e_calc_min_inline(mode, skb); - num_bytes = max_t(unsigned int, skb->len, ETH_ZLEN); - stats->packets++; - } - - stats->bytes += num_bytes; stats->xmit_more += xmit_more; - headlen = skb->len - ihs - skb->data_len; - ds_cnt += !!headlen; - ds_cnt += skb_shinfo(skb)->nr_frags; - - if (ihs) { - ihs += !!skb_vlan_tag_present(skb) * VLAN_HLEN; - - ds_cnt_inl = DIV_ROUND_UP(ihs - INL_HDR_START_SZ, MLX5_SEND_WQE_DS); - ds_cnt += ds_cnt_inl; - } - - num_wqebbs = DIV_ROUND_UP(ds_cnt, MLX5_SEND_WQEBB_NUM_DS); - contig_wqebbs_room = mlx5_wq_cyc_get_contig_wqebbs(wq, pi); - if (unlikely(contig_wqebbs_room < num_wqebbs)) { -#ifdef CONFIG_MLX5_EN_IPSEC - struct mlx5_wqe_eth_seg cur_eth = wqe->eth; -#endif -#ifdef CONFIG_MLX5_EN_TLS - struct mlx5_wqe_ctrl_seg cur_ctrl = wqe->ctrl; -#endif - mlx5e_fill_sq_frag_edge(sq, wq, pi, contig_wqebbs_room); - pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); - wqe = MLX5E_TX_FETCH_WQE(sq, pi); -#ifdef CONFIG_MLX5_EN_IPSEC - wqe->eth = cur_eth; -#endif -#ifdef CONFIG_MLX5_EN_TLS - wqe->ctrl = cur_ctrl; -#endif - } - /* fill wqe */ wi = &sq->db.wqe_info[pi]; cseg = &wqe->ctrl; eseg = &wqe->eth; dseg = wqe->data; -#if IS_ENABLED(CONFIG_GENEVE) - if (skb->encapsulation) - mlx5e_tx_tunnel_accel(skb, eseg); -#endif - mlx5e_txwqe_build_eseg_csum(sq, skb, eseg); + eseg->mss = attr->mss; - eseg->mss = mss; - - if (ihs) { - eseg->inline_hdr.sz = cpu_to_be16(ihs); + if (attr->ihs) { if (skb_vlan_tag_present(skb)) { - ihs -= VLAN_HLEN; - mlx5e_insert_vlan(eseg->inline_hdr.start, skb, ihs); + eseg->inline_hdr.sz |= cpu_to_be16(attr->ihs + VLAN_HLEN); + mlx5e_insert_vlan(eseg->inline_hdr.start, skb, attr->ihs); stats->added_vlan_packets++; } else { - memcpy(eseg->inline_hdr.start, skb->data, ihs); + eseg->inline_hdr.sz |= cpu_to_be16(attr->ihs); + memcpy(eseg->inline_hdr.start, skb->data, attr->ihs); } - dseg += ds_cnt_inl; + dseg += wqe_attr->ds_cnt_inl; } else if (skb_vlan_tag_present(skb)) { eseg->insert.type = cpu_to_be16(MLX5_ETH_WQE_INSERT_VLAN); if (skb->vlan_proto == cpu_to_be16(ETH_P_8021AD)) @@ -365,12 +442,13 @@ void mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, stats->added_vlan_packets++; } - num_dma = mlx5e_txwqe_build_dsegs(sq, skb, skb->data + ihs, headlen, dseg); + dseg += wqe_attr->ds_cnt_ids; + num_dma = mlx5e_txwqe_build_dsegs(sq, skb, skb->data + attr->ihs, + attr->headlen, dseg); if (unlikely(num_dma < 0)) goto err_drop; - mlx5e_txwqe_complete(sq, skb, opcode, ds_cnt, num_wqebbs, num_bytes, - num_dma, wi, cseg, xmit_more); + mlx5e_txwqe_complete(sq, skb, attr, wqe_attr, num_dma, wi, cseg, xmit_more); return; @@ -379,10 +457,173 @@ err_drop: dev_kfree_skb_any(skb); } +static bool mlx5e_tx_skb_supports_mpwqe(struct sk_buff *skb, struct mlx5e_tx_attr *attr) +{ + return !skb_is_nonlinear(skb) && !skb_vlan_tag_present(skb) && !attr->ihs && + !attr->insz; +} + +static bool mlx5e_tx_mpwqe_same_eseg(struct mlx5e_txqsq *sq, struct mlx5_wqe_eth_seg *eseg) +{ + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; + + /* Assumes the session is already running and has at least one packet. */ + return !memcmp(&session->wqe->eth, eseg, MLX5E_ACCEL_ESEG_LEN); +} + +static void mlx5e_tx_mpwqe_session_start(struct mlx5e_txqsq *sq, + struct mlx5_wqe_eth_seg *eseg) +{ + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; + struct mlx5e_tx_wqe *wqe; + u16 pi; + + pi = mlx5e_txqsq_get_next_pi(sq, MLX5E_TX_MPW_MAX_WQEBBS); + wqe = MLX5E_TX_FETCH_WQE(sq, pi); + prefetchw(wqe->data); + + *session = (struct mlx5e_tx_mpwqe) { + .wqe = wqe, + .bytes_count = 0, + .ds_count = MLX5E_TX_WQE_EMPTY_DS_COUNT, + .pkt_count = 0, + .inline_on = 0, + }; + + memcpy(&session->wqe->eth, eseg, MLX5E_ACCEL_ESEG_LEN); + + sq->stats->mpwqe_blks++; +} + +static bool mlx5e_tx_mpwqe_session_is_active(struct mlx5e_txqsq *sq) +{ + return sq->mpwqe.wqe; +} + +static void mlx5e_tx_mpwqe_add_dseg(struct mlx5e_txqsq *sq, struct mlx5e_xmit_data *txd) +{ + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; + struct mlx5_wqe_data_seg *dseg; + + dseg = (struct mlx5_wqe_data_seg *)session->wqe + session->ds_count; + + session->pkt_count++; + session->bytes_count += txd->len; + + dseg->addr = cpu_to_be64(txd->dma_addr); + dseg->byte_count = cpu_to_be32(txd->len); + dseg->lkey = sq->mkey_be; + session->ds_count++; + + sq->stats->mpwqe_pkts++; +} + +static struct mlx5_wqe_ctrl_seg *mlx5e_tx_mpwqe_session_complete(struct mlx5e_txqsq *sq) +{ + struct mlx5e_tx_mpwqe *session = &sq->mpwqe; + u8 ds_count = session->ds_count; + struct mlx5_wqe_ctrl_seg *cseg; + struct mlx5e_tx_wqe_info *wi; + u16 pi; + + cseg = &session->wqe->ctrl; + cseg->opmod_idx_opcode = cpu_to_be32((sq->pc << 8) | MLX5_OPCODE_ENHANCED_MPSW); + cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | ds_count); + + pi = mlx5_wq_cyc_ctr2ix(&sq->wq, sq->pc); + wi = &sq->db.wqe_info[pi]; + *wi = (struct mlx5e_tx_wqe_info) { + .skb = NULL, + .num_bytes = session->bytes_count, + .num_wqebbs = DIV_ROUND_UP(ds_count, MLX5_SEND_WQEBB_NUM_DS), + .num_dma = session->pkt_count, + .num_fifo_pkts = session->pkt_count, + }; + + sq->pc += wi->num_wqebbs; + + session->wqe = NULL; + + mlx5e_tx_check_stop(sq); + + return cseg; +} + +static void +mlx5e_sq_xmit_mpwqe(struct mlx5e_txqsq *sq, struct sk_buff *skb, + struct mlx5_wqe_eth_seg *eseg, bool xmit_more) +{ + struct mlx5_wqe_ctrl_seg *cseg; + struct mlx5e_xmit_data txd; + + if (!mlx5e_tx_mpwqe_session_is_active(sq)) { + mlx5e_tx_mpwqe_session_start(sq, eseg); + } else if (!mlx5e_tx_mpwqe_same_eseg(sq, eseg)) { + mlx5e_tx_mpwqe_session_complete(sq); + mlx5e_tx_mpwqe_session_start(sq, eseg); + } + + sq->stats->xmit_more += xmit_more; + + txd.data = skb->data; + txd.len = skb->len; + + txd.dma_addr = dma_map_single(sq->pdev, txd.data, txd.len, DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(sq->pdev, txd.dma_addr))) + goto err_unmap; + mlx5e_dma_push(sq, txd.dma_addr, txd.len, MLX5E_DMA_MAP_SINGLE); + + mlx5e_skb_fifo_push(sq, skb); + + mlx5e_tx_mpwqe_add_dseg(sq, &txd); + + mlx5e_tx_skb_update_hwts_flags(skb); + + if (unlikely(mlx5e_tx_mpwqe_is_full(&sq->mpwqe))) { + /* Might stop the queue and affect the retval of __netdev_tx_sent_queue. */ + cseg = mlx5e_tx_mpwqe_session_complete(sq); + + if (__netdev_tx_sent_queue(sq->txq, txd.len, xmit_more)) + mlx5e_notify_hw(&sq->wq, sq->pc, sq->uar_map, cseg); + } else if (__netdev_tx_sent_queue(sq->txq, txd.len, xmit_more)) { + /* Might stop the queue, but we were asked to ring the doorbell anyway. */ + cseg = mlx5e_tx_mpwqe_session_complete(sq); + + mlx5e_notify_hw(&sq->wq, sq->pc, sq->uar_map, cseg); + } + + return; + +err_unmap: + mlx5e_dma_unmap_wqe_err(sq, 1); + sq->stats->dropped++; + dev_kfree_skb_any(skb); +} + +void mlx5e_tx_mpwqe_ensure_complete(struct mlx5e_txqsq *sq) +{ + /* Unlikely in non-MPWQE workloads; not important in MPWQE workloads. */ + if (unlikely(mlx5e_tx_mpwqe_session_is_active(sq))) + mlx5e_tx_mpwqe_session_complete(sq); +} + +static bool mlx5e_txwqe_build_eseg(struct mlx5e_priv *priv, struct mlx5e_txqsq *sq, + struct sk_buff *skb, struct mlx5_wqe_eth_seg *eseg) +{ + if (unlikely(!mlx5e_accel_tx_eseg(priv, skb, eseg))) + return false; + + mlx5e_txwqe_build_eseg_csum(sq, skb, eseg); + + return true; +} + netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev) { struct mlx5e_priv *priv = netdev_priv(dev); struct mlx5e_accel_tx_state accel = {}; + struct mlx5e_tx_wqe_attr wqe_attr; + struct mlx5e_tx_attr attr; struct mlx5e_tx_wqe *wqe; struct mlx5e_txqsq *sq; u16 pi; @@ -391,21 +632,92 @@ netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev) /* May send SKBs and WQEs. */ if (unlikely(!mlx5e_accel_tx_begin(dev, sq, skb, &accel))) - goto out; + return NETDEV_TX_OK; - pi = mlx5_wq_cyc_ctr2ix(&sq->wq, sq->pc); + mlx5e_sq_xmit_prepare(sq, skb, &accel, &attr); + + if (test_bit(MLX5E_SQ_STATE_MPWQE, &sq->state)) { + if (mlx5e_tx_skb_supports_mpwqe(skb, &attr)) { + struct mlx5_wqe_eth_seg eseg = {}; + + if (unlikely(!mlx5e_txwqe_build_eseg(priv, sq, skb, &eseg))) + return NETDEV_TX_OK; + + mlx5e_sq_xmit_mpwqe(sq, skb, &eseg, netdev_xmit_more()); + return NETDEV_TX_OK; + } + + mlx5e_tx_mpwqe_ensure_complete(sq); + } + + mlx5e_sq_calc_wqe_attr(skb, &attr, &wqe_attr); + pi = mlx5e_txqsq_get_next_pi(sq, wqe_attr.num_wqebbs); wqe = MLX5E_TX_FETCH_WQE(sq, pi); /* May update the WQE, but may not post other WQEs. */ - if (unlikely(!mlx5e_accel_tx_finish(priv, sq, skb, wqe, &accel))) - goto out; + mlx5e_accel_tx_finish(sq, wqe, &accel, + (struct mlx5_wqe_inline_seg *)(wqe->data + wqe_attr.ds_cnt_inl)); + if (unlikely(!mlx5e_txwqe_build_eseg(priv, sq, skb, &wqe->eth))) + return NETDEV_TX_OK; - mlx5e_sq_xmit(sq, skb, wqe, pi, netdev_xmit_more()); + mlx5e_sq_xmit_wqe(sq, skb, &attr, &wqe_attr, wqe, pi, netdev_xmit_more()); -out: return NETDEV_TX_OK; } +void mlx5e_sq_xmit_simple(struct mlx5e_txqsq *sq, struct sk_buff *skb, bool xmit_more) +{ + struct mlx5e_tx_wqe_attr wqe_attr; + struct mlx5e_tx_attr attr; + struct mlx5e_tx_wqe *wqe; + u16 pi; + + mlx5e_sq_xmit_prepare(sq, skb, NULL, &attr); + mlx5e_sq_calc_wqe_attr(skb, &attr, &wqe_attr); + pi = mlx5e_txqsq_get_next_pi(sq, wqe_attr.num_wqebbs); + wqe = MLX5E_TX_FETCH_WQE(sq, pi); + mlx5e_txwqe_build_eseg_csum(sq, skb, &wqe->eth); + mlx5e_sq_xmit_wqe(sq, skb, &attr, &wqe_attr, wqe, pi, xmit_more); +} + +static void mlx5e_tx_wi_dma_unmap(struct mlx5e_txqsq *sq, struct mlx5e_tx_wqe_info *wi, + u32 *dma_fifo_cc) +{ + int i; + + for (i = 0; i < wi->num_dma; i++) { + struct mlx5e_sq_dma *dma = mlx5e_dma_get(sq, (*dma_fifo_cc)++); + + mlx5e_tx_dma_unmap(sq->pdev, dma); + } +} + +static void mlx5e_consume_skb(struct mlx5e_txqsq *sq, struct sk_buff *skb, + struct mlx5_cqe64 *cqe, int napi_budget) +{ + if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) { + struct skb_shared_hwtstamps hwts = {}; + u64 ts = get_cqe_ts(cqe); + + hwts.hwtstamp = mlx5_timecounter_cyc2time(sq->clock, ts); + skb_tstamp_tx(skb, &hwts); + } + + napi_consume_skb(skb, napi_budget); +} + +static void mlx5e_tx_wi_consume_fifo_skbs(struct mlx5e_txqsq *sq, struct mlx5e_tx_wqe_info *wi, + struct mlx5_cqe64 *cqe, int napi_budget) +{ + int i; + + for (i = 0; i < wi->num_fifo_pkts; i++) { + struct sk_buff *skb = mlx5e_skb_fifo_pop(sq); + + mlx5e_consume_skb(sq, skb, cqe, napi_budget); + } +} + bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) { struct mlx5e_sq_stats *stats; @@ -451,42 +763,33 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) wqe_counter = be16_to_cpu(cqe->wqe_counter); do { - struct sk_buff *skb; - int j; - last_wqe = (sqcc == wqe_counter); ci = mlx5_wq_cyc_ctr2ix(&sq->wq, sqcc); wi = &sq->db.wqe_info[ci]; - skb = wi->skb; - if (unlikely(!skb)) { - mlx5e_ktls_tx_handle_resync_dump_comp(sq, wi, &dma_fifo_cc); - sqcc += wi->num_wqebbs; - continue; - } + sqcc += wi->num_wqebbs; - if (unlikely(skb_shinfo(skb)->tx_flags & - SKBTX_HW_TSTAMP)) { - struct skb_shared_hwtstamps hwts = {}; + if (likely(wi->skb)) { + mlx5e_tx_wi_dma_unmap(sq, wi, &dma_fifo_cc); + mlx5e_consume_skb(sq, wi->skb, cqe, napi_budget); - hwts.hwtstamp = - mlx5_timecounter_cyc2time(sq->clock, - get_cqe_ts(cqe)); - skb_tstamp_tx(skb, &hwts); + npkts++; + nbytes += wi->num_bytes; + continue; } - for (j = 0; j < wi->num_dma; j++) { - struct mlx5e_sq_dma *dma = - mlx5e_dma_get(sq, dma_fifo_cc++); + if (unlikely(mlx5e_ktls_tx_try_handle_resync_dump_comp(sq, wi, + &dma_fifo_cc))) + continue; - mlx5e_tx_dma_unmap(sq->pdev, dma); - } + if (wi->num_fifo_pkts) { + mlx5e_tx_wi_dma_unmap(sq, wi, &dma_fifo_cc); + mlx5e_tx_wi_consume_fifo_skbs(sq, wi, cqe, napi_budget); - npkts++; - nbytes += wi->num_bytes; - sqcc += wi->num_wqebbs; - napi_consume_skb(skb, napi_budget); + npkts += wi->num_fifo_pkts; + nbytes += wi->num_bytes; + } } while (!last_wqe); if (unlikely(get_cqe_opcode(cqe) == MLX5_CQE_REQ_ERR)) { @@ -525,13 +828,19 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) return (i == MLX5E_TX_CQ_POLL_BUDGET); } +static void mlx5e_tx_wi_kfree_fifo_skbs(struct mlx5e_txqsq *sq, struct mlx5e_tx_wqe_info *wi) +{ + int i; + + for (i = 0; i < wi->num_fifo_pkts; i++) + dev_kfree_skb_any(mlx5e_skb_fifo_pop(sq)); +} + void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq) { struct mlx5e_tx_wqe_info *wi; u32 dma_fifo_cc, nbytes = 0; u16 ci, sqcc, npkts = 0; - struct sk_buff *skb; - int i; sqcc = sq->cc; dma_fifo_cc = sq->dma_fifo_cc; @@ -539,25 +848,28 @@ void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq) while (sqcc != sq->pc) { ci = mlx5_wq_cyc_ctr2ix(&sq->wq, sqcc); wi = &sq->db.wqe_info[ci]; - skb = wi->skb; - if (!skb) { - mlx5e_ktls_tx_handle_resync_dump_comp(sq, wi, &dma_fifo_cc); - sqcc += wi->num_wqebbs; + sqcc += wi->num_wqebbs; + + if (likely(wi->skb)) { + mlx5e_tx_wi_dma_unmap(sq, wi, &dma_fifo_cc); + dev_kfree_skb_any(wi->skb); + + npkts++; + nbytes += wi->num_bytes; continue; } - for (i = 0; i < wi->num_dma; i++) { - struct mlx5e_sq_dma *dma = - mlx5e_dma_get(sq, dma_fifo_cc++); + if (unlikely(mlx5e_ktls_tx_try_handle_resync_dump_comp(sq, wi, &dma_fifo_cc))) + continue; - mlx5e_tx_dma_unmap(sq->pdev, dma); - } + if (wi->num_fifo_pkts) { + mlx5e_tx_wi_dma_unmap(sq, wi, &dma_fifo_cc); + mlx5e_tx_wi_kfree_fifo_skbs(sq, wi); - dev_kfree_skb_any(skb); - npkts++; - nbytes += wi->num_bytes; - sqcc += wi->num_wqebbs; + npkts += wi->num_fifo_pkts; + nbytes += wi->num_bytes; + } } sq->dma_fifo_cc = dma_fifo_cc; @@ -576,9 +888,34 @@ mlx5i_txwqe_build_datagram(struct mlx5_av *av, u32 dqpn, u32 dqkey, dseg->av.key.qkey.qkey = cpu_to_be32(dqkey); } +static void mlx5i_sq_calc_wqe_attr(struct sk_buff *skb, + const struct mlx5e_tx_attr *attr, + struct mlx5e_tx_wqe_attr *wqe_attr) +{ + u16 ds_cnt = sizeof(struct mlx5i_tx_wqe) / MLX5_SEND_WQE_DS; + u16 ds_cnt_inl = 0; + + ds_cnt += !!attr->headlen + skb_shinfo(skb)->nr_frags; + + if (attr->ihs) { + u16 inl = attr->ihs - INL_HDR_START_SZ; + + ds_cnt_inl = DIV_ROUND_UP(inl, MLX5_SEND_WQE_DS); + ds_cnt += ds_cnt_inl; + } + + *wqe_attr = (struct mlx5e_tx_wqe_attr) { + .ds_cnt = ds_cnt, + .ds_cnt_inl = ds_cnt_inl, + .num_wqebbs = DIV_ROUND_UP(ds_cnt, MLX5_SEND_WQEBB_NUM_DS), + }; +} + void mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, struct mlx5_av *av, u32 dqpn, u32 dqkey, bool xmit_more) { + struct mlx5e_tx_wqe_attr wqe_attr; + struct mlx5e_tx_attr attr; struct mlx5i_tx_wqe *wqe; struct mlx5_wqe_datagram_seg *datagram; @@ -588,47 +925,17 @@ void mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, struct mlx5e_tx_wqe_info *wi; struct mlx5e_sq_stats *stats = sq->stats; - u16 ds_cnt, ds_cnt_inl = 0; - u8 num_wqebbs, opcode; - u16 headlen, ihs, pi; - u32 num_bytes; int num_dma; - __be16 mss; + u16 pi; - /* Calc ihs and ds cnt, no writes to wqe yet */ - ds_cnt = sizeof(*wqe) / MLX5_SEND_WQE_DS; - if (skb_is_gso(skb)) { - opcode = MLX5_OPCODE_LSO; - mss = cpu_to_be16(skb_shinfo(skb)->gso_size); - ihs = mlx5e_tx_get_gso_ihs(sq, skb); - num_bytes = skb->len + (skb_shinfo(skb)->gso_segs - 1) * ihs; - stats->packets += skb_shinfo(skb)->gso_segs; - } else { - u8 mode = mlx5e_tx_wqe_inline_mode(sq, NULL, skb); + mlx5e_sq_xmit_prepare(sq, skb, NULL, &attr); + mlx5i_sq_calc_wqe_attr(skb, &attr, &wqe_attr); - opcode = MLX5_OPCODE_SEND; - mss = 0; - ihs = mlx5e_calc_min_inline(mode, skb); - num_bytes = max_t(unsigned int, skb->len, ETH_ZLEN); - stats->packets++; - } + pi = mlx5e_txqsq_get_next_pi(sq, wqe_attr.num_wqebbs); + wqe = MLX5I_SQ_FETCH_WQE(sq, pi); - stats->bytes += num_bytes; stats->xmit_more += xmit_more; - headlen = skb->len - ihs - skb->data_len; - ds_cnt += !!headlen; - ds_cnt += skb_shinfo(skb)->nr_frags; - - if (ihs) { - ds_cnt_inl = DIV_ROUND_UP(ihs - INL_HDR_START_SZ, MLX5_SEND_WQE_DS); - ds_cnt += ds_cnt_inl; - } - - num_wqebbs = DIV_ROUND_UP(ds_cnt, MLX5_SEND_WQEBB_NUM_DS); - pi = mlx5e_txqsq_get_next_pi(sq, num_wqebbs); - wqe = MLX5I_SQ_FETCH_WQE(sq, pi); - /* fill wqe */ wi = &sq->db.wqe_info[pi]; cseg = &wqe->ctrl; @@ -640,20 +947,20 @@ void mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, mlx5e_txwqe_build_eseg_csum(sq, skb, eseg); - eseg->mss = mss; + eseg->mss = attr.mss; - if (ihs) { - memcpy(eseg->inline_hdr.start, skb->data, ihs); - eseg->inline_hdr.sz = cpu_to_be16(ihs); - dseg += ds_cnt_inl; + if (attr.ihs) { + memcpy(eseg->inline_hdr.start, skb->data, attr.ihs); + eseg->inline_hdr.sz = cpu_to_be16(attr.ihs); + dseg += wqe_attr.ds_cnt_inl; } - num_dma = mlx5e_txwqe_build_dsegs(sq, skb, skb->data + ihs, headlen, dseg); + num_dma = mlx5e_txwqe_build_dsegs(sq, skb, skb->data + attr.ihs, + attr.headlen, dseg); if (unlikely(num_dma < 0)) goto err_drop; - mlx5e_txwqe_complete(sq, skb, opcode, ds_cnt, num_wqebbs, num_bytes, - num_dma, wi, cseg, xmit_more); + mlx5e_txwqe_complete(sq, skb, &attr, &wqe_attr, num_dma, wi, cseg, xmit_more); return; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c index 22a19d391e17..8ebfe782f95e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c @@ -828,8 +828,7 @@ static int create_comp_eqs(struct mlx5_core_dev *dev) INIT_LIST_HEAD(&eq->tasklet_ctx.list); INIT_LIST_HEAD(&eq->tasklet_ctx.process_list); spin_lock_init(&eq->tasklet_ctx.lock); - tasklet_init(&eq->tasklet_ctx.task, mlx5_cq_tasklet_cb, - (unsigned long)&eq->tasklet_ctx); + tasklet_setup(&eq->tasklet_ctx.task, mlx5_cq_tasklet_cb); eq->irq_nb.notifier_call = mlx5_eq_comp_int; param = (struct mlx5_eq_param) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/egress_ofld.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/egress_ofld.c index 07b2acd7e6b3..c3faae67e4d6 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/egress_ofld.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/egress_ofld.c @@ -148,6 +148,11 @@ static void esw_acl_egress_ofld_groups_destroy(struct mlx5_vport *vport) esw_acl_egress_vlan_grp_destroy(vport); } +static bool esw_acl_egress_needed(const struct mlx5_eswitch *esw, u16 vport_num) +{ + return mlx5_eswitch_is_vf_vport(esw, vport_num); +} + int esw_acl_egress_ofld_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { int table_size = 0; @@ -157,6 +162,9 @@ int esw_acl_egress_ofld_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport !MLX5_CAP_GEN(esw->dev, prio_tag_required)) return 0; + if (!esw_acl_egress_needed(esw, vport->vport)) + return 0; + esw_acl_egress_ofld_rules_destroy(vport); if (mlx5_esw_acl_egress_fwd2vport_supported(esw)) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/chains.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/chains.c deleted file mode 100644 index d5bf908dfecd..000000000000 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/chains.c +++ /dev/null @@ -1,944 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB -// Copyright (c) 2020 Mellanox Technologies. - -#include <linux/mlx5/driver.h> -#include <linux/mlx5/mlx5_ifc.h> -#include <linux/mlx5/fs.h> - -#include "esw/chains.h" -#include "en/mapping.h" -#include "mlx5_core.h" -#include "fs_core.h" -#include "eswitch.h" -#include "en.h" -#include "en_tc.h" - -#define esw_chains_priv(esw) ((esw)->fdb_table.offloads.esw_chains_priv) -#define esw_chains_lock(esw) (esw_chains_priv(esw)->lock) -#define esw_chains_ht(esw) (esw_chains_priv(esw)->chains_ht) -#define esw_chains_mapping(esw) (esw_chains_priv(esw)->chains_mapping) -#define esw_prios_ht(esw) (esw_chains_priv(esw)->prios_ht) -#define fdb_pool_left(esw) (esw_chains_priv(esw)->fdb_left) -#define tc_slow_fdb(esw) ((esw)->fdb_table.offloads.slow_fdb) -#define tc_end_fdb(esw) (esw_chains_priv(esw)->tc_end_fdb) -#define fdb_ignore_flow_level_supported(esw) \ - (MLX5_CAP_ESW_FLOWTABLE_FDB((esw)->dev, ignore_flow_level)) -#define fdb_modify_header_fwd_to_table_supported(esw) \ - (MLX5_CAP_ESW_FLOWTABLE((esw)->dev, fdb_modify_header_fwd_to_table)) - -/* Firmware currently has 4 pool of 4 sizes that it supports (ESW_POOLS), - * and a virtual memory region of 16M (ESW_SIZE), this region is duplicated - * for each flow table pool. We can allocate up to 16M of each pool, - * and we keep track of how much we used via get_next_avail_sz_from_pool. - * Firmware doesn't report any of this for now. - * ESW_POOL is expected to be sorted from large to small and match firmware - * pools. - */ -#define ESW_SIZE (16 * 1024 * 1024) -static const unsigned int ESW_POOLS[] = { 4 * 1024 * 1024, - 1 * 1024 * 1024, - 64 * 1024, - 128 }; -#define ESW_FT_TBL_SZ (64 * 1024) - -struct mlx5_esw_chains_priv { - struct rhashtable chains_ht; - struct rhashtable prios_ht; - /* Protects above chains_ht and prios_ht */ - struct mutex lock; - - struct mlx5_flow_table *tc_end_fdb; - struct mapping_ctx *chains_mapping; - - int fdb_left[ARRAY_SIZE(ESW_POOLS)]; -}; - -struct fdb_chain { - struct rhash_head node; - - u32 chain; - - int ref; - int id; - - struct mlx5_eswitch *esw; - struct list_head prios_list; - struct mlx5_flow_handle *restore_rule; - struct mlx5_modify_hdr *miss_modify_hdr; -}; - -struct fdb_prio_key { - u32 chain; - u32 prio; - u32 level; -}; - -struct fdb_prio { - struct rhash_head node; - struct list_head list; - - struct fdb_prio_key key; - - int ref; - - struct fdb_chain *fdb_chain; - struct mlx5_flow_table *fdb; - struct mlx5_flow_table *next_fdb; - struct mlx5_flow_group *miss_group; - struct mlx5_flow_handle *miss_rule; -}; - -static const struct rhashtable_params chain_params = { - .head_offset = offsetof(struct fdb_chain, node), - .key_offset = offsetof(struct fdb_chain, chain), - .key_len = sizeof_field(struct fdb_chain, chain), - .automatic_shrinking = true, -}; - -static const struct rhashtable_params prio_params = { - .head_offset = offsetof(struct fdb_prio, node), - .key_offset = offsetof(struct fdb_prio, key), - .key_len = sizeof_field(struct fdb_prio, key), - .automatic_shrinking = true, -}; - -bool mlx5_esw_chains_prios_supported(struct mlx5_eswitch *esw) -{ - return esw->fdb_table.flags & ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED; -} - -bool mlx5_esw_chains_backwards_supported(struct mlx5_eswitch *esw) -{ - return mlx5_esw_chains_prios_supported(esw) && - fdb_ignore_flow_level_supported(esw); -} - -u32 mlx5_esw_chains_get_chain_range(struct mlx5_eswitch *esw) -{ - if (!mlx5_esw_chains_prios_supported(esw)) - return 1; - - if (fdb_ignore_flow_level_supported(esw)) - return UINT_MAX - 1; - - return FDB_TC_MAX_CHAIN; -} - -u32 mlx5_esw_chains_get_ft_chain(struct mlx5_eswitch *esw) -{ - return mlx5_esw_chains_get_chain_range(esw) + 1; -} - -u32 mlx5_esw_chains_get_prio_range(struct mlx5_eswitch *esw) -{ - if (!mlx5_esw_chains_prios_supported(esw)) - return 1; - - if (fdb_ignore_flow_level_supported(esw)) - return UINT_MAX; - - return FDB_TC_MAX_PRIO; -} - -static unsigned int mlx5_esw_chains_get_level_range(struct mlx5_eswitch *esw) -{ - if (fdb_ignore_flow_level_supported(esw)) - return UINT_MAX; - - return FDB_TC_LEVELS_PER_PRIO; -} - -#define POOL_NEXT_SIZE 0 -static int -mlx5_esw_chains_get_avail_sz_from_pool(struct mlx5_eswitch *esw, - int desired_size) -{ - int i, found_i = -1; - - for (i = ARRAY_SIZE(ESW_POOLS) - 1; i >= 0; i--) { - if (fdb_pool_left(esw)[i] && ESW_POOLS[i] > desired_size) { - found_i = i; - if (desired_size != POOL_NEXT_SIZE) - break; - } - } - - if (found_i != -1) { - --fdb_pool_left(esw)[found_i]; - return ESW_POOLS[found_i]; - } - - return 0; -} - -static void -mlx5_esw_chains_put_sz_to_pool(struct mlx5_eswitch *esw, int sz) -{ - int i; - - for (i = ARRAY_SIZE(ESW_POOLS) - 1; i >= 0; i--) { - if (sz == ESW_POOLS[i]) { - ++fdb_pool_left(esw)[i]; - return; - } - } - - WARN_ONCE(1, "Couldn't find size %d in fdb size pool", sz); -} - -static void -mlx5_esw_chains_init_sz_pool(struct mlx5_eswitch *esw) -{ - u32 fdb_max; - int i; - - fdb_max = 1 << MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, log_max_ft_size); - - for (i = ARRAY_SIZE(ESW_POOLS) - 1; i >= 0; i--) - fdb_pool_left(esw)[i] = - ESW_POOLS[i] <= fdb_max ? ESW_SIZE / ESW_POOLS[i] : 0; -} - -static struct mlx5_flow_table * -mlx5_esw_chains_create_fdb_table(struct mlx5_eswitch *esw, - u32 chain, u32 prio, u32 level) -{ - struct mlx5_flow_table_attr ft_attr = {}; - struct mlx5_flow_namespace *ns; - struct mlx5_flow_table *fdb; - int sz; - - if (esw->offloads.encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE) - ft_attr.flags |= (MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT | - MLX5_FLOW_TABLE_TUNNEL_EN_DECAP); - - sz = (chain == mlx5_esw_chains_get_ft_chain(esw)) ? - mlx5_esw_chains_get_avail_sz_from_pool(esw, ESW_FT_TBL_SZ) : - mlx5_esw_chains_get_avail_sz_from_pool(esw, POOL_NEXT_SIZE); - if (!sz) - return ERR_PTR(-ENOSPC); - ft_attr.max_fte = sz; - - /* We use tc_slow_fdb(esw) as the table's next_ft till - * ignore_flow_level is allowed on FT creation and not just for FTEs. - * Instead caller should add an explicit miss rule if needed. - */ - ft_attr.next_ft = tc_slow_fdb(esw); - - /* The root table(chain 0, prio 1, level 0) is required to be - * connected to the previous prio (FDB_BYPASS_PATH if exists). - * We always create it, as a managed table, in order to align with - * fs_core logic. - */ - if (!fdb_ignore_flow_level_supported(esw) || - (chain == 0 && prio == 1 && level == 0)) { - ft_attr.level = level; - ft_attr.prio = prio - 1; - ns = mlx5_get_fdb_sub_ns(esw->dev, chain); - } else { - ft_attr.flags |= MLX5_FLOW_TABLE_UNMANAGED; - ft_attr.prio = FDB_TC_OFFLOAD; - /* Firmware doesn't allow us to create another level 0 table, - * so we create all unmanaged tables as level 1. - * - * To connect them, we use explicit miss rules with - * ignore_flow_level. Caller is responsible to create - * these rules (if needed). - */ - ft_attr.level = 1; - ns = mlx5_get_flow_namespace(esw->dev, MLX5_FLOW_NAMESPACE_FDB); - } - - ft_attr.autogroup.num_reserved_entries = 2; - ft_attr.autogroup.max_num_groups = esw->params.large_group_num; - fdb = mlx5_create_auto_grouped_flow_table(ns, &ft_attr); - if (IS_ERR(fdb)) { - esw_warn(esw->dev, - "Failed to create FDB table err %d (chain: %d, prio: %d, level: %d, size: %d)\n", - (int)PTR_ERR(fdb), chain, prio, level, sz); - mlx5_esw_chains_put_sz_to_pool(esw, sz); - return fdb; - } - - return fdb; -} - -static void -mlx5_esw_chains_destroy_fdb_table(struct mlx5_eswitch *esw, - struct mlx5_flow_table *fdb) -{ - mlx5_esw_chains_put_sz_to_pool(esw, fdb->max_fte); - mlx5_destroy_flow_table(fdb); -} - -static int -create_fdb_chain_restore(struct fdb_chain *fdb_chain) -{ - char modact[MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto)]; - struct mlx5_eswitch *esw = fdb_chain->esw; - struct mlx5_modify_hdr *mod_hdr; - u32 index; - int err; - - if (fdb_chain->chain == mlx5_esw_chains_get_ft_chain(esw) || - !mlx5_esw_chains_prios_supported(esw)) - return 0; - - err = mapping_add(esw_chains_mapping(esw), &fdb_chain->chain, &index); - if (err) - return err; - if (index == MLX5_FS_DEFAULT_FLOW_TAG) { - /* we got the special default flow tag id, so we won't know - * if we actually marked the packet with the restore rule - * we create. - * - * This case isn't possible with MLX5_FS_DEFAULT_FLOW_TAG = 0. - */ - err = mapping_add(esw_chains_mapping(esw), - &fdb_chain->chain, &index); - mapping_remove(esw_chains_mapping(esw), - MLX5_FS_DEFAULT_FLOW_TAG); - if (err) - return err; - } - - fdb_chain->id = index; - - MLX5_SET(set_action_in, modact, action_type, MLX5_ACTION_TYPE_SET); - MLX5_SET(set_action_in, modact, field, - mlx5e_tc_attr_to_reg_mappings[CHAIN_TO_REG].mfield); - MLX5_SET(set_action_in, modact, offset, - mlx5e_tc_attr_to_reg_mappings[CHAIN_TO_REG].moffset * 8); - MLX5_SET(set_action_in, modact, length, - mlx5e_tc_attr_to_reg_mappings[CHAIN_TO_REG].mlen * 8); - MLX5_SET(set_action_in, modact, data, fdb_chain->id); - mod_hdr = mlx5_modify_header_alloc(esw->dev, MLX5_FLOW_NAMESPACE_FDB, - 1, modact); - if (IS_ERR(mod_hdr)) { - err = PTR_ERR(mod_hdr); - goto err_mod_hdr; - } - fdb_chain->miss_modify_hdr = mod_hdr; - - fdb_chain->restore_rule = esw_add_restore_rule(esw, fdb_chain->id); - if (IS_ERR(fdb_chain->restore_rule)) { - err = PTR_ERR(fdb_chain->restore_rule); - goto err_rule; - } - - return 0; - -err_rule: - mlx5_modify_header_dealloc(esw->dev, fdb_chain->miss_modify_hdr); -err_mod_hdr: - /* Datapath can't find this mapping, so we can safely remove it */ - mapping_remove(esw_chains_mapping(esw), fdb_chain->id); - return err; -} - -static void destroy_fdb_chain_restore(struct fdb_chain *fdb_chain) -{ - struct mlx5_eswitch *esw = fdb_chain->esw; - - if (!fdb_chain->miss_modify_hdr) - return; - - mlx5_del_flow_rules(fdb_chain->restore_rule); - mlx5_modify_header_dealloc(esw->dev, fdb_chain->miss_modify_hdr); - mapping_remove(esw_chains_mapping(esw), fdb_chain->id); -} - -static struct fdb_chain * -mlx5_esw_chains_create_fdb_chain(struct mlx5_eswitch *esw, u32 chain) -{ - struct fdb_chain *fdb_chain = NULL; - int err; - - fdb_chain = kvzalloc(sizeof(*fdb_chain), GFP_KERNEL); - if (!fdb_chain) - return ERR_PTR(-ENOMEM); - - fdb_chain->esw = esw; - fdb_chain->chain = chain; - INIT_LIST_HEAD(&fdb_chain->prios_list); - - err = create_fdb_chain_restore(fdb_chain); - if (err) - goto err_restore; - - err = rhashtable_insert_fast(&esw_chains_ht(esw), &fdb_chain->node, - chain_params); - if (err) - goto err_insert; - - return fdb_chain; - -err_insert: - destroy_fdb_chain_restore(fdb_chain); -err_restore: - kvfree(fdb_chain); - return ERR_PTR(err); -} - -static void -mlx5_esw_chains_destroy_fdb_chain(struct fdb_chain *fdb_chain) -{ - struct mlx5_eswitch *esw = fdb_chain->esw; - - rhashtable_remove_fast(&esw_chains_ht(esw), &fdb_chain->node, - chain_params); - - destroy_fdb_chain_restore(fdb_chain); - kvfree(fdb_chain); -} - -static struct fdb_chain * -mlx5_esw_chains_get_fdb_chain(struct mlx5_eswitch *esw, u32 chain) -{ - struct fdb_chain *fdb_chain; - - fdb_chain = rhashtable_lookup_fast(&esw_chains_ht(esw), &chain, - chain_params); - if (!fdb_chain) { - fdb_chain = mlx5_esw_chains_create_fdb_chain(esw, chain); - if (IS_ERR(fdb_chain)) - return fdb_chain; - } - - fdb_chain->ref++; - - return fdb_chain; -} - -static struct mlx5_flow_handle * -mlx5_esw_chains_add_miss_rule(struct fdb_chain *fdb_chain, - struct mlx5_flow_table *fdb, - struct mlx5_flow_table *next_fdb) -{ - struct mlx5_eswitch *esw = fdb_chain->esw; - struct mlx5_flow_destination dest = {}; - struct mlx5_flow_act act = {}; - - act.flags = FLOW_ACT_IGNORE_FLOW_LEVEL | FLOW_ACT_NO_APPEND; - act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; - dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; - dest.ft = next_fdb; - - if (next_fdb == tc_end_fdb(esw) && - mlx5_esw_chains_prios_supported(esw)) { - act.modify_hdr = fdb_chain->miss_modify_hdr; - act.action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; - } - - return mlx5_add_flow_rules(fdb, NULL, &act, &dest, 1); -} - -static int -mlx5_esw_chains_update_prio_prevs(struct fdb_prio *fdb_prio, - struct mlx5_flow_table *next_fdb) -{ - struct mlx5_flow_handle *miss_rules[FDB_TC_LEVELS_PER_PRIO + 1] = {}; - struct fdb_chain *fdb_chain = fdb_prio->fdb_chain; - struct fdb_prio *pos; - int n = 0, err; - - if (fdb_prio->key.level) - return 0; - - /* Iterate in reverse order until reaching the level 0 rule of - * the previous priority, adding all the miss rules first, so we can - * revert them if any of them fails. - */ - pos = fdb_prio; - list_for_each_entry_continue_reverse(pos, - &fdb_chain->prios_list, - list) { - miss_rules[n] = mlx5_esw_chains_add_miss_rule(fdb_chain, - pos->fdb, - next_fdb); - if (IS_ERR(miss_rules[n])) { - err = PTR_ERR(miss_rules[n]); - goto err_prev_rule; - } - - n++; - if (!pos->key.level) - break; - } - - /* Success, delete old miss rules, and update the pointers. */ - n = 0; - pos = fdb_prio; - list_for_each_entry_continue_reverse(pos, - &fdb_chain->prios_list, - list) { - mlx5_del_flow_rules(pos->miss_rule); - - pos->miss_rule = miss_rules[n]; - pos->next_fdb = next_fdb; - - n++; - if (!pos->key.level) - break; - } - - return 0; - -err_prev_rule: - while (--n >= 0) - mlx5_del_flow_rules(miss_rules[n]); - - return err; -} - -static void -mlx5_esw_chains_put_fdb_chain(struct fdb_chain *fdb_chain) -{ - if (--fdb_chain->ref == 0) - mlx5_esw_chains_destroy_fdb_chain(fdb_chain); -} - -static struct fdb_prio * -mlx5_esw_chains_create_fdb_prio(struct mlx5_eswitch *esw, - u32 chain, u32 prio, u32 level) -{ - int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); - struct mlx5_flow_handle *miss_rule = NULL; - struct mlx5_flow_group *miss_group; - struct fdb_prio *fdb_prio = NULL; - struct mlx5_flow_table *next_fdb; - struct fdb_chain *fdb_chain; - struct mlx5_flow_table *fdb; - struct list_head *pos; - u32 *flow_group_in; - int err; - - fdb_chain = mlx5_esw_chains_get_fdb_chain(esw, chain); - if (IS_ERR(fdb_chain)) - return ERR_CAST(fdb_chain); - - fdb_prio = kvzalloc(sizeof(*fdb_prio), GFP_KERNEL); - flow_group_in = kvzalloc(inlen, GFP_KERNEL); - if (!fdb_prio || !flow_group_in) { - err = -ENOMEM; - goto err_alloc; - } - - /* Chain's prio list is sorted by prio and level. - * And all levels of some prio point to the next prio's level 0. - * Example list (prio, level): - * (3,0)->(3,1)->(5,0)->(5,1)->(6,1)->(7,0) - * In hardware, we will we have the following pointers: - * (3,0) -> (5,0) -> (7,0) -> Slow path - * (3,1) -> (5,0) - * (5,1) -> (7,0) - * (6,1) -> (7,0) - */ - - /* Default miss for each chain: */ - next_fdb = (chain == mlx5_esw_chains_get_ft_chain(esw)) ? - tc_slow_fdb(esw) : - tc_end_fdb(esw); - list_for_each(pos, &fdb_chain->prios_list) { - struct fdb_prio *p = list_entry(pos, struct fdb_prio, list); - - /* exit on first pos that is larger */ - if (prio < p->key.prio || (prio == p->key.prio && - level < p->key.level)) { - /* Get next level 0 table */ - next_fdb = p->key.level == 0 ? p->fdb : p->next_fdb; - break; - } - } - - fdb = mlx5_esw_chains_create_fdb_table(esw, chain, prio, level); - if (IS_ERR(fdb)) { - err = PTR_ERR(fdb); - goto err_create; - } - - MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, - fdb->max_fte - 2); - MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, - fdb->max_fte - 1); - miss_group = mlx5_create_flow_group(fdb, flow_group_in); - if (IS_ERR(miss_group)) { - err = PTR_ERR(miss_group); - goto err_group; - } - - /* Add miss rule to next_fdb */ - miss_rule = mlx5_esw_chains_add_miss_rule(fdb_chain, fdb, next_fdb); - if (IS_ERR(miss_rule)) { - err = PTR_ERR(miss_rule); - goto err_miss_rule; - } - - fdb_prio->miss_group = miss_group; - fdb_prio->miss_rule = miss_rule; - fdb_prio->next_fdb = next_fdb; - fdb_prio->fdb_chain = fdb_chain; - fdb_prio->key.chain = chain; - fdb_prio->key.prio = prio; - fdb_prio->key.level = level; - fdb_prio->fdb = fdb; - - err = rhashtable_insert_fast(&esw_prios_ht(esw), &fdb_prio->node, - prio_params); - if (err) - goto err_insert; - - list_add(&fdb_prio->list, pos->prev); - - /* Table is ready, connect it */ - err = mlx5_esw_chains_update_prio_prevs(fdb_prio, fdb); - if (err) - goto err_update; - - kvfree(flow_group_in); - return fdb_prio; - -err_update: - list_del(&fdb_prio->list); - rhashtable_remove_fast(&esw_prios_ht(esw), &fdb_prio->node, - prio_params); -err_insert: - mlx5_del_flow_rules(miss_rule); -err_miss_rule: - mlx5_destroy_flow_group(miss_group); -err_group: - mlx5_esw_chains_destroy_fdb_table(esw, fdb); -err_create: -err_alloc: - kvfree(fdb_prio); - kvfree(flow_group_in); - mlx5_esw_chains_put_fdb_chain(fdb_chain); - return ERR_PTR(err); -} - -static void -mlx5_esw_chains_destroy_fdb_prio(struct mlx5_eswitch *esw, - struct fdb_prio *fdb_prio) -{ - struct fdb_chain *fdb_chain = fdb_prio->fdb_chain; - - WARN_ON(mlx5_esw_chains_update_prio_prevs(fdb_prio, - fdb_prio->next_fdb)); - - list_del(&fdb_prio->list); - rhashtable_remove_fast(&esw_prios_ht(esw), &fdb_prio->node, - prio_params); - mlx5_del_flow_rules(fdb_prio->miss_rule); - mlx5_destroy_flow_group(fdb_prio->miss_group); - mlx5_esw_chains_destroy_fdb_table(esw, fdb_prio->fdb); - mlx5_esw_chains_put_fdb_chain(fdb_chain); - kvfree(fdb_prio); -} - -struct mlx5_flow_table * -mlx5_esw_chains_get_table(struct mlx5_eswitch *esw, u32 chain, u32 prio, - u32 level) -{ - struct mlx5_flow_table *prev_fts; - struct fdb_prio *fdb_prio; - struct fdb_prio_key key; - int l = 0; - - if ((chain > mlx5_esw_chains_get_chain_range(esw) && - chain != mlx5_esw_chains_get_ft_chain(esw)) || - prio > mlx5_esw_chains_get_prio_range(esw) || - level > mlx5_esw_chains_get_level_range(esw)) - return ERR_PTR(-EOPNOTSUPP); - - /* create earlier levels for correct fs_core lookup when - * connecting tables. - */ - for (l = 0; l < level; l++) { - prev_fts = mlx5_esw_chains_get_table(esw, chain, prio, l); - if (IS_ERR(prev_fts)) { - fdb_prio = ERR_CAST(prev_fts); - goto err_get_prevs; - } - } - - key.chain = chain; - key.prio = prio; - key.level = level; - - mutex_lock(&esw_chains_lock(esw)); - fdb_prio = rhashtable_lookup_fast(&esw_prios_ht(esw), &key, - prio_params); - if (!fdb_prio) { - fdb_prio = mlx5_esw_chains_create_fdb_prio(esw, chain, - prio, level); - if (IS_ERR(fdb_prio)) - goto err_create_prio; - } - - ++fdb_prio->ref; - mutex_unlock(&esw_chains_lock(esw)); - - return fdb_prio->fdb; - -err_create_prio: - mutex_unlock(&esw_chains_lock(esw)); -err_get_prevs: - while (--l >= 0) - mlx5_esw_chains_put_table(esw, chain, prio, l); - return ERR_CAST(fdb_prio); -} - -void -mlx5_esw_chains_put_table(struct mlx5_eswitch *esw, u32 chain, u32 prio, - u32 level) -{ - struct fdb_prio *fdb_prio; - struct fdb_prio_key key; - - key.chain = chain; - key.prio = prio; - key.level = level; - - mutex_lock(&esw_chains_lock(esw)); - fdb_prio = rhashtable_lookup_fast(&esw_prios_ht(esw), &key, - prio_params); - if (!fdb_prio) - goto err_get_prio; - - if (--fdb_prio->ref == 0) - mlx5_esw_chains_destroy_fdb_prio(esw, fdb_prio); - mutex_unlock(&esw_chains_lock(esw)); - - while (level-- > 0) - mlx5_esw_chains_put_table(esw, chain, prio, level); - - return; - -err_get_prio: - mutex_unlock(&esw_chains_lock(esw)); - WARN_ONCE(1, - "Couldn't find table: (chain: %d prio: %d level: %d)", - chain, prio, level); -} - -struct mlx5_flow_table * -mlx5_esw_chains_get_tc_end_ft(struct mlx5_eswitch *esw) -{ - return tc_end_fdb(esw); -} - -struct mlx5_flow_table * -mlx5_esw_chains_create_global_table(struct mlx5_eswitch *esw) -{ - u32 chain, prio, level; - int err; - - if (!fdb_ignore_flow_level_supported(esw)) { - err = -EOPNOTSUPP; - - esw_warn(esw->dev, - "Couldn't create global flow table, ignore_flow_level not supported."); - goto err_ignore; - } - - chain = mlx5_esw_chains_get_chain_range(esw), - prio = mlx5_esw_chains_get_prio_range(esw); - level = mlx5_esw_chains_get_level_range(esw); - - return mlx5_esw_chains_create_fdb_table(esw, chain, prio, level); - -err_ignore: - return ERR_PTR(err); -} - -void -mlx5_esw_chains_destroy_global_table(struct mlx5_eswitch *esw, - struct mlx5_flow_table *ft) -{ - mlx5_esw_chains_destroy_fdb_table(esw, ft); -} - -static int -mlx5_esw_chains_init(struct mlx5_eswitch *esw) -{ - struct mlx5_esw_chains_priv *chains_priv; - struct mlx5_core_dev *dev = esw->dev; - u32 max_flow_counter, fdb_max; - struct mapping_ctx *mapping; - int err; - - chains_priv = kzalloc(sizeof(*chains_priv), GFP_KERNEL); - if (!chains_priv) - return -ENOMEM; - esw_chains_priv(esw) = chains_priv; - - max_flow_counter = (MLX5_CAP_GEN(dev, max_flow_counter_31_16) << 16) | - MLX5_CAP_GEN(dev, max_flow_counter_15_0); - fdb_max = 1 << MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size); - - esw_debug(dev, - "Init esw offloads chains, max counters(%d), groups(%d), max flow table size(%d)\n", - max_flow_counter, esw->params.large_group_num, fdb_max); - - mlx5_esw_chains_init_sz_pool(esw); - - if (!MLX5_CAP_ESW_FLOWTABLE(esw->dev, multi_fdb_encap) && - esw->offloads.encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE) { - esw->fdb_table.flags &= ~ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED; - esw_warn(dev, "Tc chains and priorities offload aren't supported, update firmware if needed\n"); - } else if (!mlx5_eswitch_reg_c1_loopback_enabled(esw)) { - esw->fdb_table.flags &= ~ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED; - esw_warn(dev, "Tc chains and priorities offload aren't supported\n"); - } else if (!fdb_modify_header_fwd_to_table_supported(esw)) { - /* Disabled when ttl workaround is needed, e.g - * when ESWITCH_IPV4_TTL_MODIFY_ENABLE = true in mlxconfig - */ - esw_warn(dev, - "Tc chains and priorities offload aren't supported, check firmware version, or mlxconfig settings\n"); - esw->fdb_table.flags &= ~ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED; - } else { - esw->fdb_table.flags |= ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED; - esw_info(dev, "Supported tc offload range - chains: %u, prios: %u\n", - mlx5_esw_chains_get_chain_range(esw), - mlx5_esw_chains_get_prio_range(esw)); - } - - err = rhashtable_init(&esw_chains_ht(esw), &chain_params); - if (err) - goto init_chains_ht_err; - - err = rhashtable_init(&esw_prios_ht(esw), &prio_params); - if (err) - goto init_prios_ht_err; - - mapping = mapping_create(sizeof(u32), esw_get_max_restore_tag(esw), - true); - if (IS_ERR(mapping)) { - err = PTR_ERR(mapping); - goto mapping_err; - } - esw_chains_mapping(esw) = mapping; - - mutex_init(&esw_chains_lock(esw)); - - return 0; - -mapping_err: - rhashtable_destroy(&esw_prios_ht(esw)); -init_prios_ht_err: - rhashtable_destroy(&esw_chains_ht(esw)); -init_chains_ht_err: - kfree(chains_priv); - return err; -} - -static void -mlx5_esw_chains_cleanup(struct mlx5_eswitch *esw) -{ - mutex_destroy(&esw_chains_lock(esw)); - mapping_destroy(esw_chains_mapping(esw)); - rhashtable_destroy(&esw_prios_ht(esw)); - rhashtable_destroy(&esw_chains_ht(esw)); - - kfree(esw_chains_priv(esw)); -} - -static int -mlx5_esw_chains_open(struct mlx5_eswitch *esw) -{ - struct mlx5_flow_table *ft; - int err; - - /* Create tc_end_fdb(esw) which is the always created ft chain */ - ft = mlx5_esw_chains_get_table(esw, mlx5_esw_chains_get_ft_chain(esw), - 1, 0); - if (IS_ERR(ft)) - return PTR_ERR(ft); - - tc_end_fdb(esw) = ft; - - /* Always open the root for fast path */ - ft = mlx5_esw_chains_get_table(esw, 0, 1, 0); - if (IS_ERR(ft)) { - err = PTR_ERR(ft); - goto level_0_err; - } - - /* Open level 1 for split rules now if prios isn't supported */ - if (!mlx5_esw_chains_prios_supported(esw)) { - err = mlx5_esw_vport_tbl_get(esw); - if (err) - goto level_1_err; - } - - return 0; - -level_1_err: - mlx5_esw_chains_put_table(esw, 0, 1, 0); -level_0_err: - mlx5_esw_chains_put_table(esw, mlx5_esw_chains_get_ft_chain(esw), 1, 0); - return err; -} - -static void -mlx5_esw_chains_close(struct mlx5_eswitch *esw) -{ - if (!mlx5_esw_chains_prios_supported(esw)) - mlx5_esw_vport_tbl_put(esw); - mlx5_esw_chains_put_table(esw, 0, 1, 0); - mlx5_esw_chains_put_table(esw, mlx5_esw_chains_get_ft_chain(esw), 1, 0); -} - -int -mlx5_esw_chains_create(struct mlx5_eswitch *esw) -{ - int err; - - err = mlx5_esw_chains_init(esw); - if (err) - return err; - - err = mlx5_esw_chains_open(esw); - if (err) - goto err_open; - - return 0; - -err_open: - mlx5_esw_chains_cleanup(esw); - return err; -} - -void -mlx5_esw_chains_destroy(struct mlx5_eswitch *esw) -{ - mlx5_esw_chains_close(esw); - mlx5_esw_chains_cleanup(esw); -} - -int -mlx5_esw_chains_get_chain_mapping(struct mlx5_eswitch *esw, u32 chain, - u32 *chain_mapping) -{ - return mapping_add(esw_chains_mapping(esw), &chain, chain_mapping); -} - -int -mlx5_esw_chains_put_chain_mapping(struct mlx5_eswitch *esw, u32 chain_mapping) -{ - return mapping_remove(esw_chains_mapping(esw), chain_mapping); -} - -int mlx5_eswitch_get_chain_for_tag(struct mlx5_eswitch *esw, u32 tag, - u32 *chain) -{ - int err; - - err = mapping_find(esw_chains_mapping(esw), tag, chain); - if (err) { - esw_warn(esw->dev, "Can't find chain for tag: %d\n", tag); - return -ENOENT; - } - - return 0; -} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/chains.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/chains.h deleted file mode 100644 index 7679ac359e31..000000000000 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/chains.h +++ /dev/null @@ -1,68 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ -/* Copyright (c) 2020 Mellanox Technologies. */ - -#ifndef __ML5_ESW_CHAINS_H__ -#define __ML5_ESW_CHAINS_H__ - -#include "eswitch.h" - -#if IS_ENABLED(CONFIG_MLX5_CLS_ACT) - -bool -mlx5_esw_chains_prios_supported(struct mlx5_eswitch *esw); -bool -mlx5_esw_chains_backwards_supported(struct mlx5_eswitch *esw); -u32 -mlx5_esw_chains_get_prio_range(struct mlx5_eswitch *esw); -u32 -mlx5_esw_chains_get_chain_range(struct mlx5_eswitch *esw); -u32 -mlx5_esw_chains_get_ft_chain(struct mlx5_eswitch *esw); - -struct mlx5_flow_table * -mlx5_esw_chains_get_table(struct mlx5_eswitch *esw, u32 chain, u32 prio, - u32 level); -void -mlx5_esw_chains_put_table(struct mlx5_eswitch *esw, u32 chain, u32 prio, - u32 level); - -struct mlx5_flow_table * -mlx5_esw_chains_get_tc_end_ft(struct mlx5_eswitch *esw); - -struct mlx5_flow_table * -mlx5_esw_chains_create_global_table(struct mlx5_eswitch *esw); -void -mlx5_esw_chains_destroy_global_table(struct mlx5_eswitch *esw, - struct mlx5_flow_table *ft); - -int -mlx5_esw_chains_get_chain_mapping(struct mlx5_eswitch *esw, u32 chain, - u32 *chain_mapping); -int -mlx5_esw_chains_put_chain_mapping(struct mlx5_eswitch *esw, - u32 chain_mapping); - -int mlx5_esw_chains_create(struct mlx5_eswitch *esw); -void mlx5_esw_chains_destroy(struct mlx5_eswitch *esw); - -int -mlx5_eswitch_get_chain_for_tag(struct mlx5_eswitch *esw, u32 tag, u32 *chain); - -#else /* CONFIG_MLX5_CLS_ACT */ - -static inline struct mlx5_flow_table * -mlx5_esw_chains_get_table(struct mlx5_eswitch *esw, u32 chain, u32 prio, - u32 level) { return ERR_PTR(-EOPNOTSUPP); } -static inline void -mlx5_esw_chains_put_table(struct mlx5_eswitch *esw, u32 chain, u32 prio, - u32 level) {} - -static inline struct mlx5_flow_table * -mlx5_esw_chains_get_tc_end_ft(struct mlx5_eswitch *esw) { return ERR_PTR(-EOPNOTSUPP); } - -static inline int mlx5_esw_chains_create(struct mlx5_eswitch *esw) { return 0; } -static inline void mlx5_esw_chains_destroy(struct mlx5_eswitch *esw) {} - -#endif /* CONFIG_MLX5_CLS_ACT */ - -#endif /* __ML5_ESW_CHAINS_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c new file mode 100644 index 000000000000..ffff11baa3d0 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c @@ -0,0 +1,124 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* Copyright (c) 2020 Mellanox Technologies Ltd. */ + +#include <linux/mlx5/driver.h> +#include "eswitch.h" + +static void +mlx5_esw_get_port_parent_id(struct mlx5_core_dev *dev, struct netdev_phys_item_id *ppid) +{ + u64 parent_id; + + parent_id = mlx5_query_nic_system_image_guid(dev); + ppid->id_len = sizeof(parent_id); + memcpy(ppid->id, &parent_id, sizeof(parent_id)); +} + +static bool +mlx5_esw_devlink_port_supported(const struct mlx5_eswitch *esw, u16 vport_num) +{ + return vport_num == MLX5_VPORT_UPLINK || + (mlx5_core_is_ecpf(esw->dev) && vport_num == MLX5_VPORT_PF) || + mlx5_eswitch_is_vf_vport(esw, vport_num); +} + +static struct devlink_port *mlx5_esw_dl_port_alloc(struct mlx5_eswitch *esw, u16 vport_num) +{ + struct mlx5_core_dev *dev = esw->dev; + struct devlink_port_attrs attrs = {}; + struct netdev_phys_item_id ppid = {}; + struct devlink_port *dl_port; + u32 controller_num = 0; + bool external; + u16 pfnum; + + dl_port = kzalloc(sizeof(*dl_port), GFP_KERNEL); + if (!dl_port) + return NULL; + + mlx5_esw_get_port_parent_id(dev, &ppid); + pfnum = PCI_FUNC(dev->pdev->devfn); + external = mlx5_core_is_ecpf_esw_manager(dev); + if (external) + controller_num = dev->priv.eswitch->offloads.host_number + 1; + + if (vport_num == MLX5_VPORT_UPLINK) { + attrs.flavour = DEVLINK_PORT_FLAVOUR_PHYSICAL; + attrs.phys.port_number = pfnum; + memcpy(attrs.switch_id.id, ppid.id, ppid.id_len); + attrs.switch_id.id_len = ppid.id_len; + devlink_port_attrs_set(dl_port, &attrs); + } else if (vport_num == MLX5_VPORT_PF) { + memcpy(dl_port->attrs.switch_id.id, ppid.id, ppid.id_len); + dl_port->attrs.switch_id.id_len = ppid.id_len; + devlink_port_attrs_pci_pf_set(dl_port, controller_num, pfnum, external); + } else if (mlx5_eswitch_is_vf_vport(esw, vport_num)) { + memcpy(dl_port->attrs.switch_id.id, ppid.id, ppid.id_len); + dl_port->attrs.switch_id.id_len = ppid.id_len; + devlink_port_attrs_pci_vf_set(dl_port, controller_num, pfnum, + vport_num - 1, external); + } + return dl_port; +} + +static void mlx5_esw_dl_port_free(struct devlink_port *dl_port) +{ + kfree(dl_port); +} + +int mlx5_esw_offloads_devlink_port_register(struct mlx5_eswitch *esw, u16 vport_num) +{ + struct mlx5_core_dev *dev = esw->dev; + struct devlink_port *dl_port; + unsigned int dl_port_index; + struct mlx5_vport *vport; + struct devlink *devlink; + int err; + + if (!mlx5_esw_devlink_port_supported(esw, vport_num)) + return 0; + + vport = mlx5_eswitch_get_vport(esw, vport_num); + if (IS_ERR(vport)) + return PTR_ERR(vport); + + dl_port = mlx5_esw_dl_port_alloc(esw, vport_num); + if (!dl_port) + return -ENOMEM; + + devlink = priv_to_devlink(dev); + dl_port_index = mlx5_esw_vport_to_devlink_port_index(dev, vport_num); + err = devlink_port_register(devlink, dl_port, dl_port_index); + if (err) + goto reg_err; + + vport->dl_port = dl_port; + return 0; + +reg_err: + mlx5_esw_dl_port_free(dl_port); + return err; +} + +void mlx5_esw_offloads_devlink_port_unregister(struct mlx5_eswitch *esw, u16 vport_num) +{ + struct mlx5_vport *vport; + + if (!mlx5_esw_devlink_port_supported(esw, vport_num)) + return; + + vport = mlx5_eswitch_get_vport(esw, vport_num); + if (IS_ERR(vport)) + return; + devlink_port_unregister(vport->dl_port); + mlx5_esw_dl_port_free(vport->dl_port); + vport->dl_port = NULL; +} + +struct devlink_port *mlx5_esw_offloads_devlink_port(struct mlx5_eswitch *esw, u16 vport_num) +{ + struct mlx5_vport *vport; + + vport = mlx5_eswitch_get_vport(esw, vport_num); + return vport->dl_port; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 867d8120b8a5..cf87de94418f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -42,6 +42,7 @@ #include <linux/mlx5/vport.h> #include <linux/mlx5/fs.h> #include "lib/mpfs.h" +#include "lib/fs_chains.h" #include "en/tc_ct.h" #ifdef CONFIG_MLX5_ESWITCH @@ -62,6 +63,9 @@ #define mlx5_esw_has_fwd_fdb(dev) \ MLX5_CAP_ESW_FLOWTABLE(dev, fdb_multi_path_to_table) +#define esw_chains(esw) \ + ((esw)->fdb_table.offloads.esw_chains_priv) + struct vport_ingress { struct mlx5_flow_table *acl; struct mlx5_flow_handle *allow_rule; @@ -152,14 +156,9 @@ struct mlx5_vport { bool enabled; enum mlx5_eswitch_vport_event enabled_events; + struct devlink_port *dl_port; }; -enum offloads_fdb_flags { - ESW_FDB_CHAINS_AND_PRIOS_SUPPORTED = BIT(0), -}; - -struct mlx5_esw_chains_priv; - struct mlx5_eswitch_fdb { union { struct legacy_fdb { @@ -183,7 +182,7 @@ struct mlx5_eswitch_fdb { struct mlx5_flow_handle *miss_rule_multi; int vlan_push_pop_refcount; - struct mlx5_esw_chains_priv *esw_chains_priv; + struct mlx5_fs_chains *esw_chains_priv; struct { DECLARE_HASHTABLE(table, 8); /* Protects vports.table */ @@ -217,6 +216,7 @@ struct mlx5_esw_offload { atomic64_t num_flows; enum devlink_eswitch_encap_mode encap; struct ida vport_metadata_ida; + unsigned int host_number; /* ECPF supports one external host */ }; /* E-Switch MC FDB table hash node */ @@ -329,7 +329,7 @@ struct mlx5_termtbl_handle; bool mlx5_eswitch_termtbl_required(struct mlx5_eswitch *esw, - struct mlx5_esw_flow_attr *attr, + struct mlx5_flow_attr *attr, struct mlx5_flow_act *flow_act, struct mlx5_flow_spec *spec); @@ -349,19 +349,19 @@ mlx5_eswitch_termtbl_put(struct mlx5_eswitch *esw, struct mlx5_flow_handle * mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, struct mlx5_flow_spec *spec, - struct mlx5_esw_flow_attr *attr); + struct mlx5_flow_attr *attr); struct mlx5_flow_handle * mlx5_eswitch_add_fwd_rule(struct mlx5_eswitch *esw, struct mlx5_flow_spec *spec, - struct mlx5_esw_flow_attr *attr); + struct mlx5_flow_attr *attr); void mlx5_eswitch_del_offloaded_rule(struct mlx5_eswitch *esw, struct mlx5_flow_handle *rule, - struct mlx5_esw_flow_attr *attr); + struct mlx5_flow_attr *attr); void mlx5_eswitch_del_fwd_rule(struct mlx5_eswitch *esw, struct mlx5_flow_handle *rule, - struct mlx5_esw_flow_attr *attr); + struct mlx5_flow_attr *attr); struct mlx5_flow_handle * mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, u16 vport, @@ -401,7 +401,6 @@ struct mlx5_esw_flow_attr { int split_count; int out_count; - int action; __be16 vlan_proto[MLX5_FS_VLAN_DEPTH]; u16 vlan_vid[MLX5_FS_VLAN_DEPTH]; u8 vlan_prio[MLX5_FS_VLAN_DEPTH]; @@ -413,19 +412,7 @@ struct mlx5_esw_flow_attr { struct mlx5_core_dev *mdev; struct mlx5_termtbl_handle *termtbl; } dests[MLX5_MAX_FLOW_FWD_VPORTS]; - struct mlx5_modify_hdr *modify_hdr; - u8 inner_match_level; - u8 outer_match_level; - struct mlx5_fc *counter; - u32 chain; - u16 prio; - u32 dest_chain; - u32 flags; - struct mlx5_flow_table *fdb; - struct mlx5_flow_table *dest_ft; - struct mlx5_ct_attr ct_attr; struct mlx5_pkt_reformat *decap_pkt_reformat; - struct mlx5e_tc_flow_parse_attr *parse_attr; }; int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode, @@ -451,9 +438,9 @@ int mlx5_devlink_port_function_hw_addr_set(struct devlink *devlink, void *mlx5_eswitch_get_uplink_priv(struct mlx5_eswitch *esw, u8 rep_type); int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw, - struct mlx5_esw_flow_attr *attr); + struct mlx5_flow_attr *attr); int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw, - struct mlx5_esw_flow_attr *attr); + struct mlx5_flow_attr *attr); int __mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw, u16 vport, u16 vlan, u8 qos, u8 set_flags); @@ -677,6 +664,9 @@ int mlx5_eswitch_load_vf_vports(struct mlx5_eswitch *esw, u16 num_vfs, enum mlx5_eswitch_vport_event enabled_events); void mlx5_eswitch_unload_vf_vports(struct mlx5_eswitch *esw, u16 num_vfs); +int mlx5_esw_offloads_devlink_port_register(struct mlx5_eswitch *esw, u16 vport_num); +void mlx5_esw_offloads_devlink_port_unregister(struct mlx5_eswitch *esw, u16 vport_num); +struct devlink_port *mlx5_esw_offloads_devlink_port(struct mlx5_eswitch *esw, u16 vport_num); #else /* CONFIG_MLX5_ESWITCH */ /* eswitch API stubs */ static inline int mlx5_eswitch_init(struct mlx5_core_dev *dev) { return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 1bcf2609dca8..c9c2962ad49f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -39,12 +39,13 @@ #include "mlx5_core.h" #include "eswitch.h" #include "esw/acl/ofld.h" -#include "esw/chains.h" #include "rdma.h" #include "en.h" #include "fs_core.h" #include "lib/devcom.h" #include "lib/eq.h" +#include "lib/fs_chains.h" +#include "en_tc.h" /* There are two match-all miss flows, one for unicast dst mac and * one for multicast. @@ -66,6 +67,12 @@ struct mlx5_vport_key { u16 vhca_id; } __packed; +struct mlx5_vport_tbl_attr { + u16 chain; + u16 prio; + u16 vport; +}; + struct mlx5_vport_table { struct hlist_node hlist; struct mlx5_flow_table *fdb; @@ -94,10 +101,10 @@ esw_vport_tbl_create(struct mlx5_eswitch *esw, struct mlx5_flow_namespace *ns) } static u32 flow_attr_to_vport_key(struct mlx5_eswitch *esw, - struct mlx5_esw_flow_attr *attr, + struct mlx5_vport_tbl_attr *attr, struct mlx5_vport_key *key) { - key->vport = attr->in_rep->vport; + key->vport = attr->vport; key->chain = attr->chain; key->prio = attr->prio; key->vhca_id = MLX5_CAP_GEN(esw->dev, vhca_id); @@ -118,7 +125,7 @@ esw_vport_tbl_lookup(struct mlx5_eswitch *esw, struct mlx5_vport_key *skey, u32 } static void -esw_vport_tbl_put(struct mlx5_eswitch *esw, struct mlx5_esw_flow_attr *attr) +esw_vport_tbl_put(struct mlx5_eswitch *esw, struct mlx5_vport_tbl_attr *attr) { struct mlx5_vport_table *e; struct mlx5_vport_key key; @@ -138,7 +145,7 @@ out: } static struct mlx5_flow_table * -esw_vport_tbl_get(struct mlx5_eswitch *esw, struct mlx5_esw_flow_attr *attr) +esw_vport_tbl_get(struct mlx5_eswitch *esw, struct mlx5_vport_tbl_attr *attr) { struct mlx5_core_dev *dev = esw->dev; struct mlx5_flow_namespace *ns; @@ -189,16 +196,15 @@ err_alloc: int mlx5_esw_vport_tbl_get(struct mlx5_eswitch *esw) { - struct mlx5_esw_flow_attr attr = {}; - struct mlx5_eswitch_rep rep = {}; + struct mlx5_vport_tbl_attr attr; struct mlx5_flow_table *fdb; struct mlx5_vport *vport; int i; + attr.chain = 0; attr.prio = 1; - attr.in_rep = &rep; mlx5_esw_for_all_vports(esw, i, vport) { - attr.in_rep->vport = vport->vport; + attr.vport = vport->vport; fdb = esw_vport_tbl_get(esw, &attr); if (IS_ERR(fdb)) goto out; @@ -212,15 +218,14 @@ out: void mlx5_esw_vport_tbl_put(struct mlx5_eswitch *esw) { - struct mlx5_esw_flow_attr attr = {}; - struct mlx5_eswitch_rep rep = {}; + struct mlx5_vport_tbl_attr attr; struct mlx5_vport *vport; int i; + attr.chain = 0; attr.prio = 1; - attr.in_rep = &rep; mlx5_esw_for_all_vports(esw, i, vport) { - attr.in_rep->vport = vport->vport; + attr.vport = vport->vport; esw_vport_tbl_put(esw, &attr); } } @@ -242,8 +247,11 @@ mlx5_eswitch_set_rule_flow_source(struct mlx5_eswitch *esw, struct mlx5_esw_flow_attr *attr) { if (MLX5_CAP_ESW_FLOWTABLE(esw->dev, flow_source) && - attr && attr->in_rep && attr->in_rep->vport == MLX5_VPORT_UPLINK) - spec->flow_context.flow_source = MLX5_FLOW_CONTEXT_FLOW_SOURCE_UPLINK; + attr && attr->in_rep) + spec->flow_context.flow_source = + attr->in_rep->vport == MLX5_VPORT_UPLINK ? + MLX5_FLOW_CONTEXT_FLOW_SOURCE_UPLINK : + MLX5_FLOW_CONTEXT_FLOW_SOURCE_LOCAL_VPORT; } static void @@ -290,11 +298,14 @@ mlx5_eswitch_set_rule_source_port(struct mlx5_eswitch *esw, struct mlx5_flow_handle * mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, struct mlx5_flow_spec *spec, - struct mlx5_esw_flow_attr *attr) + struct mlx5_flow_attr *attr) { struct mlx5_flow_destination dest[MLX5_MAX_FLOW_FWD_VPORTS + 1] = {}; struct mlx5_flow_act flow_act = { .flags = FLOW_ACT_NO_APPEND, }; - bool split = !!(attr->split_count); + struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr; + struct mlx5_fs_chains *chains = esw_chains(esw); + bool split = !!(esw_attr->split_count); + struct mlx5_vport_tbl_attr fwd_attr; struct mlx5_flow_handle *rule; struct mlx5_flow_table *fdb; int j, i = 0; @@ -308,13 +319,13 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, flow_act.action &= ~(MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH | MLX5_FLOW_CONTEXT_ACTION_VLAN_POP); else if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) { - flow_act.vlan[0].ethtype = ntohs(attr->vlan_proto[0]); - flow_act.vlan[0].vid = attr->vlan_vid[0]; - flow_act.vlan[0].prio = attr->vlan_prio[0]; + flow_act.vlan[0].ethtype = ntohs(esw_attr->vlan_proto[0]); + flow_act.vlan[0].vid = esw_attr->vlan_vid[0]; + flow_act.vlan[0].prio = esw_attr->vlan_prio[0]; if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH_2) { - flow_act.vlan[1].ethtype = ntohs(attr->vlan_proto[1]); - flow_act.vlan[1].vid = attr->vlan_vid[1]; - flow_act.vlan[1].prio = attr->vlan_prio[1]; + flow_act.vlan[1].ethtype = ntohs(esw_attr->vlan_proto[1]); + flow_act.vlan[1].vid = esw_attr->vlan_vid[1]; + flow_act.vlan[1].prio = esw_attr->vlan_prio[1]; } } @@ -329,12 +340,12 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, } else if (attr->flags & MLX5_ESW_ATTR_FLAG_SLOW_PATH) { flow_act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL; dest[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; - dest[i].ft = mlx5_esw_chains_get_tc_end_ft(esw); + dest[i].ft = mlx5_chains_get_tc_end_ft(chains); i++; } else if (attr->dest_chain) { flow_act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL; - ft = mlx5_esw_chains_get_table(esw, attr->dest_chain, - 1, 0); + ft = mlx5_chains_get_table(chains, attr->dest_chain, + 1, 0); if (IS_ERR(ft)) { rule = ERR_CAST(ft); goto err_create_goto_table; @@ -344,28 +355,29 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, dest[i].ft = ft; i++; } else { - for (j = attr->split_count; j < attr->out_count; j++) { + for (j = esw_attr->split_count; j < esw_attr->out_count; j++) { dest[i].type = MLX5_FLOW_DESTINATION_TYPE_VPORT; - dest[i].vport.num = attr->dests[j].rep->vport; + dest[i].vport.num = esw_attr->dests[j].rep->vport; dest[i].vport.vhca_id = - MLX5_CAP_GEN(attr->dests[j].mdev, vhca_id); + MLX5_CAP_GEN(esw_attr->dests[j].mdev, vhca_id); if (MLX5_CAP_ESW(esw->dev, merged_eswitch)) dest[i].vport.flags |= MLX5_FLOW_DEST_VPORT_VHCA_ID; - if (attr->dests[j].flags & MLX5_ESW_DEST_ENCAP) { + if (esw_attr->dests[j].flags & MLX5_ESW_DEST_ENCAP) { flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT; - flow_act.pkt_reformat = attr->dests[j].pkt_reformat; + flow_act.pkt_reformat = + esw_attr->dests[j].pkt_reformat; dest[i].vport.flags |= MLX5_FLOW_DEST_VPORT_REFORMAT_ID; dest[i].vport.pkt_reformat = - attr->dests[j].pkt_reformat; + esw_attr->dests[j].pkt_reformat; } i++; } } } - if (attr->decap_pkt_reformat) - flow_act.pkt_reformat = attr->decap_pkt_reformat; + if (esw_attr->decap_pkt_reformat) + flow_act.pkt_reformat = esw_attr->decap_pkt_reformat; if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_COUNT) { dest[i].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; @@ -382,26 +394,30 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, flow_act.modify_hdr = attr->modify_hdr; if (split) { - fdb = esw_vport_tbl_get(esw, attr); + fwd_attr.chain = attr->chain; + fwd_attr.prio = attr->prio; + fwd_attr.vport = esw_attr->in_rep->vport; + + fdb = esw_vport_tbl_get(esw, &fwd_attr); } else { if (attr->chain || attr->prio) - fdb = mlx5_esw_chains_get_table(esw, attr->chain, - attr->prio, 0); + fdb = mlx5_chains_get_table(chains, attr->chain, + attr->prio, 0); else - fdb = attr->fdb; + fdb = attr->ft; if (!(attr->flags & MLX5_ESW_ATTR_FLAG_NO_IN_PORT)) - mlx5_eswitch_set_rule_source_port(esw, spec, attr); + mlx5_eswitch_set_rule_source_port(esw, spec, esw_attr); } if (IS_ERR(fdb)) { rule = ERR_CAST(fdb); goto err_esw_get; } - mlx5_eswitch_set_rule_flow_source(esw, spec, attr); + mlx5_eswitch_set_rule_flow_source(esw, spec, esw_attr); if (mlx5_eswitch_termtbl_required(esw, attr, &flow_act, spec)) - rule = mlx5_eswitch_add_termtbl_rule(esw, fdb, spec, attr, + rule = mlx5_eswitch_add_termtbl_rule(esw, fdb, spec, esw_attr, &flow_act, dest, i); else rule = mlx5_add_flow_rules(fdb, spec, &flow_act, dest, i); @@ -414,12 +430,12 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, err_add_rule: if (split) - esw_vport_tbl_put(esw, attr); + esw_vport_tbl_put(esw, &fwd_attr); else if (attr->chain || attr->prio) - mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, 0); + mlx5_chains_put_table(chains, attr->chain, attr->prio, 0); err_esw_get: if (!(attr->flags & MLX5_ESW_ATTR_FLAG_SLOW_PATH) && attr->dest_chain) - mlx5_esw_chains_put_table(esw, attr->dest_chain, 1, 0); + mlx5_chains_put_table(chains, attr->dest_chain, 1, 0); err_create_goto_table: return rule; } @@ -427,46 +443,51 @@ err_create_goto_table: struct mlx5_flow_handle * mlx5_eswitch_add_fwd_rule(struct mlx5_eswitch *esw, struct mlx5_flow_spec *spec, - struct mlx5_esw_flow_attr *attr) + struct mlx5_flow_attr *attr) { struct mlx5_flow_destination dest[MLX5_MAX_FLOW_FWD_VPORTS + 1] = {}; struct mlx5_flow_act flow_act = { .flags = FLOW_ACT_NO_APPEND, }; + struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr; + struct mlx5_fs_chains *chains = esw_chains(esw); + struct mlx5_vport_tbl_attr fwd_attr; struct mlx5_flow_table *fast_fdb; struct mlx5_flow_table *fwd_fdb; struct mlx5_flow_handle *rule; int i; - fast_fdb = mlx5_esw_chains_get_table(esw, attr->chain, attr->prio, 0); + fast_fdb = mlx5_chains_get_table(chains, attr->chain, attr->prio, 0); if (IS_ERR(fast_fdb)) { rule = ERR_CAST(fast_fdb); goto err_get_fast; } - fwd_fdb = esw_vport_tbl_get(esw, attr); + fwd_attr.chain = attr->chain; + fwd_attr.prio = attr->prio; + fwd_attr.vport = esw_attr->in_rep->vport; + fwd_fdb = esw_vport_tbl_get(esw, &fwd_attr); if (IS_ERR(fwd_fdb)) { rule = ERR_CAST(fwd_fdb); goto err_get_fwd; } flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; - for (i = 0; i < attr->split_count; i++) { + for (i = 0; i < esw_attr->split_count; i++) { dest[i].type = MLX5_FLOW_DESTINATION_TYPE_VPORT; - dest[i].vport.num = attr->dests[i].rep->vport; + dest[i].vport.num = esw_attr->dests[i].rep->vport; dest[i].vport.vhca_id = - MLX5_CAP_GEN(attr->dests[i].mdev, vhca_id); + MLX5_CAP_GEN(esw_attr->dests[i].mdev, vhca_id); if (MLX5_CAP_ESW(esw->dev, merged_eswitch)) dest[i].vport.flags |= MLX5_FLOW_DEST_VPORT_VHCA_ID; - if (attr->dests[i].flags & MLX5_ESW_DEST_ENCAP) { + if (esw_attr->dests[i].flags & MLX5_ESW_DEST_ENCAP) { dest[i].vport.flags |= MLX5_FLOW_DEST_VPORT_REFORMAT_ID; - dest[i].vport.pkt_reformat = attr->dests[i].pkt_reformat; + dest[i].vport.pkt_reformat = esw_attr->dests[i].pkt_reformat; } } dest[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; dest[i].ft = fwd_fdb, i++; - mlx5_eswitch_set_rule_source_port(esw, spec, attr); - mlx5_eswitch_set_rule_flow_source(esw, spec, attr); + mlx5_eswitch_set_rule_source_port(esw, spec, esw_attr); if (attr->outer_match_level != MLX5_MATCH_NONE) spec->match_criteria_enable |= MLX5_MATCH_OUTER_HEADERS; @@ -481,9 +502,9 @@ mlx5_eswitch_add_fwd_rule(struct mlx5_eswitch *esw, return rule; add_err: - esw_vport_tbl_put(esw, attr); + esw_vport_tbl_put(esw, &fwd_attr); err_get_fwd: - mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, 0); + mlx5_chains_put_table(chains, attr->chain, attr->prio, 0); err_get_fast: return rule; } @@ -491,10 +512,13 @@ err_get_fast: static void __mlx5_eswitch_del_rule(struct mlx5_eswitch *esw, struct mlx5_flow_handle *rule, - struct mlx5_esw_flow_attr *attr, + struct mlx5_flow_attr *attr, bool fwd_rule) { - bool split = (attr->split_count > 0); + struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr; + struct mlx5_fs_chains *chains = esw_chains(esw); + bool split = (esw_attr->split_count > 0); + struct mlx5_vport_tbl_attr fwd_attr; int i; mlx5_del_flow_rules(rule); @@ -502,31 +526,36 @@ __mlx5_eswitch_del_rule(struct mlx5_eswitch *esw, if (!(attr->flags & MLX5_ESW_ATTR_FLAG_SLOW_PATH)) { /* unref the term table */ for (i = 0; i < MLX5_MAX_FLOW_FWD_VPORTS; i++) { - if (attr->dests[i].termtbl) - mlx5_eswitch_termtbl_put(esw, attr->dests[i].termtbl); + if (esw_attr->dests[i].termtbl) + mlx5_eswitch_termtbl_put(esw, esw_attr->dests[i].termtbl); } } atomic64_dec(&esw->offloads.num_flows); + if (fwd_rule || split) { + fwd_attr.chain = attr->chain; + fwd_attr.prio = attr->prio; + fwd_attr.vport = esw_attr->in_rep->vport; + } + if (fwd_rule) { - esw_vport_tbl_put(esw, attr); - mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, 0); + esw_vport_tbl_put(esw, &fwd_attr); + mlx5_chains_put_table(chains, attr->chain, attr->prio, 0); } else { if (split) - esw_vport_tbl_put(esw, attr); + esw_vport_tbl_put(esw, &fwd_attr); else if (attr->chain || attr->prio) - mlx5_esw_chains_put_table(esw, attr->chain, attr->prio, - 0); + mlx5_chains_put_table(chains, attr->chain, attr->prio, 0); if (attr->dest_chain) - mlx5_esw_chains_put_table(esw, attr->dest_chain, 1, 0); + mlx5_chains_put_table(chains, attr->dest_chain, 1, 0); } } void mlx5_eswitch_del_offloaded_rule(struct mlx5_eswitch *esw, struct mlx5_flow_handle *rule, - struct mlx5_esw_flow_attr *attr) + struct mlx5_flow_attr *attr) { __mlx5_eswitch_del_rule(esw, rule, attr, false); } @@ -534,7 +563,7 @@ mlx5_eswitch_del_offloaded_rule(struct mlx5_eswitch *esw, void mlx5_eswitch_del_fwd_rule(struct mlx5_eswitch *esw, struct mlx5_flow_handle *rule, - struct mlx5_esw_flow_attr *attr) + struct mlx5_flow_attr *attr) { __mlx5_eswitch_del_rule(esw, rule, attr, true); } @@ -611,9 +640,10 @@ out_notsupp: } int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw, - struct mlx5_esw_flow_attr *attr) + struct mlx5_flow_attr *attr) { struct offloads_fdb *offloads = &esw->fdb_table.offloads; + struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr; struct mlx5_eswitch_rep *vport = NULL; bool push, pop, fwd; int err = 0; @@ -629,17 +659,17 @@ int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw, mutex_lock(&esw->state_lock); - err = esw_add_vlan_action_check(attr, push, pop, fwd); + err = esw_add_vlan_action_check(esw_attr, push, pop, fwd); if (err) goto unlock; attr->flags &= ~MLX5_ESW_ATTR_FLAG_VLAN_HANDLED; - vport = esw_vlan_action_get_vport(attr, push, pop); + vport = esw_vlan_action_get_vport(esw_attr, push, pop); if (!push && !pop && fwd) { /* tracks VF --> wire rules without vlan push action */ - if (attr->dests[0].rep->vport == MLX5_VPORT_UPLINK) { + if (esw_attr->dests[0].rep->vport == MLX5_VPORT_UPLINK) { vport->vlan_refcount++; attr->flags |= MLX5_ESW_ATTR_FLAG_VLAN_HANDLED; } @@ -662,11 +692,11 @@ int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw, if (vport->vlan_refcount) goto skip_set_push; - err = __mlx5_eswitch_set_vport_vlan(esw, vport->vport, attr->vlan_vid[0], 0, - SET_VLAN_INSERT | SET_VLAN_STRIP); + err = __mlx5_eswitch_set_vport_vlan(esw, vport->vport, esw_attr->vlan_vid[0], + 0, SET_VLAN_INSERT | SET_VLAN_STRIP); if (err) goto out; - vport->vlan = attr->vlan_vid[0]; + vport->vlan = esw_attr->vlan_vid[0]; skip_set_push: vport->vlan_refcount++; } @@ -679,9 +709,10 @@ unlock: } int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw, - struct mlx5_esw_flow_attr *attr) + struct mlx5_flow_attr *attr) { struct offloads_fdb *offloads = &esw->fdb_table.offloads; + struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr; struct mlx5_eswitch_rep *vport = NULL; bool push, pop, fwd; int err = 0; @@ -699,11 +730,11 @@ int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw, mutex_lock(&esw->state_lock); - vport = esw_vlan_action_get_vport(attr, push, pop); + vport = esw_vlan_action_get_vport(esw_attr, push, pop); if (!push && !pop && fwd) { /* tracks VF --> wire rules without vlan push action */ - if (attr->dests[0].rep->vport == MLX5_VPORT_UPLINK) + if (esw_attr->dests[0].rep->vport == MLX5_VPORT_UPLINK) vport->vlan_refcount--; goto out; @@ -1137,6 +1168,126 @@ static void esw_set_flow_group_source_port(struct mlx5_eswitch *esw, } } +#if IS_ENABLED(CONFIG_MLX5_CLS_ACT) +#define fdb_modify_header_fwd_to_table_supported(esw) \ + (MLX5_CAP_ESW_FLOWTABLE((esw)->dev, fdb_modify_header_fwd_to_table)) +static void esw_init_chains_offload_flags(struct mlx5_eswitch *esw, u32 *flags) +{ + struct mlx5_core_dev *dev = esw->dev; + + if (MLX5_CAP_ESW_FLOWTABLE_FDB(dev, ignore_flow_level)) + *flags |= MLX5_CHAINS_IGNORE_FLOW_LEVEL_SUPPORTED; + + if (!MLX5_CAP_ESW_FLOWTABLE(dev, multi_fdb_encap) && + esw->offloads.encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE) { + *flags &= ~MLX5_CHAINS_AND_PRIOS_SUPPORTED; + esw_warn(dev, "Tc chains and priorities offload aren't supported, update firmware if needed\n"); + } else if (!mlx5_eswitch_reg_c1_loopback_enabled(esw)) { + *flags &= ~MLX5_CHAINS_AND_PRIOS_SUPPORTED; + esw_warn(dev, "Tc chains and priorities offload aren't supported\n"); + } else if (!fdb_modify_header_fwd_to_table_supported(esw)) { + /* Disabled when ttl workaround is needed, e.g + * when ESWITCH_IPV4_TTL_MODIFY_ENABLE = true in mlxconfig + */ + esw_warn(dev, + "Tc chains and priorities offload aren't supported, check firmware version, or mlxconfig settings\n"); + *flags &= ~MLX5_CHAINS_AND_PRIOS_SUPPORTED; + } else { + *flags |= MLX5_CHAINS_AND_PRIOS_SUPPORTED; + esw_info(dev, "Supported tc chains and prios offload\n"); + } + + if (esw->offloads.encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE) + *flags |= MLX5_CHAINS_FT_TUNNEL_SUPPORTED; +} + +static int +esw_chains_create(struct mlx5_eswitch *esw, struct mlx5_flow_table *miss_fdb) +{ + struct mlx5_core_dev *dev = esw->dev; + struct mlx5_flow_table *nf_ft, *ft; + struct mlx5_chains_attr attr = {}; + struct mlx5_fs_chains *chains; + u32 fdb_max; + int err; + + fdb_max = 1 << MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size); + + esw_init_chains_offload_flags(esw, &attr.flags); + attr.ns = MLX5_FLOW_NAMESPACE_FDB; + attr.max_ft_sz = fdb_max; + attr.max_grp_num = esw->params.large_group_num; + attr.default_ft = miss_fdb; + attr.max_restore_tag = esw_get_max_restore_tag(esw); + + chains = mlx5_chains_create(dev, &attr); + if (IS_ERR(chains)) { + err = PTR_ERR(chains); + esw_warn(dev, "Failed to create fdb chains err(%d)\n", err); + return err; + } + + esw->fdb_table.offloads.esw_chains_priv = chains; + + /* Create tc_end_ft which is the always created ft chain */ + nf_ft = mlx5_chains_get_table(chains, mlx5_chains_get_nf_ft_chain(chains), + 1, 0); + if (IS_ERR(nf_ft)) { + err = PTR_ERR(nf_ft); + goto nf_ft_err; + } + + /* Always open the root for fast path */ + ft = mlx5_chains_get_table(chains, 0, 1, 0); + if (IS_ERR(ft)) { + err = PTR_ERR(ft); + goto level_0_err; + } + + /* Open level 1 for split fdb rules now if prios isn't supported */ + if (!mlx5_chains_prios_supported(chains)) { + err = mlx5_esw_vport_tbl_get(esw); + if (err) + goto level_1_err; + } + + mlx5_chains_set_end_ft(chains, nf_ft); + + return 0; + +level_1_err: + mlx5_chains_put_table(chains, 0, 1, 0); +level_0_err: + mlx5_chains_put_table(chains, mlx5_chains_get_nf_ft_chain(chains), 1, 0); +nf_ft_err: + mlx5_chains_destroy(chains); + esw->fdb_table.offloads.esw_chains_priv = NULL; + + return err; +} + +static void +esw_chains_destroy(struct mlx5_eswitch *esw, struct mlx5_fs_chains *chains) +{ + if (!mlx5_chains_prios_supported(chains)) + mlx5_esw_vport_tbl_put(esw); + mlx5_chains_put_table(chains, 0, 1, 0); + mlx5_chains_put_table(chains, mlx5_chains_get_nf_ft_chain(chains), 1, 0); + mlx5_chains_destroy(chains); +} + +#else /* CONFIG_MLX5_CLS_ACT */ + +static int +esw_chains_create(struct mlx5_eswitch *esw, struct mlx5_flow_table *miss_fdb) +{ return 0; } + +static void +esw_chains_destroy(struct mlx5_eswitch *esw, struct mlx5_fs_chains *chains) +{} + +#endif + static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw) { int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); @@ -1192,9 +1343,9 @@ static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw) } esw->fdb_table.offloads.slow_fdb = fdb; - err = mlx5_esw_chains_create(esw); + err = esw_chains_create(esw, fdb); if (err) { - esw_warn(dev, "Failed to create fdb chains err(%d)\n", err); + esw_warn(dev, "Failed to open fdb chains err(%d)\n", err); goto fdb_chains_err; } @@ -1288,7 +1439,7 @@ miss_err: peer_miss_err: mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_grp); send_vport_err: - mlx5_esw_chains_destroy(esw); + esw_chains_destroy(esw, esw_chains(esw)); fdb_chains_err: mlx5_destroy_flow_table(esw->fdb_table.offloads.slow_fdb); slow_fdb_err: @@ -1312,7 +1463,8 @@ static void esw_destroy_offloads_fdb_tables(struct mlx5_eswitch *esw) mlx5_destroy_flow_group(esw->fdb_table.offloads.peer_miss_grp); mlx5_destroy_flow_group(esw->fdb_table.offloads.miss_grp); - mlx5_esw_chains_destroy(esw); + esw_chains_destroy(esw, esw_chains(esw)); + mlx5_destroy_flow_table(esw->fdb_table.offloads.slow_fdb); /* Holds true only as long as DMFS is the default */ mlx5_flow_namespace_set_mode(esw->fdb_table.offloads.ns, @@ -1671,15 +1823,12 @@ static void __unload_reps_all_vport(struct mlx5_eswitch *esw, u8 rep_type) __esw_offloads_unload_rep(esw, rep, rep_type); } -int esw_offloads_load_rep(struct mlx5_eswitch *esw, u16 vport_num) +static int mlx5_esw_offloads_rep_load(struct mlx5_eswitch *esw, u16 vport_num) { struct mlx5_eswitch_rep *rep; int rep_type; int err; - if (esw->mode != MLX5_ESWITCH_OFFLOADS) - return 0; - rep = mlx5_eswitch_get_rep(esw, vport_num); for (rep_type = 0; rep_type < NUM_REP_TYPES; rep_type++) if (atomic_cmpxchg(&rep->rep_data[rep_type].state, @@ -1698,19 +1847,46 @@ err_reps: return err; } -void esw_offloads_unload_rep(struct mlx5_eswitch *esw, u16 vport_num) +static void mlx5_esw_offloads_rep_unload(struct mlx5_eswitch *esw, u16 vport_num) { struct mlx5_eswitch_rep *rep; int rep_type; - if (esw->mode != MLX5_ESWITCH_OFFLOADS) - return; - rep = mlx5_eswitch_get_rep(esw, vport_num); for (rep_type = NUM_REP_TYPES - 1; rep_type >= 0; rep_type--) __esw_offloads_unload_rep(esw, rep, rep_type); } +int esw_offloads_load_rep(struct mlx5_eswitch *esw, u16 vport_num) +{ + int err; + + if (esw->mode != MLX5_ESWITCH_OFFLOADS) + return 0; + + err = mlx5_esw_offloads_devlink_port_register(esw, vport_num); + if (err) + return err; + + err = mlx5_esw_offloads_rep_load(esw, vport_num); + if (err) + goto load_err; + return err; + +load_err: + mlx5_esw_offloads_devlink_port_unregister(esw, vport_num); + return err; +} + +void esw_offloads_unload_rep(struct mlx5_eswitch *esw, u16 vport_num) +{ + if (esw->mode != MLX5_ESWITCH_OFFLOADS) + return; + + mlx5_esw_offloads_rep_unload(esw, vport_num); + mlx5_esw_offloads_devlink_port_unregister(esw, vport_num); +} + #define ESW_OFFLOADS_DEVCOM_PAIR (0) #define ESW_OFFLOADS_DEVCOM_UNPAIR (1) @@ -1868,53 +2044,38 @@ esw_check_vport_match_metadata_supported(const struct mlx5_eswitch *esw) return true; } -static bool -esw_check_vport_match_metadata_mandatory(const struct mlx5_eswitch *esw) -{ - return mlx5_core_mp_enabled(esw->dev); -} - -static bool esw_use_vport_metadata(const struct mlx5_eswitch *esw) -{ - return esw_check_vport_match_metadata_mandatory(esw) && - esw_check_vport_match_metadata_supported(esw); -} - u32 mlx5_esw_match_metadata_alloc(struct mlx5_eswitch *esw) { - u32 num_vports = GENMASK(ESW_VPORT_BITS - 1, 0) - 1; - u32 vhca_id_mask = GENMASK(ESW_VHCA_ID_BITS - 1, 0); - u32 vhca_id = MLX5_CAP_GEN(esw->dev, vhca_id); - u32 start; - u32 end; + u32 vport_end_ida = (1 << ESW_VPORT_BITS) - 1; + u32 max_pf_num = (1 << ESW_PFNUM_BITS) - 1; + u32 pf_num; int id; - /* Make sure the vhca_id fits the ESW_VHCA_ID_BITS */ - WARN_ON_ONCE(vhca_id >= BIT(ESW_VHCA_ID_BITS)); - - /* Trim vhca_id to ESW_VHCA_ID_BITS */ - vhca_id &= vhca_id_mask; - - start = (vhca_id << ESW_VPORT_BITS); - end = start + num_vports; - if (!vhca_id) - start += 1; /* zero is reserved/invalid metadata */ - id = ida_alloc_range(&esw->offloads.vport_metadata_ida, start, end, GFP_KERNEL); + /* Only 4 bits of pf_num */ + pf_num = PCI_FUNC(esw->dev->pdev->devfn); + if (pf_num > max_pf_num) + return 0; - return (id < 0) ? 0 : id; + /* Metadata is 4 bits of PFNUM and 12 bits of unique id */ + /* Use only non-zero vport_id (1-4095) for all PF's */ + id = ida_alloc_range(&esw->offloads.vport_metadata_ida, 1, vport_end_ida, GFP_KERNEL); + if (id < 0) + return 0; + id = (pf_num << ESW_VPORT_BITS) | id; + return id; } void mlx5_esw_match_metadata_free(struct mlx5_eswitch *esw, u32 metadata) { - ida_free(&esw->offloads.vport_metadata_ida, metadata); + u32 vport_bit_mask = (1 << ESW_VPORT_BITS) - 1; + + /* Metadata contains only 12 bits of actual ida id */ + ida_free(&esw->offloads.vport_metadata_ida, metadata & vport_bit_mask); } static int esw_offloads_vport_metadata_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { - if (vport->vport == MLX5_VPORT_UPLINK) - return 0; - vport->default_metadata = mlx5_esw_match_metadata_alloc(esw); vport->metadata = vport->default_metadata; return vport->metadata ? 0 : -ENOSPC; @@ -1923,40 +2084,65 @@ static int esw_offloads_vport_metadata_setup(struct mlx5_eswitch *esw, static void esw_offloads_vport_metadata_cleanup(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { - if (vport->vport == MLX5_VPORT_UPLINK || !vport->default_metadata) + if (!vport->default_metadata) return; WARN_ON(vport->metadata != vport->default_metadata); mlx5_esw_match_metadata_free(esw, vport->default_metadata); } +static void esw_offloads_metadata_uninit(struct mlx5_eswitch *esw) +{ + struct mlx5_vport *vport; + int i; + + if (!mlx5_eswitch_vport_match_metadata_enabled(esw)) + return; + + mlx5_esw_for_all_vports_reverse(esw, i, vport) + esw_offloads_vport_metadata_cleanup(esw, vport); +} + +static int esw_offloads_metadata_init(struct mlx5_eswitch *esw) +{ + struct mlx5_vport *vport; + int err; + int i; + + if (!mlx5_eswitch_vport_match_metadata_enabled(esw)) + return 0; + + mlx5_esw_for_all_vports(esw, i, vport) { + err = esw_offloads_vport_metadata_setup(esw, vport); + if (err) + goto metadata_err; + } + + return 0; + +metadata_err: + esw_offloads_metadata_uninit(esw); + return err; +} + int esw_vport_create_offloads_acl_tables(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { int err; - err = esw_offloads_vport_metadata_setup(esw, vport); - if (err) - goto metadata_err; - err = esw_acl_ingress_ofld_setup(esw, vport); if (err) - goto ingress_err; + return err; - if (mlx5_eswitch_is_vf_vport(esw, vport->vport)) { - err = esw_acl_egress_ofld_setup(esw, vport); - if (err) - goto egress_err; - } + err = esw_acl_egress_ofld_setup(esw, vport); + if (err) + goto egress_err; return 0; egress_err: esw_acl_ingress_ofld_cleanup(esw, vport); -ingress_err: - esw_offloads_vport_metadata_cleanup(esw, vport); -metadata_err: return err; } @@ -1966,22 +2152,14 @@ esw_vport_destroy_offloads_acl_tables(struct mlx5_eswitch *esw, { esw_acl_egress_ofld_cleanup(vport); esw_acl_ingress_ofld_cleanup(esw, vport); - esw_offloads_vport_metadata_cleanup(esw, vport); } static int esw_create_uplink_offloads_acl_tables(struct mlx5_eswitch *esw) { struct mlx5_vport *vport; - int err; - - if (esw_use_vport_metadata(esw)) - esw->flags |= MLX5_ESWITCH_VPORT_MATCH_METADATA; vport = mlx5_eswitch_get_vport(esw, MLX5_VPORT_UPLINK); - err = esw_vport_create_offloads_acl_tables(esw, vport); - if (err) - esw->flags &= ~MLX5_ESWITCH_VPORT_MATCH_METADATA; - return err; + return esw_vport_create_offloads_acl_tables(esw, vport); } static void esw_destroy_uplink_offloads_acl_tables(struct mlx5_eswitch *esw) @@ -1990,7 +2168,6 @@ static void esw_destroy_uplink_offloads_acl_tables(struct mlx5_eswitch *esw) vport = mlx5_eswitch_get_vport(esw, MLX5_VPORT_UPLINK); esw_vport_destroy_offloads_acl_tables(esw, vport); - esw->flags &= ~MLX5_ESWITCH_VPORT_MATCH_METADATA; } static int esw_offloads_steering_init(struct mlx5_eswitch *esw) @@ -2114,6 +2291,24 @@ int mlx5_esw_funcs_changed_handler(struct notifier_block *nb, unsigned long type return NOTIFY_OK; } +static int mlx5_esw_host_number_init(struct mlx5_eswitch *esw) +{ + const u32 *query_host_out; + + if (!mlx5_core_is_ecpf_esw_manager(esw->dev)) + return 0; + + query_host_out = mlx5_esw_query_functions(esw->dev); + if (IS_ERR(query_host_out)) + return PTR_ERR(query_host_out); + + /* Mark non local controller with non zero controller number. */ + esw->offloads.host_number = MLX5_GET(query_esw_functions_out, query_host_out, + host_params_context.host_number); + kvfree(query_host_out); + return 0; +} + int esw_offloads_enable(struct mlx5_eswitch *esw) { struct mlx5_vport *vport; @@ -2128,6 +2323,17 @@ int esw_offloads_enable(struct mlx5_eswitch *esw) mutex_init(&esw->offloads.termtbl_mutex); mlx5_rdma_enable_roce(esw->dev); + err = mlx5_esw_host_number_init(esw); + if (err) + goto err_metadata; + + if (esw_check_vport_match_metadata_supported(esw)) + esw->flags |= MLX5_ESWITCH_VPORT_MATCH_METADATA; + + err = esw_offloads_metadata_init(esw); + if (err) + goto err_metadata; + err = esw_set_passing_vport_metadata(esw, true); if (err) goto err_vport_metadata; @@ -2160,6 +2366,9 @@ err_uplink: err_steering_init: esw_set_passing_vport_metadata(esw, false); err_vport_metadata: + esw_offloads_metadata_uninit(esw); +err_metadata: + esw->flags &= ~MLX5_ESWITCH_VPORT_MATCH_METADATA; mlx5_rdma_disable_roce(esw->dev); mutex_destroy(&esw->offloads.termtbl_mutex); return err; @@ -2193,6 +2402,8 @@ void esw_offloads_disable(struct mlx5_eswitch *esw) esw_offloads_unload_rep(esw, MLX5_VPORT_UPLINK); esw_set_passing_vport_metadata(esw, false); esw_offloads_steering_cleanup(esw); + esw_offloads_metadata_uninit(esw); + esw->flags &= ~MLX5_ESWITCH_VPORT_MATCH_METADATA; mlx5_rdma_disable_roce(esw->dev); mutex_destroy(&esw->offloads.termtbl_mutex); esw->offloads.encap = DEVLINK_ESWITCH_ENCAP_MODE_NONE; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c index 17a0d2bc102b..ec679560a95d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c @@ -3,6 +3,7 @@ #include <linux/mlx5/fs.h> #include "eswitch.h" +#include "en_tc.h" #include "fs_core.h" struct mlx5_termtbl_handle { @@ -228,10 +229,11 @@ static bool mlx5_eswitch_offload_is_uplink_port(const struct mlx5_eswitch *esw, bool mlx5_eswitch_termtbl_required(struct mlx5_eswitch *esw, - struct mlx5_esw_flow_attr *attr, + struct mlx5_flow_attr *attr, struct mlx5_flow_act *flow_act, struct mlx5_flow_spec *spec) { + struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr; int i; if (!MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, termination_table) || @@ -244,8 +246,8 @@ mlx5_eswitch_termtbl_required(struct mlx5_eswitch *esw, return true; /* hairpin */ - for (i = attr->split_count; i < attr->out_count; i++) - if (attr->dests[i].rep->vport == MLX5_VPORT_UPLINK) + for (i = esw_attr->split_count; i < esw_attr->out_count; i++) + if (esw_attr->dests[i].rep->vport == MLX5_VPORT_UPLINK) return true; return false; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/conn.c b/drivers/net/ethernet/mellanox/mlx5/core/fpga/conn.c index 831d2c39e153..80da50e12915 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/conn.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/conn.c @@ -54,7 +54,7 @@ static int mlx5_fpga_conn_map_buf(struct mlx5_fpga_conn *conn, if (unlikely(!buf->sg[0].data)) goto out; - dma_device = &conn->fdev->mdev->pdev->dev; + dma_device = mlx5_core_dma_dev(conn->fdev->mdev); buf->sg[0].dma_addr = dma_map_single(dma_device, buf->sg[0].data, buf->sg[0].size, buf->dma_dir); err = dma_mapping_error(dma_device, buf->sg[0].dma_addr); @@ -86,7 +86,7 @@ static void mlx5_fpga_conn_unmap_buf(struct mlx5_fpga_conn *conn, { struct device *dma_device; - dma_device = &conn->fdev->mdev->pdev->dev; + dma_device = mlx5_core_dma_dev(conn->fdev->mdev); if (buf->sg[1].data) dma_unmap_single(dma_device, buf->sg[1].dma_addr, buf->sg[1].size, buf->dma_dir); @@ -388,9 +388,9 @@ static inline void mlx5_fpga_conn_cqes(struct mlx5_fpga_conn *conn, mlx5_fpga_conn_arm_cq(conn); } -static void mlx5_fpga_conn_cq_tasklet(unsigned long data) +static void mlx5_fpga_conn_cq_tasklet(struct tasklet_struct *t) { - struct mlx5_fpga_conn *conn = (void *)data; + struct mlx5_fpga_conn *conn = from_tasklet(conn, t, cq.tasklet); if (unlikely(!conn->qp.active)) return; @@ -478,8 +478,7 @@ static int mlx5_fpga_conn_create_cq(struct mlx5_fpga_conn *conn, int cq_size) conn->cq.mcq.comp = mlx5_fpga_conn_cq_complete; conn->cq.mcq.irqn = irqn; conn->cq.mcq.uar = fdev->conn_res.uar; - tasklet_init(&conn->cq.tasklet, mlx5_fpga_conn_cq_tasklet, - (unsigned long)conn); + tasklet_setup(&conn->cq.tasklet, mlx5_fpga_conn_cq_tasklet); mlx5_fpga_dbg(fdev, "Created CQ #0x%x\n", conn->cq.mcq.cqn); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c index fee169732de7..babe3405132a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c @@ -776,6 +776,9 @@ static int mlx5_cmd_modify_header_alloc(struct mlx5_flow_root_namespace *ns, table_type = FS_FT_NIC_RX; break; case MLX5_FLOW_NAMESPACE_EGRESS: +#ifdef CONFIG_MLX5_IPSEC + case MLX5_FLOW_NAMESPACE_EGRESS_KERNEL: +#endif max_actions = MLX5_CAP_FLOWTABLE_NIC_TX(dev, max_modify_header_actions); table_type = FS_FT_NIC_TX; break; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 75fa44eee434..16091838bfcf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -126,6 +126,10 @@ #define LAG_NUM_PRIOS 1 #define LAG_MIN_LEVEL (OFFLOADS_MIN_LEVEL + 1) +#define KERNEL_TX_IPSEC_NUM_PRIOS 1 +#define KERNEL_TX_IPSEC_NUM_LEVELS 1 +#define KERNEL_TX_MIN_LEVEL (KERNEL_TX_IPSEC_NUM_LEVELS) + struct node_caps { size_t arr_sz; long *caps; @@ -180,13 +184,24 @@ static struct init_tree_node { static struct init_tree_node egress_root_fs = { .type = FS_TYPE_NAMESPACE, +#ifdef CONFIG_MLX5_IPSEC + .ar_size = 2, +#else .ar_size = 1, +#endif .children = (struct init_tree_node[]) { ADD_PRIO(0, MLX5_BY_PASS_NUM_PRIOS, 0, FS_CHAINING_CAPS_EGRESS, ADD_NS(MLX5_FLOW_TABLE_MISS_ACTION_DEF, ADD_MULTIPLE_PRIO(MLX5_BY_PASS_NUM_PRIOS, BY_PASS_PRIO_NUM_LEVELS))), +#ifdef CONFIG_MLX5_IPSEC + ADD_PRIO(0, KERNEL_TX_MIN_LEVEL, 0, + FS_CHAINING_CAPS_EGRESS, + ADD_NS(MLX5_FLOW_TABLE_MISS_ACTION_DEF, + ADD_MULTIPLE_PRIO(KERNEL_TX_IPSEC_NUM_PRIOS, + KERNEL_TX_IPSEC_NUM_LEVELS))), +#endif } }; @@ -1595,11 +1610,12 @@ static bool dest_is_valid(struct mlx5_flow_destination *dest, return true; if (ignore_level) { - if (ft->type != FS_FT_FDB) + if (ft->type != FS_FT_FDB && + ft->type != FS_FT_NIC_RX) return false; if (dest->type == MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE && - dest->ft->type != FS_FT_FDB) + ft->type != dest->ft->type) return false; } @@ -2164,8 +2180,10 @@ struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, break; } - if (type == MLX5_FLOW_NAMESPACE_EGRESS) { + if (type == MLX5_FLOW_NAMESPACE_EGRESS || + type == MLX5_FLOW_NAMESPACE_EGRESS_KERNEL) { root_ns = steering->egress_root_ns; + prio = type - MLX5_FLOW_NAMESPACE_EGRESS; } else if (type == MLX5_FLOW_NAMESPACE_RDMA_RX) { root_ns = steering->rdma_rx_root_ns; prio = RDMA_RX_BYPASS_PRIO; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c new file mode 100644 index 000000000000..f9042e147c7f --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c @@ -0,0 +1,463 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* Copyright (c) 2020, Mellanox Technologies inc. All rights reserved. */ + +#include "fw_reset.h" +#include "diag/fw_tracer.h" + +enum { + MLX5_FW_RESET_FLAGS_RESET_REQUESTED, + MLX5_FW_RESET_FLAGS_NACK_RESET_REQUEST, + MLX5_FW_RESET_FLAGS_PENDING_COMP +}; + +struct mlx5_fw_reset { + struct mlx5_core_dev *dev; + struct mlx5_nb nb; + struct workqueue_struct *wq; + struct work_struct fw_live_patch_work; + struct work_struct reset_request_work; + struct work_struct reset_reload_work; + struct work_struct reset_now_work; + struct work_struct reset_abort_work; + unsigned long reset_flags; + struct timer_list timer; + struct completion done; + int ret; +}; + +void mlx5_fw_reset_enable_remote_dev_reset_set(struct mlx5_core_dev *dev, bool enable) +{ + struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset; + + if (enable) + clear_bit(MLX5_FW_RESET_FLAGS_NACK_RESET_REQUEST, &fw_reset->reset_flags); + else + set_bit(MLX5_FW_RESET_FLAGS_NACK_RESET_REQUEST, &fw_reset->reset_flags); +} + +bool mlx5_fw_reset_enable_remote_dev_reset_get(struct mlx5_core_dev *dev) +{ + struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset; + + return !test_bit(MLX5_FW_RESET_FLAGS_NACK_RESET_REQUEST, &fw_reset->reset_flags); +} + +static int mlx5_reg_mfrl_set(struct mlx5_core_dev *dev, u8 reset_level, + u8 reset_type_sel, u8 sync_resp, bool sync_start) +{ + u32 out[MLX5_ST_SZ_DW(mfrl_reg)] = {}; + u32 in[MLX5_ST_SZ_DW(mfrl_reg)] = {}; + + MLX5_SET(mfrl_reg, in, reset_level, reset_level); + MLX5_SET(mfrl_reg, in, rst_type_sel, reset_type_sel); + MLX5_SET(mfrl_reg, in, pci_sync_for_fw_update_resp, sync_resp); + MLX5_SET(mfrl_reg, in, pci_sync_for_fw_update_start, sync_start); + + return mlx5_core_access_reg(dev, in, sizeof(in), out, sizeof(out), MLX5_REG_MFRL, 0, 1); +} + +static int mlx5_reg_mfrl_query(struct mlx5_core_dev *dev, u8 *reset_level, u8 *reset_type) +{ + u32 out[MLX5_ST_SZ_DW(mfrl_reg)] = {}; + u32 in[MLX5_ST_SZ_DW(mfrl_reg)] = {}; + int err; + + err = mlx5_core_access_reg(dev, in, sizeof(in), out, sizeof(out), MLX5_REG_MFRL, 0, 0); + if (err) + return err; + + if (reset_level) + *reset_level = MLX5_GET(mfrl_reg, out, reset_level); + if (reset_type) + *reset_type = MLX5_GET(mfrl_reg, out, reset_type); + + return 0; +} + +int mlx5_fw_reset_query(struct mlx5_core_dev *dev, u8 *reset_level, u8 *reset_type) +{ + return mlx5_reg_mfrl_query(dev, reset_level, reset_type); +} + +int mlx5_fw_reset_set_reset_sync(struct mlx5_core_dev *dev, u8 reset_type_sel) +{ + struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset; + int err; + + set_bit(MLX5_FW_RESET_FLAGS_PENDING_COMP, &fw_reset->reset_flags); + err = mlx5_reg_mfrl_set(dev, MLX5_MFRL_REG_RESET_LEVEL3, reset_type_sel, 0, true); + if (err) + clear_bit(MLX5_FW_RESET_FLAGS_PENDING_COMP, &fw_reset->reset_flags); + return err; +} + +int mlx5_fw_reset_set_live_patch(struct mlx5_core_dev *dev) +{ + return mlx5_reg_mfrl_set(dev, MLX5_MFRL_REG_RESET_LEVEL0, 0, 0, false); +} + +static void mlx5_fw_reset_complete_reload(struct mlx5_core_dev *dev) +{ + struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset; + + /* if this is the driver that initiated the fw reset, devlink completed the reload */ + if (test_bit(MLX5_FW_RESET_FLAGS_PENDING_COMP, &fw_reset->reset_flags)) { + complete(&fw_reset->done); + } else { + mlx5_load_one(dev, false); + devlink_remote_reload_actions_performed(priv_to_devlink(dev), 0, + BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT) | + BIT(DEVLINK_RELOAD_ACTION_FW_ACTIVATE)); + } +} + +static void mlx5_sync_reset_reload_work(struct work_struct *work) +{ + struct mlx5_fw_reset *fw_reset = container_of(work, struct mlx5_fw_reset, + reset_reload_work); + struct mlx5_core_dev *dev = fw_reset->dev; + int err; + + mlx5_enter_error_state(dev, true); + mlx5_unload_one(dev, false); + err = mlx5_health_wait_pci_up(dev); + if (err) + mlx5_core_err(dev, "reset reload flow aborted, PCI reads still not working\n"); + fw_reset->ret = err; + mlx5_fw_reset_complete_reload(dev); +} + +static void mlx5_stop_sync_reset_poll(struct mlx5_core_dev *dev) +{ + struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset; + + del_timer(&fw_reset->timer); +} + +static void mlx5_sync_reset_clear_reset_requested(struct mlx5_core_dev *dev, bool poll_health) +{ + struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset; + + mlx5_stop_sync_reset_poll(dev); + clear_bit(MLX5_FW_RESET_FLAGS_RESET_REQUESTED, &fw_reset->reset_flags); + if (poll_health) + mlx5_start_health_poll(dev); +} + +#define MLX5_RESET_POLL_INTERVAL (HZ / 10) +static void poll_sync_reset(struct timer_list *t) +{ + struct mlx5_fw_reset *fw_reset = from_timer(fw_reset, t, timer); + struct mlx5_core_dev *dev = fw_reset->dev; + u32 fatal_error; + + if (!test_bit(MLX5_FW_RESET_FLAGS_RESET_REQUESTED, &fw_reset->reset_flags)) + return; + + fatal_error = mlx5_health_check_fatal_sensors(dev); + + if (fatal_error) { + mlx5_core_warn(dev, "Got Device Reset\n"); + mlx5_sync_reset_clear_reset_requested(dev, false); + queue_work(fw_reset->wq, &fw_reset->reset_reload_work); + return; + } + + mod_timer(&fw_reset->timer, round_jiffies(jiffies + MLX5_RESET_POLL_INTERVAL)); +} + +static void mlx5_start_sync_reset_poll(struct mlx5_core_dev *dev) +{ + struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset; + + timer_setup(&fw_reset->timer, poll_sync_reset, 0); + fw_reset->timer.expires = round_jiffies(jiffies + MLX5_RESET_POLL_INTERVAL); + add_timer(&fw_reset->timer); +} + +static int mlx5_fw_reset_set_reset_sync_ack(struct mlx5_core_dev *dev) +{ + return mlx5_reg_mfrl_set(dev, MLX5_MFRL_REG_RESET_LEVEL3, 0, 1, false); +} + +static int mlx5_fw_reset_set_reset_sync_nack(struct mlx5_core_dev *dev) +{ + return mlx5_reg_mfrl_set(dev, MLX5_MFRL_REG_RESET_LEVEL3, 0, 2, false); +} + +static void mlx5_sync_reset_set_reset_requested(struct mlx5_core_dev *dev) +{ + struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset; + + mlx5_stop_health_poll(dev, true); + set_bit(MLX5_FW_RESET_FLAGS_RESET_REQUESTED, &fw_reset->reset_flags); + mlx5_start_sync_reset_poll(dev); +} + +static void mlx5_fw_live_patch_event(struct work_struct *work) +{ + struct mlx5_fw_reset *fw_reset = container_of(work, struct mlx5_fw_reset, + fw_live_patch_work); + struct mlx5_core_dev *dev = fw_reset->dev; + struct mlx5_fw_tracer *tracer; + + mlx5_core_info(dev, "Live patch updated firmware version: %d.%d.%d\n", fw_rev_maj(dev), + fw_rev_min(dev), fw_rev_sub(dev)); + + tracer = dev->tracer; + if (IS_ERR_OR_NULL(tracer)) + return; + + if (mlx5_fw_tracer_reload(tracer)) + mlx5_core_err(dev, "Failed to reload FW tracer\n"); +} + +static void mlx5_sync_reset_request_event(struct work_struct *work) +{ + struct mlx5_fw_reset *fw_reset = container_of(work, struct mlx5_fw_reset, + reset_request_work); + struct mlx5_core_dev *dev = fw_reset->dev; + int err; + + if (test_bit(MLX5_FW_RESET_FLAGS_NACK_RESET_REQUEST, &fw_reset->reset_flags)) { + err = mlx5_fw_reset_set_reset_sync_nack(dev); + mlx5_core_warn(dev, "PCI Sync FW Update Reset Nack %s", + err ? "Failed" : "Sent"); + return; + } + mlx5_sync_reset_set_reset_requested(dev); + err = mlx5_fw_reset_set_reset_sync_ack(dev); + if (err) + mlx5_core_warn(dev, "PCI Sync FW Update Reset Ack Failed. Error code: %d\n", err); + else + mlx5_core_warn(dev, "PCI Sync FW Update Reset Ack. Device reset is expected.\n"); +} + +#define MLX5_PCI_LINK_UP_TIMEOUT 2000 + +static int mlx5_pci_link_toggle(struct mlx5_core_dev *dev) +{ + struct pci_bus *bridge_bus = dev->pdev->bus; + struct pci_dev *bridge = bridge_bus->self; + u16 reg16, dev_id, sdev_id; + unsigned long timeout; + struct pci_dev *sdev; + int cap, err; + u32 reg32; + + /* Check that all functions under the pci bridge are PFs of + * this device otherwise fail this function. + */ + err = pci_read_config_word(dev->pdev, PCI_DEVICE_ID, &dev_id); + if (err) + return err; + list_for_each_entry(sdev, &bridge_bus->devices, bus_list) { + err = pci_read_config_word(sdev, PCI_DEVICE_ID, &sdev_id); + if (err) + return err; + if (sdev_id != dev_id) + return -EPERM; + } + + cap = pci_find_capability(bridge, PCI_CAP_ID_EXP); + if (!cap) + return -EOPNOTSUPP; + + list_for_each_entry(sdev, &bridge_bus->devices, bus_list) { + pci_save_state(sdev); + pci_cfg_access_lock(sdev); + } + /* PCI link toggle */ + err = pci_read_config_word(bridge, cap + PCI_EXP_LNKCTL, ®16); + if (err) + return err; + reg16 |= PCI_EXP_LNKCTL_LD; + err = pci_write_config_word(bridge, cap + PCI_EXP_LNKCTL, reg16); + if (err) + return err; + msleep(500); + reg16 &= ~PCI_EXP_LNKCTL_LD; + err = pci_write_config_word(bridge, cap + PCI_EXP_LNKCTL, reg16); + if (err) + return err; + + /* Check link */ + err = pci_read_config_dword(bridge, cap + PCI_EXP_LNKCAP, ®32); + if (err) + return err; + if (!(reg32 & PCI_EXP_LNKCAP_DLLLARC)) { + mlx5_core_warn(dev, "No PCI link reporting capability (0x%08x)\n", reg32); + msleep(1000); + goto restore; + } + + timeout = jiffies + msecs_to_jiffies(MLX5_PCI_LINK_UP_TIMEOUT); + do { + err = pci_read_config_word(bridge, cap + PCI_EXP_LNKSTA, ®16); + if (err) + return err; + if (reg16 & PCI_EXP_LNKSTA_DLLLA) + break; + msleep(20); + } while (!time_after(jiffies, timeout)); + + if (reg16 & PCI_EXP_LNKSTA_DLLLA) { + mlx5_core_info(dev, "PCI Link up\n"); + } else { + mlx5_core_err(dev, "PCI link not ready (0x%04x) after %d ms\n", + reg16, MLX5_PCI_LINK_UP_TIMEOUT); + err = -ETIMEDOUT; + } + +restore: + list_for_each_entry(sdev, &bridge_bus->devices, bus_list) { + pci_cfg_access_unlock(sdev); + pci_restore_state(sdev); + } + + return err; +} + +static void mlx5_sync_reset_now_event(struct work_struct *work) +{ + struct mlx5_fw_reset *fw_reset = container_of(work, struct mlx5_fw_reset, + reset_now_work); + struct mlx5_core_dev *dev = fw_reset->dev; + int err; + + mlx5_sync_reset_clear_reset_requested(dev, false); + + mlx5_core_warn(dev, "Sync Reset now. Device is going to reset.\n"); + + err = mlx5_cmd_fast_teardown_hca(dev); + if (err) { + mlx5_core_warn(dev, "Fast teardown failed, no reset done, err %d\n", err); + goto done; + } + + err = mlx5_pci_link_toggle(dev); + if (err) { + mlx5_core_warn(dev, "mlx5_pci_link_toggle failed, no reset done, err %d\n", err); + goto done; + } + + mlx5_enter_error_state(dev, true); + mlx5_unload_one(dev, false); +done: + fw_reset->ret = err; + mlx5_fw_reset_complete_reload(dev); +} + +static void mlx5_sync_reset_abort_event(struct work_struct *work) +{ + struct mlx5_fw_reset *fw_reset = container_of(work, struct mlx5_fw_reset, + reset_abort_work); + struct mlx5_core_dev *dev = fw_reset->dev; + + mlx5_sync_reset_clear_reset_requested(dev, true); + mlx5_core_warn(dev, "PCI Sync FW Update Reset Aborted.\n"); +} + +static void mlx5_sync_reset_events_handle(struct mlx5_fw_reset *fw_reset, struct mlx5_eqe *eqe) +{ + struct mlx5_eqe_sync_fw_update *sync_fw_update_eqe; + u8 sync_event_rst_type; + + sync_fw_update_eqe = &eqe->data.sync_fw_update; + sync_event_rst_type = sync_fw_update_eqe->sync_rst_state & SYNC_RST_STATE_MASK; + switch (sync_event_rst_type) { + case MLX5_SYNC_RST_STATE_RESET_REQUEST: + queue_work(fw_reset->wq, &fw_reset->reset_request_work); + break; + case MLX5_SYNC_RST_STATE_RESET_NOW: + queue_work(fw_reset->wq, &fw_reset->reset_now_work); + break; + case MLX5_SYNC_RST_STATE_RESET_ABORT: + queue_work(fw_reset->wq, &fw_reset->reset_abort_work); + break; + } +} + +static int fw_reset_event_notifier(struct notifier_block *nb, unsigned long action, void *data) +{ + struct mlx5_fw_reset *fw_reset = mlx5_nb_cof(nb, struct mlx5_fw_reset, nb); + struct mlx5_eqe *eqe = data; + + switch (eqe->sub_type) { + case MLX5_GENERAL_SUBTYPE_FW_LIVE_PATCH_EVENT: + queue_work(fw_reset->wq, &fw_reset->fw_live_patch_work); + break; + case MLX5_GENERAL_SUBTYPE_PCI_SYNC_FOR_FW_UPDATE_EVENT: + mlx5_sync_reset_events_handle(fw_reset, eqe); + break; + default: + return NOTIFY_DONE; + } + + return NOTIFY_OK; +} + +#define MLX5_FW_RESET_TIMEOUT_MSEC 5000 +int mlx5_fw_reset_wait_reset_done(struct mlx5_core_dev *dev) +{ + unsigned long timeout = msecs_to_jiffies(MLX5_FW_RESET_TIMEOUT_MSEC); + struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset; + int err; + + if (!wait_for_completion_timeout(&fw_reset->done, timeout)) { + mlx5_core_warn(dev, "FW sync reset timeout after %d seconds\n", + MLX5_FW_RESET_TIMEOUT_MSEC / 1000); + err = -ETIMEDOUT; + goto out; + } + err = fw_reset->ret; +out: + clear_bit(MLX5_FW_RESET_FLAGS_PENDING_COMP, &fw_reset->reset_flags); + return err; +} + +void mlx5_fw_reset_events_start(struct mlx5_core_dev *dev) +{ + struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset; + + MLX5_NB_INIT(&fw_reset->nb, fw_reset_event_notifier, GENERAL_EVENT); + mlx5_eq_notifier_register(dev, &fw_reset->nb); +} + +void mlx5_fw_reset_events_stop(struct mlx5_core_dev *dev) +{ + mlx5_eq_notifier_unregister(dev, &dev->priv.fw_reset->nb); +} + +int mlx5_fw_reset_init(struct mlx5_core_dev *dev) +{ + struct mlx5_fw_reset *fw_reset = kzalloc(sizeof(*fw_reset), GFP_KERNEL); + + if (!fw_reset) + return -ENOMEM; + fw_reset->wq = create_singlethread_workqueue("mlx5_fw_reset_events"); + if (!fw_reset->wq) { + kfree(fw_reset); + return -ENOMEM; + } + + fw_reset->dev = dev; + dev->priv.fw_reset = fw_reset; + + INIT_WORK(&fw_reset->fw_live_patch_work, mlx5_fw_live_patch_event); + INIT_WORK(&fw_reset->reset_request_work, mlx5_sync_reset_request_event); + INIT_WORK(&fw_reset->reset_reload_work, mlx5_sync_reset_reload_work); + INIT_WORK(&fw_reset->reset_now_work, mlx5_sync_reset_now_event); + INIT_WORK(&fw_reset->reset_abort_work, mlx5_sync_reset_abort_event); + + init_completion(&fw_reset->done); + return 0; +} + +void mlx5_fw_reset_cleanup(struct mlx5_core_dev *dev) +{ + struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset; + + destroy_workqueue(fw_reset->wq); + kfree(dev->priv.fw_reset); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.h b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.h new file mode 100644 index 000000000000..7761ee5fc7d0 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.h @@ -0,0 +1,21 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ +/* Copyright (c) 2020, Mellanox Technologies inc. All rights reserved. */ + +#ifndef __MLX5_FW_RESET_H +#define __MLX5_FW_RESET_H + +#include "mlx5_core.h" + +void mlx5_fw_reset_enable_remote_dev_reset_set(struct mlx5_core_dev *dev, bool enable); +bool mlx5_fw_reset_enable_remote_dev_reset_get(struct mlx5_core_dev *dev); +int mlx5_fw_reset_query(struct mlx5_core_dev *dev, u8 *reset_level, u8 *reset_type); +int mlx5_fw_reset_set_reset_sync(struct mlx5_core_dev *dev, u8 reset_type_sel); +int mlx5_fw_reset_set_live_patch(struct mlx5_core_dev *dev); + +int mlx5_fw_reset_wait_reset_done(struct mlx5_core_dev *dev); +void mlx5_fw_reset_events_start(struct mlx5_core_dev *dev); +void mlx5_fw_reset_events_stop(struct mlx5_core_dev *dev); +int mlx5_fw_reset_init(struct mlx5_core_dev *dev); +void mlx5_fw_reset_cleanup(struct mlx5_core_dev *dev); + +#endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/health.c b/drivers/net/ethernet/mellanox/mlx5/core/health.c index b31f769d2df9..54523bed16cd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/health.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/health.c @@ -110,7 +110,7 @@ static bool sensor_fw_synd_rfr(struct mlx5_core_dev *dev) return rfr && synd; } -static u32 check_fatal_sensors(struct mlx5_core_dev *dev) +u32 mlx5_health_check_fatal_sensors(struct mlx5_core_dev *dev) { if (sensor_pci_not_working(dev)) return MLX5_SENSOR_PCI_COMM_ERR; @@ -173,7 +173,7 @@ static bool reset_fw_if_needed(struct mlx5_core_dev *dev) * Check again to avoid a redundant 2nd reset. If the fatal erros was * PCI related a reset won't help. */ - fatal_error = check_fatal_sensors(dev); + fatal_error = mlx5_health_check_fatal_sensors(dev); if (fatal_error == MLX5_SENSOR_PCI_COMM_ERR || fatal_error == MLX5_SENSOR_NIC_DISABLED || fatal_error == MLX5_SENSOR_NIC_SW_RESET) { @@ -195,7 +195,7 @@ void mlx5_enter_error_state(struct mlx5_core_dev *dev, bool force) bool err_detected = false; /* Mark the device as fatal in order to abort FW commands */ - if ((check_fatal_sensors(dev) || force) && + if ((mlx5_health_check_fatal_sensors(dev) || force) && dev->state == MLX5_DEVICE_STATE_UP) { dev->state = MLX5_DEVICE_STATE_INTERNAL_ERROR; err_detected = true; @@ -208,7 +208,7 @@ void mlx5_enter_error_state(struct mlx5_core_dev *dev, bool force) goto unlock; } - if (check_fatal_sensors(dev) || force) { /* protected state setting */ + if (mlx5_health_check_fatal_sensors(dev) || force) { /* protected state setting */ dev->state = MLX5_DEVICE_STATE_INTERNAL_ERROR; mlx5_cmd_flush(dev); } @@ -231,7 +231,7 @@ void mlx5_error_sw_reset(struct mlx5_core_dev *dev) mlx5_core_err(dev, "start\n"); - if (check_fatal_sensors(dev) == MLX5_SENSOR_FW_SYND_RFR) { + if (mlx5_health_check_fatal_sensors(dev) == MLX5_SENSOR_FW_SYND_RFR) { /* Get cr-dump and reset FW semaphore */ lock = lock_sem_sw_reset(dev, true); @@ -308,26 +308,31 @@ static void mlx5_handle_bad_state(struct mlx5_core_dev *dev) /* How much time to wait until health resetting the driver (in msecs) */ #define MLX5_RECOVERY_WAIT_MSECS 60000 -static int mlx5_health_try_recover(struct mlx5_core_dev *dev) +int mlx5_health_wait_pci_up(struct mlx5_core_dev *dev) { unsigned long end; - mlx5_core_warn(dev, "handling bad device here\n"); - mlx5_handle_bad_state(dev); end = jiffies + msecs_to_jiffies(MLX5_RECOVERY_WAIT_MSECS); while (sensor_pci_not_working(dev)) { - if (time_after(jiffies, end)) { - mlx5_core_err(dev, - "health recovery flow aborted, PCI reads still not working\n"); - return -EIO; - } + if (time_after(jiffies, end)) + return -ETIMEDOUT; msleep(100); } + return 0; +} +static int mlx5_health_try_recover(struct mlx5_core_dev *dev) +{ + mlx5_core_warn(dev, "handling bad device here\n"); + mlx5_handle_bad_state(dev); + if (mlx5_health_wait_pci_up(dev)) { + mlx5_core_err(dev, "health recovery flow aborted, PCI reads still not working\n"); + return -EIO; + } mlx5_core_err(dev, "starting health recovery flow\n"); mlx5_recover_device(dev); if (!test_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state) || - check_fatal_sensors(dev)) { + mlx5_health_check_fatal_sensors(dev)) { mlx5_core_err(dev, "health recovery failed\n"); return -EIO; } @@ -696,7 +701,7 @@ static void poll_health(struct timer_list *t) if (dev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) goto out; - fatal_error = check_fatal_sensors(dev); + fatal_error = mlx5_health_check_fatal_sensors(dev); if (fatal_error && !health->fatal_error) { mlx5_core_err(dev, "Fatal error %u detected\n", fatal_error); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag.c b/drivers/net/ethernet/mellanox/mlx5/core/lag.c index 874c70e8cc54..33081b24f10a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lag.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lag.c @@ -102,7 +102,7 @@ int mlx5_lag_dev_get_netdev_idx(struct mlx5_lag *ldev, if (ldev->pf[i].netdev == ndev) return i; - return -1; + return -ENOENT; } static bool __mlx5_lag_is_roce(struct mlx5_lag *ldev) @@ -271,7 +271,7 @@ static void mlx5_do_bond(struct mlx5_lag *ldev) bool do_bond, roce_lag; int err; - if (!dev0 || !dev1) + if (!mlx5_lag_is_ready(ldev)) return; spin_lock(&lag_lock); @@ -355,7 +355,7 @@ static int mlx5_handle_changeupper_event(struct mlx5_lag *ldev, { struct net_device *upper = info->upper_dev, *ndev_tmp; struct netdev_lag_upper_info *lag_upper_info = NULL; - bool is_bonded; + bool is_bonded, is_in_lag, mode_supported; int bond_status = 0; int num_slaves = 0; int idx; @@ -374,7 +374,7 @@ static int mlx5_handle_changeupper_event(struct mlx5_lag *ldev, rcu_read_lock(); for_each_netdev_in_bond_rcu(upper, ndev_tmp) { idx = mlx5_lag_dev_get_netdev_idx(ldev, ndev_tmp); - if (idx > -1) + if (idx >= 0) bond_status |= (1 << idx); num_slaves++; @@ -391,13 +391,24 @@ static int mlx5_handle_changeupper_event(struct mlx5_lag *ldev, /* Determine bonding status: * A device is considered bonded if both its physical ports are slaves * of the same lag master, and only them. - * Lag mode must be activebackup or hash. */ - is_bonded = (num_slaves == MLX5_MAX_PORTS) && - (bond_status == 0x3) && - ((tracker->tx_type == NETDEV_LAG_TX_TYPE_ACTIVEBACKUP) || - (tracker->tx_type == NETDEV_LAG_TX_TYPE_HASH)); + is_in_lag = num_slaves == MLX5_MAX_PORTS && bond_status == 0x3; + if (!mlx5_lag_is_ready(ldev) && is_in_lag) { + NL_SET_ERR_MSG_MOD(info->info.extack, + "Can't activate LAG offload, PF is configured with more than 64 VFs"); + return 0; + } + + /* Lag mode must be activebackup or hash. */ + mode_supported = tracker->tx_type == NETDEV_LAG_TX_TYPE_ACTIVEBACKUP || + tracker->tx_type == NETDEV_LAG_TX_TYPE_HASH; + + if (is_in_lag && !mode_supported) + NL_SET_ERR_MSG_MOD(info->info.extack, + "Can't activate LAG offload, TX type isn't supported"); + + is_bonded = is_in_lag && mode_supported; if (tracker->is_bonded != is_bonded) { tracker->is_bonded = is_bonded; return 1; @@ -418,7 +429,7 @@ static int mlx5_handle_changelowerstate_event(struct mlx5_lag *ldev, return 0; idx = mlx5_lag_dev_get_netdev_idx(ldev, ndev); - if (idx == -1) + if (idx < 0) return 0; /* This information is used to determine virtual to physical @@ -445,6 +456,10 @@ static int mlx5_lag_netdev_event(struct notifier_block *this, return NOTIFY_DONE; ldev = container_of(this, struct mlx5_lag, nb); + + if (!mlx5_lag_is_ready(ldev) && event == NETDEV_CHANGELOWERSTATE) + return NOTIFY_DONE; + tracker = ldev->tracker; switch (event) { @@ -493,14 +508,14 @@ static void mlx5_lag_dev_free(struct mlx5_lag *ldev) kfree(ldev); } -static void mlx5_lag_dev_add_pf(struct mlx5_lag *ldev, - struct mlx5_core_dev *dev, - struct net_device *netdev) +static int mlx5_lag_dev_add_pf(struct mlx5_lag *ldev, + struct mlx5_core_dev *dev, + struct net_device *netdev) { unsigned int fn = PCI_FUNC(dev->pdev->devfn); if (fn >= MLX5_MAX_PORTS) - return; + return -EPERM; spin_lock(&lag_lock); ldev->pf[fn].dev = dev; @@ -511,6 +526,8 @@ static void mlx5_lag_dev_add_pf(struct mlx5_lag *ldev, dev->priv.lag = ldev; spin_unlock(&lag_lock); + + return fn; } static void mlx5_lag_dev_remove_pf(struct mlx5_lag *ldev, @@ -537,11 +554,9 @@ void mlx5_lag_add(struct mlx5_core_dev *dev, struct net_device *netdev) { struct mlx5_lag *ldev = NULL; struct mlx5_core_dev *tmp_dev; - int err; + int i, err; - if (!MLX5_CAP_GEN(dev, vport_group_manager) || - !MLX5_CAP_GEN(dev, lag_master) || - (MLX5_CAP_GEN(dev, num_lag_ports) != MLX5_MAX_PORTS)) + if (!MLX5_CAP_GEN(dev, vport_group_manager)) return; tmp_dev = mlx5_get_next_phys_dev(dev); @@ -556,7 +571,18 @@ void mlx5_lag_add(struct mlx5_core_dev *dev, struct net_device *netdev) } } - mlx5_lag_dev_add_pf(ldev, dev, netdev); + if (mlx5_lag_dev_add_pf(ldev, dev, netdev) < 0) + return; + + for (i = 0; i < MLX5_MAX_PORTS; i++) { + tmp_dev = ldev->pf[i].dev; + if (!tmp_dev || !MLX5_CAP_GEN(tmp_dev, lag_master) || + MLX5_CAP_GEN(tmp_dev, num_lag_ports) != MLX5_MAX_PORTS) + break; + } + + if (i >= MLX5_MAX_PORTS) + ldev->flags |= MLX5_LAG_FLAG_READY; if (!ldev->nb.notifier_call) { ldev->nb.notifier_call = mlx5_lag_netdev_event; @@ -587,6 +613,8 @@ void mlx5_lag_remove(struct mlx5_core_dev *dev) mlx5_lag_dev_remove_pf(ldev, dev); + ldev->flags &= ~MLX5_LAG_FLAG_READY; + for (i = 0; i < MLX5_MAX_PORTS; i++) if (ldev->pf[i].dev) break; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag.h b/drivers/net/ethernet/mellanox/mlx5/core/lag.h index f1068aac6406..8d8cf2d0bc6d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lag.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/lag.h @@ -16,6 +16,7 @@ enum { MLX5_LAG_FLAG_ROCE = 1 << 0, MLX5_LAG_FLAG_SRIOV = 1 << 1, MLX5_LAG_FLAG_MULTIPATH = 1 << 2, + MLX5_LAG_FLAG_READY = 1 << 3, }; #define MLX5_LAG_MODE_FLAGS (MLX5_LAG_FLAG_ROCE | MLX5_LAG_FLAG_SRIOV |\ @@ -59,6 +60,12 @@ __mlx5_lag_is_active(struct mlx5_lag *ldev) return !!(ldev->flags & MLX5_LAG_MODE_FLAGS); } +static inline bool +mlx5_lag_is_ready(struct mlx5_lag *ldev) +{ + return ldev->flags & MLX5_LAG_FLAG_READY; +} + void mlx5_modify_lag(struct mlx5_lag *ldev, struct lag_tracker *tracker); int mlx5_activate_lag(struct mlx5_lag *ldev, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c b/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c index 9e68f5926ab6..88e58ac902de 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c @@ -11,7 +11,7 @@ static bool mlx5_lag_multipath_check_prereq(struct mlx5_lag *ldev) { - if (!ldev->pf[MLX5_LAG_P1].dev || !ldev->pf[MLX5_LAG_P2].dev) + if (!mlx5_lag_is_ready(ldev)) return false; return mlx5_esw_multipath_prereq(ldev->pf[MLX5_LAG_P1].dev, @@ -131,7 +131,12 @@ static void mlx5_lag_fib_route_event(struct mlx5_lag *ldev, struct net_device *nh_dev = nh->fib_nh_dev; int i = mlx5_lag_dev_get_netdev_idx(ldev, nh_dev); - mlx5_lag_set_port_affinity(ldev, ++i); + if (i < 0) + i = MLX5_LAG_NORMAL_AFFINITY; + else + ++i; + + mlx5_lag_set_port_affinity(ldev, i); } return; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c index 2d55b7c22c03..c70c1f0ca0c1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c @@ -150,28 +150,30 @@ static void mlx5_pps_out(struct work_struct *work) static void mlx5_timestamp_overflow(struct work_struct *work) { struct delayed_work *dwork = to_delayed_work(work); - struct mlx5_clock *clock = container_of(dwork, struct mlx5_clock, - overflow_work); + struct mlx5_core_dev *mdev; + struct mlx5_clock *clock; unsigned long flags; + clock = container_of(dwork, struct mlx5_clock, overflow_work); + mdev = container_of(clock, struct mlx5_core_dev, clock); write_seqlock_irqsave(&clock->lock, flags); timecounter_read(&clock->tc); - mlx5_update_clock_info_page(clock->mdev); + mlx5_update_clock_info_page(mdev); write_sequnlock_irqrestore(&clock->lock, flags); schedule_delayed_work(&clock->overflow_work, clock->overflow_period); } -static int mlx5_ptp_settime(struct ptp_clock_info *ptp, - const struct timespec64 *ts) +static int mlx5_ptp_settime(struct ptp_clock_info *ptp, const struct timespec64 *ts) { - struct mlx5_clock *clock = container_of(ptp, struct mlx5_clock, - ptp_info); + struct mlx5_clock *clock = container_of(ptp, struct mlx5_clock, ptp_info); u64 ns = timespec64_to_ns(ts); + struct mlx5_core_dev *mdev; unsigned long flags; + mdev = container_of(clock, struct mlx5_core_dev, clock); write_seqlock_irqsave(&clock->lock, flags); timecounter_init(&clock->tc, &clock->cycles, ns); - mlx5_update_clock_info_page(clock->mdev); + mlx5_update_clock_info_page(mdev); write_sequnlock_irqrestore(&clock->lock, flags); return 0; @@ -180,13 +182,12 @@ static int mlx5_ptp_settime(struct ptp_clock_info *ptp, static int mlx5_ptp_gettimex(struct ptp_clock_info *ptp, struct timespec64 *ts, struct ptp_system_timestamp *sts) { - struct mlx5_clock *clock = container_of(ptp, struct mlx5_clock, - ptp_info); - struct mlx5_core_dev *mdev = container_of(clock, struct mlx5_core_dev, - clock); + struct mlx5_clock *clock = container_of(ptp, struct mlx5_clock, ptp_info); + struct mlx5_core_dev *mdev; unsigned long flags; u64 cycles, ns; + mdev = container_of(clock, struct mlx5_core_dev, clock); write_seqlock_irqsave(&clock->lock, flags); cycles = mlx5_read_internal_timer(mdev, sts); ns = timecounter_cyc2time(&clock->tc, cycles); @@ -199,13 +200,14 @@ static int mlx5_ptp_gettimex(struct ptp_clock_info *ptp, struct timespec64 *ts, static int mlx5_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta) { - struct mlx5_clock *clock = container_of(ptp, struct mlx5_clock, - ptp_info); + struct mlx5_clock *clock = container_of(ptp, struct mlx5_clock, ptp_info); + struct mlx5_core_dev *mdev; unsigned long flags; + mdev = container_of(clock, struct mlx5_core_dev, clock); write_seqlock_irqsave(&clock->lock, flags); timecounter_adjtime(&clock->tc, delta); - mlx5_update_clock_info_page(clock->mdev); + mlx5_update_clock_info_page(mdev); write_sequnlock_irqrestore(&clock->lock, flags); return 0; @@ -213,12 +215,13 @@ static int mlx5_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta) static int mlx5_ptp_adjfreq(struct ptp_clock_info *ptp, s32 delta) { - u64 adj; - u32 diff; + struct mlx5_clock *clock = container_of(ptp, struct mlx5_clock, ptp_info); + struct mlx5_core_dev *mdev; unsigned long flags; int neg_adj = 0; - struct mlx5_clock *clock = container_of(ptp, struct mlx5_clock, - ptp_info); + u32 diff; + u64 adj; + if (delta < 0) { neg_adj = 1; @@ -229,11 +232,12 @@ static int mlx5_ptp_adjfreq(struct ptp_clock_info *ptp, s32 delta) adj *= delta; diff = div_u64(adj, 1000000000ULL); + mdev = container_of(clock, struct mlx5_core_dev, clock); write_seqlock_irqsave(&clock->lock, flags); timecounter_read(&clock->tc); clock->cycles.mult = neg_adj ? clock->nominal_c_mult - diff : clock->nominal_c_mult + diff; - mlx5_update_clock_info_page(clock->mdev); + mlx5_update_clock_info_page(mdev); write_sequnlock_irqrestore(&clock->lock, flags); return 0; @@ -431,13 +435,11 @@ static int mlx5_ptp_verify(struct ptp_clock_info *ptp, unsigned int pin, default: return -EOPNOTSUPP; } - - return -EOPNOTSUPP; } static const struct ptp_clock_info mlx5_ptp_clock_info = { .owner = THIS_MODULE, - .name = "mlx5_p2p", + .name = "mlx5_ptp", .max_adj = 100000000, .n_alarm = 0, .n_ext_ts = 0, @@ -465,7 +467,8 @@ static int mlx5_query_mtpps_pin_mode(struct mlx5_core_dev *mdev, u8 pin, static int mlx5_get_pps_pin_mode(struct mlx5_clock *clock, u8 pin) { - struct mlx5_core_dev *mdev = clock->mdev; + struct mlx5_core_dev *mdev = container_of(clock, struct mlx5_core_dev, clock); + u32 out[MLX5_ST_SZ_DW(mtpps_reg)] = {}; u8 mode; int err; @@ -538,20 +541,23 @@ static int mlx5_pps_event(struct notifier_block *nb, unsigned long type, void *data) { struct mlx5_clock *clock = mlx5_nb_cof(nb, struct mlx5_clock, pps_nb); - struct mlx5_core_dev *mdev = clock->mdev; struct ptp_clock_event ptp_event; u64 cycles_now, cycles_delta; u64 nsec_now, nsec_delta, ns; struct mlx5_eqe *eqe = data; int pin = eqe->data.pps.pin; + struct mlx5_core_dev *mdev; struct timespec64 ts; unsigned long flags; + mdev = container_of(clock, struct mlx5_core_dev, clock); + switch (clock->ptp_info.pin_config[pin].func) { case PTP_PF_EXTTS: ptp_event.index = pin; - ptp_event.timestamp = timecounter_cyc2time(&clock->tc, - be64_to_cpu(eqe->data.pps.time_stamp)); + ptp_event.timestamp = + mlx5_timecounter_cyc2time(clock, + be64_to_cpu(eqe->data.pps.time_stamp)); if (clock->pps_info.enabled) { ptp_event.type = PTP_CLOCK_PPSUSR; ptp_event.pps_times.ts_real = @@ -574,8 +580,8 @@ static int mlx5_pps_event(struct notifier_block *nb, cycles_delta = div64_u64(nsec_delta << clock->cycles.shift, clock->cycles.mult); clock->pps_info.start[pin] = cycles_now + cycles_delta; - schedule_work(&clock->pps_info.out_work); write_sequnlock_irqrestore(&clock->lock, flags); + schedule_work(&clock->pps_info.out_work); break; default: mlx5_core_err(mdev, " Unhandled clock PPS event, func %d\n", @@ -605,7 +611,6 @@ void mlx5_init_clock(struct mlx5_core_dev *mdev) clock->cycles.shift); clock->nominal_c_mult = clock->cycles.mult; clock->cycles.mask = CLOCKSOURCE_MASK(41); - clock->mdev = mdev; timecounter_init(&clock->tc, &clock->cycles, ktime_to_ns(ktime_get_real())); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h index 5c681e31983b..81f2cc4ca1da 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h @@ -78,7 +78,7 @@ int mlx5_eq_add_cq(struct mlx5_eq *eq, struct mlx5_core_cq *cq); void mlx5_eq_del_cq(struct mlx5_eq *eq, struct mlx5_core_cq *cq); struct mlx5_eq_comp *mlx5_eqn2comp_eq(struct mlx5_core_dev *dev, int eqn); struct mlx5_eq *mlx5_get_async_eq(struct mlx5_core_dev *dev); -void mlx5_cq_tasklet_cb(unsigned long data); +void mlx5_cq_tasklet_cb(struct tasklet_struct *t); struct cpumask *mlx5_eq_comp_cpumask(struct mlx5_core_dev *dev, int ix); u32 mlx5_eq_poll_irq_disabled(struct mlx5_eq_comp *eq); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c new file mode 100644 index 000000000000..947f346bdc2d --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c @@ -0,0 +1,911 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +// Copyright (c) 2020 Mellanox Technologies. + +#include <linux/mlx5/driver.h> +#include <linux/mlx5/mlx5_ifc.h> +#include <linux/mlx5/fs.h> + +#include "lib/fs_chains.h" +#include "en/mapping.h" +#include "mlx5_core.h" +#include "fs_core.h" +#include "eswitch.h" +#include "en.h" +#include "en_tc.h" + +#define chains_lock(chains) ((chains)->lock) +#define chains_ht(chains) ((chains)->chains_ht) +#define chains_mapping(chains) ((chains)->chains_mapping) +#define prios_ht(chains) ((chains)->prios_ht) +#define ft_pool_left(chains) ((chains)->ft_left) +#define tc_default_ft(chains) ((chains)->tc_default_ft) +#define tc_end_ft(chains) ((chains)->tc_end_ft) +#define ns_to_chains_fs_prio(ns) ((ns) == MLX5_FLOW_NAMESPACE_FDB ? \ + FDB_TC_OFFLOAD : MLX5E_TC_PRIO) + +/* Firmware currently has 4 pool of 4 sizes that it supports (FT_POOLS), + * and a virtual memory region of 16M (MLX5_FT_SIZE), this region is duplicated + * for each flow table pool. We can allocate up to 16M of each pool, + * and we keep track of how much we used via get_next_avail_sz_from_pool. + * Firmware doesn't report any of this for now. + * ESW_POOL is expected to be sorted from large to small and match firmware + * pools. + */ +#define FT_SIZE (16 * 1024 * 1024) +static const unsigned int FT_POOLS[] = { 4 * 1024 * 1024, + 1 * 1024 * 1024, + 64 * 1024, + 128 }; +#define FT_TBL_SZ (64 * 1024) + +struct mlx5_fs_chains { + struct mlx5_core_dev *dev; + + struct rhashtable chains_ht; + struct rhashtable prios_ht; + /* Protects above chains_ht and prios_ht */ + struct mutex lock; + + struct mlx5_flow_table *tc_default_ft; + struct mlx5_flow_table *tc_end_ft; + struct mapping_ctx *chains_mapping; + + enum mlx5_flow_namespace_type ns; + u32 group_num; + u32 flags; + + int ft_left[ARRAY_SIZE(FT_POOLS)]; +}; + +struct fs_chain { + struct rhash_head node; + + u32 chain; + + int ref; + int id; + + struct mlx5_fs_chains *chains; + struct list_head prios_list; + struct mlx5_flow_handle *restore_rule; + struct mlx5_modify_hdr *miss_modify_hdr; +}; + +struct prio_key { + u32 chain; + u32 prio; + u32 level; +}; + +struct prio { + struct rhash_head node; + struct list_head list; + + struct prio_key key; + + int ref; + + struct fs_chain *chain; + struct mlx5_flow_table *ft; + struct mlx5_flow_table *next_ft; + struct mlx5_flow_group *miss_group; + struct mlx5_flow_handle *miss_rule; +}; + +static const struct rhashtable_params chain_params = { + .head_offset = offsetof(struct fs_chain, node), + .key_offset = offsetof(struct fs_chain, chain), + .key_len = sizeof_field(struct fs_chain, chain), + .automatic_shrinking = true, +}; + +static const struct rhashtable_params prio_params = { + .head_offset = offsetof(struct prio, node), + .key_offset = offsetof(struct prio, key), + .key_len = sizeof_field(struct prio, key), + .automatic_shrinking = true, +}; + +bool mlx5_chains_prios_supported(struct mlx5_fs_chains *chains) +{ + return chains->flags & MLX5_CHAINS_AND_PRIOS_SUPPORTED; +} + +static bool mlx5_chains_ignore_flow_level_supported(struct mlx5_fs_chains *chains) +{ + return chains->flags & MLX5_CHAINS_IGNORE_FLOW_LEVEL_SUPPORTED; +} + +bool mlx5_chains_backwards_supported(struct mlx5_fs_chains *chains) +{ + return mlx5_chains_prios_supported(chains) && + mlx5_chains_ignore_flow_level_supported(chains); +} + +u32 mlx5_chains_get_chain_range(struct mlx5_fs_chains *chains) +{ + if (!mlx5_chains_prios_supported(chains)) + return 1; + + if (mlx5_chains_ignore_flow_level_supported(chains)) + return UINT_MAX - 1; + + /* We should get here only for eswitch case */ + return FDB_TC_MAX_CHAIN; +} + +u32 mlx5_chains_get_nf_ft_chain(struct mlx5_fs_chains *chains) +{ + return mlx5_chains_get_chain_range(chains) + 1; +} + +u32 mlx5_chains_get_prio_range(struct mlx5_fs_chains *chains) +{ + if (!mlx5_chains_prios_supported(chains)) + return 1; + + if (mlx5_chains_ignore_flow_level_supported(chains)) + return UINT_MAX; + + /* We should get here only for eswitch case */ + return FDB_TC_MAX_PRIO; +} + +static unsigned int mlx5_chains_get_level_range(struct mlx5_fs_chains *chains) +{ + if (mlx5_chains_ignore_flow_level_supported(chains)) + return UINT_MAX; + + /* Same value for FDB and NIC RX tables */ + return FDB_TC_LEVELS_PER_PRIO; +} + +void +mlx5_chains_set_end_ft(struct mlx5_fs_chains *chains, + struct mlx5_flow_table *ft) +{ + tc_end_ft(chains) = ft; +} + +#define POOL_NEXT_SIZE 0 +static int +mlx5_chains_get_avail_sz_from_pool(struct mlx5_fs_chains *chains, + int desired_size) +{ + int i, found_i = -1; + + for (i = ARRAY_SIZE(FT_POOLS) - 1; i >= 0; i--) { + if (ft_pool_left(chains)[i] && FT_POOLS[i] > desired_size) { + found_i = i; + if (desired_size != POOL_NEXT_SIZE) + break; + } + } + + if (found_i != -1) { + --ft_pool_left(chains)[found_i]; + return FT_POOLS[found_i]; + } + + return 0; +} + +static void +mlx5_chains_put_sz_to_pool(struct mlx5_fs_chains *chains, int sz) +{ + int i; + + for (i = ARRAY_SIZE(FT_POOLS) - 1; i >= 0; i--) { + if (sz == FT_POOLS[i]) { + ++ft_pool_left(chains)[i]; + return; + } + } + + WARN_ONCE(1, "Couldn't find size %d in flow table size pool", sz); +} + +static void +mlx5_chains_init_sz_pool(struct mlx5_fs_chains *chains, u32 ft_max) +{ + int i; + + for (i = ARRAY_SIZE(FT_POOLS) - 1; i >= 0; i--) + ft_pool_left(chains)[i] = + FT_POOLS[i] <= ft_max ? FT_SIZE / FT_POOLS[i] : 0; +} + +static struct mlx5_flow_table * +mlx5_chains_create_table(struct mlx5_fs_chains *chains, + u32 chain, u32 prio, u32 level) +{ + struct mlx5_flow_table_attr ft_attr = {}; + struct mlx5_flow_namespace *ns; + struct mlx5_flow_table *ft; + int sz; + + if (chains->flags & MLX5_CHAINS_FT_TUNNEL_SUPPORTED) + ft_attr.flags |= (MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT | + MLX5_FLOW_TABLE_TUNNEL_EN_DECAP); + + sz = (chain == mlx5_chains_get_nf_ft_chain(chains)) ? + mlx5_chains_get_avail_sz_from_pool(chains, FT_TBL_SZ) : + mlx5_chains_get_avail_sz_from_pool(chains, POOL_NEXT_SIZE); + if (!sz) + return ERR_PTR(-ENOSPC); + ft_attr.max_fte = sz; + + /* We use tc_default_ft(chains) as the table's next_ft till + * ignore_flow_level is allowed on FT creation and not just for FTEs. + * Instead caller should add an explicit miss rule if needed. + */ + ft_attr.next_ft = tc_default_ft(chains); + + /* The root table(chain 0, prio 1, level 0) is required to be + * connected to the previous fs_core managed prio. + * We always create it, as a managed table, in order to align with + * fs_core logic. + */ + if (!mlx5_chains_ignore_flow_level_supported(chains) || + (chain == 0 && prio == 1 && level == 0)) { + ft_attr.level = level; + ft_attr.prio = prio - 1; + ns = (chains->ns == MLX5_FLOW_NAMESPACE_FDB) ? + mlx5_get_fdb_sub_ns(chains->dev, chain) : + mlx5_get_flow_namespace(chains->dev, chains->ns); + } else { + ft_attr.flags |= MLX5_FLOW_TABLE_UNMANAGED; + ft_attr.prio = ns_to_chains_fs_prio(chains->ns); + /* Firmware doesn't allow us to create another level 0 table, + * so we create all unmanaged tables as level 1. + * + * To connect them, we use explicit miss rules with + * ignore_flow_level. Caller is responsible to create + * these rules (if needed). + */ + ft_attr.level = 1; + ns = mlx5_get_flow_namespace(chains->dev, chains->ns); + } + + ft_attr.autogroup.num_reserved_entries = 2; + ft_attr.autogroup.max_num_groups = chains->group_num; + ft = mlx5_create_auto_grouped_flow_table(ns, &ft_attr); + if (IS_ERR(ft)) { + mlx5_core_warn(chains->dev, "Failed to create chains table err %d (chain: %d, prio: %d, level: %d, size: %d)\n", + (int)PTR_ERR(ft), chain, prio, level, sz); + mlx5_chains_put_sz_to_pool(chains, sz); + return ft; + } + + return ft; +} + +static void +mlx5_chains_destroy_table(struct mlx5_fs_chains *chains, + struct mlx5_flow_table *ft) +{ + mlx5_chains_put_sz_to_pool(chains, ft->max_fte); + mlx5_destroy_flow_table(ft); +} + +static int +create_chain_restore(struct fs_chain *chain) +{ + struct mlx5_eswitch *esw = chain->chains->dev->priv.eswitch; + char modact[MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto)]; + struct mlx5_fs_chains *chains = chain->chains; + enum mlx5e_tc_attr_to_reg chain_to_reg; + struct mlx5_modify_hdr *mod_hdr; + u32 index; + int err; + + if (chain->chain == mlx5_chains_get_nf_ft_chain(chains) || + !mlx5_chains_prios_supported(chains)) + return 0; + + err = mapping_add(chains_mapping(chains), &chain->chain, &index); + if (err) + return err; + if (index == MLX5_FS_DEFAULT_FLOW_TAG) { + /* we got the special default flow tag id, so we won't know + * if we actually marked the packet with the restore rule + * we create. + * + * This case isn't possible with MLX5_FS_DEFAULT_FLOW_TAG = 0. + */ + err = mapping_add(chains_mapping(chains), + &chain->chain, &index); + mapping_remove(chains_mapping(chains), + MLX5_FS_DEFAULT_FLOW_TAG); + if (err) + return err; + } + + chain->id = index; + + if (chains->ns == MLX5_FLOW_NAMESPACE_FDB) { + chain_to_reg = CHAIN_TO_REG; + chain->restore_rule = esw_add_restore_rule(esw, chain->id); + if (IS_ERR(chain->restore_rule)) { + err = PTR_ERR(chain->restore_rule); + goto err_rule; + } + } else if (chains->ns == MLX5_FLOW_NAMESPACE_KERNEL) { + /* For NIC RX we don't need a restore rule + * since we write the metadata to reg_b + * that is passed to SW directly. + */ + chain_to_reg = NIC_CHAIN_TO_REG; + } else { + err = -EINVAL; + goto err_rule; + } + + MLX5_SET(set_action_in, modact, action_type, MLX5_ACTION_TYPE_SET); + MLX5_SET(set_action_in, modact, field, + mlx5e_tc_attr_to_reg_mappings[chain_to_reg].mfield); + MLX5_SET(set_action_in, modact, offset, + mlx5e_tc_attr_to_reg_mappings[chain_to_reg].moffset * 8); + MLX5_SET(set_action_in, modact, length, + mlx5e_tc_attr_to_reg_mappings[chain_to_reg].mlen * 8); + MLX5_SET(set_action_in, modact, data, chain->id); + mod_hdr = mlx5_modify_header_alloc(chains->dev, chains->ns, + 1, modact); + if (IS_ERR(mod_hdr)) { + err = PTR_ERR(mod_hdr); + goto err_mod_hdr; + } + chain->miss_modify_hdr = mod_hdr; + + return 0; + +err_mod_hdr: + if (!IS_ERR_OR_NULL(chain->restore_rule)) + mlx5_del_flow_rules(chain->restore_rule); +err_rule: + /* Datapath can't find this mapping, so we can safely remove it */ + mapping_remove(chains_mapping(chains), chain->id); + return err; +} + +static void destroy_chain_restore(struct fs_chain *chain) +{ + struct mlx5_fs_chains *chains = chain->chains; + + if (!chain->miss_modify_hdr) + return; + + if (chain->restore_rule) + mlx5_del_flow_rules(chain->restore_rule); + + mlx5_modify_header_dealloc(chains->dev, chain->miss_modify_hdr); + mapping_remove(chains_mapping(chains), chain->id); +} + +static struct fs_chain * +mlx5_chains_create_chain(struct mlx5_fs_chains *chains, u32 chain) +{ + struct fs_chain *chain_s = NULL; + int err; + + chain_s = kvzalloc(sizeof(*chain_s), GFP_KERNEL); + if (!chain_s) + return ERR_PTR(-ENOMEM); + + chain_s->chains = chains; + chain_s->chain = chain; + INIT_LIST_HEAD(&chain_s->prios_list); + + err = create_chain_restore(chain_s); + if (err) + goto err_restore; + + err = rhashtable_insert_fast(&chains_ht(chains), &chain_s->node, + chain_params); + if (err) + goto err_insert; + + return chain_s; + +err_insert: + destroy_chain_restore(chain_s); +err_restore: + kvfree(chain_s); + return ERR_PTR(err); +} + +static void +mlx5_chains_destroy_chain(struct fs_chain *chain) +{ + struct mlx5_fs_chains *chains = chain->chains; + + rhashtable_remove_fast(&chains_ht(chains), &chain->node, + chain_params); + + destroy_chain_restore(chain); + kvfree(chain); +} + +static struct fs_chain * +mlx5_chains_get_chain(struct mlx5_fs_chains *chains, u32 chain) +{ + struct fs_chain *chain_s; + + chain_s = rhashtable_lookup_fast(&chains_ht(chains), &chain, + chain_params); + if (!chain_s) { + chain_s = mlx5_chains_create_chain(chains, chain); + if (IS_ERR(chain_s)) + return chain_s; + } + + chain_s->ref++; + + return chain_s; +} + +static struct mlx5_flow_handle * +mlx5_chains_add_miss_rule(struct fs_chain *chain, + struct mlx5_flow_table *ft, + struct mlx5_flow_table *next_ft) +{ + struct mlx5_fs_chains *chains = chain->chains; + struct mlx5_flow_destination dest = {}; + struct mlx5_flow_act act = {}; + + act.flags = FLOW_ACT_NO_APPEND; + if (mlx5_chains_ignore_flow_level_supported(chain->chains)) + act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL; + + act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; + dest.ft = next_ft; + + if (next_ft == tc_end_ft(chains) && + chain->chain != mlx5_chains_get_nf_ft_chain(chains) && + mlx5_chains_prios_supported(chains)) { + act.modify_hdr = chain->miss_modify_hdr; + act.action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; + } + + return mlx5_add_flow_rules(ft, NULL, &act, &dest, 1); +} + +static int +mlx5_chains_update_prio_prevs(struct prio *prio, + struct mlx5_flow_table *next_ft) +{ + struct mlx5_flow_handle *miss_rules[FDB_TC_LEVELS_PER_PRIO + 1] = {}; + struct fs_chain *chain = prio->chain; + struct prio *pos; + int n = 0, err; + + if (prio->key.level) + return 0; + + /* Iterate in reverse order until reaching the level 0 rule of + * the previous priority, adding all the miss rules first, so we can + * revert them if any of them fails. + */ + pos = prio; + list_for_each_entry_continue_reverse(pos, + &chain->prios_list, + list) { + miss_rules[n] = mlx5_chains_add_miss_rule(chain, + pos->ft, + next_ft); + if (IS_ERR(miss_rules[n])) { + err = PTR_ERR(miss_rules[n]); + goto err_prev_rule; + } + + n++; + if (!pos->key.level) + break; + } + + /* Success, delete old miss rules, and update the pointers. */ + n = 0; + pos = prio; + list_for_each_entry_continue_reverse(pos, + &chain->prios_list, + list) { + mlx5_del_flow_rules(pos->miss_rule); + + pos->miss_rule = miss_rules[n]; + pos->next_ft = next_ft; + + n++; + if (!pos->key.level) + break; + } + + return 0; + +err_prev_rule: + while (--n >= 0) + mlx5_del_flow_rules(miss_rules[n]); + + return err; +} + +static void +mlx5_chains_put_chain(struct fs_chain *chain) +{ + if (--chain->ref == 0) + mlx5_chains_destroy_chain(chain); +} + +static struct prio * +mlx5_chains_create_prio(struct mlx5_fs_chains *chains, + u32 chain, u32 prio, u32 level) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_flow_handle *miss_rule = NULL; + struct mlx5_flow_group *miss_group; + struct mlx5_flow_table *next_ft; + struct mlx5_flow_table *ft; + struct prio *prio_s = NULL; + struct fs_chain *chain_s; + struct list_head *pos; + u32 *flow_group_in; + int err; + + chain_s = mlx5_chains_get_chain(chains, chain); + if (IS_ERR(chain_s)) + return ERR_CAST(chain_s); + + prio_s = kvzalloc(sizeof(*prio_s), GFP_KERNEL); + flow_group_in = kvzalloc(inlen, GFP_KERNEL); + if (!prio_s || !flow_group_in) { + err = -ENOMEM; + goto err_alloc; + } + + /* Chain's prio list is sorted by prio and level. + * And all levels of some prio point to the next prio's level 0. + * Example list (prio, level): + * (3,0)->(3,1)->(5,0)->(5,1)->(6,1)->(7,0) + * In hardware, we will we have the following pointers: + * (3,0) -> (5,0) -> (7,0) -> Slow path + * (3,1) -> (5,0) + * (5,1) -> (7,0) + * (6,1) -> (7,0) + */ + + /* Default miss for each chain: */ + next_ft = (chain == mlx5_chains_get_nf_ft_chain(chains)) ? + tc_default_ft(chains) : + tc_end_ft(chains); + list_for_each(pos, &chain_s->prios_list) { + struct prio *p = list_entry(pos, struct prio, list); + + /* exit on first pos that is larger */ + if (prio < p->key.prio || (prio == p->key.prio && + level < p->key.level)) { + /* Get next level 0 table */ + next_ft = p->key.level == 0 ? p->ft : p->next_ft; + break; + } + } + + ft = mlx5_chains_create_table(chains, chain, prio, level); + if (IS_ERR(ft)) { + err = PTR_ERR(ft); + goto err_create; + } + + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, + ft->max_fte - 2); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, + ft->max_fte - 1); + miss_group = mlx5_create_flow_group(ft, flow_group_in); + if (IS_ERR(miss_group)) { + err = PTR_ERR(miss_group); + goto err_group; + } + + /* Add miss rule to next_ft */ + miss_rule = mlx5_chains_add_miss_rule(chain_s, ft, next_ft); + if (IS_ERR(miss_rule)) { + err = PTR_ERR(miss_rule); + goto err_miss_rule; + } + + prio_s->miss_group = miss_group; + prio_s->miss_rule = miss_rule; + prio_s->next_ft = next_ft; + prio_s->chain = chain_s; + prio_s->key.chain = chain; + prio_s->key.prio = prio; + prio_s->key.level = level; + prio_s->ft = ft; + + err = rhashtable_insert_fast(&prios_ht(chains), &prio_s->node, + prio_params); + if (err) + goto err_insert; + + list_add(&prio_s->list, pos->prev); + + /* Table is ready, connect it */ + err = mlx5_chains_update_prio_prevs(prio_s, ft); + if (err) + goto err_update; + + kvfree(flow_group_in); + return prio_s; + +err_update: + list_del(&prio_s->list); + rhashtable_remove_fast(&prios_ht(chains), &prio_s->node, + prio_params); +err_insert: + mlx5_del_flow_rules(miss_rule); +err_miss_rule: + mlx5_destroy_flow_group(miss_group); +err_group: + mlx5_chains_destroy_table(chains, ft); +err_create: +err_alloc: + kvfree(prio_s); + kvfree(flow_group_in); + mlx5_chains_put_chain(chain_s); + return ERR_PTR(err); +} + +static void +mlx5_chains_destroy_prio(struct mlx5_fs_chains *chains, + struct prio *prio) +{ + struct fs_chain *chain = prio->chain; + + WARN_ON(mlx5_chains_update_prio_prevs(prio, + prio->next_ft)); + + list_del(&prio->list); + rhashtable_remove_fast(&prios_ht(chains), &prio->node, + prio_params); + mlx5_del_flow_rules(prio->miss_rule); + mlx5_destroy_flow_group(prio->miss_group); + mlx5_chains_destroy_table(chains, prio->ft); + mlx5_chains_put_chain(chain); + kvfree(prio); +} + +struct mlx5_flow_table * +mlx5_chains_get_table(struct mlx5_fs_chains *chains, u32 chain, u32 prio, + u32 level) +{ + struct mlx5_flow_table *prev_fts; + struct prio *prio_s; + struct prio_key key; + int l = 0; + + if ((chain > mlx5_chains_get_chain_range(chains) && + chain != mlx5_chains_get_nf_ft_chain(chains)) || + prio > mlx5_chains_get_prio_range(chains) || + level > mlx5_chains_get_level_range(chains)) + return ERR_PTR(-EOPNOTSUPP); + + /* create earlier levels for correct fs_core lookup when + * connecting tables. + */ + for (l = 0; l < level; l++) { + prev_fts = mlx5_chains_get_table(chains, chain, prio, l); + if (IS_ERR(prev_fts)) { + prio_s = ERR_CAST(prev_fts); + goto err_get_prevs; + } + } + + key.chain = chain; + key.prio = prio; + key.level = level; + + mutex_lock(&chains_lock(chains)); + prio_s = rhashtable_lookup_fast(&prios_ht(chains), &key, + prio_params); + if (!prio_s) { + prio_s = mlx5_chains_create_prio(chains, chain, + prio, level); + if (IS_ERR(prio_s)) + goto err_create_prio; + } + + ++prio_s->ref; + mutex_unlock(&chains_lock(chains)); + + return prio_s->ft; + +err_create_prio: + mutex_unlock(&chains_lock(chains)); +err_get_prevs: + while (--l >= 0) + mlx5_chains_put_table(chains, chain, prio, l); + return ERR_CAST(prio_s); +} + +void +mlx5_chains_put_table(struct mlx5_fs_chains *chains, u32 chain, u32 prio, + u32 level) +{ + struct prio *prio_s; + struct prio_key key; + + key.chain = chain; + key.prio = prio; + key.level = level; + + mutex_lock(&chains_lock(chains)); + prio_s = rhashtable_lookup_fast(&prios_ht(chains), &key, + prio_params); + if (!prio_s) + goto err_get_prio; + + if (--prio_s->ref == 0) + mlx5_chains_destroy_prio(chains, prio_s); + mutex_unlock(&chains_lock(chains)); + + while (level-- > 0) + mlx5_chains_put_table(chains, chain, prio, level); + + return; + +err_get_prio: + mutex_unlock(&chains_lock(chains)); + WARN_ONCE(1, + "Couldn't find table: (chain: %d prio: %d level: %d)", + chain, prio, level); +} + +struct mlx5_flow_table * +mlx5_chains_get_tc_end_ft(struct mlx5_fs_chains *chains) +{ + return tc_end_ft(chains); +} + +struct mlx5_flow_table * +mlx5_chains_create_global_table(struct mlx5_fs_chains *chains) +{ + u32 chain, prio, level; + int err; + + if (!mlx5_chains_ignore_flow_level_supported(chains)) { + err = -EOPNOTSUPP; + + mlx5_core_warn(chains->dev, + "Couldn't create global flow table, ignore_flow_level not supported."); + goto err_ignore; + } + + chain = mlx5_chains_get_chain_range(chains), + prio = mlx5_chains_get_prio_range(chains); + level = mlx5_chains_get_level_range(chains); + + return mlx5_chains_create_table(chains, chain, prio, level); + +err_ignore: + return ERR_PTR(err); +} + +void +mlx5_chains_destroy_global_table(struct mlx5_fs_chains *chains, + struct mlx5_flow_table *ft) +{ + mlx5_chains_destroy_table(chains, ft); +} + +static struct mlx5_fs_chains * +mlx5_chains_init(struct mlx5_core_dev *dev, struct mlx5_chains_attr *attr) +{ + struct mlx5_fs_chains *chains_priv; + struct mapping_ctx *mapping; + u32 max_flow_counter; + int err; + + chains_priv = kzalloc(sizeof(*chains_priv), GFP_KERNEL); + if (!chains_priv) + return ERR_PTR(-ENOMEM); + + max_flow_counter = (MLX5_CAP_GEN(dev, max_flow_counter_31_16) << 16) | + MLX5_CAP_GEN(dev, max_flow_counter_15_0); + + mlx5_core_dbg(dev, + "Init flow table chains, max counters(%d), groups(%d), max flow table size(%d)\n", + max_flow_counter, attr->max_grp_num, attr->max_ft_sz); + + chains_priv->dev = dev; + chains_priv->flags = attr->flags; + chains_priv->ns = attr->ns; + chains_priv->group_num = attr->max_grp_num; + tc_default_ft(chains_priv) = tc_end_ft(chains_priv) = attr->default_ft; + + mlx5_core_info(dev, "Supported tc offload range - chains: %u, prios: %u\n", + mlx5_chains_get_chain_range(chains_priv), + mlx5_chains_get_prio_range(chains_priv)); + + mlx5_chains_init_sz_pool(chains_priv, attr->max_ft_sz); + + err = rhashtable_init(&chains_ht(chains_priv), &chain_params); + if (err) + goto init_chains_ht_err; + + err = rhashtable_init(&prios_ht(chains_priv), &prio_params); + if (err) + goto init_prios_ht_err; + + mapping = mapping_create(sizeof(u32), attr->max_restore_tag, + true); + if (IS_ERR(mapping)) { + err = PTR_ERR(mapping); + goto mapping_err; + } + chains_mapping(chains_priv) = mapping; + + mutex_init(&chains_lock(chains_priv)); + + return chains_priv; + +mapping_err: + rhashtable_destroy(&prios_ht(chains_priv)); +init_prios_ht_err: + rhashtable_destroy(&chains_ht(chains_priv)); +init_chains_ht_err: + kfree(chains_priv); + return ERR_PTR(err); +} + +static void +mlx5_chains_cleanup(struct mlx5_fs_chains *chains) +{ + mutex_destroy(&chains_lock(chains)); + mapping_destroy(chains_mapping(chains)); + rhashtable_destroy(&prios_ht(chains)); + rhashtable_destroy(&chains_ht(chains)); + + kfree(chains); +} + +struct mlx5_fs_chains * +mlx5_chains_create(struct mlx5_core_dev *dev, struct mlx5_chains_attr *attr) +{ + struct mlx5_fs_chains *chains; + + chains = mlx5_chains_init(dev, attr); + + return chains; +} + +void +mlx5_chains_destroy(struct mlx5_fs_chains *chains) +{ + mlx5_chains_cleanup(chains); +} + +int +mlx5_chains_get_chain_mapping(struct mlx5_fs_chains *chains, u32 chain, + u32 *chain_mapping) +{ + return mapping_add(chains_mapping(chains), &chain, chain_mapping); +} + +int +mlx5_chains_put_chain_mapping(struct mlx5_fs_chains *chains, u32 chain_mapping) +{ + return mapping_remove(chains_mapping(chains), chain_mapping); +} + +int mlx5_get_chain_for_tag(struct mlx5_fs_chains *chains, u32 tag, + u32 *chain) +{ + int err; + + err = mapping_find(chains_mapping(chains), tag, chain); + if (err) { + mlx5_core_warn(chains->dev, "Can't find chain for tag: %d\n", tag); + return -ENOENT; + } + + return 0; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.h new file mode 100644 index 000000000000..6d5be31b05dd --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.h @@ -0,0 +1,93 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ +/* Copyright (c) 2020 Mellanox Technologies. */ + +#ifndef __ML5_ESW_CHAINS_H__ +#define __ML5_ESW_CHAINS_H__ + +#include <linux/mlx5/fs.h> + +struct mlx5_fs_chains; + +enum mlx5_chains_flags { + MLX5_CHAINS_AND_PRIOS_SUPPORTED = BIT(0), + MLX5_CHAINS_IGNORE_FLOW_LEVEL_SUPPORTED = BIT(1), + MLX5_CHAINS_FT_TUNNEL_SUPPORTED = BIT(2), +}; + +struct mlx5_chains_attr { + enum mlx5_flow_namespace_type ns; + u32 flags; + u32 max_ft_sz; + u32 max_grp_num; + struct mlx5_flow_table *default_ft; + u32 max_restore_tag; +}; + +#if IS_ENABLED(CONFIG_MLX5_CLS_ACT) + +bool +mlx5_chains_prios_supported(struct mlx5_fs_chains *chains); +bool +mlx5_chains_backwards_supported(struct mlx5_fs_chains *chains); +u32 +mlx5_chains_get_prio_range(struct mlx5_fs_chains *chains); +u32 +mlx5_chains_get_chain_range(struct mlx5_fs_chains *chains); +u32 +mlx5_chains_get_nf_ft_chain(struct mlx5_fs_chains *chains); + +struct mlx5_flow_table * +mlx5_chains_get_table(struct mlx5_fs_chains *chains, u32 chain, u32 prio, + u32 level); +void +mlx5_chains_put_table(struct mlx5_fs_chains *chains, u32 chain, u32 prio, + u32 level); + +struct mlx5_flow_table * +mlx5_chains_get_tc_end_ft(struct mlx5_fs_chains *chains); + +struct mlx5_flow_table * +mlx5_chains_create_global_table(struct mlx5_fs_chains *chains); +void +mlx5_chains_destroy_global_table(struct mlx5_fs_chains *chains, + struct mlx5_flow_table *ft); + +int +mlx5_chains_get_chain_mapping(struct mlx5_fs_chains *chains, u32 chain, + u32 *chain_mapping); +int +mlx5_chains_put_chain_mapping(struct mlx5_fs_chains *chains, + u32 chain_mapping); + +struct mlx5_fs_chains * +mlx5_chains_create(struct mlx5_core_dev *dev, struct mlx5_chains_attr *attr); +void mlx5_chains_destroy(struct mlx5_fs_chains *chains); + +int +mlx5_get_chain_for_tag(struct mlx5_fs_chains *chains, u32 tag, u32 *chain); + +void +mlx5_chains_set_end_ft(struct mlx5_fs_chains *chains, + struct mlx5_flow_table *ft); + +#else /* CONFIG_MLX5_CLS_ACT */ + +static inline struct mlx5_flow_table * +mlx5_chains_get_table(struct mlx5_fs_chains *chains, u32 chain, u32 prio, + u32 level) { return ERR_PTR(-EOPNOTSUPP); } +static inline void +mlx5_chains_put_table(struct mlx5_fs_chains *chains, u32 chain, u32 prio, + u32 level) {}; + +static inline struct mlx5_flow_table * +mlx5_chains_get_tc_end_ft(struct mlx5_fs_chains *chains) { return ERR_PTR(-EOPNOTSUPP); } + +static inline struct mlx5_fs_chains * +mlx5_chains_create(struct mlx5_core_dev *dev, struct mlx5_chains_attr *attr) +{ return NULL; } +static inline void +mlx5_chains_destroy(struct mlx5_fs_chains *chains) {}; + +#endif /* CONFIG_MLX5_CLS_ACT */ + +#endif /* __ML5_ESW_CHAINS_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index ce43e3feccd9..8ff207aa1479 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -57,6 +57,7 @@ #include "lib/mpfs.h" #include "eswitch.h" #include "devlink.h" +#include "fw_reset.h" #include "lib/mlx5.h" #include "fpga/core.h" #include "fpga/ipsec.h" @@ -548,6 +549,9 @@ static int handle_hca_cap(struct mlx5_core_dev *dev, void *set_ctx) if (MLX5_CAP_GEN_MAX(dev, dct)) MLX5_SET(cmd_hca_cap, set_hca_cap, dct, 1); + if (MLX5_CAP_GEN_MAX(dev, pci_sync_for_fw_update_event)) + MLX5_SET(cmd_hca_cap, set_hca_cap, pci_sync_for_fw_update_event, 1); + if (MLX5_CAP_GEN_MAX(dev, num_vhca_ports)) MLX5_SET(cmd_hca_cap, set_hca_cap, @@ -739,7 +743,7 @@ static int mlx5_pci_init(struct mlx5_core_dev *dev, struct pci_dev *pdev, pci_set_drvdata(dev->pdev, dev); dev->bar_addr = pci_resource_start(pdev, 0); - priv->numa_node = dev_to_node(&dev->pdev->dev); + priv->numa_node = dev_to_node(mlx5_core_dma_dev(dev)); err = mlx5_pci_enable_device(dev); if (err) { @@ -832,6 +836,12 @@ static int mlx5_init_once(struct mlx5_core_dev *dev) goto err_eq_cleanup; } + err = mlx5_fw_reset_init(dev); + if (err) { + mlx5_core_err(dev, "failed to initialize fw reset events\n"); + goto err_events_cleanup; + } + mlx5_cq_debugfs_init(dev); mlx5_init_reserved_gids(dev); @@ -893,6 +903,8 @@ err_tables_cleanup: mlx5_geneve_destroy(dev->geneve); mlx5_vxlan_destroy(dev->vxlan); mlx5_cq_debugfs_cleanup(dev); + mlx5_fw_reset_cleanup(dev); +err_events_cleanup: mlx5_events_cleanup(dev); err_eq_cleanup: mlx5_eq_table_cleanup(dev); @@ -920,6 +932,7 @@ static void mlx5_cleanup_once(struct mlx5_core_dev *dev) mlx5_cleanup_clock(dev); mlx5_cleanup_reserved_gids(dev); mlx5_cq_debugfs_cleanup(dev); + mlx5_fw_reset_cleanup(dev); mlx5_events_cleanup(dev); mlx5_eq_table_cleanup(dev); mlx5_irq_table_cleanup(dev); @@ -1078,6 +1091,7 @@ static int mlx5_load(struct mlx5_core_dev *dev) goto err_fw_tracer; } + mlx5_fw_reset_events_start(dev); mlx5_hv_vhca_init(dev->hv_vhca); err = mlx5_rsc_dump_init(dev); @@ -1139,6 +1153,7 @@ err_fpga_start: mlx5_rsc_dump_cleanup(dev); err_rsc_dump: mlx5_hv_vhca_cleanup(dev->hv_vhca); + mlx5_fw_reset_events_stop(dev); mlx5_fw_tracer_cleanup(dev->tracer); err_fw_tracer: mlx5_eq_table_destroy(dev); @@ -1161,6 +1176,7 @@ static void mlx5_unload(struct mlx5_core_dev *dev) mlx5_fpga_device_stop(dev); mlx5_rsc_dump_cleanup(dev); mlx5_hv_vhca_cleanup(dev->hv_vhca); + mlx5_fw_reset_events_stop(dev); mlx5_fw_tracer_cleanup(dev->tracer); mlx5_eq_table_destroy(dev); mlx5_irq_table_destroy(dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index fc1649dac11b..8cec85ab419d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -100,6 +100,11 @@ do { \ __func__, __LINE__, current->pid, \ ##__VA_ARGS__) +static inline struct device *mlx5_core_dma_dev(struct mlx5_core_dev *dev) +{ + return &dev->pdev->dev; +} + enum { MLX5_CMD_DATA, /* print command payload only */ MLX5_CMD_TIME, /* print command execution time */ @@ -123,6 +128,8 @@ int mlx5_cmd_force_teardown_hca(struct mlx5_core_dev *dev); int mlx5_cmd_fast_teardown_hca(struct mlx5_core_dev *dev); void mlx5_enter_error_state(struct mlx5_core_dev *dev, bool force); void mlx5_error_sw_reset(struct mlx5_core_dev *dev); +u32 mlx5_health_check_fatal_sensors(struct mlx5_core_dev *dev); +int mlx5_health_wait_pci_up(struct mlx5_core_dev *dev); void mlx5_disable_device(struct mlx5_core_dev *dev); void mlx5_recover_device(struct mlx5_core_dev *dev); int mlx5_sriov_init(struct mlx5_core_dev *dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c index c0e18f2ade99..150638814517 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c @@ -238,7 +238,7 @@ static void free_fwp(struct mlx5_core_dev *dev, struct fw_page *fwp, rb_erase(&fwp->rb_node, root); if (in_free_list) list_del(&fwp->list); - dma_unmap_page(dev->device, fwp->addr & MLX5_U64_4K_PAGE_MASK, + dma_unmap_page(mlx5_core_dma_dev(dev), fwp->addr & MLX5_U64_4K_PAGE_MASK, PAGE_SIZE, DMA_BIDIRECTIONAL); __free_page(fwp->page); kfree(fwp); @@ -265,7 +265,7 @@ static void free_4k(struct mlx5_core_dev *dev, u64 addr, u32 func_id) static int alloc_system_page(struct mlx5_core_dev *dev, u16 func_id) { - struct device *device = dev->device; + struct device *device = mlx5_core_dma_dev(dev); int nid = dev_to_node(device); struct page *page; u64 zero_addr = 1; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c index c63f727273d8..7df883686d46 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c @@ -203,7 +203,6 @@ static int dr_matcher_set_ste_builders(struct mlx5dr_matcher *matcher, struct mlx5dr_domain_rx_tx *nic_dmn = nic_matcher->nic_tbl->nic_dmn; struct mlx5dr_domain *dmn = matcher->tbl->dmn; struct mlx5dr_match_param mask = {}; - struct mlx5dr_match_misc3 *misc3; struct mlx5dr_ste_build *sb; bool inner, rx; int idx = 0; @@ -252,18 +251,14 @@ static int dr_matcher_set_ste_builders(struct mlx5dr_matcher *matcher, if (dr_mask_is_gvmi_or_qpn_set(&mask.misc) && (dmn->type == MLX5DR_DOMAIN_TYPE_FDB || dmn->type == MLX5DR_DOMAIN_TYPE_NIC_RX)) { - ret = mlx5dr_ste_build_src_gvmi_qpn(&sb[idx++], &mask, - dmn, inner, rx); - if (ret) - return ret; + mlx5dr_ste_build_src_gvmi_qpn(&sb[idx++], &mask, + dmn, inner, rx); } if (dr_mask_is_smac_set(&mask.outer) && dr_mask_is_dmac_set(&mask.outer)) { - ret = mlx5dr_ste_build_eth_l2_src_des(&sb[idx++], &mask, - inner, rx); - if (ret) - return ret; + mlx5dr_ste_build_eth_l2_src_des(&sb[idx++], &mask, + inner, rx); } if (dr_mask_is_smac_set(&mask.outer)) @@ -313,8 +308,7 @@ static int dr_matcher_set_ste_builders(struct mlx5dr_matcher *matcher, mlx5dr_ste_build_flex_parser_0(&sb[idx++], &mask, inner, rx); - misc3 = &mask.misc3; - if ((DR_MASK_IS_FLEX_PARSER_ICMPV4_SET(misc3) && + if ((DR_MASK_IS_FLEX_PARSER_ICMPV4_SET(&mask.misc3) && mlx5dr_matcher_supp_flex_parser_icmp_v4(&dmn->info.caps)) || (dr_mask_is_flex_parser_icmpv6_set(&mask.misc3) && mlx5dr_matcher_supp_flex_parser_icmp_v6(&dmn->info.caps))) { @@ -340,10 +334,8 @@ static int dr_matcher_set_ste_builders(struct mlx5dr_matcher *matcher, if (dr_mask_is_smac_set(&mask.inner) && dr_mask_is_dmac_set(&mask.inner)) { - ret = mlx5dr_ste_build_eth_l2_src_des(&sb[idx++], - &mask, inner, rx); - if (ret) - return ret; + mlx5dr_ste_build_eth_l2_src_des(&sb[idx++], + &mask, inner, rx); } if (dr_mask_is_smac_set(&mask.inner)) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c index 6ec5106bc472..b3c9dc032026 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c @@ -242,7 +242,7 @@ dr_rule_rehash_copy_ste(struct mlx5dr_matcher *matcher, new_idx = mlx5dr_ste_calc_hash_index(hw_ste, new_htbl); new_ste = &new_htbl->ste_arr[new_idx]; - if (mlx5dr_ste_not_used_ste(new_ste)) { + if (mlx5dr_ste_is_not_used(new_ste)) { mlx5dr_htbl_get(new_htbl); list_add_tail(&new_ste->miss_list_node, mlx5dr_ste_get_miss_list(new_ste)); @@ -335,7 +335,7 @@ static int dr_rule_rehash_copy_htbl(struct mlx5dr_matcher *matcher, for (i = 0; i < cur_entries; i++) { cur_ste = &cur_htbl->ste_arr[i]; - if (mlx5dr_ste_not_used_ste(cur_ste)) /* Empty, nothing to copy */ + if (mlx5dr_ste_is_not_used(cur_ste)) /* Empty, nothing to copy */ continue; err = dr_rule_rehash_copy_miss_list(matcher, @@ -791,7 +791,7 @@ again: miss_list = &cur_htbl->chunk->miss_list[index]; ste = &cur_htbl->ste_arr[index]; - if (mlx5dr_ste_not_used_ste(ste)) { + if (mlx5dr_ste_is_not_used(ste)) { if (dr_rule_handle_empty_entry(matcher, nic_matcher, cur_htbl, ste, ste_location, hw_ste, miss_list, @@ -985,31 +985,28 @@ static enum mlx5dr_ipv dr_rule_get_ipv(struct mlx5dr_match_spec *spec) static bool dr_rule_skip(enum mlx5dr_domain_type domain, enum mlx5dr_ste_entry_type ste_type, struct mlx5dr_match_param *mask, - struct mlx5dr_match_param *value) + struct mlx5dr_match_param *value, + u32 flow_source) { + bool rx = ste_type == MLX5DR_STE_TYPE_RX; + if (domain != MLX5DR_DOMAIN_TYPE_FDB) return false; if (mask->misc.source_port) { - if (ste_type == MLX5DR_STE_TYPE_RX) - if (value->misc.source_port != WIRE_PORT) - return true; + if (rx && value->misc.source_port != WIRE_PORT) + return true; - if (ste_type == MLX5DR_STE_TYPE_TX) - if (value->misc.source_port == WIRE_PORT) - return true; + if (!rx && value->misc.source_port == WIRE_PORT) + return true; } - /* Metadata C can be used to describe the source vport */ - if (mask->misc2.metadata_reg_c_0) { - if (ste_type == MLX5DR_STE_TYPE_RX) - if ((value->misc2.metadata_reg_c_0 & WIRE_PORT) != WIRE_PORT) - return true; + if (rx && flow_source == MLX5_FLOW_CONTEXT_FLOW_SOURCE_LOCAL_VPORT) + return true; + + if (!rx && flow_source == MLX5_FLOW_CONTEXT_FLOW_SOURCE_UPLINK) + return true; - if (ste_type == MLX5DR_STE_TYPE_TX) - if ((value->misc2.metadata_reg_c_0 & WIRE_PORT) == WIRE_PORT) - return true; - } return false; } @@ -1038,7 +1035,8 @@ dr_rule_create_rule_nic(struct mlx5dr_rule *rule, INIT_LIST_HEAD(&nic_rule->rule_members_list); - if (dr_rule_skip(dmn->type, nic_dmn->ste_type, &matcher->mask, param)) + if (dr_rule_skip(dmn->type, nic_dmn->ste_type, &matcher->mask, param, + rule->flow_source)) return 0; hw_ste_arr = kzalloc(DR_RULE_MAX_STE_CHAIN * DR_STE_SIZE, GFP_KERNEL); @@ -1173,7 +1171,8 @@ static struct mlx5dr_rule * dr_rule_create_rule(struct mlx5dr_matcher *matcher, struct mlx5dr_match_parameters *value, size_t num_actions, - struct mlx5dr_action *actions[]) + struct mlx5dr_action *actions[], + u32 flow_source) { struct mlx5dr_domain *dmn = matcher->tbl->dmn; struct mlx5dr_match_param param = {}; @@ -1188,6 +1187,7 @@ dr_rule_create_rule(struct mlx5dr_matcher *matcher, return NULL; rule->matcher = matcher; + rule->flow_source = flow_source; INIT_LIST_HEAD(&rule->rule_actions_list); ret = dr_rule_add_action_members(rule, num_actions, actions); @@ -1232,13 +1232,14 @@ free_rule: struct mlx5dr_rule *mlx5dr_rule_create(struct mlx5dr_matcher *matcher, struct mlx5dr_match_parameters *value, size_t num_actions, - struct mlx5dr_action *actions[]) + struct mlx5dr_action *actions[], + u32 flow_source) { struct mlx5dr_rule *rule; refcount_inc(&matcher->refcount); - rule = dr_rule_create_rule(matcher, value, num_actions, actions); + rule = dr_rule_create_rule(matcher, value, num_actions, actions, flow_source); if (!rule) refcount_dec(&matcher->refcount); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c index 2ca79b9bde1f..24dede1b0a20 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c @@ -466,10 +466,10 @@ int mlx5dr_send_postsend_htbl(struct mlx5dr_domain *dmn, * need to add the bit_mask */ for (j = 0; j < num_stes_per_iter; j++) { - u8 *hw_ste = htbl->ste_arr[ste_index + j].hw_ste; + struct mlx5dr_ste *ste = &htbl->ste_arr[ste_index + j]; u32 ste_off = j * DR_STE_SIZE; - if (mlx5dr_ste_is_not_valid_entry(hw_ste)) { + if (mlx5dr_ste_is_not_used(ste)) { memcpy(data + ste_off, formatted_ste, DR_STE_SIZE); } else { @@ -831,7 +831,7 @@ static struct mlx5dr_mr *dr_reg_mr(struct mlx5_core_dev *mdev, if (!mr) return NULL; - dma_device = &mdev->pdev->dev; + dma_device = mlx5_core_dma_dev(mdev); dma_addr = dma_map_single(dma_device, buf, size, DMA_BIDIRECTIONAL); err = dma_mapping_error(dma_device, dma_addr); @@ -860,7 +860,7 @@ static struct mlx5dr_mr *dr_reg_mr(struct mlx5_core_dev *mdev, static void dr_dereg_mr(struct mlx5_core_dev *mdev, struct mlx5dr_mr *mr) { mlx5_core_destroy_mkey(mdev, &mr->mkey); - dma_unmap_single(&mdev->pdev->dev, mr->dma_addr, mr->size, + dma_unmap_single(mlx5_core_dma_dev(mdev), mr->dma_addr, mr->size, DMA_BIDIRECTIONAL); kfree(mr); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c index 00c2f598f034..b01aaec75622 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c @@ -155,6 +155,13 @@ static u16 dr_ste_conv_bit_to_byte_mask(u8 *bit_mask) return byte_mask; } +static u8 *mlx5dr_ste_get_tag(u8 *hw_ste_p) +{ + struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; + + return hw_ste->tag; +} + void mlx5dr_ste_set_bit_mask(u8 *hw_ste_p, u8 *bit_mask) { struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; @@ -549,25 +556,6 @@ void mlx5dr_ste_always_miss_addr(struct mlx5dr_ste *ste, u64 miss_addr) dr_ste_set_always_miss((struct dr_hw_ste_format *)ste->hw_ste); } -/* The assumption here is that we don't update the ste->hw_ste if it is not - * used ste, so it will be all zero, checking the next_lu_type. - */ -bool mlx5dr_ste_is_not_valid_entry(u8 *p_hw_ste) -{ - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)p_hw_ste; - - if (MLX5_GET(ste_general, hw_ste, next_lu_type) == - MLX5DR_STE_LU_TYPE_NOP) - return true; - - return false; -} - -bool mlx5dr_ste_not_used_ste(struct mlx5dr_ste *ste) -{ - return !ste->refcount; -} - /* Init one ste as a pattern for ste data array */ void mlx5dr_ste_set_formatted_ste(u16 gvmi, struct mlx5dr_domain_rx_tx *nic_dmn, @@ -728,7 +716,14 @@ int mlx5dr_ste_build_pre_check(struct mlx5dr_domain *dmn, { if (!value && (match_criteria & DR_MATCHER_CRITERIA_MISC)) { if (mask->misc.source_port && mask->misc.source_port != 0xffff) { - mlx5dr_err(dmn, "Partial mask source_port is not supported\n"); + mlx5dr_err(dmn, + "Partial mask source_port is not supported\n"); + return -EINVAL; + } + if (mask->misc.source_eswitch_owner_vhca_id && + mask->misc.source_eswitch_owner_vhca_id != 0xffff) { + mlx5dr_err(dmn, + "Partial mask source_eswitch_owner_vhca_id is not supported\n"); return -EINVAL; } } @@ -760,7 +755,7 @@ int mlx5dr_ste_build_ste_arr(struct mlx5dr_matcher *matcher, mlx5dr_ste_set_bit_mask(ste_arr, sb->bit_mask); - ret = sb->ste_build_tag_func(value, sb, ste_arr); + ret = sb->ste_build_tag_func(value, sb, mlx5dr_ste_get_tag(ste_arr)); if (ret) return ret; @@ -778,8 +773,8 @@ int mlx5dr_ste_build_ste_arr(struct mlx5dr_matcher *matcher, return 0; } -static int dr_ste_build_eth_l2_src_des_bit_mask(struct mlx5dr_match_param *value, - bool inner, u8 *bit_mask) +static void dr_ste_build_eth_l2_src_des_bit_mask(struct mlx5dr_match_param *value, + bool inner, u8 *bit_mask) { struct mlx5dr_match_spec *mask = inner ? &value->inner : &value->outer; @@ -807,13 +802,6 @@ static int dr_ste_build_eth_l2_src_des_bit_mask(struct mlx5dr_match_param *value MLX5_SET(ste_eth_l2_src_dst, bit_mask, first_vlan_qualifier, -1); mask->svlan_tag = 0; } - - if (mask->cvlan_tag || mask->svlan_tag) { - pr_info("Invalid c/svlan mask configuration\n"); - return -EINVAL; - } - - return 0; } static void dr_ste_copy_mask_misc(char *mask, struct mlx5dr_match_misc *spec) @@ -1059,11 +1047,9 @@ void mlx5dr_ste_copy_param(u8 match_criteria, static int dr_ste_build_eth_l2_src_des_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { struct mlx5dr_match_spec *spec = sb->inner ? &value->inner : &value->outer; - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(eth_l2_src_dst, tag, dmac_47_16, spec, dmac_47_16); DR_STE_SET_TAG(eth_l2_src_dst, tag, dmac_15_0, spec, dmac_15_0); @@ -1104,23 +1090,17 @@ static int dr_ste_build_eth_l2_src_des_tag(struct mlx5dr_match_param *value, return 0; } -int mlx5dr_ste_build_eth_l2_src_des(struct mlx5dr_ste_build *sb, - struct mlx5dr_match_param *mask, - bool inner, bool rx) +void mlx5dr_ste_build_eth_l2_src_des(struct mlx5dr_ste_build *sb, + struct mlx5dr_match_param *mask, + bool inner, bool rx) { - int ret; - - ret = dr_ste_build_eth_l2_src_des_bit_mask(mask, inner, sb->bit_mask); - if (ret) - return ret; + dr_ste_build_eth_l2_src_des_bit_mask(mask, inner, sb->bit_mask); sb->rx = rx; sb->inner = inner; sb->lu_type = DR_STE_CALC_LU_TYPE(ETHL2_SRC_DST, rx, inner); sb->byte_mask = dr_ste_conv_bit_to_byte_mask(sb->bit_mask); sb->ste_build_tag_func = &dr_ste_build_eth_l2_src_des_tag; - - return 0; } static void dr_ste_build_eth_l3_ipv6_dst_bit_mask(struct mlx5dr_match_param *value, @@ -1136,11 +1116,9 @@ static void dr_ste_build_eth_l3_ipv6_dst_bit_mask(struct mlx5dr_match_param *val static int dr_ste_build_eth_l3_ipv6_dst_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_spec *spec = sb->inner ? &value->inner : &value->outer; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(eth_l3_ipv6_dst, tag, dst_ip_127_96, spec, dst_ip_127_96); DR_STE_SET_TAG(eth_l3_ipv6_dst, tag, dst_ip_95_64, spec, dst_ip_95_64); @@ -1176,11 +1154,9 @@ static void dr_ste_build_eth_l3_ipv6_src_bit_mask(struct mlx5dr_match_param *val static int dr_ste_build_eth_l3_ipv6_src_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { struct mlx5dr_match_spec *spec = sb->inner ? &value->inner : &value->outer; - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(eth_l3_ipv6_src, tag, src_ip_127_96, spec, src_ip_127_96); DR_STE_SET_TAG(eth_l3_ipv6_src, tag, src_ip_95_64, spec, src_ip_95_64); @@ -1238,11 +1214,9 @@ static void dr_ste_build_eth_l3_ipv4_5_tuple_bit_mask(struct mlx5dr_match_param static int dr_ste_build_eth_l3_ipv4_5_tuple_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_spec *spec = sb->inner ? &value->inner : &value->outer; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(eth_l3_ipv4_5_tuple, tag, destination_address, spec, dst_ip_31_0); DR_STE_SET_TAG(eth_l3_ipv4_5_tuple, tag, source_address, spec, src_ip_31_0); @@ -1328,12 +1302,10 @@ dr_ste_build_eth_l2_src_or_dst_bit_mask(struct mlx5dr_match_param *value, } static int dr_ste_build_eth_l2_src_or_dst_tag(struct mlx5dr_match_param *value, - bool inner, u8 *hw_ste_p) + bool inner, u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_spec *spec = inner ? &value->inner : &value->outer; struct mlx5dr_match_misc *misc_spec = &value->misc; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(eth_l2_src, tag, first_vlan_id, spec, first_vid); DR_STE_SET_TAG(eth_l2_src, tag, first_cfi, spec, first_cfi); @@ -1403,16 +1375,14 @@ static void dr_ste_build_eth_l2_src_bit_mask(struct mlx5dr_match_param *value, static int dr_ste_build_eth_l2_src_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_spec *spec = sb->inner ? &value->inner : &value->outer; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(eth_l2_src, tag, smac_47_16, spec, smac_47_16); DR_STE_SET_TAG(eth_l2_src, tag, smac_15_0, spec, smac_15_0); - return dr_ste_build_eth_l2_src_or_dst_tag(value, sb->inner, hw_ste_p); + return dr_ste_build_eth_l2_src_or_dst_tag(value, sb->inner, tag); } void mlx5dr_ste_build_eth_l2_src(struct mlx5dr_ste_build *sb, @@ -1440,16 +1410,14 @@ static void dr_ste_build_eth_l2_dst_bit_mask(struct mlx5dr_match_param *value, static int dr_ste_build_eth_l2_dst_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_spec *spec = sb->inner ? &value->inner : &value->outer; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(eth_l2_dst, tag, dmac_47_16, spec, dmac_47_16); DR_STE_SET_TAG(eth_l2_dst, tag, dmac_15_0, spec, dmac_15_0); - return dr_ste_build_eth_l2_src_or_dst_tag(value, sb->inner, hw_ste_p); + return dr_ste_build_eth_l2_src_or_dst_tag(value, sb->inner, tag); } void mlx5dr_ste_build_eth_l2_dst(struct mlx5dr_ste_build *sb, @@ -1495,12 +1463,10 @@ static void dr_ste_build_eth_l2_tnl_bit_mask(struct mlx5dr_match_param *value, static int dr_ste_build_eth_l2_tnl_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { struct mlx5dr_match_spec *spec = sb->inner ? &value->inner : &value->outer; - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_misc *misc = &value->misc; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(eth_l2_tnl, tag, dmac_47_16, spec, dmac_47_16); DR_STE_SET_TAG(eth_l2_tnl, tag, dmac_15_0, spec, dmac_15_0); @@ -1561,11 +1527,9 @@ static void dr_ste_build_eth_l3_ipv4_misc_bit_mask(struct mlx5dr_match_param *va static int dr_ste_build_eth_l3_ipv4_misc_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_spec *spec = sb->inner ? &value->inner : &value->outer; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(eth_l3_ipv4_misc, tag, time_to_live, spec, ttl_hoplimit); @@ -1608,11 +1572,9 @@ static void dr_ste_build_ipv6_l3_l4_bit_mask(struct mlx5dr_match_param *value, static int dr_ste_build_ipv6_l3_l4_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { struct mlx5dr_match_spec *spec = sb->inner ? &value->inner : &value->outer; - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(eth_l4, tag, dst_port, spec, tcp_dport); DR_STE_SET_TAG(eth_l4, tag, src_port, spec, tcp_sport); @@ -1647,7 +1609,7 @@ void mlx5dr_ste_build_ipv6_l3_l4(struct mlx5dr_ste_build *sb, static int dr_ste_build_empty_always_hit_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { return 0; } @@ -1673,11 +1635,9 @@ static void dr_ste_build_mpls_bit_mask(struct mlx5dr_match_param *value, static int dr_ste_build_mpls_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_misc2 *misc2_mask = &value->misc2; - u8 *tag = hw_ste->tag; if (sb->inner) DR_STE_SET_MPLS_TAG(mpls, misc2_mask, inner, tag); @@ -1716,11 +1676,9 @@ static void dr_ste_build_gre_bit_mask(struct mlx5dr_match_param *value, static int dr_ste_build_gre_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_misc *misc = &value->misc; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(gre, tag, gre_protocol, misc, gre_protocol); @@ -1781,11 +1739,9 @@ static void dr_ste_build_flex_parser_0_bit_mask(struct mlx5dr_match_param *value static int dr_ste_build_flex_parser_0_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_misc2 *misc_2_mask = &value->misc2; - u8 *tag = hw_ste->tag; if (DR_STE_IS_OUTER_MPLS_OVER_GRE_SET(misc_2_mask)) { DR_STE_SET_TAG(flex_parser_0, tag, parser_3_label, @@ -1903,11 +1859,9 @@ static int dr_ste_build_flex_parser_1_bit_mask(struct mlx5dr_match_param *mask, static int dr_ste_build_flex_parser_1_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_misc3 *misc_3 = &value->misc3; - u8 *tag = hw_ste->tag; u32 icmp_header_data; int dw0_location; int dw1_location; @@ -2007,11 +1961,9 @@ static void dr_ste_build_general_purpose_bit_mask(struct mlx5dr_match_param *val static int dr_ste_build_general_purpose_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_misc2 *misc_2_mask = &value->misc2; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(general_purpose, tag, general_purpose_lookup_field, misc_2_mask, metadata_reg_a); @@ -2052,11 +2004,9 @@ static void dr_ste_build_eth_l4_misc_bit_mask(struct mlx5dr_match_param *value, static int dr_ste_build_eth_l4_misc_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_misc3 *misc3 = &value->misc3; - u8 *tag = hw_ste->tag; if (sb->inner) { DR_STE_SET_TAG(eth_l4_misc, tag, seq_num, misc3, inner_tcp_seq_num); @@ -2102,11 +2052,9 @@ dr_ste_build_flex_parser_tnl_vxlan_gpe_bit_mask(struct mlx5dr_match_param *value static int dr_ste_build_flex_parser_tnl_vxlan_gpe_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_misc3 *misc3 = &value->misc3; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(flex_parser_tnl_vxlan_gpe, tag, outer_vxlan_gpe_flags, misc3, @@ -2158,11 +2106,9 @@ dr_ste_build_flex_parser_tnl_geneve_bit_mask(struct mlx5dr_match_param *value, static int dr_ste_build_flex_parser_tnl_geneve_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_misc *misc = &value->misc; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(flex_parser_tnl_geneve, tag, geneve_protocol_type, misc, geneve_protocol_type); @@ -2205,11 +2151,9 @@ static void dr_ste_build_register_0_bit_mask(struct mlx5dr_match_param *value, static int dr_ste_build_register_0_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_misc2 *misc2 = &value->misc2; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(register_0, tag, register_0_h, misc2, metadata_reg_c_0); DR_STE_SET_TAG(register_0, tag, register_0_l, misc2, metadata_reg_c_1); @@ -2249,11 +2193,9 @@ static void dr_ste_build_register_1_bit_mask(struct mlx5dr_match_param *value, static int dr_ste_build_register_1_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_misc2 *misc2 = &value->misc2; - u8 *tag = hw_ste->tag; DR_STE_SET_TAG(register_1, tag, register_2_h, misc2, metadata_reg_c_4); DR_STE_SET_TAG(register_1, tag, register_2_l, misc2, metadata_reg_c_5); @@ -2276,38 +2218,25 @@ void mlx5dr_ste_build_register_1(struct mlx5dr_ste_build *sb, sb->ste_build_tag_func = &dr_ste_build_register_1_tag; } -static int dr_ste_build_src_gvmi_qpn_bit_mask(struct mlx5dr_match_param *value, - u8 *bit_mask) +static void dr_ste_build_src_gvmi_qpn_bit_mask(struct mlx5dr_match_param *value, + u8 *bit_mask) { struct mlx5dr_match_misc *misc_mask = &value->misc; - /* Partial misc source_port is not supported */ - if (misc_mask->source_port && misc_mask->source_port != 0xffff) - return -EINVAL; - - /* Partial misc source_eswitch_owner_vhca_id is not supported */ - if (misc_mask->source_eswitch_owner_vhca_id && - misc_mask->source_eswitch_owner_vhca_id != 0xffff) - return -EINVAL; - DR_STE_SET_MASK(src_gvmi_qp, bit_mask, source_gvmi, misc_mask, source_port); DR_STE_SET_MASK(src_gvmi_qp, bit_mask, source_qp, misc_mask, source_sqn); misc_mask->source_eswitch_owner_vhca_id = 0; - - return 0; } static int dr_ste_build_src_gvmi_qpn_tag(struct mlx5dr_match_param *value, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p) + u8 *tag) { - struct dr_hw_ste_format *hw_ste = (struct dr_hw_ste_format *)hw_ste_p; struct mlx5dr_match_misc *misc = &value->misc; struct mlx5dr_cmd_vport_cap *vport_cap; struct mlx5dr_domain *dmn = sb->dmn; struct mlx5dr_cmd_caps *caps; u8 *bit_mask = sb->bit_mask; - u8 *tag = hw_ste->tag; bool source_gvmi_set; DR_STE_SET_TAG(src_gvmi_qp, tag, source_qp, misc, source_sqn); @@ -2339,19 +2268,15 @@ static int dr_ste_build_src_gvmi_qpn_tag(struct mlx5dr_match_param *value, return 0; } -int mlx5dr_ste_build_src_gvmi_qpn(struct mlx5dr_ste_build *sb, - struct mlx5dr_match_param *mask, - struct mlx5dr_domain *dmn, - bool inner, bool rx) +void mlx5dr_ste_build_src_gvmi_qpn(struct mlx5dr_ste_build *sb, + struct mlx5dr_match_param *mask, + struct mlx5dr_domain *dmn, + bool inner, bool rx) { - int ret; - /* Set vhca_id_valid before we reset source_eswitch_owner_vhca_id */ sb->vhca_id_valid = mask->misc.source_eswitch_owner_vhca_id; - ret = dr_ste_build_src_gvmi_qpn_bit_mask(mask, sb->bit_mask); - if (ret) - return ret; + dr_ste_build_src_gvmi_qpn_bit_mask(mask, sb->bit_mask); sb->rx = rx; sb->dmn = dmn; @@ -2359,6 +2284,4 @@ int mlx5dr_ste_build_src_gvmi_qpn(struct mlx5dr_ste_build *sb, sb->lu_type = MLX5DR_STE_LU_TYPE_SRC_GVMI_AND_QP; sb->byte_mask = dr_ste_conv_bit_to_byte_mask(sb->bit_mask); sb->ste_build_tag_func = &dr_ste_build_src_gvmi_qpn_tag; - - return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h index 0883956c58c0..f50f3b107aa3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h @@ -194,7 +194,7 @@ struct mlx5dr_ste_build { u8 bit_mask[DR_STE_SIZE_MASK]; int (*ste_build_tag_func)(struct mlx5dr_match_param *spec, struct mlx5dr_ste_build *sb, - u8 *hw_ste_p); + u8 *tag); }; struct mlx5dr_ste_htbl * @@ -227,7 +227,6 @@ void mlx5dr_ste_set_hit_gvmi(u8 *hw_ste_p, u16 gvmi); void mlx5dr_ste_set_hit_addr(u8 *hw_ste, u64 icm_addr, u32 ht_size); void mlx5dr_ste_always_miss_addr(struct mlx5dr_ste *ste, u64 miss_addr); void mlx5dr_ste_set_bit_mask(u8 *hw_ste_p, u8 *bit_mask); -bool mlx5dr_ste_not_used_ste(struct mlx5dr_ste *ste); bool mlx5dr_ste_is_last_in_rule(struct mlx5dr_matcher_rx_tx *nic_matcher, u8 ste_location); void mlx5dr_ste_rx_set_flow_tag(u8 *hw_ste_p, u32 flow_tag); @@ -266,6 +265,11 @@ static inline void mlx5dr_ste_get(struct mlx5dr_ste *ste) ste->refcount++; } +static inline bool mlx5dr_ste_is_not_used(struct mlx5dr_ste *ste) +{ + return !ste->refcount; +} + void mlx5dr_ste_set_hit_addr_by_next_htbl(u8 *hw_ste, struct mlx5dr_ste_htbl *next_htbl); bool mlx5dr_ste_equal_tag(void *src, void *dst); @@ -284,9 +288,9 @@ int mlx5dr_ste_build_ste_arr(struct mlx5dr_matcher *matcher, struct mlx5dr_matcher_rx_tx *nic_matcher, struct mlx5dr_match_param *value, u8 *ste_arr); -int mlx5dr_ste_build_eth_l2_src_des(struct mlx5dr_ste_build *builder, - struct mlx5dr_match_param *mask, - bool inner, bool rx); +void mlx5dr_ste_build_eth_l2_src_des(struct mlx5dr_ste_build *builder, + struct mlx5dr_match_param *mask, + bool inner, bool rx); void mlx5dr_ste_build_eth_l3_ipv4_5_tuple(struct mlx5dr_ste_build *sb, struct mlx5dr_match_param *mask, bool inner, bool rx); @@ -342,10 +346,10 @@ void mlx5dr_ste_build_register_0(struct mlx5dr_ste_build *sb, void mlx5dr_ste_build_register_1(struct mlx5dr_ste_build *sb, struct mlx5dr_match_param *mask, bool inner, bool rx); -int mlx5dr_ste_build_src_gvmi_qpn(struct mlx5dr_ste_build *sb, - struct mlx5dr_match_param *mask, - struct mlx5dr_domain *dmn, - bool inner, bool rx); +void mlx5dr_ste_build_src_gvmi_qpn(struct mlx5dr_ste_build *sb, + struct mlx5dr_match_param *mask, + struct mlx5dr_domain *dmn, + bool inner, bool rx); void mlx5dr_ste_build_empty_always_hit(struct mlx5dr_ste_build *sb, bool rx); /* Actions utils */ @@ -793,6 +797,7 @@ struct mlx5dr_rule { struct mlx5dr_rule_rx_tx rx; struct mlx5dr_rule_rx_tx tx; struct list_head rule_actions_list; + u32 flow_source; }; void mlx5dr_rule_update_rule_member(struct mlx5dr_ste *new_ste, @@ -991,7 +996,6 @@ struct mlx5dr_icm_chunk * mlx5dr_icm_alloc_chunk(struct mlx5dr_icm_pool *pool, enum mlx5dr_icm_chunk_size chunk_size); void mlx5dr_icm_free_chunk(struct mlx5dr_icm_chunk *chunk); -bool mlx5dr_ste_is_not_valid_entry(u8 *p_hw_ste); int mlx5dr_ste_htbl_init_and_postsend(struct mlx5dr_domain *dmn, struct mlx5dr_domain_rx_tx *nic_dmn, struct mlx5dr_ste_htbl *htbl, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c index 9b08eb557a31..96c39a17d026 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c @@ -487,7 +487,8 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns, rule = mlx5dr_rule_create(group->fs_dr_matcher.dr_matcher, ¶ms, num_actions, - actions); + actions, + fte->flow_context.flow_source); if (!rule) { err = -EINVAL; goto free_actions; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h index 7deaca9ade3b..7914fe3fc68d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h @@ -67,7 +67,8 @@ struct mlx5dr_rule * mlx5dr_rule_create(struct mlx5dr_matcher *matcher, struct mlx5dr_match_parameters *value, size_t num_actions, - struct mlx5dr_action *actions[]); + struct mlx5dr_action *actions[], + u32 flow_source); int mlx5dr_rule_destroy(struct mlx5dr_rule *rule); diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c index ec45a03140d7..7f77c2a71d1c 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core.c @@ -20,11 +20,13 @@ #include <linux/rcupdate.h> #include <linux/slab.h> #include <linux/workqueue.h> +#include <linux/firmware.h> #include <asm/byteorder.h> #include <net/devlink.h> #include <trace/events/devlink.h> #include "core.h" +#include "core_env.h" #include "item.h" #include "cmd.h" #include "port.h" @@ -32,6 +34,7 @@ #include "emad.h" #include "reg.h" #include "resources.h" +#include "../mlxfw/mlxfw.h" static LIST_HEAD(mlxsw_core_driver_list); static DEFINE_SPINLOCK(mlxsw_core_driver_list_lock); @@ -82,6 +85,11 @@ struct mlxsw_core { struct mlxsw_core_port *ports; unsigned int max_ports; bool fw_flash_in_progress; + struct { + struct devlink_health_reporter *fw_fatal; + } health; + struct mlxsw_env *env; + bool is_initialized; /* Denotes if core was already initialized. */ unsigned long driver_priv[]; /* driver_priv has to be always the last item */ }; @@ -128,6 +136,11 @@ bool mlxsw_core_res_query_enabled(const struct mlxsw_core *mlxsw_core) } EXPORT_SYMBOL(mlxsw_core_res_query_enabled); +bool mlxsw_core_temp_warn_enabled(const struct mlxsw_core *mlxsw_core) +{ + return mlxsw_core->driver->temp_warn_enabled; +} + bool mlxsw_core_fw_rev_minor_subminor_validate(const struct mlxsw_fw_rev *rev, const struct mlxsw_fw_rev *req_rev) @@ -864,6 +877,294 @@ static struct mlxsw_driver *mlxsw_core_driver_get(const char *kind) return mlxsw_driver; } +struct mlxsw_core_fw_info { + struct mlxfw_dev mlxfw_dev; + struct mlxsw_core *mlxsw_core; +}; + +static int mlxsw_core_fw_component_query(struct mlxfw_dev *mlxfw_dev, + u16 component_index, u32 *p_max_size, + u8 *p_align_bits, u16 *p_max_write_size) +{ + struct mlxsw_core_fw_info *mlxsw_core_fw_info = + container_of(mlxfw_dev, struct mlxsw_core_fw_info, mlxfw_dev); + struct mlxsw_core *mlxsw_core = mlxsw_core_fw_info->mlxsw_core; + char mcqi_pl[MLXSW_REG_MCQI_LEN]; + int err; + + mlxsw_reg_mcqi_pack(mcqi_pl, component_index); + err = mlxsw_reg_query(mlxsw_core, MLXSW_REG(mcqi), mcqi_pl); + if (err) + return err; + mlxsw_reg_mcqi_unpack(mcqi_pl, p_max_size, p_align_bits, p_max_write_size); + + *p_align_bits = max_t(u8, *p_align_bits, 2); + *p_max_write_size = min_t(u16, *p_max_write_size, MLXSW_REG_MCDA_MAX_DATA_LEN); + return 0; +} + +static int mlxsw_core_fw_fsm_lock(struct mlxfw_dev *mlxfw_dev, u32 *fwhandle) +{ + struct mlxsw_core_fw_info *mlxsw_core_fw_info = + container_of(mlxfw_dev, struct mlxsw_core_fw_info, mlxfw_dev); + struct mlxsw_core *mlxsw_core = mlxsw_core_fw_info->mlxsw_core; + char mcc_pl[MLXSW_REG_MCC_LEN]; + u8 control_state; + int err; + + mlxsw_reg_mcc_pack(mcc_pl, 0, 0, 0, 0); + err = mlxsw_reg_query(mlxsw_core, MLXSW_REG(mcc), mcc_pl); + if (err) + return err; + + mlxsw_reg_mcc_unpack(mcc_pl, fwhandle, NULL, &control_state); + if (control_state != MLXFW_FSM_STATE_IDLE) + return -EBUSY; + + mlxsw_reg_mcc_pack(mcc_pl, MLXSW_REG_MCC_INSTRUCTION_LOCK_UPDATE_HANDLE, 0, *fwhandle, 0); + return mlxsw_reg_write(mlxsw_core, MLXSW_REG(mcc), mcc_pl); +} + +static int mlxsw_core_fw_fsm_component_update(struct mlxfw_dev *mlxfw_dev, u32 fwhandle, + u16 component_index, u32 component_size) +{ + struct mlxsw_core_fw_info *mlxsw_core_fw_info = + container_of(mlxfw_dev, struct mlxsw_core_fw_info, mlxfw_dev); + struct mlxsw_core *mlxsw_core = mlxsw_core_fw_info->mlxsw_core; + char mcc_pl[MLXSW_REG_MCC_LEN]; + + mlxsw_reg_mcc_pack(mcc_pl, MLXSW_REG_MCC_INSTRUCTION_UPDATE_COMPONENT, + component_index, fwhandle, component_size); + return mlxsw_reg_write(mlxsw_core, MLXSW_REG(mcc), mcc_pl); +} + +static int mlxsw_core_fw_fsm_block_download(struct mlxfw_dev *mlxfw_dev, u32 fwhandle, + u8 *data, u16 size, u32 offset) +{ + struct mlxsw_core_fw_info *mlxsw_core_fw_info = + container_of(mlxfw_dev, struct mlxsw_core_fw_info, mlxfw_dev); + struct mlxsw_core *mlxsw_core = mlxsw_core_fw_info->mlxsw_core; + char mcda_pl[MLXSW_REG_MCDA_LEN]; + + mlxsw_reg_mcda_pack(mcda_pl, fwhandle, offset, size, data); + return mlxsw_reg_write(mlxsw_core, MLXSW_REG(mcda), mcda_pl); +} + +static int mlxsw_core_fw_fsm_component_verify(struct mlxfw_dev *mlxfw_dev, u32 fwhandle, + u16 component_index) +{ + struct mlxsw_core_fw_info *mlxsw_core_fw_info = + container_of(mlxfw_dev, struct mlxsw_core_fw_info, mlxfw_dev); + struct mlxsw_core *mlxsw_core = mlxsw_core_fw_info->mlxsw_core; + char mcc_pl[MLXSW_REG_MCC_LEN]; + + mlxsw_reg_mcc_pack(mcc_pl, MLXSW_REG_MCC_INSTRUCTION_VERIFY_COMPONENT, + component_index, fwhandle, 0); + return mlxsw_reg_write(mlxsw_core, MLXSW_REG(mcc), mcc_pl); +} + +static int mlxsw_core_fw_fsm_activate(struct mlxfw_dev *mlxfw_dev, u32 fwhandle) +{ + struct mlxsw_core_fw_info *mlxsw_core_fw_info = + container_of(mlxfw_dev, struct mlxsw_core_fw_info, mlxfw_dev); + struct mlxsw_core *mlxsw_core = mlxsw_core_fw_info->mlxsw_core; + char mcc_pl[MLXSW_REG_MCC_LEN]; + + mlxsw_reg_mcc_pack(mcc_pl, MLXSW_REG_MCC_INSTRUCTION_ACTIVATE, 0, fwhandle, 0); + return mlxsw_reg_write(mlxsw_core, MLXSW_REG(mcc), mcc_pl); +} + +static int mlxsw_core_fw_fsm_query_state(struct mlxfw_dev *mlxfw_dev, u32 fwhandle, + enum mlxfw_fsm_state *fsm_state, + enum mlxfw_fsm_state_err *fsm_state_err) +{ + struct mlxsw_core_fw_info *mlxsw_core_fw_info = + container_of(mlxfw_dev, struct mlxsw_core_fw_info, mlxfw_dev); + struct mlxsw_core *mlxsw_core = mlxsw_core_fw_info->mlxsw_core; + char mcc_pl[MLXSW_REG_MCC_LEN]; + u8 control_state; + u8 error_code; + int err; + + mlxsw_reg_mcc_pack(mcc_pl, 0, 0, fwhandle, 0); + err = mlxsw_reg_query(mlxsw_core, MLXSW_REG(mcc), mcc_pl); + if (err) + return err; + + mlxsw_reg_mcc_unpack(mcc_pl, NULL, &error_code, &control_state); + *fsm_state = control_state; + *fsm_state_err = min_t(enum mlxfw_fsm_state_err, error_code, MLXFW_FSM_STATE_ERR_MAX); + return 0; +} + +static void mlxsw_core_fw_fsm_cancel(struct mlxfw_dev *mlxfw_dev, u32 fwhandle) +{ + struct mlxsw_core_fw_info *mlxsw_core_fw_info = + container_of(mlxfw_dev, struct mlxsw_core_fw_info, mlxfw_dev); + struct mlxsw_core *mlxsw_core = mlxsw_core_fw_info->mlxsw_core; + char mcc_pl[MLXSW_REG_MCC_LEN]; + + mlxsw_reg_mcc_pack(mcc_pl, MLXSW_REG_MCC_INSTRUCTION_CANCEL, 0, fwhandle, 0); + mlxsw_reg_write(mlxsw_core, MLXSW_REG(mcc), mcc_pl); +} + +static void mlxsw_core_fw_fsm_release(struct mlxfw_dev *mlxfw_dev, u32 fwhandle) +{ + struct mlxsw_core_fw_info *mlxsw_core_fw_info = + container_of(mlxfw_dev, struct mlxsw_core_fw_info, mlxfw_dev); + struct mlxsw_core *mlxsw_core = mlxsw_core_fw_info->mlxsw_core; + char mcc_pl[MLXSW_REG_MCC_LEN]; + + mlxsw_reg_mcc_pack(mcc_pl, MLXSW_REG_MCC_INSTRUCTION_RELEASE_UPDATE_HANDLE, 0, fwhandle, 0); + mlxsw_reg_write(mlxsw_core, MLXSW_REG(mcc), mcc_pl); +} + +static const struct mlxfw_dev_ops mlxsw_core_fw_mlxsw_dev_ops = { + .component_query = mlxsw_core_fw_component_query, + .fsm_lock = mlxsw_core_fw_fsm_lock, + .fsm_component_update = mlxsw_core_fw_fsm_component_update, + .fsm_block_download = mlxsw_core_fw_fsm_block_download, + .fsm_component_verify = mlxsw_core_fw_fsm_component_verify, + .fsm_activate = mlxsw_core_fw_fsm_activate, + .fsm_query_state = mlxsw_core_fw_fsm_query_state, + .fsm_cancel = mlxsw_core_fw_fsm_cancel, + .fsm_release = mlxsw_core_fw_fsm_release, +}; + +static int mlxsw_core_fw_flash(struct mlxsw_core *mlxsw_core, const struct firmware *firmware, + struct netlink_ext_ack *extack) +{ + struct mlxsw_core_fw_info mlxsw_core_fw_info = { + .mlxfw_dev = { + .ops = &mlxsw_core_fw_mlxsw_dev_ops, + .psid = mlxsw_core->bus_info->psid, + .psid_size = strlen(mlxsw_core->bus_info->psid), + .devlink = priv_to_devlink(mlxsw_core), + }, + .mlxsw_core = mlxsw_core + }; + int err; + + mlxsw_core->fw_flash_in_progress = true; + err = mlxfw_firmware_flash(&mlxsw_core_fw_info.mlxfw_dev, firmware, extack); + mlxsw_core->fw_flash_in_progress = false; + + return err; +} + +static int mlxsw_core_fw_rev_validate(struct mlxsw_core *mlxsw_core, + const struct mlxsw_bus_info *mlxsw_bus_info, + const struct mlxsw_fw_rev *req_rev, + const char *filename) +{ + const struct mlxsw_fw_rev *rev = &mlxsw_bus_info->fw_rev; + union devlink_param_value value; + const struct firmware *firmware; + int err; + + /* Don't check if driver does not require it */ + if (!req_rev || !filename) + return 0; + + /* Don't check if devlink 'fw_load_policy' param is 'flash' */ + err = devlink_param_driverinit_value_get(priv_to_devlink(mlxsw_core), + DEVLINK_PARAM_GENERIC_ID_FW_LOAD_POLICY, + &value); + if (err) + return err; + if (value.vu8 == DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_FLASH) + return 0; + + /* Validate driver & FW are compatible */ + if (rev->major != req_rev->major) { + WARN(1, "Mismatch in major FW version [%d:%d] is never expected; Please contact support\n", + rev->major, req_rev->major); + return -EINVAL; + } + if (mlxsw_core_fw_rev_minor_subminor_validate(rev, req_rev)) + return 0; + + dev_err(mlxsw_bus_info->dev, "The firmware version %d.%d.%d is incompatible with the driver (required >= %d.%d.%d)\n", + rev->major, rev->minor, rev->subminor, req_rev->major, + req_rev->minor, req_rev->subminor); + dev_info(mlxsw_bus_info->dev, "Flashing firmware using file %s\n", filename); + + err = request_firmware_direct(&firmware, filename, mlxsw_bus_info->dev); + if (err) { + dev_err(mlxsw_bus_info->dev, "Could not request firmware file %s\n", filename); + return err; + } + + err = mlxsw_core_fw_flash(mlxsw_core, firmware, NULL); + release_firmware(firmware); + if (err) + dev_err(mlxsw_bus_info->dev, "Could not upgrade firmware\n"); + + /* On FW flash success, tell the caller FW reset is needed + * if current FW supports it. + */ + if (rev->minor >= req_rev->can_reset_minor) + return err ? err : -EAGAIN; + else + return 0; +} + +static int mlxsw_core_fw_flash_update(struct mlxsw_core *mlxsw_core, + struct devlink_flash_update_params *params, + struct netlink_ext_ack *extack) +{ + const struct firmware *firmware; + int err; + + err = request_firmware_direct(&firmware, params->file_name, mlxsw_core->bus_info->dev); + if (err) + return err; + err = mlxsw_core_fw_flash(mlxsw_core, firmware, extack); + release_firmware(firmware); + + return err; +} + +static int mlxsw_core_devlink_param_fw_load_policy_validate(struct devlink *devlink, u32 id, + union devlink_param_value val, + struct netlink_ext_ack *extack) +{ + if (val.vu8 != DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_DRIVER && + val.vu8 != DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_FLASH) { + NL_SET_ERR_MSG_MOD(extack, "'fw_load_policy' must be 'driver' or 'flash'"); + return -EINVAL; + } + + return 0; +} + +static const struct devlink_param mlxsw_core_fw_devlink_params[] = { + DEVLINK_PARAM_GENERIC(FW_LOAD_POLICY, BIT(DEVLINK_PARAM_CMODE_DRIVERINIT), NULL, NULL, + mlxsw_core_devlink_param_fw_load_policy_validate), +}; + +static int mlxsw_core_fw_params_register(struct mlxsw_core *mlxsw_core) +{ + struct devlink *devlink = priv_to_devlink(mlxsw_core); + union devlink_param_value value; + int err; + + err = devlink_params_register(devlink, mlxsw_core_fw_devlink_params, + ARRAY_SIZE(mlxsw_core_fw_devlink_params)); + if (err) + return err; + + value.vu8 = DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_DRIVER; + devlink_param_driverinit_value_set(devlink, DEVLINK_PARAM_GENERIC_ID_FW_LOAD_POLICY, value); + return 0; +} + +static void mlxsw_core_fw_params_unregister(struct mlxsw_core *mlxsw_core) +{ + devlink_params_unregister(priv_to_devlink(mlxsw_core), mlxsw_core_fw_devlink_params, + ARRAY_SIZE(mlxsw_core_fw_devlink_params)); +} + static int mlxsw_devlink_port_split(struct devlink *devlink, unsigned int port_index, unsigned int count, @@ -1113,7 +1414,8 @@ mlxsw_devlink_info_get(struct devlink *devlink, struct devlink_info_req *req, static int mlxsw_devlink_core_bus_device_reload_down(struct devlink *devlink, - bool netns_change, + bool netns_change, enum devlink_reload_action action, + enum devlink_reload_limit limit, struct netlink_ext_ack *extack) { struct mlxsw_core *mlxsw_core = devlink_priv(devlink); @@ -1126,11 +1428,14 @@ mlxsw_devlink_core_bus_device_reload_down(struct devlink *devlink, } static int -mlxsw_devlink_core_bus_device_reload_up(struct devlink *devlink, +mlxsw_devlink_core_bus_device_reload_up(struct devlink *devlink, enum devlink_reload_action action, + enum devlink_reload_limit limit, u32 *actions_performed, struct netlink_ext_ack *extack) { struct mlxsw_core *mlxsw_core = devlink_priv(devlink); + *actions_performed = BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT) | + BIT(DEVLINK_RELOAD_ACTION_FW_ACTIVATE); return mlxsw_core_bus_device_register(mlxsw_core->bus_info, mlxsw_core->bus, mlxsw_core->bus_priv, true, @@ -1138,17 +1443,12 @@ mlxsw_devlink_core_bus_device_reload_up(struct devlink *devlink, } static int mlxsw_devlink_flash_update(struct devlink *devlink, - const char *file_name, - const char *component, + struct devlink_flash_update_params *params, struct netlink_ext_ack *extack) { struct mlxsw_core *mlxsw_core = devlink_priv(devlink); - struct mlxsw_driver *mlxsw_driver = mlxsw_core->driver; - if (!mlxsw_driver->flash_update) - return -EOPNOTSUPP; - return mlxsw_driver->flash_update(mlxsw_core, file_name, - component, extack); + return mlxsw_core_fw_flash_update(mlxsw_core, params, extack); } static int mlxsw_devlink_trap_init(struct devlink *devlink, @@ -1268,6 +1568,8 @@ mlxsw_devlink_trap_policer_counter_get(struct devlink *devlink, } static const struct devlink_ops mlxsw_devlink_ops = { + .reload_actions = BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT) | + BIT(DEVLINK_RELOAD_ACTION_FW_ACTIVATE), .reload_down = mlxsw_devlink_core_bus_device_reload_down, .reload_up = mlxsw_devlink_core_bus_device_reload_up, .port_type_set = mlxsw_devlink_port_type_set, @@ -1296,6 +1598,263 @@ static const struct devlink_ops mlxsw_devlink_ops = { .trap_policer_counter_get = mlxsw_devlink_trap_policer_counter_get, }; +static int mlxsw_core_params_register(struct mlxsw_core *mlxsw_core) +{ + int err; + + err = mlxsw_core_fw_params_register(mlxsw_core); + if (err) + return err; + + if (mlxsw_core->driver->params_register) { + err = mlxsw_core->driver->params_register(mlxsw_core); + if (err) + goto err_params_register; + } + return 0; + +err_params_register: + mlxsw_core_fw_params_unregister(mlxsw_core); + return err; +} + +static void mlxsw_core_params_unregister(struct mlxsw_core *mlxsw_core) +{ + mlxsw_core_fw_params_unregister(mlxsw_core); + if (mlxsw_core->driver->params_register) + mlxsw_core->driver->params_unregister(mlxsw_core); +} + +struct mlxsw_core_health_event { + struct mlxsw_core *mlxsw_core; + char mfde_pl[MLXSW_REG_MFDE_LEN]; + struct work_struct work; +}; + +static void mlxsw_core_health_event_work(struct work_struct *work) +{ + struct mlxsw_core_health_event *event; + struct mlxsw_core *mlxsw_core; + + event = container_of(work, struct mlxsw_core_health_event, work); + mlxsw_core = event->mlxsw_core; + devlink_health_report(mlxsw_core->health.fw_fatal, "FW fatal event occurred", + event->mfde_pl); + kfree(event); +} + +static void mlxsw_core_health_listener_func(const struct mlxsw_reg_info *reg, + char *mfde_pl, void *priv) +{ + struct mlxsw_core_health_event *event; + struct mlxsw_core *mlxsw_core = priv; + + event = kmalloc(sizeof(*event), GFP_ATOMIC); + if (!event) + return; + event->mlxsw_core = mlxsw_core; + memcpy(event->mfde_pl, mfde_pl, sizeof(event->mfde_pl)); + INIT_WORK(&event->work, mlxsw_core_health_event_work); + mlxsw_core_schedule_work(&event->work); +} + +static const struct mlxsw_listener mlxsw_core_health_listener = + MLXSW_EVENTL(mlxsw_core_health_listener_func, MFDE, MFDE); + +static int mlxsw_core_health_fw_fatal_dump(struct devlink_health_reporter *reporter, + struct devlink_fmsg *fmsg, void *priv_ctx, + struct netlink_ext_ack *extack) +{ + char *mfde_pl = priv_ctx; + char *val_str; + u8 event_id; + u32 val; + int err; + + if (!priv_ctx) + /* User-triggered dumps are not possible */ + return -EOPNOTSUPP; + + val = mlxsw_reg_mfde_irisc_id_get(mfde_pl); + err = devlink_fmsg_u8_pair_put(fmsg, "irisc_id", val); + if (err) + return err; + err = devlink_fmsg_arr_pair_nest_start(fmsg, "event"); + if (err) + return err; + + event_id = mlxsw_reg_mfde_event_id_get(mfde_pl); + err = devlink_fmsg_u8_pair_put(fmsg, "id", event_id); + if (err) + return err; + switch (event_id) { + case MLXSW_REG_MFDE_EVENT_ID_CRSPACE_TO: + val_str = "CR space timeout"; + break; + case MLXSW_REG_MFDE_EVENT_ID_KVD_IM_STOP: + val_str = "KVD insertion machine stopped"; + break; + default: + val_str = NULL; + } + if (val_str) { + err = devlink_fmsg_string_pair_put(fmsg, "desc", val_str); + if (err) + return err; + } + err = devlink_fmsg_arr_pair_nest_end(fmsg); + if (err) + return err; + + val = mlxsw_reg_mfde_method_get(mfde_pl); + switch (val) { + case MLXSW_REG_MFDE_METHOD_QUERY: + val_str = "query"; + break; + case MLXSW_REG_MFDE_METHOD_WRITE: + val_str = "write"; + break; + default: + val_str = NULL; + } + if (val_str) { + err = devlink_fmsg_string_pair_put(fmsg, "method", val_str); + if (err) + return err; + } + + val = mlxsw_reg_mfde_long_process_get(mfde_pl); + err = devlink_fmsg_bool_pair_put(fmsg, "long_process", val); + if (err) + return err; + + val = mlxsw_reg_mfde_command_type_get(mfde_pl); + switch (val) { + case MLXSW_REG_MFDE_COMMAND_TYPE_MAD: + val_str = "mad"; + break; + case MLXSW_REG_MFDE_COMMAND_TYPE_EMAD: + val_str = "emad"; + break; + case MLXSW_REG_MFDE_COMMAND_TYPE_CMDIF: + val_str = "cmdif"; + break; + default: + val_str = NULL; + } + if (val_str) { + err = devlink_fmsg_string_pair_put(fmsg, "command_type", val_str); + if (err) + return err; + } + + val = mlxsw_reg_mfde_reg_attr_id_get(mfde_pl); + err = devlink_fmsg_u32_pair_put(fmsg, "reg_attr_id", val); + if (err) + return err; + + if (event_id == MLXSW_REG_MFDE_EVENT_ID_CRSPACE_TO) { + val = mlxsw_reg_mfde_log_address_get(mfde_pl); + err = devlink_fmsg_u32_pair_put(fmsg, "log_address", val); + if (err) + return err; + val = mlxsw_reg_mfde_log_id_get(mfde_pl); + err = devlink_fmsg_u8_pair_put(fmsg, "log_irisc_id", val); + if (err) + return err; + } else if (event_id == MLXSW_REG_MFDE_EVENT_ID_KVD_IM_STOP) { + val = mlxsw_reg_mfde_pipes_mask_get(mfde_pl); + err = devlink_fmsg_u32_pair_put(fmsg, "pipes_mask", val); + if (err) + return err; + } + + return 0; +} + +static int +mlxsw_core_health_fw_fatal_test(struct devlink_health_reporter *reporter, + struct netlink_ext_ack *extack) +{ + struct mlxsw_core *mlxsw_core = devlink_health_reporter_priv(reporter); + char mfgd_pl[MLXSW_REG_MFGD_LEN]; + int err; + + /* Read the register first to make sure no other bits are changed. */ + err = mlxsw_reg_query(mlxsw_core, MLXSW_REG(mfgd), mfgd_pl); + if (err) + return err; + mlxsw_reg_mfgd_trigger_test_set(mfgd_pl, true); + return mlxsw_reg_write(mlxsw_core, MLXSW_REG(mfgd), mfgd_pl); +} + +static const struct devlink_health_reporter_ops +mlxsw_core_health_fw_fatal_ops = { + .name = "fw_fatal", + .dump = mlxsw_core_health_fw_fatal_dump, + .test = mlxsw_core_health_fw_fatal_test, +}; + +static int mlxsw_core_health_fw_fatal_config(struct mlxsw_core *mlxsw_core, + bool enable) +{ + char mfgd_pl[MLXSW_REG_MFGD_LEN]; + int err; + + /* Read the register first to make sure no other bits are changed. */ + err = mlxsw_reg_query(mlxsw_core, MLXSW_REG(mfgd), mfgd_pl); + if (err) + return err; + mlxsw_reg_mfgd_fatal_event_mode_set(mfgd_pl, enable); + return mlxsw_reg_write(mlxsw_core, MLXSW_REG(mfgd), mfgd_pl); +} + +static int mlxsw_core_health_init(struct mlxsw_core *mlxsw_core) +{ + struct devlink *devlink = priv_to_devlink(mlxsw_core); + struct devlink_health_reporter *fw_fatal; + int err; + + if (!mlxsw_core->driver->fw_fatal_enabled) + return 0; + + fw_fatal = devlink_health_reporter_create(devlink, &mlxsw_core_health_fw_fatal_ops, + 0, mlxsw_core); + if (IS_ERR(fw_fatal)) { + dev_err(mlxsw_core->bus_info->dev, "Failed to create fw fatal reporter"); + return PTR_ERR(fw_fatal); + } + mlxsw_core->health.fw_fatal = fw_fatal; + + err = mlxsw_core_trap_register(mlxsw_core, &mlxsw_core_health_listener, mlxsw_core); + if (err) + goto err_trap_register; + + err = mlxsw_core_health_fw_fatal_config(mlxsw_core, true); + if (err) + goto err_fw_fatal_config; + + return 0; + +err_fw_fatal_config: + mlxsw_core_trap_unregister(mlxsw_core, &mlxsw_core_health_listener, mlxsw_core); +err_trap_register: + devlink_health_reporter_destroy(mlxsw_core->health.fw_fatal); + return err; +} + +static void mlxsw_core_health_fini(struct mlxsw_core *mlxsw_core) +{ + if (!mlxsw_core->driver->fw_fatal_enabled) + return; + + mlxsw_core_health_fw_fatal_config(mlxsw_core, false); + mlxsw_core_trap_unregister(mlxsw_core, &mlxsw_core_health_listener, mlxsw_core); + /* Make sure there is no more event work scheduled */ + mlxsw_core_flush_owq(); + devlink_health_reporter_destroy(mlxsw_core->health.fw_fatal); +} + static int __mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info, const struct mlxsw_bus *mlxsw_bus, @@ -1368,12 +1927,21 @@ __mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info, goto err_devlink_register; } - if (mlxsw_driver->params_register && !reload) { - err = mlxsw_driver->params_register(mlxsw_core); + if (!reload) { + err = mlxsw_core_params_register(mlxsw_core); if (err) goto err_register_params; } + err = mlxsw_core_fw_rev_validate(mlxsw_core, mlxsw_bus_info, mlxsw_driver->fw_req_rev, + mlxsw_driver->fw_filename); + if (err) + goto err_fw_rev_validate; + + err = mlxsw_core_health_init(mlxsw_core); + if (err) + goto err_health_init; + if (mlxsw_driver->init) { err = mlxsw_driver->init(mlxsw_core, mlxsw_bus_info, extack); if (err) @@ -1389,22 +1957,31 @@ __mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info, if (err) goto err_thermal_init; - if (mlxsw_driver->params_register) - devlink_params_publish(devlink); + err = mlxsw_env_init(mlxsw_core, &mlxsw_core->env); + if (err) + goto err_env_init; + + mlxsw_core->is_initialized = true; + devlink_params_publish(devlink); if (!reload) devlink_reload_enable(devlink); return 0; +err_env_init: + mlxsw_thermal_fini(mlxsw_core->thermal); err_thermal_init: mlxsw_hwmon_fini(mlxsw_core->hwmon); err_hwmon_init: if (mlxsw_core->driver->fini) mlxsw_core->driver->fini(mlxsw_core); err_driver_init: - if (mlxsw_driver->params_unregister && !reload) - mlxsw_driver->params_unregister(mlxsw_core); + mlxsw_core_health_fini(mlxsw_core); +err_health_init: +err_fw_rev_validate: + if (!reload) + mlxsw_core_params_unregister(mlxsw_core); err_register_params: if (!reload) devlink_unregister(devlink); @@ -1469,14 +2046,16 @@ void mlxsw_core_bus_device_unregister(struct mlxsw_core *mlxsw_core, return; } - if (mlxsw_core->driver->params_unregister) - devlink_params_unpublish(devlink); + devlink_params_unpublish(devlink); + mlxsw_core->is_initialized = false; + mlxsw_env_fini(mlxsw_core->env); mlxsw_thermal_fini(mlxsw_core->thermal); mlxsw_hwmon_fini(mlxsw_core->hwmon); if (mlxsw_core->driver->fini) mlxsw_core->driver->fini(mlxsw_core); - if (mlxsw_core->driver->params_unregister && !reload) - mlxsw_core->driver->params_unregister(mlxsw_core); + mlxsw_core_health_fini(mlxsw_core); + if (!reload) + mlxsw_core_params_unregister(mlxsw_core); if (!reload) devlink_unregister(devlink); mlxsw_emad_fini(mlxsw_core); @@ -1489,8 +2068,7 @@ void mlxsw_core_bus_device_unregister(struct mlxsw_core *mlxsw_core, return; reload_fail_deinit: - if (mlxsw_core->driver->params_unregister) - mlxsw_core->driver->params_unregister(mlxsw_core); + mlxsw_core_params_unregister(mlxsw_core); devlink_unregister(devlink); devlink_resources_unregister(devlink, NULL); devlink_free(devlink); @@ -2274,6 +2852,16 @@ mlxsw_core_port_devlink_port_get(struct mlxsw_core *mlxsw_core, } EXPORT_SYMBOL(mlxsw_core_port_devlink_port_get); +struct mlxsw_env *mlxsw_core_env(const struct mlxsw_core *mlxsw_core) +{ + return mlxsw_core->env; +} + +bool mlxsw_core_is_initialized(const struct mlxsw_core *mlxsw_core) +{ + return mlxsw_core->is_initialized; +} + int mlxsw_core_module_max_width(struct mlxsw_core *mlxsw_core, u8 module) { enum mlxsw_reg_pmtm_module_type module_type; @@ -2410,18 +2998,6 @@ int mlxsw_core_kvd_sizes_get(struct mlxsw_core *mlxsw_core, } EXPORT_SYMBOL(mlxsw_core_kvd_sizes_get); -void mlxsw_core_fw_flash_start(struct mlxsw_core *mlxsw_core) -{ - mlxsw_core->fw_flash_in_progress = true; -} -EXPORT_SYMBOL(mlxsw_core_fw_flash_start); - -void mlxsw_core_fw_flash_end(struct mlxsw_core *mlxsw_core) -{ - mlxsw_core->fw_flash_in_progress = false; -} -EXPORT_SYMBOL(mlxsw_core_fw_flash_end); - int mlxsw_core_resources_query(struct mlxsw_core *mlxsw_core, char *mbox, struct mlxsw_res *res) { diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.h b/drivers/net/ethernet/mellanox/mlxsw/core.h index 11af3308f8cc..92f7398287be 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.h +++ b/drivers/net/ethernet/mellanox/mlxsw/core.h @@ -32,6 +32,8 @@ void *mlxsw_core_driver_priv(struct mlxsw_core *mlxsw_core); bool mlxsw_core_res_query_enabled(const struct mlxsw_core *mlxsw_core); +bool mlxsw_core_temp_warn_enabled(const struct mlxsw_core *mlxsw_core); + bool mlxsw_core_fw_rev_minor_subminor_validate(const struct mlxsw_fw_rev *rev, const struct mlxsw_fw_rev *req_rev); @@ -221,6 +223,8 @@ enum devlink_port_type mlxsw_core_port_type_get(struct mlxsw_core *mlxsw_core, struct devlink_port * mlxsw_core_port_devlink_port_get(struct mlxsw_core *mlxsw_core, u8 local_port); +struct mlxsw_env *mlxsw_core_env(const struct mlxsw_core *mlxsw_core); +bool mlxsw_core_is_initialized(const struct mlxsw_core *mlxsw_core); int mlxsw_core_module_max_width(struct mlxsw_core *mlxsw_core, u8 module); int mlxsw_core_schedule_dw(struct delayed_work *dwork, unsigned long delay); @@ -280,6 +284,8 @@ struct mlxsw_driver { struct list_head list; const char *kind; size_t priv_size; + const struct mlxsw_fw_rev *fw_req_rev; + const char *fw_filename; int (*init)(struct mlxsw_core *mlxsw_core, const struct mlxsw_bus_info *mlxsw_bus_info, struct netlink_ext_ack *extack); @@ -324,9 +330,6 @@ struct mlxsw_driver { unsigned int sb_index, u16 tc_index, enum devlink_sb_pool_type pool_type, u32 *p_cur, u32 *p_max); - int (*flash_update)(struct mlxsw_core *mlxsw_core, - const char *file_name, const char *component, - struct netlink_ext_ack *extack); int (*trap_init)(struct mlxsw_core *mlxsw_core, const struct devlink_trap *trap, void *trap_ctx); void (*trap_fini)(struct mlxsw_core *mlxsw_core, @@ -371,6 +374,8 @@ struct mlxsw_driver { u8 txhdr_len; const struct mlxsw_config_profile *profile; bool res_query_enabled; + bool fw_fatal_enabled; + bool temp_warn_enabled; }; int mlxsw_core_kvd_sizes_get(struct mlxsw_core *mlxsw_core, @@ -378,9 +383,6 @@ int mlxsw_core_kvd_sizes_get(struct mlxsw_core *mlxsw_core, u64 *p_single_size, u64 *p_double_size, u64 *p_linear_size); -void mlxsw_core_fw_flash_start(struct mlxsw_core *mlxsw_core); -void mlxsw_core_fw_flash_end(struct mlxsw_core *mlxsw_core); - u32 mlxsw_core_read_frc_h(struct mlxsw_core *mlxsw_core); u32 mlxsw_core_read_frc_l(struct mlxsw_core *mlxsw_core); diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_env.c b/drivers/net/ethernet/mellanox/mlxsw/core_env.c index 056eeb85be60..dd26865bd587 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core_env.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core_env.c @@ -10,6 +10,18 @@ #include "item.h" #include "reg.h" +struct mlxsw_env_module_info { + u64 module_overheat_counter; + bool is_overheat; +}; + +struct mlxsw_env { + struct mlxsw_core *core; + u8 module_count; + spinlock_t module_info_lock; /* Protects 'module_info'. */ + struct mlxsw_env_module_info module_info[]; +}; + static int mlxsw_env_validate_cable_ident(struct mlxsw_core *core, int id, bool *qsfp, bool *cmis) { @@ -293,3 +305,359 @@ int mlxsw_env_get_module_eeprom(struct net_device *netdev, return 0; } EXPORT_SYMBOL(mlxsw_env_get_module_eeprom); + +static int mlxsw_env_module_has_temp_sensor(struct mlxsw_core *mlxsw_core, + u8 module, + bool *p_has_temp_sensor) +{ + char mtbr_pl[MLXSW_REG_MTBR_LEN]; + u16 temp; + int err; + + mlxsw_reg_mtbr_pack(mtbr_pl, MLXSW_REG_MTBR_BASE_MODULE_INDEX + module, + 1); + err = mlxsw_reg_query(mlxsw_core, MLXSW_REG(mtbr), mtbr_pl); + if (err) + return err; + + mlxsw_reg_mtbr_temp_unpack(mtbr_pl, 0, &temp, NULL); + + switch (temp) { + case MLXSW_REG_MTBR_BAD_SENS_INFO: + case MLXSW_REG_MTBR_NO_CONN: + case MLXSW_REG_MTBR_NO_TEMP_SENS: + case MLXSW_REG_MTBR_INDEX_NA: + *p_has_temp_sensor = false; + break; + default: + *p_has_temp_sensor = temp ? true : false; + } + return 0; +} + +static int mlxsw_env_temp_event_set(struct mlxsw_core *mlxsw_core, + u16 sensor_index, bool enable) +{ + char mtmp_pl[MLXSW_REG_MTMP_LEN] = {0}; + enum mlxsw_reg_mtmp_tee tee; + int err, threshold_hi; + + mlxsw_reg_mtmp_sensor_index_set(mtmp_pl, sensor_index); + err = mlxsw_reg_query(mlxsw_core, MLXSW_REG(mtmp), mtmp_pl); + if (err) + return err; + + if (enable) { + err = mlxsw_env_module_temp_thresholds_get(mlxsw_core, + sensor_index - + MLXSW_REG_MTMP_MODULE_INDEX_MIN, + SFP_TEMP_HIGH_WARN, + &threshold_hi); + /* In case it is not possible to query the module's threshold, + * use the default value. + */ + if (err) + threshold_hi = MLXSW_REG_MTMP_THRESH_HI; + else + /* mlxsw_env_module_temp_thresholds_get() multiplies + * Celsius degrees by 1000 whereas MTMP expects + * temperature in 0.125 Celsius degrees units. + * Convert threshold_hi to correct units. + */ + threshold_hi = threshold_hi / 1000 * 8; + + mlxsw_reg_mtmp_temperature_threshold_hi_set(mtmp_pl, threshold_hi); + mlxsw_reg_mtmp_temperature_threshold_lo_set(mtmp_pl, threshold_hi - + MLXSW_REG_MTMP_HYSTERESIS_TEMP); + } + tee = enable ? MLXSW_REG_MTMP_TEE_GENERATE_EVENT : MLXSW_REG_MTMP_TEE_NO_EVENT; + mlxsw_reg_mtmp_tee_set(mtmp_pl, tee); + return mlxsw_reg_write(mlxsw_core, MLXSW_REG(mtmp), mtmp_pl); +} + +static int mlxsw_env_module_temp_event_enable(struct mlxsw_core *mlxsw_core, + u8 module_count) +{ + int i, err, sensor_index; + bool has_temp_sensor; + + for (i = 0; i < module_count; i++) { + err = mlxsw_env_module_has_temp_sensor(mlxsw_core, i, + &has_temp_sensor); + if (err) + return err; + + if (!has_temp_sensor) + continue; + + sensor_index = i + MLXSW_REG_MTMP_MODULE_INDEX_MIN; + err = mlxsw_env_temp_event_set(mlxsw_core, sensor_index, true); + if (err) + return err; + } + + return 0; +} + +static void mlxsw_env_mtwe_event_func(const struct mlxsw_reg_info *reg, + char *mtwe_pl, void *priv) +{ + struct mlxsw_env *mlxsw_env = priv; + int i, sensor_warning; + bool is_overheat; + + for (i = 0; i < mlxsw_env->module_count; i++) { + /* 64-127 of sensor_index are mapped to the port modules + * sequentially (module 0 is mapped to sensor_index 64, + * module 1 to sensor_index 65 and so on) + */ + sensor_warning = + mlxsw_reg_mtwe_sensor_warning_get(mtwe_pl, + i + MLXSW_REG_MTMP_MODULE_INDEX_MIN); + spin_lock(&mlxsw_env->module_info_lock); + is_overheat = + mlxsw_env->module_info[i].is_overheat; + + if ((is_overheat && sensor_warning) || + (!is_overheat && !sensor_warning)) { + /* Current state is "warning" and MTWE still reports + * warning OR current state in "no warning" and MTWE + * does not report warning. + */ + spin_unlock(&mlxsw_env->module_info_lock); + continue; + } else if (is_overheat && !sensor_warning) { + /* MTWE reports "no warning", turn is_overheat off. + */ + mlxsw_env->module_info[i].is_overheat = false; + spin_unlock(&mlxsw_env->module_info_lock); + } else { + /* Current state is "no warning" and MTWE reports + * "warning", increase the counter and turn is_overheat + * on. + */ + mlxsw_env->module_info[i].is_overheat = true; + mlxsw_env->module_info[i].module_overheat_counter++; + spin_unlock(&mlxsw_env->module_info_lock); + } + } +} + +static const struct mlxsw_listener mlxsw_env_temp_warn_listener = + MLXSW_EVENTL(mlxsw_env_mtwe_event_func, MTWE, MTWE); + +static int mlxsw_env_temp_warn_event_register(struct mlxsw_core *mlxsw_core) +{ + struct mlxsw_env *mlxsw_env = mlxsw_core_env(mlxsw_core); + + if (!mlxsw_core_temp_warn_enabled(mlxsw_core)) + return 0; + + return mlxsw_core_trap_register(mlxsw_core, + &mlxsw_env_temp_warn_listener, + mlxsw_env); +} + +static void mlxsw_env_temp_warn_event_unregister(struct mlxsw_env *mlxsw_env) +{ + if (!mlxsw_core_temp_warn_enabled(mlxsw_env->core)) + return; + + mlxsw_core_trap_unregister(mlxsw_env->core, + &mlxsw_env_temp_warn_listener, mlxsw_env); +} + +struct mlxsw_env_module_plug_unplug_event { + struct mlxsw_env *mlxsw_env; + u8 module; + struct work_struct work; +}; + +static void mlxsw_env_pmpe_event_work(struct work_struct *work) +{ + struct mlxsw_env_module_plug_unplug_event *event; + struct mlxsw_env *mlxsw_env; + bool has_temp_sensor; + u16 sensor_index; + int err; + + event = container_of(work, struct mlxsw_env_module_plug_unplug_event, + work); + mlxsw_env = event->mlxsw_env; + + spin_lock_bh(&mlxsw_env->module_info_lock); + mlxsw_env->module_info[event->module].is_overheat = false; + spin_unlock_bh(&mlxsw_env->module_info_lock); + + err = mlxsw_env_module_has_temp_sensor(mlxsw_env->core, event->module, + &has_temp_sensor); + /* Do not disable events on modules without sensors or faulty sensors + * because FW returns errors. + */ + if (err) + goto out; + + if (!has_temp_sensor) + goto out; + + sensor_index = event->module + MLXSW_REG_MTMP_MODULE_INDEX_MIN; + mlxsw_env_temp_event_set(mlxsw_env->core, sensor_index, true); + +out: + kfree(event); +} + +static void +mlxsw_env_pmpe_listener_func(const struct mlxsw_reg_info *reg, char *pmpe_pl, + void *priv) +{ + struct mlxsw_env_module_plug_unplug_event *event; + enum mlxsw_reg_pmpe_module_status module_status; + u8 module = mlxsw_reg_pmpe_module_get(pmpe_pl); + struct mlxsw_env *mlxsw_env = priv; + + if (WARN_ON_ONCE(module >= mlxsw_env->module_count)) + return; + + module_status = mlxsw_reg_pmpe_module_status_get(pmpe_pl); + if (module_status != MLXSW_REG_PMPE_MODULE_STATUS_PLUGGED_ENABLED) + return; + + event = kmalloc(sizeof(*event), GFP_ATOMIC); + if (!event) + return; + + event->mlxsw_env = mlxsw_env; + event->module = module; + INIT_WORK(&event->work, mlxsw_env_pmpe_event_work); + mlxsw_core_schedule_work(&event->work); +} + +static const struct mlxsw_listener mlxsw_env_module_plug_listener = + MLXSW_EVENTL(mlxsw_env_pmpe_listener_func, PMPE, PMPE); + +static int +mlxsw_env_module_plug_event_register(struct mlxsw_core *mlxsw_core) +{ + struct mlxsw_env *mlxsw_env = mlxsw_core_env(mlxsw_core); + + if (!mlxsw_core_temp_warn_enabled(mlxsw_core)) + return 0; + + return mlxsw_core_trap_register(mlxsw_core, + &mlxsw_env_module_plug_listener, + mlxsw_env); +} + +static void +mlxsw_env_module_plug_event_unregister(struct mlxsw_env *mlxsw_env) +{ + if (!mlxsw_core_temp_warn_enabled(mlxsw_env->core)) + return; + + mlxsw_core_trap_unregister(mlxsw_env->core, + &mlxsw_env_module_plug_listener, + mlxsw_env); +} + +static int +mlxsw_env_module_oper_state_event_enable(struct mlxsw_core *mlxsw_core, + u8 module_count) +{ + int i, err; + + for (i = 0; i < module_count; i++) { + char pmaos_pl[MLXSW_REG_PMAOS_LEN]; + + mlxsw_reg_pmaos_pack(pmaos_pl, i, + MLXSW_REG_PMAOS_E_GENERATE_EVENT); + err = mlxsw_reg_write(mlxsw_core, MLXSW_REG(pmaos), pmaos_pl); + if (err) + return err; + } + return 0; +} + +int +mlxsw_env_module_overheat_counter_get(struct mlxsw_core *mlxsw_core, u8 module, + u64 *p_counter) +{ + struct mlxsw_env *mlxsw_env = mlxsw_core_env(mlxsw_core); + + /* Prevent switch driver from accessing uninitialized data. */ + if (!mlxsw_core_is_initialized(mlxsw_core)) { + *p_counter = 0; + return 0; + } + + if (WARN_ON_ONCE(module >= mlxsw_env->module_count)) + return -EINVAL; + + spin_lock_bh(&mlxsw_env->module_info_lock); + *p_counter = mlxsw_env->module_info[module].module_overheat_counter; + spin_unlock_bh(&mlxsw_env->module_info_lock); + + return 0; +} +EXPORT_SYMBOL(mlxsw_env_module_overheat_counter_get); + +int mlxsw_env_init(struct mlxsw_core *mlxsw_core, struct mlxsw_env **p_env) +{ + char mgpir_pl[MLXSW_REG_MGPIR_LEN]; + struct mlxsw_env *env; + u8 module_count; + int err; + + mlxsw_reg_mgpir_pack(mgpir_pl); + err = mlxsw_reg_query(mlxsw_core, MLXSW_REG(mgpir), mgpir_pl); + if (err) + return err; + + mlxsw_reg_mgpir_unpack(mgpir_pl, NULL, NULL, NULL, &module_count); + + env = kzalloc(struct_size(env, module_info, module_count), GFP_KERNEL); + if (!env) + return -ENOMEM; + + spin_lock_init(&env->module_info_lock); + env->core = mlxsw_core; + env->module_count = module_count; + *p_env = env; + + err = mlxsw_env_temp_warn_event_register(mlxsw_core); + if (err) + goto err_temp_warn_event_register; + + err = mlxsw_env_module_plug_event_register(mlxsw_core); + if (err) + goto err_module_plug_event_register; + + err = mlxsw_env_module_oper_state_event_enable(mlxsw_core, + env->module_count); + if (err) + goto err_oper_state_event_enable; + + err = mlxsw_env_module_temp_event_enable(mlxsw_core, env->module_count); + if (err) + goto err_temp_event_enable; + + return 0; + +err_temp_event_enable: +err_oper_state_event_enable: + mlxsw_env_module_plug_event_unregister(env); +err_module_plug_event_register: + mlxsw_env_temp_warn_event_unregister(env); +err_temp_warn_event_register: + kfree(env); + return err; +} + +void mlxsw_env_fini(struct mlxsw_env *env) +{ + mlxsw_env_module_plug_event_unregister(env); + /* Make sure there is no more event work scheduled. */ + mlxsw_core_flush_owq(); + mlxsw_env_temp_warn_event_unregister(env); + kfree(env); +} diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_env.h b/drivers/net/ethernet/mellanox/mlxsw/core_env.h index 064d0e770c01..8e36a2634ef5 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core_env.h +++ b/drivers/net/ethernet/mellanox/mlxsw/core_env.h @@ -14,4 +14,10 @@ int mlxsw_env_get_module_eeprom(struct net_device *netdev, struct mlxsw_core *mlxsw_core, int module, struct ethtool_eeprom *ee, u8 *data); +int +mlxsw_env_module_overheat_counter_get(struct mlxsw_core *mlxsw_core, u8 module, + u64 *p_counter); +int mlxsw_env_init(struct mlxsw_core *core, struct mlxsw_env **p_env); +void mlxsw_env_fini(struct mlxsw_env *env); + #endif diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c b/drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c index 61719ec89808..2196c946698a 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c @@ -12,8 +12,17 @@ #include "core.h" #include "core_env.h" -#define MLXSW_HWMON_TEMP_SENSOR_MAX_COUNT 127 -#define MLXSW_HWMON_ATTR_COUNT (MLXSW_HWMON_TEMP_SENSOR_MAX_COUNT * 4 + \ +#define MLXSW_HWMON_SENSORS_MAX_COUNT 64 +#define MLXSW_HWMON_MODULES_MAX_COUNT 64 +#define MLXSW_HWMON_GEARBOXES_MAX_COUNT 32 + +#define MLXSW_HWMON_ATTR_PER_SENSOR 3 +#define MLXSW_HWMON_ATTR_PER_MODULE 7 +#define MLXSW_HWMON_ATTR_PER_GEARBOX 4 + +#define MLXSW_HWMON_ATTR_COUNT (MLXSW_HWMON_SENSORS_MAX_COUNT * MLXSW_HWMON_ATTR_PER_SENSOR + \ + MLXSW_HWMON_MODULES_MAX_COUNT * MLXSW_HWMON_ATTR_PER_MODULE + \ + MLXSW_HWMON_GEARBOXES_MAX_COUNT * MLXSW_HWMON_ATTR_PER_GEARBOX + \ MLXSW_MFCR_TACHOS_MAX + MLXSW_MFCR_PWMS_MAX) struct mlxsw_hwmon_attr { @@ -97,7 +106,7 @@ static ssize_t mlxsw_hwmon_temp_rst_store(struct device *dev, struct mlxsw_hwmon_attr *mlwsw_hwmon_attr = container_of(attr, struct mlxsw_hwmon_attr, dev_attr); struct mlxsw_hwmon *mlxsw_hwmon = mlwsw_hwmon_attr->hwmon; - char mtmp_pl[MLXSW_REG_MTMP_LEN]; + char mtmp_pl[MLXSW_REG_MTMP_LEN] = {0}; unsigned long val; int index; int err; @@ -110,7 +119,13 @@ static ssize_t mlxsw_hwmon_temp_rst_store(struct device *dev, index = mlxsw_hwmon_get_attr_index(mlwsw_hwmon_attr->type_index, mlxsw_hwmon->module_sensor_max); - mlxsw_reg_mtmp_pack(mtmp_pl, index, true, true); + + mlxsw_reg_mtmp_sensor_index_set(mtmp_pl, index); + err = mlxsw_reg_query(mlxsw_hwmon->core, MLXSW_REG(mtmp), mtmp_pl); + if (err) + return err; + mlxsw_reg_mtmp_mte_set(mtmp_pl, true); + mlxsw_reg_mtmp_mtr_set(mtmp_pl, true); err = mlxsw_reg_write(mlxsw_hwmon->core, MLXSW_REG(mtmp), mtmp_pl); if (err) { dev_err(mlxsw_hwmon->bus_info->dev, "Failed to reset temp sensor history\n"); @@ -205,25 +220,39 @@ static ssize_t mlxsw_hwmon_pwm_store(struct device *dev, return len; } -static ssize_t mlxsw_hwmon_module_temp_show(struct device *dev, - struct device_attribute *attr, - char *buf) +static int mlxsw_hwmon_module_temp_get(struct device *dev, + struct device_attribute *attr, + int *p_temp) { struct mlxsw_hwmon_attr *mlwsw_hwmon_attr = container_of(attr, struct mlxsw_hwmon_attr, dev_attr); struct mlxsw_hwmon *mlxsw_hwmon = mlwsw_hwmon_attr->hwmon; char mtmp_pl[MLXSW_REG_MTMP_LEN]; u8 module; - int temp; int err; module = mlwsw_hwmon_attr->type_index - mlxsw_hwmon->sensor_count; mlxsw_reg_mtmp_pack(mtmp_pl, MLXSW_REG_MTMP_MODULE_INDEX_MIN + module, false, false); err = mlxsw_reg_query(mlxsw_hwmon->core, MLXSW_REG(mtmp), mtmp_pl); + if (err) { + dev_err(dev, "Failed to query module temperature\n"); + return err; + } + mlxsw_reg_mtmp_unpack(mtmp_pl, p_temp, NULL, NULL); + + return 0; +} + +static ssize_t mlxsw_hwmon_module_temp_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + int err, temp; + + err = mlxsw_hwmon_module_temp_get(dev, attr, &temp); if (err) return err; - mlxsw_reg_mtmp_unpack(mtmp_pl, &temp, NULL, NULL); return sprintf(buf, "%d\n", temp); } @@ -270,48 +299,72 @@ static ssize_t mlxsw_hwmon_module_temp_fault_show(struct device *dev, return sprintf(buf, "%u\n", fault); } -static ssize_t -mlxsw_hwmon_module_temp_critical_show(struct device *dev, - struct device_attribute *attr, char *buf) +static int mlxsw_hwmon_module_temp_critical_get(struct device *dev, + struct device_attribute *attr, + int *p_temp) { struct mlxsw_hwmon_attr *mlwsw_hwmon_attr = container_of(attr, struct mlxsw_hwmon_attr, dev_attr); struct mlxsw_hwmon *mlxsw_hwmon = mlwsw_hwmon_attr->hwmon; - int temp; u8 module; int err; module = mlwsw_hwmon_attr->type_index - mlxsw_hwmon->sensor_count; err = mlxsw_env_module_temp_thresholds_get(mlxsw_hwmon->core, module, - SFP_TEMP_HIGH_WARN, &temp); + SFP_TEMP_HIGH_WARN, p_temp); if (err) { dev_err(dev, "Failed to query module temperature thresholds\n"); return err; } - return sprintf(buf, "%u\n", temp); + return 0; } static ssize_t -mlxsw_hwmon_module_temp_emergency_show(struct device *dev, - struct device_attribute *attr, - char *buf) +mlxsw_hwmon_module_temp_critical_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int err, temp; + + err = mlxsw_hwmon_module_temp_critical_get(dev, attr, &temp); + if (err) + return err; + + return sprintf(buf, "%u\n", temp); +} + +static int mlxsw_hwmon_module_temp_emergency_get(struct device *dev, + struct device_attribute *attr, + int *p_temp) { struct mlxsw_hwmon_attr *mlwsw_hwmon_attr = container_of(attr, struct mlxsw_hwmon_attr, dev_attr); struct mlxsw_hwmon *mlxsw_hwmon = mlwsw_hwmon_attr->hwmon; u8 module; - int temp; int err; module = mlwsw_hwmon_attr->type_index - mlxsw_hwmon->sensor_count; err = mlxsw_env_module_temp_thresholds_get(mlxsw_hwmon->core, module, - SFP_TEMP_HIGH_ALARM, &temp); + SFP_TEMP_HIGH_ALARM, p_temp); if (err) { dev_err(dev, "Failed to query module temperature thresholds\n"); return err; } + return 0; +} + +static ssize_t +mlxsw_hwmon_module_temp_emergency_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + int err, temp; + + err = mlxsw_hwmon_module_temp_emergency_get(dev, attr, &temp); + if (err) + return err; + return sprintf(buf, "%u\n", temp); } @@ -341,6 +394,53 @@ mlxsw_hwmon_gbox_temp_label_show(struct device *dev, return sprintf(buf, "gearbox %03u\n", index); } +static ssize_t mlxsw_hwmon_temp_critical_alarm_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + int err, temp, emergency_temp, critic_temp; + + err = mlxsw_hwmon_module_temp_get(dev, attr, &temp); + if (err) + return err; + + if (temp <= 0) + return sprintf(buf, "%d\n", false); + + err = mlxsw_hwmon_module_temp_emergency_get(dev, attr, &emergency_temp); + if (err) + return err; + + if (temp >= emergency_temp) + return sprintf(buf, "%d\n", false); + + err = mlxsw_hwmon_module_temp_critical_get(dev, attr, &critic_temp); + if (err) + return err; + + return sprintf(buf, "%d\n", temp >= critic_temp); +} + +static ssize_t mlxsw_hwmon_temp_emergency_alarm_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + int err, temp, emergency_temp; + + err = mlxsw_hwmon_module_temp_get(dev, attr, &temp); + if (err) + return err; + + if (temp <= 0) + return sprintf(buf, "%d\n", false); + + err = mlxsw_hwmon_module_temp_emergency_get(dev, attr, &emergency_temp); + if (err) + return err; + + return sprintf(buf, "%d\n", temp >= emergency_temp); +} + enum mlxsw_hwmon_attr_type { MLXSW_HWMON_ATTR_TYPE_TEMP, MLXSW_HWMON_ATTR_TYPE_TEMP_MAX, @@ -354,6 +454,8 @@ enum mlxsw_hwmon_attr_type { MLXSW_HWMON_ATTR_TYPE_TEMP_MODULE_EMERG, MLXSW_HWMON_ATTR_TYPE_TEMP_MODULE_LABEL, MLXSW_HWMON_ATTR_TYPE_TEMP_GBOX_LABEL, + MLXSW_HWMON_ATTR_TYPE_TEMP_CRIT_ALARM, + MLXSW_HWMON_ATTR_TYPE_TEMP_EMERGENCY_ALARM, }; static void mlxsw_hwmon_attr_add(struct mlxsw_hwmon *mlxsw_hwmon, @@ -444,6 +546,20 @@ static void mlxsw_hwmon_attr_add(struct mlxsw_hwmon *mlxsw_hwmon, snprintf(mlxsw_hwmon_attr->name, sizeof(mlxsw_hwmon_attr->name), "temp%u_label", num + 1); break; + case MLXSW_HWMON_ATTR_TYPE_TEMP_CRIT_ALARM: + mlxsw_hwmon_attr->dev_attr.show = + mlxsw_hwmon_temp_critical_alarm_show; + mlxsw_hwmon_attr->dev_attr.attr.mode = 0444; + snprintf(mlxsw_hwmon_attr->name, sizeof(mlxsw_hwmon_attr->name), + "temp%u_crit_alarm", num + 1); + break; + case MLXSW_HWMON_ATTR_TYPE_TEMP_EMERGENCY_ALARM: + mlxsw_hwmon_attr->dev_attr.show = + mlxsw_hwmon_temp_emergency_alarm_show; + mlxsw_hwmon_attr->dev_attr.attr.mode = 0444; + snprintf(mlxsw_hwmon_attr->name, sizeof(mlxsw_hwmon_attr->name), + "temp%u_emergency_alarm", num + 1); + break; default: WARN_ON(1); } @@ -460,7 +576,6 @@ static void mlxsw_hwmon_attr_add(struct mlxsw_hwmon *mlxsw_hwmon, static int mlxsw_hwmon_temp_init(struct mlxsw_hwmon *mlxsw_hwmon) { char mtcap_pl[MLXSW_REG_MTCAP_LEN] = {0}; - char mtmp_pl[MLXSW_REG_MTMP_LEN]; int i; int err; @@ -471,7 +586,15 @@ static int mlxsw_hwmon_temp_init(struct mlxsw_hwmon *mlxsw_hwmon) } mlxsw_hwmon->sensor_count = mlxsw_reg_mtcap_sensor_count_get(mtcap_pl); for (i = 0; i < mlxsw_hwmon->sensor_count; i++) { - mlxsw_reg_mtmp_pack(mtmp_pl, i, true, true); + char mtmp_pl[MLXSW_REG_MTMP_LEN] = {0}; + + mlxsw_reg_mtmp_sensor_index_set(mtmp_pl, i); + err = mlxsw_reg_query(mlxsw_hwmon->core, MLXSW_REG(mtmp), + mtmp_pl); + if (err) + return err; + mlxsw_reg_mtmp_mte_set(mtmp_pl, true); + mlxsw_reg_mtmp_mtr_set(mtmp_pl, true); err = mlxsw_reg_write(mlxsw_hwmon->core, MLXSW_REG(mtmp), mtmp_pl); if (err) { @@ -566,6 +689,12 @@ static int mlxsw_hwmon_module_init(struct mlxsw_hwmon *mlxsw_hwmon) mlxsw_hwmon_attr_add(mlxsw_hwmon, MLXSW_HWMON_ATTR_TYPE_TEMP_MODULE_LABEL, i, i); + mlxsw_hwmon_attr_add(mlxsw_hwmon, + MLXSW_HWMON_ATTR_TYPE_TEMP_CRIT_ALARM, + i, i); + mlxsw_hwmon_attr_add(mlxsw_hwmon, + MLXSW_HWMON_ATTR_TYPE_TEMP_EMERGENCY_ALARM, + i, i); } return 0; diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c index 1c64b03ff48e..641cdd81882b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/pci.c +++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c @@ -620,9 +620,9 @@ static char *mlxsw_pci_cq_sw_cqe_get(struct mlxsw_pci_queue *q) return elem; } -static void mlxsw_pci_cq_tasklet(unsigned long data) +static void mlxsw_pci_cq_tasklet(struct tasklet_struct *t) { - struct mlxsw_pci_queue *q = (struct mlxsw_pci_queue *) data; + struct mlxsw_pci_queue *q = from_tasklet(q, t, tasklet); struct mlxsw_pci *mlxsw_pci = q->pci; char *cqe; int items = 0; @@ -733,9 +733,9 @@ static char *mlxsw_pci_eq_sw_eqe_get(struct mlxsw_pci_queue *q) return elem; } -static void mlxsw_pci_eq_tasklet(unsigned long data) +static void mlxsw_pci_eq_tasklet(struct tasklet_struct *t) { - struct mlxsw_pci_queue *q = (struct mlxsw_pci_queue *) data; + struct mlxsw_pci_queue *q = from_tasklet(q, t, tasklet); struct mlxsw_pci *mlxsw_pci = q->pci; u8 cq_count = mlxsw_pci_cq_count(mlxsw_pci); unsigned long active_cqns[BITS_TO_LONGS(MLXSW_PCI_CQS_MAX)]; @@ -792,7 +792,7 @@ struct mlxsw_pci_queue_ops { struct mlxsw_pci_queue *q); void (*fini)(struct mlxsw_pci *mlxsw_pci, struct mlxsw_pci_queue *q); - void (*tasklet)(unsigned long data); + void (*tasklet)(struct tasklet_struct *t); u16 (*elem_count_f)(const struct mlxsw_pci_queue *q); u8 (*elem_size_f)(const struct mlxsw_pci_queue *q); u16 elem_count; @@ -855,7 +855,7 @@ static int mlxsw_pci_queue_init(struct mlxsw_pci *mlxsw_pci, char *mbox, q->pci = mlxsw_pci; if (q_ops->tasklet) - tasklet_init(&q->tasklet, q_ops->tasklet, (unsigned long) q); + tasklet_setup(&q->tasklet, q_ops->tasklet); mem_item->size = MLXSW_PCI_AQ_SIZE; mem_item->buf = pci_alloc_consistent(mlxsw_pci->pdev, diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index 079b080de7f7..39eff6a57ba2 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -4174,7 +4174,6 @@ MLXSW_ITEM32(reg, ptys, an_status, 0x04, 28, 4); #define MLXSW_REG_PTYS_EXT_ETH_SPEED_SGMII_100M BIT(0) #define MLXSW_REG_PTYS_EXT_ETH_SPEED_1000BASE_X_SGMII BIT(1) -#define MLXSW_REG_PTYS_EXT_ETH_SPEED_2_5GBASE_X_2_5GMII BIT(2) #define MLXSW_REG_PTYS_EXT_ETH_SPEED_5GBASE_R BIT(3) #define MLXSW_REG_PTYS_EXT_ETH_SPEED_XFI_XAUI_1_10G BIT(4) #define MLXSW_REG_PTYS_EXT_ETH_SPEED_XLAUI_4_XLPPI_4_40G BIT(5) @@ -4197,7 +4196,6 @@ MLXSW_ITEM32(reg, ptys, ext_eth_proto_cap, 0x08, 0, 32); #define MLXSW_REG_PTYS_ETH_SPEED_10GBASE_CX4 BIT(2) #define MLXSW_REG_PTYS_ETH_SPEED_10GBASE_KX4 BIT(3) #define MLXSW_REG_PTYS_ETH_SPEED_10GBASE_KR BIT(4) -#define MLXSW_REG_PTYS_ETH_SPEED_20GBASE_KR2 BIT(5) #define MLXSW_REG_PTYS_ETH_SPEED_40GBASE_CR4 BIT(6) #define MLXSW_REG_PTYS_ETH_SPEED_40GBASE_KR4 BIT(7) #define MLXSW_REG_PTYS_ETH_SPEED_10GBASE_CR BIT(12) @@ -4210,10 +4208,6 @@ MLXSW_ITEM32(reg, ptys, ext_eth_proto_cap, 0x08, 0, 32); #define MLXSW_REG_PTYS_ETH_SPEED_100GBASE_CR4 BIT(20) #define MLXSW_REG_PTYS_ETH_SPEED_100GBASE_SR4 BIT(21) #define MLXSW_REG_PTYS_ETH_SPEED_100GBASE_KR4 BIT(22) -#define MLXSW_REG_PTYS_ETH_SPEED_100GBASE_LR4_ER4 BIT(23) -#define MLXSW_REG_PTYS_ETH_SPEED_100BASE_TX BIT(24) -#define MLXSW_REG_PTYS_ETH_SPEED_100BASE_T BIT(25) -#define MLXSW_REG_PTYS_ETH_SPEED_10GBASE_T BIT(26) #define MLXSW_REG_PTYS_ETH_SPEED_25GBASE_CR BIT(27) #define MLXSW_REG_PTYS_ETH_SPEED_25GBASE_KR BIT(28) #define MLXSW_REG_PTYS_ETH_SPEED_25GBASE_SR BIT(29) @@ -5411,6 +5405,64 @@ static inline void mlxsw_reg_pspa_pack(char *payload, u8 swid, u8 local_port) mlxsw_reg_pspa_sub_port_set(payload, 0); } +/* PMAOS - Ports Module Administrative and Operational Status + * ---------------------------------------------------------- + * This register configures and retrieves the per module status. + */ +#define MLXSW_REG_PMAOS_ID 0x5012 +#define MLXSW_REG_PMAOS_LEN 0x10 + +MLXSW_REG_DEFINE(pmaos, MLXSW_REG_PMAOS_ID, MLXSW_REG_PMAOS_LEN); + +/* reg_slot_index + * Slot index. + * Access: Index + */ +MLXSW_ITEM32(reg, pmaos, slot_index, 0x00, 24, 4); + +/* reg_pmaos_module + * Module number. + * Access: Index + */ +MLXSW_ITEM32(reg, pmaos, module, 0x00, 16, 8); + +/* reg_pmaos_ase + * Admin state update enable. + * If this bit is set, admin state will be updated based on admin_state field. + * Only relevant on Set() operations. + * Access: WO + */ +MLXSW_ITEM32(reg, pmaos, ase, 0x04, 31, 1); + +/* reg_pmaos_ee + * Event update enable. + * If this bit is set, event generation will be updated based on the e field. + * Only relevant on Set operations. + * Access: WO + */ +MLXSW_ITEM32(reg, pmaos, ee, 0x04, 30, 1); + +enum mlxsw_reg_pmaos_e { + MLXSW_REG_PMAOS_E_DO_NOT_GENERATE_EVENT, + MLXSW_REG_PMAOS_E_GENERATE_EVENT, + MLXSW_REG_PMAOS_E_GENERATE_SINGLE_EVENT, +}; + +/* reg_pmaos_e + * Event Generation on operational state change. + * Access: RW + */ +MLXSW_ITEM32(reg, pmaos, e, 0x04, 0, 2); + +static inline void mlxsw_reg_pmaos_pack(char *payload, u8 module, + enum mlxsw_reg_pmaos_e e) +{ + MLXSW_REG_ZERO(pmaos, payload); + mlxsw_reg_pmaos_module_set(payload, module); + mlxsw_reg_pmaos_e_set(payload, e); + mlxsw_reg_pmaos_ee_set(payload, true); +} + /* PPLR - Port Physical Loopback Register * -------------------------------------- * This register allows configuration of the port's loopback mode. @@ -5447,6 +5499,50 @@ static inline void mlxsw_reg_pplr_pack(char *payload, u8 local_port, MLXSW_REG_PPLR_LB_TYPE_BIT_PHY_LOCAL : 0); } +/* PMPE - Port Module Plug/Unplug Event Register + * --------------------------------------------- + * This register reports any operational status change of a module. + * A change in the module’s state will generate an event only if the change + * happens after arming the event mechanism. Any changes to the module state + * while the event mechanism is not armed will not be reported. Software can + * query the PMPE register for module status. + */ +#define MLXSW_REG_PMPE_ID 0x5024 +#define MLXSW_REG_PMPE_LEN 0x10 + +MLXSW_REG_DEFINE(pmpe, MLXSW_REG_PMPE_ID, MLXSW_REG_PMPE_LEN); + +/* reg_pmpe_slot_index + * Slot index. + * Access: Index + */ +MLXSW_ITEM32(reg, pmpe, slot_index, 0x00, 24, 4); + +/* reg_pmpe_module + * Module number. + * Access: Index + */ +MLXSW_ITEM32(reg, pmpe, module, 0x00, 16, 8); + +enum mlxsw_reg_pmpe_module_status { + MLXSW_REG_PMPE_MODULE_STATUS_PLUGGED_ENABLED = 1, + MLXSW_REG_PMPE_MODULE_STATUS_UNPLUGGED, + MLXSW_REG_PMPE_MODULE_STATUS_PLUGGED_ERROR, + MLXSW_REG_PMPE_MODULE_STATUS_PLUGGED_DISABLED, +}; + +/* reg_pmpe_module_status + * Module status. + * Access: RO + */ +MLXSW_ITEM32(reg, pmpe, module_status, 0x00, 0, 4); + +/* reg_pmpe_error_type + * Module error details. + * Access: RO + */ +MLXSW_ITEM32(reg, pmpe, error_type, 0x04, 8, 4); + /* PDDR - Port Diagnostics Database Register * ----------------------------------------- * The PDDR enables to read the Phy debug database @@ -5585,6 +5681,9 @@ MLXSW_ITEM32(reg, htgt, type, 0x00, 8, 4); enum mlxsw_reg_htgt_trap_group { MLXSW_REG_HTGT_TRAP_GROUP_EMAD, + MLXSW_REG_HTGT_TRAP_GROUP_MFDE, + MLXSW_REG_HTGT_TRAP_GROUP_MTWE, + MLXSW_REG_HTGT_TRAP_GROUP_PMPE, MLXSW_REG_HTGT_TRAP_GROUP_SP_STP, MLXSW_REG_HTGT_TRAP_GROUP_SP_LACP, MLXSW_REG_HTGT_TRAP_GROUP_SP_LLDP, @@ -8418,6 +8517,13 @@ MLXSW_ITEM32(reg, mtmp, max_temperature, 0x08, 0, 16); * 2 - Generate single event * Access: RW */ + +enum mlxsw_reg_mtmp_tee { + MLXSW_REG_MTMP_TEE_NO_EVENT, + MLXSW_REG_MTMP_TEE_GENERATE_EVENT, + MLXSW_REG_MTMP_TEE_GENERATE_SINGLE_EVENT, +}; + MLXSW_ITEM32(reg, mtmp, tee, 0x0C, 30, 2); #define MLXSW_REG_MTMP_THRESH_HI 0x348 /* 105 Celsius */ @@ -8428,6 +8534,7 @@ MLXSW_ITEM32(reg, mtmp, tee, 0x0C, 30, 2); */ MLXSW_ITEM32(reg, mtmp, temperature_threshold_hi, 0x0C, 0, 16); +#define MLXSW_REG_MTMP_HYSTERESIS_TEMP 0x28 /* 5 Celsius */ /* reg_mtmp_temperature_threshold_lo * Low threshold for Temperature Warning Event. In 0.125 Celsius. * Access: RW @@ -8471,6 +8578,23 @@ static inline void mlxsw_reg_mtmp_unpack(char *payload, int *p_temp, mlxsw_reg_mtmp_sensor_name_memcpy_from(payload, sensor_name); } +/* MTWE - Management Temperature Warning Event + * ------------------------------------------- + * This register is used for over temperature warning. + */ +#define MLXSW_REG_MTWE_ID 0x900B +#define MLXSW_REG_MTWE_LEN 0x10 + +MLXSW_REG_DEFINE(mtwe, MLXSW_REG_MTWE_ID, MLXSW_REG_MTWE_LEN); + +/* reg_mtwe_sensor_warning + * Bit vector indicating which of the sensor reading is above threshold. + * Address 00h bit31 is sensor_warning[127]. + * Address 0Ch bit0 is sensor_warning[0]. + * Access: RO + */ +MLXSW_ITEM_BIT_ARRAY(reg, mtwe, sensor_warning, 0x0, 0x10, 1); + /* MTBR - Management Temperature Bulk Register * ------------------------------------------- * This register is used for bulk temperature reading. @@ -9827,6 +9951,26 @@ static inline void mlxsw_reg_mtptptp_pack(char *payload, mlxsw_reg_mtptpt_message_type_set(payload, message_type); } +/* MFGD - Monitoring FW General Debug Register + * ------------------------------------------- + */ +#define MLXSW_REG_MFGD_ID 0x90F0 +#define MLXSW_REG_MFGD_LEN 0x0C + +MLXSW_REG_DEFINE(mfgd, MLXSW_REG_MFGD_ID, MLXSW_REG_MFGD_LEN); + +/* reg_mfgd_fw_fatal_event_mode + * 0 - don't check FW fatal (default) + * 1 - check FW fatal - enable MFDE trap + * Access: RW + */ +MLXSW_ITEM32(reg, mfgd, fatal_event_mode, 0x00, 9, 2); + +/* reg_mfgd_trigger_test + * Access: WO + */ +MLXSW_ITEM32(reg, mfgd, trigger_test, 0x00, 11, 1); + /* MGPIR - Management General Peripheral Information Register * ---------------------------------------------------------- * MGPIR register allows software to query the hardware and @@ -9886,6 +10030,84 @@ mlxsw_reg_mgpir_unpack(char *payload, u8 *num_of_devices, *num_of_modules = mlxsw_reg_mgpir_num_of_modules_get(payload); } +/* MFDE - Monitoring FW Debug Register + * ----------------------------------- + */ +#define MLXSW_REG_MFDE_ID 0x9200 +#define MLXSW_REG_MFDE_LEN 0x18 + +MLXSW_REG_DEFINE(mfde, MLXSW_REG_MFDE_ID, MLXSW_REG_MFDE_LEN); + +/* reg_mfde_irisc_id + * Which irisc triggered the event + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, irisc_id, 0x00, 8, 4); + +enum mlxsw_reg_mfde_event_id { + MLXSW_REG_MFDE_EVENT_ID_CRSPACE_TO = 1, + /* KVD insertion machine stopped */ + MLXSW_REG_MFDE_EVENT_ID_KVD_IM_STOP, +}; + +/* reg_mfde_event_id + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, event_id, 0x00, 0, 8); + +enum mlxsw_reg_mfde_method { + MLXSW_REG_MFDE_METHOD_QUERY, + MLXSW_REG_MFDE_METHOD_WRITE, +}; + +/* reg_mfde_method + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, method, 0x04, 29, 1); + +/* reg_mfde_long_process + * Indicates if the command is in long_process mode. + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, long_process, 0x04, 28, 1); + +enum mlxsw_reg_mfde_command_type { + MLXSW_REG_MFDE_COMMAND_TYPE_MAD, + MLXSW_REG_MFDE_COMMAND_TYPE_EMAD, + MLXSW_REG_MFDE_COMMAND_TYPE_CMDIF, +}; + +/* reg_mfde_command_type + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, command_type, 0x04, 24, 2); + +/* reg_mfde_reg_attr_id + * EMAD - register id, MAD - attibute id + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, reg_attr_id, 0x04, 0, 16); + +/* reg_mfde_log_address + * crspace address accessed, which resulted in timeout. + * Valid in case event_id == MLXSW_REG_MFDE_EVENT_ID_CRSPACE_TO + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, log_address, 0x10, 0, 32); + +/* reg_mfde_log_id + * Which irisc triggered the timeout. + * Valid in case event_id == MLXSW_REG_MFDE_EVENT_ID_CRSPACE_TO + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, log_id, 0x14, 0, 4); + +/* reg_mfde_pipes_mask + * Bit per kvh pipe. + * Access: RO + */ +MLXSW_ITEM32(reg, mfde, pipes_mask, 0x10, 0, 16); + /* TNGCR - Tunneling NVE General Configuration Register * ---------------------------------------------------- * The TNGCR register is used for setting up the NVE Tunneling configuration. @@ -10948,7 +11170,9 @@ static const struct mlxsw_reg_info *mlxsw_reg_infos[] = { MLXSW_REG(pptb), MLXSW_REG(pbmc), MLXSW_REG(pspa), + MLXSW_REG(pmaos), MLXSW_REG(pplr), + MLXSW_REG(pmpe), MLXSW_REG(pddr), MLXSW_REG(pmtm), MLXSW_REG(htgt), @@ -10978,6 +11202,7 @@ static const struct mlxsw_reg_info *mlxsw_reg_infos[] = { MLXSW_REG(fore), MLXSW_REG(mtcap), MLXSW_REG(mtmp), + MLXSW_REG(mtwe), MLXSW_REG(mtbr), MLXSW_REG(mcia), MLXSW_REG(mpat), @@ -10999,7 +11224,9 @@ static const struct mlxsw_reg_info *mlxsw_reg_infos[] = { MLXSW_REG(mtpppc), MLXSW_REG(mtpptr), MLXSW_REG(mtptpt), + MLXSW_REG(mfgd), MLXSW_REG(mgpir), + MLXSW_REG(mfde), MLXSW_REG(tngcr), MLXSW_REG(tnumt), MLXSW_REG(tnqcr), diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index f3c0e241e1b4..16b47fce540b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -42,11 +42,10 @@ #include "spectrum_span.h" #include "spectrum_ptp.h" #include "spectrum_trap.h" -#include "../mlxfw/mlxfw.h" #define MLXSW_SP1_FWREV_MAJOR 13 -#define MLXSW_SP1_FWREV_MINOR 2007 -#define MLXSW_SP1_FWREV_SUBMINOR 1168 +#define MLXSW_SP1_FWREV_MINOR 2008 +#define MLXSW_SP1_FWREV_SUBMINOR 1310 #define MLXSW_SP1_FWREV_CAN_RESET_MINOR 1702 static const struct mlxsw_fw_rev mlxsw_sp1_fw_rev = { @@ -62,8 +61,8 @@ static const struct mlxsw_fw_rev mlxsw_sp1_fw_rev = { "." __stringify(MLXSW_SP1_FWREV_SUBMINOR) ".mfa2" #define MLXSW_SP2_FWREV_MAJOR 29 -#define MLXSW_SP2_FWREV_MINOR 2007 -#define MLXSW_SP2_FWREV_SUBMINOR 1168 +#define MLXSW_SP2_FWREV_MINOR 2008 +#define MLXSW_SP2_FWREV_SUBMINOR 1310 static const struct mlxsw_fw_rev mlxsw_sp2_fw_rev = { .major = MLXSW_SP2_FWREV_MAJOR, @@ -77,8 +76,8 @@ static const struct mlxsw_fw_rev mlxsw_sp2_fw_rev = { "." __stringify(MLXSW_SP2_FWREV_SUBMINOR) ".mfa2" #define MLXSW_SP3_FWREV_MAJOR 30 -#define MLXSW_SP3_FWREV_MINOR 2007 -#define MLXSW_SP3_FWREV_SUBMINOR 1168 +#define MLXSW_SP3_FWREV_MINOR 2008 +#define MLXSW_SP3_FWREV_SUBMINOR 1310 static const struct mlxsw_fw_rev mlxsw_sp3_fw_rev = { .major = MLXSW_SP3_FWREV_MAJOR, @@ -170,274 +169,6 @@ MLXSW_ITEM32(tx, hdr, fid, 0x08, 0, 16); */ MLXSW_ITEM32(tx, hdr, type, 0x0C, 0, 4); -struct mlxsw_sp_mlxfw_dev { - struct mlxfw_dev mlxfw_dev; - struct mlxsw_sp *mlxsw_sp; -}; - -static int mlxsw_sp_component_query(struct mlxfw_dev *mlxfw_dev, - u16 component_index, u32 *p_max_size, - u8 *p_align_bits, u16 *p_max_write_size) -{ - struct mlxsw_sp_mlxfw_dev *mlxsw_sp_mlxfw_dev = - container_of(mlxfw_dev, struct mlxsw_sp_mlxfw_dev, mlxfw_dev); - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_mlxfw_dev->mlxsw_sp; - char mcqi_pl[MLXSW_REG_MCQI_LEN]; - int err; - - mlxsw_reg_mcqi_pack(mcqi_pl, component_index); - err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(mcqi), mcqi_pl); - if (err) - return err; - mlxsw_reg_mcqi_unpack(mcqi_pl, p_max_size, p_align_bits, - p_max_write_size); - - *p_align_bits = max_t(u8, *p_align_bits, 2); - *p_max_write_size = min_t(u16, *p_max_write_size, - MLXSW_REG_MCDA_MAX_DATA_LEN); - return 0; -} - -static int mlxsw_sp_fsm_lock(struct mlxfw_dev *mlxfw_dev, u32 *fwhandle) -{ - struct mlxsw_sp_mlxfw_dev *mlxsw_sp_mlxfw_dev = - container_of(mlxfw_dev, struct mlxsw_sp_mlxfw_dev, mlxfw_dev); - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_mlxfw_dev->mlxsw_sp; - char mcc_pl[MLXSW_REG_MCC_LEN]; - u8 control_state; - int err; - - mlxsw_reg_mcc_pack(mcc_pl, 0, 0, 0, 0); - err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(mcc), mcc_pl); - if (err) - return err; - - mlxsw_reg_mcc_unpack(mcc_pl, fwhandle, NULL, &control_state); - if (control_state != MLXFW_FSM_STATE_IDLE) - return -EBUSY; - - mlxsw_reg_mcc_pack(mcc_pl, - MLXSW_REG_MCC_INSTRUCTION_LOCK_UPDATE_HANDLE, - 0, *fwhandle, 0); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mcc), mcc_pl); -} - -static int mlxsw_sp_fsm_component_update(struct mlxfw_dev *mlxfw_dev, - u32 fwhandle, u16 component_index, - u32 component_size) -{ - struct mlxsw_sp_mlxfw_dev *mlxsw_sp_mlxfw_dev = - container_of(mlxfw_dev, struct mlxsw_sp_mlxfw_dev, mlxfw_dev); - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_mlxfw_dev->mlxsw_sp; - char mcc_pl[MLXSW_REG_MCC_LEN]; - - mlxsw_reg_mcc_pack(mcc_pl, MLXSW_REG_MCC_INSTRUCTION_UPDATE_COMPONENT, - component_index, fwhandle, component_size); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mcc), mcc_pl); -} - -static int mlxsw_sp_fsm_block_download(struct mlxfw_dev *mlxfw_dev, - u32 fwhandle, u8 *data, u16 size, - u32 offset) -{ - struct mlxsw_sp_mlxfw_dev *mlxsw_sp_mlxfw_dev = - container_of(mlxfw_dev, struct mlxsw_sp_mlxfw_dev, mlxfw_dev); - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_mlxfw_dev->mlxsw_sp; - char mcda_pl[MLXSW_REG_MCDA_LEN]; - - mlxsw_reg_mcda_pack(mcda_pl, fwhandle, offset, size, data); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mcda), mcda_pl); -} - -static int mlxsw_sp_fsm_component_verify(struct mlxfw_dev *mlxfw_dev, - u32 fwhandle, u16 component_index) -{ - struct mlxsw_sp_mlxfw_dev *mlxsw_sp_mlxfw_dev = - container_of(mlxfw_dev, struct mlxsw_sp_mlxfw_dev, mlxfw_dev); - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_mlxfw_dev->mlxsw_sp; - char mcc_pl[MLXSW_REG_MCC_LEN]; - - mlxsw_reg_mcc_pack(mcc_pl, MLXSW_REG_MCC_INSTRUCTION_VERIFY_COMPONENT, - component_index, fwhandle, 0); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mcc), mcc_pl); -} - -static int mlxsw_sp_fsm_activate(struct mlxfw_dev *mlxfw_dev, u32 fwhandle) -{ - struct mlxsw_sp_mlxfw_dev *mlxsw_sp_mlxfw_dev = - container_of(mlxfw_dev, struct mlxsw_sp_mlxfw_dev, mlxfw_dev); - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_mlxfw_dev->mlxsw_sp; - char mcc_pl[MLXSW_REG_MCC_LEN]; - - mlxsw_reg_mcc_pack(mcc_pl, MLXSW_REG_MCC_INSTRUCTION_ACTIVATE, 0, - fwhandle, 0); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mcc), mcc_pl); -} - -static int mlxsw_sp_fsm_query_state(struct mlxfw_dev *mlxfw_dev, u32 fwhandle, - enum mlxfw_fsm_state *fsm_state, - enum mlxfw_fsm_state_err *fsm_state_err) -{ - struct mlxsw_sp_mlxfw_dev *mlxsw_sp_mlxfw_dev = - container_of(mlxfw_dev, struct mlxsw_sp_mlxfw_dev, mlxfw_dev); - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_mlxfw_dev->mlxsw_sp; - char mcc_pl[MLXSW_REG_MCC_LEN]; - u8 control_state; - u8 error_code; - int err; - - mlxsw_reg_mcc_pack(mcc_pl, 0, 0, fwhandle, 0); - err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(mcc), mcc_pl); - if (err) - return err; - - mlxsw_reg_mcc_unpack(mcc_pl, NULL, &error_code, &control_state); - *fsm_state = control_state; - *fsm_state_err = min_t(enum mlxfw_fsm_state_err, error_code, - MLXFW_FSM_STATE_ERR_MAX); - return 0; -} - -static void mlxsw_sp_fsm_cancel(struct mlxfw_dev *mlxfw_dev, u32 fwhandle) -{ - struct mlxsw_sp_mlxfw_dev *mlxsw_sp_mlxfw_dev = - container_of(mlxfw_dev, struct mlxsw_sp_mlxfw_dev, mlxfw_dev); - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_mlxfw_dev->mlxsw_sp; - char mcc_pl[MLXSW_REG_MCC_LEN]; - - mlxsw_reg_mcc_pack(mcc_pl, MLXSW_REG_MCC_INSTRUCTION_CANCEL, 0, - fwhandle, 0); - mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mcc), mcc_pl); -} - -static void mlxsw_sp_fsm_release(struct mlxfw_dev *mlxfw_dev, u32 fwhandle) -{ - struct mlxsw_sp_mlxfw_dev *mlxsw_sp_mlxfw_dev = - container_of(mlxfw_dev, struct mlxsw_sp_mlxfw_dev, mlxfw_dev); - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_mlxfw_dev->mlxsw_sp; - char mcc_pl[MLXSW_REG_MCC_LEN]; - - mlxsw_reg_mcc_pack(mcc_pl, - MLXSW_REG_MCC_INSTRUCTION_RELEASE_UPDATE_HANDLE, 0, - fwhandle, 0); - mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mcc), mcc_pl); -} - -static const struct mlxfw_dev_ops mlxsw_sp_mlxfw_dev_ops = { - .component_query = mlxsw_sp_component_query, - .fsm_lock = mlxsw_sp_fsm_lock, - .fsm_component_update = mlxsw_sp_fsm_component_update, - .fsm_block_download = mlxsw_sp_fsm_block_download, - .fsm_component_verify = mlxsw_sp_fsm_component_verify, - .fsm_activate = mlxsw_sp_fsm_activate, - .fsm_query_state = mlxsw_sp_fsm_query_state, - .fsm_cancel = mlxsw_sp_fsm_cancel, - .fsm_release = mlxsw_sp_fsm_release, -}; - -static int mlxsw_sp_firmware_flash(struct mlxsw_sp *mlxsw_sp, - const struct firmware *firmware, - struct netlink_ext_ack *extack) -{ - struct mlxsw_sp_mlxfw_dev mlxsw_sp_mlxfw_dev = { - .mlxfw_dev = { - .ops = &mlxsw_sp_mlxfw_dev_ops, - .psid = mlxsw_sp->bus_info->psid, - .psid_size = strlen(mlxsw_sp->bus_info->psid), - .devlink = priv_to_devlink(mlxsw_sp->core), - }, - .mlxsw_sp = mlxsw_sp - }; - int err; - - mlxsw_core_fw_flash_start(mlxsw_sp->core); - err = mlxfw_firmware_flash(&mlxsw_sp_mlxfw_dev.mlxfw_dev, - firmware, extack); - mlxsw_core_fw_flash_end(mlxsw_sp->core); - - return err; -} - -static int mlxsw_sp_fw_rev_validate(struct mlxsw_sp *mlxsw_sp) -{ - const struct mlxsw_fw_rev *rev = &mlxsw_sp->bus_info->fw_rev; - const struct mlxsw_fw_rev *req_rev = mlxsw_sp->req_rev; - const char *fw_filename = mlxsw_sp->fw_filename; - union devlink_param_value value; - const struct firmware *firmware; - int err; - - /* Don't check if driver does not require it */ - if (!req_rev || !fw_filename) - return 0; - - /* Don't check if devlink 'fw_load_policy' param is 'flash' */ - err = devlink_param_driverinit_value_get(priv_to_devlink(mlxsw_sp->core), - DEVLINK_PARAM_GENERIC_ID_FW_LOAD_POLICY, - &value); - if (err) - return err; - if (value.vu8 == DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_FLASH) - return 0; - - /* Validate driver & FW are compatible */ - if (rev->major != req_rev->major) { - WARN(1, "Mismatch in major FW version [%d:%d] is never expected; Please contact support\n", - rev->major, req_rev->major); - return -EINVAL; - } - if (mlxsw_core_fw_rev_minor_subminor_validate(rev, req_rev)) - return 0; - - dev_err(mlxsw_sp->bus_info->dev, "The firmware version %d.%d.%d is incompatible with the driver (required >= %d.%d.%d)\n", - rev->major, rev->minor, rev->subminor, req_rev->major, - req_rev->minor, req_rev->subminor); - dev_info(mlxsw_sp->bus_info->dev, "Flashing firmware using file %s\n", - fw_filename); - - err = request_firmware_direct(&firmware, fw_filename, - mlxsw_sp->bus_info->dev); - if (err) { - dev_err(mlxsw_sp->bus_info->dev, "Could not request firmware file %s\n", - fw_filename); - return err; - } - - err = mlxsw_sp_firmware_flash(mlxsw_sp, firmware, NULL); - release_firmware(firmware); - if (err) - dev_err(mlxsw_sp->bus_info->dev, "Could not upgrade firmware\n"); - - /* On FW flash success, tell the caller FW reset is needed - * if current FW supports it. - */ - if (rev->minor >= req_rev->can_reset_minor) - return err ? err : -EAGAIN; - else - return 0; -} - -static int mlxsw_sp_flash_update(struct mlxsw_core *mlxsw_core, - const char *file_name, const char *component, - struct netlink_ext_ack *extack) -{ - struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); - const struct firmware *firmware; - int err; - - if (component) - return -EOPNOTSUPP; - - err = request_firmware_direct(&firmware, file_name, - mlxsw_sp->bus_info->dev); - if (err) - return err; - err = mlxsw_sp_firmware_flash(mlxsw_sp, firmware, extack); - release_firmware(firmware); - - return err; -} - int mlxsw_sp_flow_counter_get(struct mlxsw_sp *mlxsw_sp, unsigned int counter_index, u64 *packets, u64 *bytes) @@ -590,21 +321,28 @@ static int mlxsw_sp_port_dev_addr_init(struct mlxsw_sp_port *mlxsw_sp_port) return mlxsw_sp_port_dev_addr_set(mlxsw_sp_port, addr); } -static int mlxsw_sp_port_mtu_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu) +static int mlxsw_sp_port_max_mtu_get(struct mlxsw_sp_port *mlxsw_sp_port, int *p_max_mtu) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; char pmtu_pl[MLXSW_REG_PMTU_LEN]; - int max_mtu; int err; - mtu += MLXSW_TXHDR_LEN + ETH_HLEN; mlxsw_reg_pmtu_pack(pmtu_pl, mlxsw_sp_port->local_port, 0); err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(pmtu), pmtu_pl); if (err) return err; - max_mtu = mlxsw_reg_pmtu_max_mtu_get(pmtu_pl); - if (mtu > max_mtu) + *p_max_mtu = mlxsw_reg_pmtu_max_mtu_get(pmtu_pl); + return 0; +} + +static int mlxsw_sp_port_mtu_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char pmtu_pl[MLXSW_REG_PMTU_LEN]; + + mtu += MLXSW_TXHDR_LEN + ETH_HLEN; + if (mtu > mlxsw_sp_port->max_mtu) return -EINVAL; mlxsw_reg_pmtu_pack(pmtu_pl, mlxsw_sp_port->local_port, mtu); @@ -872,133 +610,25 @@ static int mlxsw_sp_port_set_mac_address(struct net_device *dev, void *p) return 0; } -static u16 mlxsw_sp_pg_buf_threshold_get(const struct mlxsw_sp *mlxsw_sp, - int mtu) -{ - return 2 * mlxsw_sp_bytes_cells(mlxsw_sp, mtu); -} - -#define MLXSW_SP_CELL_FACTOR 2 /* 2 * cell_size / (IPG + cell_size + 1) */ - -static u16 mlxsw_sp_pfc_delay_get(const struct mlxsw_sp *mlxsw_sp, int mtu, - u16 delay) -{ - delay = mlxsw_sp_bytes_cells(mlxsw_sp, DIV_ROUND_UP(delay, - BITS_PER_BYTE)); - return MLXSW_SP_CELL_FACTOR * delay + mlxsw_sp_bytes_cells(mlxsw_sp, - mtu); -} - -/* Maximum delay buffer needed in case of PAUSE frames, in bytes. - * Assumes 100m cable and maximum MTU. - */ -#define MLXSW_SP_PAUSE_DELAY 58752 - -static u16 mlxsw_sp_pg_buf_delay_get(const struct mlxsw_sp *mlxsw_sp, int mtu, - u16 delay, bool pfc, bool pause) -{ - if (pfc) - return mlxsw_sp_pfc_delay_get(mlxsw_sp, mtu, delay); - else if (pause) - return mlxsw_sp_bytes_cells(mlxsw_sp, MLXSW_SP_PAUSE_DELAY); - else - return 0; -} - -static void mlxsw_sp_pg_buf_pack(char *pbmc_pl, int index, u16 size, u16 thres, - bool lossy) +static int mlxsw_sp_port_change_mtu(struct net_device *dev, int mtu) { - if (lossy) - mlxsw_reg_pbmc_lossy_buffer_pack(pbmc_pl, index, size); - else - mlxsw_reg_pbmc_lossless_buffer_pack(pbmc_pl, index, size, - thres); -} + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct mlxsw_sp_hdroom orig_hdroom; + struct mlxsw_sp_hdroom hdroom; + int err; -int __mlxsw_sp_port_headroom_set(struct mlxsw_sp_port *mlxsw_sp_port, int mtu, - u8 *prio_tc, bool pause_en, - struct ieee_pfc *my_pfc) -{ - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - u8 pfc_en = !!my_pfc ? my_pfc->pfc_en : 0; - u16 delay = !!my_pfc ? my_pfc->delay : 0; - char pbmc_pl[MLXSW_REG_PBMC_LEN]; - u32 taken_headroom_cells = 0; - u32 max_headroom_cells; - int i, j, err; + orig_hdroom = *mlxsw_sp_port->hdroom; - max_headroom_cells = mlxsw_sp_sb_max_headroom_cells(mlxsw_sp); + hdroom = orig_hdroom; + hdroom.mtu = mtu; + mlxsw_sp_hdroom_bufs_reset_sizes(mlxsw_sp_port, &hdroom); - mlxsw_reg_pbmc_pack(pbmc_pl, mlxsw_sp_port->local_port, 0, 0); - err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(pbmc), pbmc_pl); - if (err) + err = mlxsw_sp_hdroom_configure(mlxsw_sp_port, &hdroom); + if (err) { + netdev_err(dev, "Failed to configure port's headroom\n"); return err; - - for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { - bool configure = false; - bool pfc = false; - u16 thres_cells; - u16 delay_cells; - u16 total_cells; - bool lossy; - - for (j = 0; j < IEEE_8021QAZ_MAX_TCS; j++) { - if (prio_tc[j] == i) { - pfc = pfc_en & BIT(j); - configure = true; - break; - } - } - - if (!configure) - continue; - - lossy = !(pfc || pause_en); - thres_cells = mlxsw_sp_pg_buf_threshold_get(mlxsw_sp, mtu); - thres_cells = mlxsw_sp_port_headroom_8x_adjust(mlxsw_sp_port, thres_cells); - delay_cells = mlxsw_sp_pg_buf_delay_get(mlxsw_sp, mtu, delay, - pfc, pause_en); - delay_cells = mlxsw_sp_port_headroom_8x_adjust(mlxsw_sp_port, delay_cells); - total_cells = thres_cells + delay_cells; - - taken_headroom_cells += total_cells; - if (taken_headroom_cells > max_headroom_cells) - return -ENOBUFS; - - mlxsw_sp_pg_buf_pack(pbmc_pl, i, total_cells, - thres_cells, lossy); } - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(pbmc), pbmc_pl); -} - -int mlxsw_sp_port_headroom_set(struct mlxsw_sp_port *mlxsw_sp_port, - int mtu, bool pause_en) -{ - u8 def_prio_tc[IEEE_8021QAZ_MAX_TCS] = {0}; - bool dcb_en = !!mlxsw_sp_port->dcb.ets; - struct ieee_pfc *my_pfc; - u8 *prio_tc; - - prio_tc = dcb_en ? mlxsw_sp_port->dcb.ets->prio_tc : def_prio_tc; - my_pfc = dcb_en ? mlxsw_sp_port->dcb.pfc : NULL; - - return __mlxsw_sp_port_headroom_set(mlxsw_sp_port, mtu, prio_tc, - pause_en, my_pfc); -} - -static int mlxsw_sp_port_change_mtu(struct net_device *dev, int mtu) -{ - struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); - bool pause_en = mlxsw_sp_port_is_pause_en(mlxsw_sp_port); - int err; - - err = mlxsw_sp_port_headroom_set(mlxsw_sp_port, mtu, pause_en); - if (err) - return err; - err = mlxsw_sp_span_port_mtu_update(mlxsw_sp_port, mtu); - if (err) - goto err_span_port_mtu_update; err = mlxsw_sp_port_mtu_set(mlxsw_sp_port, mtu); if (err) goto err_port_mtu_set; @@ -1006,9 +636,7 @@ static int mlxsw_sp_port_change_mtu(struct net_device *dev, int mtu) return 0; err_port_mtu_set: - mlxsw_sp_span_port_mtu_update(mlxsw_sp_port, dev->mtu); -err_span_port_mtu_update: - mlxsw_sp_port_headroom_set(mlxsw_sp_port, dev->mtu, pause_en); + mlxsw_sp_hdroom_configure(mlxsw_sp_port, &orig_hdroom); return err; } @@ -1737,6 +1365,22 @@ static int mlxsw_sp_port_tc_mc_mode_set(struct mlxsw_sp_port *mlxsw_sp_port, return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(qtctm), qtctm_pl); } +static int mlxsw_sp_port_overheat_init_val_set(struct mlxsw_sp_port *mlxsw_sp_port) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + u8 module = mlxsw_sp_port->mapping.module; + u64 overheat_counter; + int err; + + err = mlxsw_env_module_overheat_counter_get(mlxsw_sp->core, module, + &overheat_counter); + if (err) + return err; + + mlxsw_sp_port->module_overheat_initial_val = overheat_counter; + return 0; +} + static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port, u8 split_base_local_port, struct mlxsw_sp_port_mapping *port_mapping) @@ -1842,6 +1486,21 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port, goto err_port_speed_by_width_set; } + err = mlxsw_sp->port_type_speed_ops->ptys_max_speed(mlxsw_sp_port, + &mlxsw_sp_port->max_speed); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to get maximum speed\n", + mlxsw_sp_port->local_port); + goto err_max_speed_get; + } + + err = mlxsw_sp_port_max_mtu_get(mlxsw_sp_port, &mlxsw_sp_port->max_mtu); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to get maximum MTU\n", + mlxsw_sp_port->local_port); + goto err_port_max_mtu_get; + } + err = mlxsw_sp_port_mtu_set(mlxsw_sp_port, ETH_DATA_LEN); if (err) { dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to set MTU\n", @@ -1930,10 +1589,16 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port, INIT_DELAYED_WORK(&mlxsw_sp_port->ptp.shaper_dw, mlxsw_sp->ptp_ops->shaper_work); - INIT_DELAYED_WORK(&mlxsw_sp_port->span.speed_update_dw, - mlxsw_sp_span_speed_update_work); mlxsw_sp->ports[local_port] = mlxsw_sp_port; + + err = mlxsw_sp_port_overheat_init_val_set(mlxsw_sp_port); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to set overheat initial value\n", + mlxsw_sp_port->local_port); + goto err_port_overheat_init_val_set; + } + err = register_netdev(dev); if (err) { dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to register netdev\n", @@ -1947,6 +1612,7 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port, return 0; err_register_netdev: +err_port_overheat_init_val_set: mlxsw_sp->ports[local_port] = NULL; mlxsw_sp_port_vlan_destroy(mlxsw_sp_port_vlan); err_port_vlan_create: @@ -1963,9 +1629,12 @@ err_port_dcb_init: mlxsw_sp_port_tc_mc_mode_set(mlxsw_sp_port, false); err_port_tc_mc_mode: err_port_ets_init: + mlxsw_sp_port_buffers_fini(mlxsw_sp_port); err_port_buffers_init: err_port_admin_status_set: err_port_mtu_set: +err_port_max_mtu_get: +err_max_speed_get: err_port_speed_by_width_set: err_port_system_port_mapping_set: err_dev_addr_init: @@ -1986,7 +1655,6 @@ static void mlxsw_sp_port_remove(struct mlxsw_sp *mlxsw_sp, u8 local_port) struct mlxsw_sp_port *mlxsw_sp_port = mlxsw_sp->ports[local_port]; cancel_delayed_work_sync(&mlxsw_sp_port->periodic_hw_stats.update_dw); - cancel_delayed_work_sync(&mlxsw_sp_port->span.speed_update_dw); cancel_delayed_work_sync(&mlxsw_sp_port->ptp.shaper_dw); mlxsw_sp_port_ptp_clear(mlxsw_sp_port); mlxsw_core_port_clear(mlxsw_sp->core, local_port, mlxsw_sp); @@ -1998,6 +1666,7 @@ static void mlxsw_sp_port_remove(struct mlxsw_sp *mlxsw_sp, u8 local_port) mlxsw_sp_port_fids_fini(mlxsw_sp_port); mlxsw_sp_port_dcb_fini(mlxsw_sp_port); mlxsw_sp_port_tc_mc_mode_set(mlxsw_sp_port, false); + mlxsw_sp_port_buffers_fini(mlxsw_sp_port); mlxsw_sp_port_swid_set(mlxsw_sp_port, MLXSW_PORT_SWID_DISABLED_PORT); mlxsw_sp_port_module_unmap(mlxsw_sp_port); free_percpu(mlxsw_sp_port->pcpu_stats); @@ -2390,7 +2059,6 @@ static void mlxsw_sp_pude_event_func(const struct mlxsw_reg_info *reg, netdev_info(mlxsw_sp_port->dev, "link up\n"); netif_carrier_on(mlxsw_sp_port->dev); mlxsw_core_schedule_dw(&mlxsw_sp_port->ptp.shaper_dw, 0); - mlxsw_core_schedule_dw(&mlxsw_sp_port->span.speed_update_dw, 0); } else { netdev_info(mlxsw_sp_port->dev, "link down\n"); netif_carrier_off(mlxsw_sp_port->dev); @@ -2783,11 +2451,36 @@ static void mlxsw_sp_lag_fini(struct mlxsw_sp *mlxsw_sp) static int mlxsw_sp_basic_trap_groups_set(struct mlxsw_core *mlxsw_core) { char htgt_pl[MLXSW_REG_HTGT_LEN]; + int err; mlxsw_reg_htgt_pack(htgt_pl, MLXSW_REG_HTGT_TRAP_GROUP_EMAD, MLXSW_REG_HTGT_INVALID_POLICER, MLXSW_REG_HTGT_DEFAULT_PRIORITY, MLXSW_REG_HTGT_DEFAULT_TC); + err = mlxsw_reg_write(mlxsw_core, MLXSW_REG(htgt), htgt_pl); + if (err) + return err; + + mlxsw_reg_htgt_pack(htgt_pl, MLXSW_REG_HTGT_TRAP_GROUP_MFDE, + MLXSW_REG_HTGT_INVALID_POLICER, + MLXSW_REG_HTGT_DEFAULT_PRIORITY, + MLXSW_REG_HTGT_DEFAULT_TC); + err = mlxsw_reg_write(mlxsw_core, MLXSW_REG(htgt), htgt_pl); + if (err) + return err; + + mlxsw_reg_htgt_pack(htgt_pl, MLXSW_REG_HTGT_TRAP_GROUP_MTWE, + MLXSW_REG_HTGT_INVALID_POLICER, + MLXSW_REG_HTGT_DEFAULT_PRIORITY, + MLXSW_REG_HTGT_DEFAULT_TC); + err = mlxsw_reg_write(mlxsw_core, MLXSW_REG(htgt), htgt_pl); + if (err) + return err; + + mlxsw_reg_htgt_pack(htgt_pl, MLXSW_REG_HTGT_TRAP_GROUP_PMPE, + MLXSW_REG_HTGT_INVALID_POLICER, + MLXSW_REG_HTGT_DEFAULT_PRIORITY, + MLXSW_REG_HTGT_DEFAULT_TC); return mlxsw_reg_write(mlxsw_core, MLXSW_REG(htgt), htgt_pl); } @@ -2836,10 +2529,6 @@ static int mlxsw_sp_init(struct mlxsw_core *mlxsw_core, mlxsw_sp->core = mlxsw_core; mlxsw_sp->bus_info = mlxsw_bus_info; - err = mlxsw_sp_fw_rev_validate(mlxsw_sp); - if (err) - return err; - mlxsw_core_emad_string_tlv_enable(mlxsw_core); err = mlxsw_sp_base_mac_get(mlxsw_sp); @@ -3039,8 +2728,6 @@ static int mlxsw_sp1_init(struct mlxsw_core *mlxsw_core, { struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); - mlxsw_sp->req_rev = &mlxsw_sp1_fw_rev; - mlxsw_sp->fw_filename = MLXSW_SP1_FW_FILENAME; mlxsw_sp->kvdl_ops = &mlxsw_sp1_kvdl_ops; mlxsw_sp->afa_ops = &mlxsw_sp1_act_afa_ops; mlxsw_sp->afk_ops = &mlxsw_sp1_afk_ops; @@ -3051,6 +2738,7 @@ static int mlxsw_sp1_init(struct mlxsw_core *mlxsw_core, mlxsw_sp->mac_mask = mlxsw_sp1_mac_mask; mlxsw_sp->rif_ops_arr = mlxsw_sp1_rif_ops_arr; mlxsw_sp->sb_vals = &mlxsw_sp1_sb_vals; + mlxsw_sp->sb_ops = &mlxsw_sp1_sb_ops; mlxsw_sp->port_type_speed_ops = &mlxsw_sp1_port_type_speed_ops; mlxsw_sp->ptp_ops = &mlxsw_sp1_ptp_ops; mlxsw_sp->span_ops = &mlxsw_sp1_span_ops; @@ -3069,8 +2757,6 @@ static int mlxsw_sp2_init(struct mlxsw_core *mlxsw_core, { struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); - mlxsw_sp->req_rev = &mlxsw_sp2_fw_rev; - mlxsw_sp->fw_filename = MLXSW_SP2_FW_FILENAME; mlxsw_sp->kvdl_ops = &mlxsw_sp2_kvdl_ops; mlxsw_sp->afa_ops = &mlxsw_sp2_act_afa_ops; mlxsw_sp->afk_ops = &mlxsw_sp2_afk_ops; @@ -3081,6 +2767,7 @@ static int mlxsw_sp2_init(struct mlxsw_core *mlxsw_core, mlxsw_sp->mac_mask = mlxsw_sp2_mac_mask; mlxsw_sp->rif_ops_arr = mlxsw_sp2_rif_ops_arr; mlxsw_sp->sb_vals = &mlxsw_sp2_sb_vals; + mlxsw_sp->sb_ops = &mlxsw_sp2_sb_ops; mlxsw_sp->port_type_speed_ops = &mlxsw_sp2_port_type_speed_ops; mlxsw_sp->ptp_ops = &mlxsw_sp2_ptp_ops; mlxsw_sp->span_ops = &mlxsw_sp2_span_ops; @@ -3097,8 +2784,6 @@ static int mlxsw_sp3_init(struct mlxsw_core *mlxsw_core, { struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); - mlxsw_sp->req_rev = &mlxsw_sp3_fw_rev; - mlxsw_sp->fw_filename = MLXSW_SP3_FW_FILENAME; mlxsw_sp->kvdl_ops = &mlxsw_sp2_kvdl_ops; mlxsw_sp->afa_ops = &mlxsw_sp2_act_afa_ops; mlxsw_sp->afk_ops = &mlxsw_sp2_afk_ops; @@ -3109,6 +2794,7 @@ static int mlxsw_sp3_init(struct mlxsw_core *mlxsw_core, mlxsw_sp->mac_mask = mlxsw_sp2_mac_mask; mlxsw_sp->rif_ops_arr = mlxsw_sp2_rif_ops_arr; mlxsw_sp->sb_vals = &mlxsw_sp2_sb_vals; + mlxsw_sp->sb_ops = &mlxsw_sp3_sb_ops; mlxsw_sp->port_type_speed_ops = &mlxsw_sp2_port_type_speed_ops; mlxsw_sp->ptp_ops = &mlxsw_sp2_ptp_ops; mlxsw_sp->span_ops = &mlxsw_sp3_span_ops; @@ -3451,52 +3137,6 @@ static int mlxsw_sp_kvd_sizes_get(struct mlxsw_core *mlxsw_core, } static int -mlxsw_sp_devlink_param_fw_load_policy_validate(struct devlink *devlink, u32 id, - union devlink_param_value val, - struct netlink_ext_ack *extack) -{ - if ((val.vu8 != DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_DRIVER) && - (val.vu8 != DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_FLASH)) { - NL_SET_ERR_MSG_MOD(extack, "'fw_load_policy' must be 'driver' or 'flash'"); - return -EINVAL; - } - - return 0; -} - -static const struct devlink_param mlxsw_sp_devlink_params[] = { - DEVLINK_PARAM_GENERIC(FW_LOAD_POLICY, - BIT(DEVLINK_PARAM_CMODE_DRIVERINIT), - NULL, NULL, - mlxsw_sp_devlink_param_fw_load_policy_validate), -}; - -static int mlxsw_sp_params_register(struct mlxsw_core *mlxsw_core) -{ - struct devlink *devlink = priv_to_devlink(mlxsw_core); - union devlink_param_value value; - int err; - - err = devlink_params_register(devlink, mlxsw_sp_devlink_params, - ARRAY_SIZE(mlxsw_sp_devlink_params)); - if (err) - return err; - - value.vu8 = DEVLINK_PARAM_FW_LOAD_POLICY_VALUE_DRIVER; - devlink_param_driverinit_value_set(devlink, - DEVLINK_PARAM_GENERIC_ID_FW_LOAD_POLICY, - value); - return 0; -} - -static void mlxsw_sp_params_unregister(struct mlxsw_core *mlxsw_core) -{ - devlink_params_unregister(priv_to_devlink(mlxsw_core), - mlxsw_sp_devlink_params, - ARRAY_SIZE(mlxsw_sp_devlink_params)); -} - -static int mlxsw_sp_params_acl_region_rehash_intrvl_get(struct devlink *devlink, u32 id, struct devlink_param_gset_ctx *ctx) { @@ -3533,24 +3173,16 @@ static int mlxsw_sp2_params_register(struct mlxsw_core *mlxsw_core) union devlink_param_value value; int err; - err = mlxsw_sp_params_register(mlxsw_core); - if (err) - return err; - err = devlink_params_register(devlink, mlxsw_sp2_devlink_params, ARRAY_SIZE(mlxsw_sp2_devlink_params)); if (err) - goto err_devlink_params_register; + return err; value.vu32 = 0; devlink_param_driverinit_value_set(devlink, MLXSW_DEVLINK_PARAM_ID_ACL_REGION_REHASH_INTERVAL, value); return 0; - -err_devlink_params_register: - mlxsw_sp_params_unregister(mlxsw_core); - return err; } static void mlxsw_sp2_params_unregister(struct mlxsw_core *mlxsw_core) @@ -3558,7 +3190,6 @@ static void mlxsw_sp2_params_unregister(struct mlxsw_core *mlxsw_core) devlink_params_unregister(priv_to_devlink(mlxsw_core), mlxsw_sp2_devlink_params, ARRAY_SIZE(mlxsw_sp2_devlink_params)); - mlxsw_sp_params_unregister(mlxsw_core); } static void mlxsw_sp_ptp_transmitted(struct mlxsw_core *mlxsw_core, @@ -3573,6 +3204,8 @@ static void mlxsw_sp_ptp_transmitted(struct mlxsw_core *mlxsw_core, static struct mlxsw_driver mlxsw_sp1_driver = { .kind = mlxsw_sp1_driver_name, .priv_size = sizeof(struct mlxsw_sp), + .fw_req_rev = &mlxsw_sp1_fw_rev, + .fw_filename = MLXSW_SP1_FW_FILENAME, .init = mlxsw_sp1_init, .fini = mlxsw_sp_fini, .basic_trap_groups_set = mlxsw_sp_basic_trap_groups_set, @@ -3588,7 +3221,6 @@ static struct mlxsw_driver mlxsw_sp1_driver = { .sb_occ_max_clear = mlxsw_sp_sb_occ_max_clear, .sb_occ_port_pool_get = mlxsw_sp_sb_occ_port_pool_get, .sb_occ_tc_port_bind_get = mlxsw_sp_sb_occ_tc_port_bind_get, - .flash_update = mlxsw_sp_flash_update, .trap_init = mlxsw_sp_trap_init, .trap_fini = mlxsw_sp_trap_fini, .trap_action_set = mlxsw_sp_trap_action_set, @@ -3601,17 +3233,19 @@ static struct mlxsw_driver mlxsw_sp1_driver = { .txhdr_construct = mlxsw_sp_txhdr_construct, .resources_register = mlxsw_sp1_resources_register, .kvd_sizes_get = mlxsw_sp_kvd_sizes_get, - .params_register = mlxsw_sp_params_register, - .params_unregister = mlxsw_sp_params_unregister, .ptp_transmitted = mlxsw_sp_ptp_transmitted, .txhdr_len = MLXSW_TXHDR_LEN, .profile = &mlxsw_sp1_config_profile, .res_query_enabled = true, + .fw_fatal_enabled = true, + .temp_warn_enabled = true, }; static struct mlxsw_driver mlxsw_sp2_driver = { .kind = mlxsw_sp2_driver_name, .priv_size = sizeof(struct mlxsw_sp), + .fw_req_rev = &mlxsw_sp2_fw_rev, + .fw_filename = MLXSW_SP2_FW_FILENAME, .init = mlxsw_sp2_init, .fini = mlxsw_sp_fini, .basic_trap_groups_set = mlxsw_sp_basic_trap_groups_set, @@ -3627,7 +3261,6 @@ static struct mlxsw_driver mlxsw_sp2_driver = { .sb_occ_max_clear = mlxsw_sp_sb_occ_max_clear, .sb_occ_port_pool_get = mlxsw_sp_sb_occ_port_pool_get, .sb_occ_tc_port_bind_get = mlxsw_sp_sb_occ_tc_port_bind_get, - .flash_update = mlxsw_sp_flash_update, .trap_init = mlxsw_sp_trap_init, .trap_fini = mlxsw_sp_trap_fini, .trap_action_set = mlxsw_sp_trap_action_set, @@ -3645,11 +3278,15 @@ static struct mlxsw_driver mlxsw_sp2_driver = { .txhdr_len = MLXSW_TXHDR_LEN, .profile = &mlxsw_sp2_config_profile, .res_query_enabled = true, + .fw_fatal_enabled = true, + .temp_warn_enabled = true, }; static struct mlxsw_driver mlxsw_sp3_driver = { .kind = mlxsw_sp3_driver_name, .priv_size = sizeof(struct mlxsw_sp), + .fw_req_rev = &mlxsw_sp3_fw_rev, + .fw_filename = MLXSW_SP3_FW_FILENAME, .init = mlxsw_sp3_init, .fini = mlxsw_sp_fini, .basic_trap_groups_set = mlxsw_sp_basic_trap_groups_set, @@ -3665,7 +3302,6 @@ static struct mlxsw_driver mlxsw_sp3_driver = { .sb_occ_max_clear = mlxsw_sp_sb_occ_max_clear, .sb_occ_port_pool_get = mlxsw_sp_sb_occ_port_pool_get, .sb_occ_tc_port_bind_get = mlxsw_sp_sb_occ_tc_port_bind_get, - .flash_update = mlxsw_sp_flash_update, .trap_init = mlxsw_sp_trap_init, .trap_fini = mlxsw_sp_trap_fini, .trap_action_set = mlxsw_sp_trap_action_set, @@ -3683,6 +3319,8 @@ static struct mlxsw_driver mlxsw_sp3_driver = { .txhdr_len = MLXSW_TXHDR_LEN, .profile = &mlxsw_sp2_config_profile, .res_query_enabled = true, + .fw_fatal_enabled = true, + .temp_warn_enabled = true, }; bool mlxsw_sp_port_dev_check(const struct net_device *dev) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 5240bf11b6c4..3e26eb6cb140 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -125,6 +125,7 @@ struct mlxsw_sp_mr_tcam_ops; struct mlxsw_sp_acl_rulei_ops; struct mlxsw_sp_acl_tcam_ops; struct mlxsw_sp_nve_ops; +struct mlxsw_sp_sb_ops; struct mlxsw_sp_sb_vals; struct mlxsw_sp_port_type_speed_ops; struct mlxsw_sp_ptp_state; @@ -162,8 +163,6 @@ struct mlxsw_sp { struct mlxsw_sp_counter_pool *counter_pool; struct mlxsw_sp_span *span; struct mlxsw_sp_trap *trap; - const struct mlxsw_fw_rev *req_rev; - const char *fw_filename; const struct mlxsw_sp_kvdl_ops *kvdl_ops; const struct mlxsw_afa_ops *afa_ops; const struct mlxsw_afk_ops *afk_ops; @@ -173,6 +172,7 @@ struct mlxsw_sp { const struct mlxsw_sp_nve_ops **nve_ops_arr; const struct mlxsw_sp_rif_ops **rif_ops_arr; const struct mlxsw_sp_sb_vals *sb_vals; + const struct mlxsw_sp_sb_ops *sb_ops; const struct mlxsw_sp_port_type_speed_ops *port_type_speed_ops; const struct mlxsw_sp_ptp_ops *ptp_ops; const struct mlxsw_sp_span_ops *span_ops; @@ -316,9 +316,10 @@ struct mlxsw_sp_port { struct mlxsw_sp_ptp_port_stats stats; } ptp; u8 split_base_local_port; - struct { - struct delayed_work speed_update_dw; - } span; + int max_mtu; + u32 max_speed; + struct mlxsw_sp_hdroom *hdroom; + u64 module_overheat_initial_val; }; struct mlxsw_sp_port_type_speed_ops { @@ -331,6 +332,7 @@ struct mlxsw_sp_port_type_speed_ops { void (*from_ptys_speed_duplex)(struct mlxsw_sp *mlxsw_sp, bool carrier_ok, u32 ptys_eth_proto, struct ethtool_link_ksettings *cmd); + int (*ptys_max_speed)(struct mlxsw_sp_port *mlxsw_sp_port, u32 *p_max_speed); u32 (*to_ptys_advert_link)(struct mlxsw_sp *mlxsw_sp, u8 width, const struct ethtool_link_ksettings *cmd); u32 (*to_ptys_speed)(struct mlxsw_sp *mlxsw_sp, u8 width, u32 speed); @@ -414,34 +416,73 @@ mlxsw_sp_port_vlan_find_by_vid(const struct mlxsw_sp_port *mlxsw_sp_port, return NULL; } -static inline u32 -mlxsw_sp_port_headroom_8x_adjust(const struct mlxsw_sp_port *mlxsw_sp_port, - u32 size_cells) -{ - /* Ports with eight lanes use two headroom buffers between which the - * configured headroom size is split. Therefore, multiply the calculated - * headroom size by two. - */ - return mlxsw_sp_port->mapping.width == 8 ? 2 * size_cells : size_cells; -} - enum mlxsw_sp_flood_type { MLXSW_SP_FLOOD_TYPE_UC, MLXSW_SP_FLOOD_TYPE_BC, MLXSW_SP_FLOOD_TYPE_MC, }; -int mlxsw_sp_port_headroom_set(struct mlxsw_sp_port *mlxsw_sp_port, - int mtu, bool pause_en); int mlxsw_sp_port_get_stats_raw(struct net_device *dev, int grp, int prio, char *ppcnt_pl); int mlxsw_sp_port_admin_status_set(struct mlxsw_sp_port *mlxsw_sp_port, bool is_up); /* spectrum_buffers.c */ +struct mlxsw_sp_hdroom_prio { + /* Number of port buffer associated with this priority. This is the + * actually configured value. + */ + u8 buf_idx; + /* Value of buf_idx deduced from the DCB ETS configuration. */ + u8 ets_buf_idx; + /* Value of buf_idx taken from the dcbnl_setbuffer configuration. */ + u8 set_buf_idx; + bool lossy; +}; + +struct mlxsw_sp_hdroom_buf { + u32 thres_cells; + u32 size_cells; + /* Size requirement form dcbnl_setbuffer. */ + u32 set_size_cells; + bool lossy; +}; + +enum mlxsw_sp_hdroom_mode { + MLXSW_SP_HDROOM_MODE_DCB, + MLXSW_SP_HDROOM_MODE_TC, +}; + +#define MLXSW_SP_PB_COUNT 10 + +struct mlxsw_sp_hdroom { + enum mlxsw_sp_hdroom_mode mode; + + struct { + struct mlxsw_sp_hdroom_prio prio[IEEE_8021Q_MAX_PRIORITIES]; + } prios; + struct { + struct mlxsw_sp_hdroom_buf buf[MLXSW_SP_PB_COUNT]; + } bufs; + struct { + /* Size actually configured for the internal buffer. Equal to + * reserve when internal buffer is enabled. + */ + u32 size_cells; + /* Space reserved in the headroom for the internal buffer. Port + * buffers are not allowed to grow into this space. + */ + u32 reserve_cells; + bool enable; + } int_buf; + int delay_bytes; + int mtu; +}; + int mlxsw_sp_buffers_init(struct mlxsw_sp *mlxsw_sp); void mlxsw_sp_buffers_fini(struct mlxsw_sp *mlxsw_sp); int mlxsw_sp_port_buffers_init(struct mlxsw_sp_port *mlxsw_sp_port); +void mlxsw_sp_port_buffers_fini(struct mlxsw_sp_port *mlxsw_sp_port); int mlxsw_sp_sb_pool_get(struct mlxsw_core *mlxsw_core, unsigned int sb_index, u16 pool_index, struct devlink_sb_pool_info *pool_info); @@ -477,11 +518,20 @@ int mlxsw_sp_sb_occ_tc_port_bind_get(struct mlxsw_core_port *mlxsw_core_port, u32 *p_cur, u32 *p_max); u32 mlxsw_sp_cells_bytes(const struct mlxsw_sp *mlxsw_sp, u32 cells); u32 mlxsw_sp_bytes_cells(const struct mlxsw_sp *mlxsw_sp, u32 bytes); -u32 mlxsw_sp_sb_max_headroom_cells(const struct mlxsw_sp *mlxsw_sp); +void mlxsw_sp_hdroom_prios_reset_buf_idx(struct mlxsw_sp_hdroom *hdroom); +void mlxsw_sp_hdroom_bufs_reset_lossiness(struct mlxsw_sp_hdroom *hdroom); +void mlxsw_sp_hdroom_bufs_reset_sizes(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_hdroom *hdroom); +int mlxsw_sp_hdroom_configure(struct mlxsw_sp_port *mlxsw_sp_port, + const struct mlxsw_sp_hdroom *hdroom); extern const struct mlxsw_sp_sb_vals mlxsw_sp1_sb_vals; extern const struct mlxsw_sp_sb_vals mlxsw_sp2_sb_vals; +extern const struct mlxsw_sp_sb_ops mlxsw_sp1_sb_ops; +extern const struct mlxsw_sp_sb_ops mlxsw_sp2_sb_ops; +extern const struct mlxsw_sp_sb_ops mlxsw_sp3_sb_ops; + /* spectrum_switchdev.c */ int mlxsw_sp_switchdev_init(struct mlxsw_sp *mlxsw_sp); void mlxsw_sp_switchdev_fini(struct mlxsw_sp *mlxsw_sp); @@ -519,9 +569,6 @@ int mlxsw_sp_port_ets_set(struct mlxsw_sp_port *mlxsw_sp_port, bool dwrr, u8 dwrr_weight); int mlxsw_sp_port_prio_tc_set(struct mlxsw_sp_port *mlxsw_sp_port, u8 switch_prio, u8 tclass); -int __mlxsw_sp_port_headroom_set(struct mlxsw_sp_port *mlxsw_sp_port, int mtu, - u8 *prio_tc, bool pause_en, - struct ieee_pfc *my_pfc); int mlxsw_sp_port_ets_maxrate_set(struct mlxsw_sp_port *mlxsw_sp_port, enum mlxsw_reg_qeec_hr hr, u8 index, u8 next_index, u32 maxrate, u8 burst_size); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c index 6f84557a5a6f..37ff29a1686e 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c @@ -121,6 +121,10 @@ struct mlxsw_sp_sb_vals { unsigned int cms_cpu_count; }; +struct mlxsw_sp_sb_ops { + u32 (*int_buf_size_get)(int mtu, u32 speed); +}; + u32 mlxsw_sp_cells_bytes(const struct mlxsw_sp *mlxsw_sp, u32 cells) { return mlxsw_sp->sb->cell_size * cells; @@ -131,9 +135,14 @@ u32 mlxsw_sp_bytes_cells(const struct mlxsw_sp *mlxsw_sp, u32 bytes) return DIV_ROUND_UP(bytes, mlxsw_sp->sb->cell_size); } -u32 mlxsw_sp_sb_max_headroom_cells(const struct mlxsw_sp *mlxsw_sp) +static u32 mlxsw_sp_port_headroom_8x_adjust(const struct mlxsw_sp_port *mlxsw_sp_port, + u32 size_cells) { - return mlxsw_sp->sb->max_headroom_cells; + /* Ports with eight lanes use two headroom buffers between which the + * configured headroom size is split. Therefore, multiply the calculated + * headroom size by two. + */ + return mlxsw_sp_port->mapping.width == 8 ? 2 * size_cells : size_cells; } static struct mlxsw_sp_sb_pr *mlxsw_sp_sb_pr_get(struct mlxsw_sp *mlxsw_sp, @@ -291,55 +300,308 @@ static int mlxsw_sp_sb_pm_occ_query(struct mlxsw_sp *mlxsw_sp, u8 local_port, (unsigned long) pm); } -/* 1/4 of a headroom necessary for 100Gbps port and 100m cable. */ -#define MLXSW_SP_PB_HEADROOM 25632 +void mlxsw_sp_hdroom_prios_reset_buf_idx(struct mlxsw_sp_hdroom *hdroom) +{ + int prio; + + for (prio = 0; prio < IEEE_8021QAZ_MAX_TCS; prio++) { + switch (hdroom->mode) { + case MLXSW_SP_HDROOM_MODE_DCB: + hdroom->prios.prio[prio].buf_idx = hdroom->prios.prio[prio].ets_buf_idx; + break; + case MLXSW_SP_HDROOM_MODE_TC: + hdroom->prios.prio[prio].buf_idx = hdroom->prios.prio[prio].set_buf_idx; + break; + } + } +} + +void mlxsw_sp_hdroom_bufs_reset_lossiness(struct mlxsw_sp_hdroom *hdroom) +{ + int prio; + int i; + + for (i = 0; i < DCBX_MAX_BUFFERS; i++) + hdroom->bufs.buf[i].lossy = true; + + for (prio = 0; prio < IEEE_8021Q_MAX_PRIORITIES; prio++) { + if (!hdroom->prios.prio[prio].lossy) + hdroom->bufs.buf[hdroom->prios.prio[prio].buf_idx].lossy = false; + } +} + +static u16 mlxsw_sp_hdroom_buf_threshold_get(const struct mlxsw_sp *mlxsw_sp, int mtu) +{ + return 2 * mlxsw_sp_bytes_cells(mlxsw_sp, mtu); +} + +static void mlxsw_sp_hdroom_buf_pack(char *pbmc_pl, int index, u16 size, u16 thres, bool lossy) +{ + if (lossy) + mlxsw_reg_pbmc_lossy_buffer_pack(pbmc_pl, index, size); + else + mlxsw_reg_pbmc_lossless_buffer_pack(pbmc_pl, index, size, + thres); +} + +static u16 mlxsw_sp_hdroom_buf_delay_get(const struct mlxsw_sp *mlxsw_sp, + const struct mlxsw_sp_hdroom *hdroom) +{ + u16 delay_cells; + + delay_cells = mlxsw_sp_bytes_cells(mlxsw_sp, hdroom->delay_bytes); + + /* In the worst case scenario the delay will be made up of packets that + * are all of size CELL_SIZE + 1, which means each packet will require + * almost twice its true size when buffered in the switch. We therefore + * multiply this value by the "cell factor", which is close to 2. + * + * Another MTU is added in case the transmitting host already started + * transmitting a maximum length frame when the PFC packet was received. + */ + return 2 * delay_cells + mlxsw_sp_bytes_cells(mlxsw_sp, hdroom->mtu); +} + +static u32 mlxsw_sp_hdroom_int_buf_size_get(struct mlxsw_sp *mlxsw_sp, int mtu, u32 speed) +{ + u32 buffsize = mlxsw_sp->sb_ops->int_buf_size_get(speed, mtu); + + return mlxsw_sp_bytes_cells(mlxsw_sp, buffsize) + 1; +} + +static bool mlxsw_sp_hdroom_buf_is_used(const struct mlxsw_sp_hdroom *hdroom, int buf) +{ + int prio; + + for (prio = 0; prio < IEEE_8021QAZ_MAX_TCS; prio++) { + if (hdroom->prios.prio[prio].buf_idx == buf) + return true; + } + return false; +} + +void mlxsw_sp_hdroom_bufs_reset_sizes(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_hdroom *hdroom) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + u16 reserve_cells; + int i; + + /* Internal buffer. */ + reserve_cells = mlxsw_sp_hdroom_int_buf_size_get(mlxsw_sp, mlxsw_sp_port->max_speed, + mlxsw_sp_port->max_mtu); + reserve_cells = mlxsw_sp_port_headroom_8x_adjust(mlxsw_sp_port, reserve_cells); + hdroom->int_buf.reserve_cells = reserve_cells; + + if (hdroom->int_buf.enable) + hdroom->int_buf.size_cells = reserve_cells; + else + hdroom->int_buf.size_cells = 0; + + /* PG buffers. */ + for (i = 0; i < DCBX_MAX_BUFFERS; i++) { + struct mlxsw_sp_hdroom_buf *buf = &hdroom->bufs.buf[i]; + u16 thres_cells; + u16 delay_cells; + + if (!mlxsw_sp_hdroom_buf_is_used(hdroom, i)) { + thres_cells = 0; + delay_cells = 0; + } else if (buf->lossy) { + thres_cells = mlxsw_sp_hdroom_buf_threshold_get(mlxsw_sp, hdroom->mtu); + delay_cells = 0; + } else { + thres_cells = mlxsw_sp_hdroom_buf_threshold_get(mlxsw_sp, hdroom->mtu); + delay_cells = mlxsw_sp_hdroom_buf_delay_get(mlxsw_sp, hdroom); + } + + thres_cells = mlxsw_sp_port_headroom_8x_adjust(mlxsw_sp_port, thres_cells); + delay_cells = mlxsw_sp_port_headroom_8x_adjust(mlxsw_sp_port, delay_cells); + + buf->thres_cells = thres_cells; + if (hdroom->mode == MLXSW_SP_HDROOM_MODE_DCB) { + buf->size_cells = thres_cells + delay_cells; + } else { + /* Do not allow going below the minimum size, even if + * the user requested it. + */ + buf->size_cells = max(buf->set_size_cells, buf->thres_cells); + } + } +} + #define MLXSW_SP_PB_UNUSED 8 -static int mlxsw_sp_port_pb_init(struct mlxsw_sp_port *mlxsw_sp_port) +static int mlxsw_sp_hdroom_configure_buffers(struct mlxsw_sp_port *mlxsw_sp_port, + const struct mlxsw_sp_hdroom *hdroom, bool force) { - const u32 pbs[] = { - [0] = MLXSW_SP_PB_HEADROOM * mlxsw_sp_port->mapping.width, - [9] = MLXSW_PORT_MAX_MTU, - }; struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; char pbmc_pl[MLXSW_REG_PBMC_LEN]; + bool dirty; + int err; int i; - mlxsw_reg_pbmc_pack(pbmc_pl, mlxsw_sp_port->local_port, - 0xffff, 0xffff / 2); - for (i = 0; i < ARRAY_SIZE(pbs); i++) { - u16 size = mlxsw_sp_bytes_cells(mlxsw_sp, pbs[i]); + dirty = memcmp(&mlxsw_sp_port->hdroom->bufs, &hdroom->bufs, sizeof(hdroom->bufs)); + if (!dirty && !force) + return 0; + + mlxsw_reg_pbmc_pack(pbmc_pl, mlxsw_sp_port->local_port, 0xffff, 0xffff / 2); + for (i = 0; i < MLXSW_SP_PB_COUNT; i++) { + const struct mlxsw_sp_hdroom_buf *buf = &hdroom->bufs.buf[i]; if (i == MLXSW_SP_PB_UNUSED) continue; - size = mlxsw_sp_port_headroom_8x_adjust(mlxsw_sp_port, size); - mlxsw_reg_pbmc_lossy_buffer_pack(pbmc_pl, i, size); + + mlxsw_sp_hdroom_buf_pack(pbmc_pl, i, buf->size_cells, buf->thres_cells, buf->lossy); } - mlxsw_reg_pbmc_lossy_buffer_pack(pbmc_pl, - MLXSW_REG_PBMC_PORT_SHARED_BUF_IDX, 0); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(pbmc), pbmc_pl); + + mlxsw_reg_pbmc_lossy_buffer_pack(pbmc_pl, MLXSW_REG_PBMC_PORT_SHARED_BUF_IDX, 0); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(pbmc), pbmc_pl); + if (err) + return err; + + mlxsw_sp_port->hdroom->bufs = hdroom->bufs; + return 0; } -static int mlxsw_sp_port_pb_prio_init(struct mlxsw_sp_port *mlxsw_sp_port) +static int mlxsw_sp_hdroom_configure_priomap(struct mlxsw_sp_port *mlxsw_sp_port, + const struct mlxsw_sp_hdroom *hdroom, bool force) { char pptb_pl[MLXSW_REG_PPTB_LEN]; - int i; + bool dirty; + int prio; + int err; + + dirty = memcmp(&mlxsw_sp_port->hdroom->prios, &hdroom->prios, sizeof(hdroom->prios)); + if (!dirty && !force) + return 0; mlxsw_reg_pptb_pack(pptb_pl, mlxsw_sp_port->local_port); - for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) - mlxsw_reg_pptb_prio_to_buff_pack(pptb_pl, i, 0); - return mlxsw_reg_write(mlxsw_sp_port->mlxsw_sp->core, MLXSW_REG(pptb), - pptb_pl); + for (prio = 0; prio < IEEE_8021QAZ_MAX_TCS; prio++) + mlxsw_reg_pptb_prio_to_buff_pack(pptb_pl, prio, hdroom->prios.prio[prio].buf_idx); + + err = mlxsw_reg_write(mlxsw_sp_port->mlxsw_sp->core, MLXSW_REG(pptb), pptb_pl); + if (err) + return err; + + mlxsw_sp_port->hdroom->prios = hdroom->prios; + return 0; } -static int mlxsw_sp_port_headroom_init(struct mlxsw_sp_port *mlxsw_sp_port) +static int mlxsw_sp_hdroom_configure_int_buf(struct mlxsw_sp_port *mlxsw_sp_port, + const struct mlxsw_sp_hdroom *hdroom, bool force) +{ + char sbib_pl[MLXSW_REG_SBIB_LEN]; + bool dirty; + int err; + + dirty = memcmp(&mlxsw_sp_port->hdroom->int_buf, &hdroom->int_buf, sizeof(hdroom->int_buf)); + if (!dirty && !force) + return 0; + + mlxsw_reg_sbib_pack(sbib_pl, mlxsw_sp_port->local_port, hdroom->int_buf.size_cells); + err = mlxsw_reg_write(mlxsw_sp_port->mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl); + if (err) + return err; + + mlxsw_sp_port->hdroom->int_buf = hdroom->int_buf; + return 0; +} + +static bool mlxsw_sp_hdroom_bufs_fit(struct mlxsw_sp *mlxsw_sp, + const struct mlxsw_sp_hdroom *hdroom) { + u32 taken_headroom_cells = 0; + int i; + + for (i = 0; i < MLXSW_SP_PB_COUNT; i++) + taken_headroom_cells += hdroom->bufs.buf[i].size_cells; + + taken_headroom_cells += hdroom->int_buf.reserve_cells; + return taken_headroom_cells <= mlxsw_sp->sb->max_headroom_cells; +} + +static int __mlxsw_sp_hdroom_configure(struct mlxsw_sp_port *mlxsw_sp_port, + const struct mlxsw_sp_hdroom *hdroom, bool force) +{ + struct mlxsw_sp_hdroom orig_hdroom; + struct mlxsw_sp_hdroom tmp_hdroom; int err; + int i; + + /* Port buffers need to be configured in three steps. First, all buffers + * with non-zero size are configured. Then, prio-to-buffer map is + * updated, allowing traffic to flow to the now non-zero buffers. + * Finally, zero-sized buffers are configured, because now no traffic + * should be directed to them anymore. This way, in a non-congested + * system, no packet drops are introduced by the reconfiguration. + */ - err = mlxsw_sp_port_pb_init(mlxsw_sp_port); + orig_hdroom = *mlxsw_sp_port->hdroom; + tmp_hdroom = orig_hdroom; + for (i = 0; i < MLXSW_SP_PB_COUNT; i++) { + if (hdroom->bufs.buf[i].size_cells) + tmp_hdroom.bufs.buf[i] = hdroom->bufs.buf[i]; + } + + if (!mlxsw_sp_hdroom_bufs_fit(mlxsw_sp_port->mlxsw_sp, &tmp_hdroom) || + !mlxsw_sp_hdroom_bufs_fit(mlxsw_sp_port->mlxsw_sp, hdroom)) + return -ENOBUFS; + + err = mlxsw_sp_hdroom_configure_buffers(mlxsw_sp_port, &tmp_hdroom, force); if (err) return err; - return mlxsw_sp_port_pb_prio_init(mlxsw_sp_port); + + err = mlxsw_sp_hdroom_configure_priomap(mlxsw_sp_port, hdroom, force); + if (err) + goto err_configure_priomap; + + err = mlxsw_sp_hdroom_configure_buffers(mlxsw_sp_port, hdroom, false); + if (err) + goto err_configure_buffers; + + err = mlxsw_sp_hdroom_configure_int_buf(mlxsw_sp_port, hdroom, false); + if (err) + goto err_configure_int_buf; + + *mlxsw_sp_port->hdroom = *hdroom; + return 0; + +err_configure_int_buf: + mlxsw_sp_hdroom_configure_buffers(mlxsw_sp_port, &tmp_hdroom, false); +err_configure_buffers: + mlxsw_sp_hdroom_configure_priomap(mlxsw_sp_port, &tmp_hdroom, false); +err_configure_priomap: + mlxsw_sp_hdroom_configure_buffers(mlxsw_sp_port, &orig_hdroom, false); + return err; +} + +int mlxsw_sp_hdroom_configure(struct mlxsw_sp_port *mlxsw_sp_port, + const struct mlxsw_sp_hdroom *hdroom) +{ + return __mlxsw_sp_hdroom_configure(mlxsw_sp_port, hdroom, false); +} + +static int mlxsw_sp_port_headroom_init(struct mlxsw_sp_port *mlxsw_sp_port) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_hdroom hdroom = {}; + u32 size9; + int prio; + + hdroom.mtu = mlxsw_sp_port->dev->mtu; + hdroom.mode = MLXSW_SP_HDROOM_MODE_DCB; + for (prio = 0; prio < IEEE_8021QAZ_MAX_TCS; prio++) + hdroom.prios.prio[prio].lossy = true; + + mlxsw_sp_hdroom_bufs_reset_lossiness(&hdroom); + mlxsw_sp_hdroom_bufs_reset_sizes(mlxsw_sp_port, &hdroom); + + /* Buffer 9 is used for control traffic. */ + size9 = mlxsw_sp_port_headroom_8x_adjust(mlxsw_sp_port, mlxsw_sp_port->max_mtu); + hdroom.bufs.buf[9].size_cells = mlxsw_sp_bytes_cells(mlxsw_sp, size9); + + return __mlxsw_sp_hdroom_configure(mlxsw_sp_port, &hdroom, true); } static int mlxsw_sp_sb_port_init(struct mlxsw_sp *mlxsw_sp, @@ -916,6 +1178,46 @@ const struct mlxsw_sp_sb_vals mlxsw_sp2_sb_vals = { .cms_cpu_count = ARRAY_SIZE(mlxsw_sp_cpu_port_sb_cms), }; +static u32 mlxsw_sp1_pb_int_buf_size_get(int mtu, u32 speed) +{ + return mtu * 5 / 2; +} + +static u32 __mlxsw_sp_pb_int_buf_size_get(int mtu, u32 speed, u32 buffer_factor) +{ + return 3 * mtu + buffer_factor * speed / 1000; +} + +#define MLXSW_SP2_SPAN_EG_MIRROR_BUFFER_FACTOR 38 + +static u32 mlxsw_sp2_pb_int_buf_size_get(int mtu, u32 speed) +{ + int factor = MLXSW_SP2_SPAN_EG_MIRROR_BUFFER_FACTOR; + + return __mlxsw_sp_pb_int_buf_size_get(mtu, speed, factor); +} + +#define MLXSW_SP3_SPAN_EG_MIRROR_BUFFER_FACTOR 50 + +static u32 mlxsw_sp3_pb_int_buf_size_get(int mtu, u32 speed) +{ + int factor = MLXSW_SP3_SPAN_EG_MIRROR_BUFFER_FACTOR; + + return __mlxsw_sp_pb_int_buf_size_get(mtu, speed, factor); +} + +const struct mlxsw_sp_sb_ops mlxsw_sp1_sb_ops = { + .int_buf_size_get = mlxsw_sp1_pb_int_buf_size_get, +}; + +const struct mlxsw_sp_sb_ops mlxsw_sp2_sb_ops = { + .int_buf_size_get = mlxsw_sp2_pb_int_buf_size_get, +}; + +const struct mlxsw_sp_sb_ops mlxsw_sp3_sb_ops = { + .int_buf_size_get = mlxsw_sp3_pb_int_buf_size_get, +}; + int mlxsw_sp_buffers_init(struct mlxsw_sp *mlxsw_sp) { u32 max_headroom_size; @@ -995,17 +1297,34 @@ int mlxsw_sp_port_buffers_init(struct mlxsw_sp_port *mlxsw_sp_port) { int err; + mlxsw_sp_port->hdroom = kzalloc(sizeof(*mlxsw_sp_port->hdroom), GFP_KERNEL); + if (!mlxsw_sp_port->hdroom) + return -ENOMEM; + mlxsw_sp_port->hdroom->mtu = mlxsw_sp_port->dev->mtu; + err = mlxsw_sp_port_headroom_init(mlxsw_sp_port); if (err) - return err; + goto err_headroom_init; err = mlxsw_sp_port_sb_cms_init(mlxsw_sp_port); if (err) - return err; + goto err_port_sb_cms_init; err = mlxsw_sp_port_sb_pms_init(mlxsw_sp_port); + if (err) + goto err_port_sb_pms_init; + return 0; +err_port_sb_pms_init: +err_port_sb_cms_init: +err_headroom_init: + kfree(mlxsw_sp_port->hdroom); return err; } +void mlxsw_sp_port_buffers_fini(struct mlxsw_sp_port *mlxsw_sp_port) +{ + kfree(mlxsw_sp_port->hdroom); +} + int mlxsw_sp_sb_pool_get(struct mlxsw_core *mlxsw_core, unsigned int sb_index, u16 pool_index, struct devlink_sb_pool_info *pool_info) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c index 0d3fb2e51ea5..5f92b1691360 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c @@ -64,87 +64,28 @@ static int mlxsw_sp_port_ets_validate(struct mlxsw_sp_port *mlxsw_sp_port, return 0; } -static int mlxsw_sp_port_pg_prio_map(struct mlxsw_sp_port *mlxsw_sp_port, - u8 *prio_tc) -{ - char pptb_pl[MLXSW_REG_PPTB_LEN]; - int i; - - mlxsw_reg_pptb_pack(pptb_pl, mlxsw_sp_port->local_port); - for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) - mlxsw_reg_pptb_prio_to_buff_pack(pptb_pl, i, prio_tc[i]); - - return mlxsw_reg_write(mlxsw_sp_port->mlxsw_sp->core, MLXSW_REG(pptb), - pptb_pl); -} - -static bool mlxsw_sp_ets_has_pg(u8 *prio_tc, u8 pg) -{ - int i; - - for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) - if (prio_tc[i] == pg) - return true; - return false; -} - -static int mlxsw_sp_port_pg_destroy(struct mlxsw_sp_port *mlxsw_sp_port, - u8 *old_prio_tc, u8 *new_prio_tc) -{ - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - char pbmc_pl[MLXSW_REG_PBMC_LEN]; - int err, i; - - mlxsw_reg_pbmc_pack(pbmc_pl, mlxsw_sp_port->local_port, 0, 0); - err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(pbmc), pbmc_pl); - if (err) - return err; - - for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { - u8 pg = old_prio_tc[i]; - - if (!mlxsw_sp_ets_has_pg(new_prio_tc, pg)) - mlxsw_reg_pbmc_lossy_buffer_pack(pbmc_pl, pg, 0); - } - - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(pbmc), pbmc_pl); -} - static int mlxsw_sp_port_headroom_ets_set(struct mlxsw_sp_port *mlxsw_sp_port, struct ieee_ets *ets) { - bool pause_en = mlxsw_sp_port_is_pause_en(mlxsw_sp_port); - struct ieee_ets *my_ets = mlxsw_sp_port->dcb.ets; struct net_device *dev = mlxsw_sp_port->dev; + struct mlxsw_sp_hdroom hdroom; + int prio; int err; - /* Create the required PGs, but don't destroy existing ones, as - * traffic is still directed to them. - */ - err = __mlxsw_sp_port_headroom_set(mlxsw_sp_port, dev->mtu, - ets->prio_tc, pause_en, - mlxsw_sp_port->dcb.pfc); + hdroom = *mlxsw_sp_port->hdroom; + for (prio = 0; prio < IEEE_8021QAZ_MAX_TCS; prio++) + hdroom.prios.prio[prio].ets_buf_idx = ets->prio_tc[prio]; + mlxsw_sp_hdroom_prios_reset_buf_idx(&hdroom); + mlxsw_sp_hdroom_bufs_reset_lossiness(&hdroom); + mlxsw_sp_hdroom_bufs_reset_sizes(mlxsw_sp_port, &hdroom); + + err = mlxsw_sp_hdroom_configure(mlxsw_sp_port, &hdroom); if (err) { netdev_err(dev, "Failed to configure port's headroom\n"); return err; } - err = mlxsw_sp_port_pg_prio_map(mlxsw_sp_port, ets->prio_tc); - if (err) { - netdev_err(dev, "Failed to set PG-priority mapping\n"); - goto err_port_prio_pg_map; - } - - err = mlxsw_sp_port_pg_destroy(mlxsw_sp_port, my_ets->prio_tc, - ets->prio_tc); - if (err) - netdev_warn(dev, "Failed to remove unused PGs\n"); - return 0; - -err_port_prio_pg_map: - mlxsw_sp_port_pg_destroy(mlxsw_sp_port, ets->prio_tc, my_ets->prio_tc); - return err; } static int __mlxsw_sp_dcbnl_ieee_setets(struct mlxsw_sp_port *mlxsw_sp_port, @@ -605,6 +546,9 @@ static int mlxsw_sp_dcbnl_ieee_setpfc(struct net_device *dev, { struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); bool pause_en = mlxsw_sp_port_is_pause_en(mlxsw_sp_port); + struct mlxsw_sp_hdroom orig_hdroom; + struct mlxsw_sp_hdroom hdroom; + int prio; int err; if (pause_en && pfc->pfc_en) { @@ -612,9 +556,21 @@ static int mlxsw_sp_dcbnl_ieee_setpfc(struct net_device *dev, return -EINVAL; } - err = __mlxsw_sp_port_headroom_set(mlxsw_sp_port, dev->mtu, - mlxsw_sp_port->dcb.ets->prio_tc, - pause_en, pfc); + orig_hdroom = *mlxsw_sp_port->hdroom; + + hdroom = orig_hdroom; + if (pfc->pfc_en) + hdroom.delay_bytes = DIV_ROUND_UP(pfc->delay, BITS_PER_BYTE); + else + hdroom.delay_bytes = 0; + + for (prio = 0; prio < IEEE_8021QAZ_MAX_TCS; prio++) + hdroom.prios.prio[prio].lossy = !(pfc->pfc_en & BIT(prio)); + + mlxsw_sp_hdroom_bufs_reset_lossiness(&hdroom); + mlxsw_sp_hdroom_bufs_reset_sizes(mlxsw_sp_port, &hdroom); + + err = mlxsw_sp_hdroom_configure(mlxsw_sp_port, &hdroom); if (err) { netdev_err(dev, "Failed to configure port's headroom for PFC\n"); return err; @@ -632,12 +588,66 @@ static int mlxsw_sp_dcbnl_ieee_setpfc(struct net_device *dev, return 0; err_port_pfc_set: - __mlxsw_sp_port_headroom_set(mlxsw_sp_port, dev->mtu, - mlxsw_sp_port->dcb.ets->prio_tc, pause_en, - mlxsw_sp_port->dcb.pfc); + mlxsw_sp_hdroom_configure(mlxsw_sp_port, &orig_hdroom); return err; } +static int mlxsw_sp_dcbnl_getbuffer(struct net_device *dev, struct dcbnl_buffer *buf) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct mlxsw_sp_hdroom *hdroom = mlxsw_sp_port->hdroom; + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + int prio; + int i; + + buf->total_size = 0; + + BUILD_BUG_ON(DCBX_MAX_BUFFERS > MLXSW_SP_PB_COUNT); + for (i = 0; i < MLXSW_SP_PB_COUNT; i++) { + u32 bytes = mlxsw_sp_cells_bytes(mlxsw_sp, hdroom->bufs.buf[i].size_cells); + + if (i < DCBX_MAX_BUFFERS) + buf->buffer_size[i] = bytes; + buf->total_size += bytes; + } + + buf->total_size += mlxsw_sp_cells_bytes(mlxsw_sp, hdroom->int_buf.size_cells); + + for (prio = 0; prio < IEEE_8021Q_MAX_PRIORITIES; prio++) + buf->prio2buffer[prio] = hdroom->prios.prio[prio].buf_idx; + + return 0; +} + +static int mlxsw_sp_dcbnl_setbuffer(struct net_device *dev, struct dcbnl_buffer *buf) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_hdroom hdroom; + int prio; + int i; + + hdroom = *mlxsw_sp_port->hdroom; + + if (hdroom.mode != MLXSW_SP_HDROOM_MODE_TC) { + netdev_err(dev, "The use of dcbnl_setbuffer is only allowed if egress is configured using TC\n"); + return -EINVAL; + } + + for (prio = 0; prio < IEEE_8021Q_MAX_PRIORITIES; prio++) + hdroom.prios.prio[prio].set_buf_idx = buf->prio2buffer[prio]; + + BUILD_BUG_ON(DCBX_MAX_BUFFERS > MLXSW_SP_PB_COUNT); + for (i = 0; i < DCBX_MAX_BUFFERS; i++) + hdroom.bufs.buf[i].set_size_cells = mlxsw_sp_bytes_cells(mlxsw_sp, + buf->buffer_size[i]); + + mlxsw_sp_hdroom_prios_reset_buf_idx(&hdroom); + mlxsw_sp_hdroom_bufs_reset_lossiness(&hdroom); + mlxsw_sp_hdroom_bufs_reset_sizes(mlxsw_sp_port, &hdroom); + return mlxsw_sp_hdroom_configure(mlxsw_sp_port, &hdroom); +} + static const struct dcbnl_rtnl_ops mlxsw_sp_dcbnl_ops = { .ieee_getets = mlxsw_sp_dcbnl_ieee_getets, .ieee_setets = mlxsw_sp_dcbnl_ieee_setets, @@ -650,6 +660,9 @@ static const struct dcbnl_rtnl_ops mlxsw_sp_dcbnl_ops = { .getdcbx = mlxsw_sp_dcbnl_getdcbx, .setdcbx = mlxsw_sp_dcbnl_setdcbx, + + .dcbnl_getbuffer = mlxsw_sp_dcbnl_getbuffer, + .dcbnl_setbuffer = mlxsw_sp_dcbnl_setbuffer, }; static int mlxsw_sp_port_ets_init(struct mlxsw_sp_port *mlxsw_sp_port) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c index 14c78f73bb65..2096b6478958 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c @@ -2,6 +2,7 @@ /* Copyright (c) 2020 Mellanox Technologies. All rights reserved */ #include "reg.h" +#include "core.h" #include "spectrum.h" #include "core_env.h" @@ -192,11 +193,19 @@ static int mlxsw_sp_port_pause_set(struct mlxsw_sp_port *mlxsw_sp_port, pfcc_pl); } +/* Maximum delay buffer needed in case of PAUSE frames. Similar to PFC delay, but is + * measured in bytes. Assumes 100m cable and does not take into account MTU. + */ +#define MLXSW_SP_PAUSE_DELAY_BYTES 19476 + static int mlxsw_sp_port_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam *pause) { struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); bool pause_en = pause->tx_pause || pause->rx_pause; + struct mlxsw_sp_hdroom orig_hdroom; + struct mlxsw_sp_hdroom hdroom; + int prio; int err; if (mlxsw_sp_port->dcb.pfc && mlxsw_sp_port->dcb.pfc->pfc_en) { @@ -209,7 +218,21 @@ static int mlxsw_sp_port_set_pauseparam(struct net_device *dev, return -EINVAL; } - err = mlxsw_sp_port_headroom_set(mlxsw_sp_port, dev->mtu, pause_en); + orig_hdroom = *mlxsw_sp_port->hdroom; + + hdroom = orig_hdroom; + if (pause_en) + hdroom.delay_bytes = MLXSW_SP_PAUSE_DELAY_BYTES; + else + hdroom.delay_bytes = 0; + + for (prio = 0; prio < IEEE_8021QAZ_MAX_TCS; prio++) + hdroom.prios.prio[prio].lossy = !pause_en; + + mlxsw_sp_hdroom_bufs_reset_lossiness(&hdroom); + mlxsw_sp_hdroom_bufs_reset_sizes(mlxsw_sp_port, &hdroom); + + err = mlxsw_sp_hdroom_configure(mlxsw_sp_port, &hdroom); if (err) { netdev_err(dev, "Failed to configure port's headroom\n"); return err; @@ -227,8 +250,7 @@ static int mlxsw_sp_port_set_pauseparam(struct net_device *dev, return 0; err_port_pause_configure: - pause_en = mlxsw_sp_port_is_pause_en(mlxsw_sp_port); - mlxsw_sp_port_headroom_set(mlxsw_sp_port, dev->mtu, pause_en); + mlxsw_sp_hdroom_configure(mlxsw_sp_port, &orig_hdroom); return err; } @@ -531,6 +553,37 @@ static struct mlxsw_sp_port_hw_stats mlxsw_sp_port_hw_tc_stats[] = { #define MLXSW_SP_PORT_HW_TC_STATS_LEN ARRAY_SIZE(mlxsw_sp_port_hw_tc_stats) +struct mlxsw_sp_port_stats { + char str[ETH_GSTRING_LEN]; + u64 (*getter)(struct mlxsw_sp_port *mlxsw_sp_port); +}; + +static u64 +mlxsw_sp_port_get_transceiver_overheat_stats(struct mlxsw_sp_port *mlxsw_sp_port) +{ + struct mlxsw_sp_port_mapping port_mapping = mlxsw_sp_port->mapping; + struct mlxsw_core *mlxsw_core = mlxsw_sp_port->mlxsw_sp->core; + u64 stats; + int err; + + err = mlxsw_env_module_overheat_counter_get(mlxsw_core, + port_mapping.module, + &stats); + if (err) + return mlxsw_sp_port->module_overheat_initial_val; + + return stats - mlxsw_sp_port->module_overheat_initial_val; +} + +static struct mlxsw_sp_port_stats mlxsw_sp_port_transceiver_stats[] = { + { + .str = "transceiver_overheat", + .getter = mlxsw_sp_port_get_transceiver_overheat_stats, + }, +}; + +#define MLXSW_SP_PORT_HW_TRANSCEIVER_STATS_LEN ARRAY_SIZE(mlxsw_sp_port_transceiver_stats) + #define MLXSW_SP_PORT_ETHTOOL_STATS_LEN (MLXSW_SP_PORT_HW_STATS_LEN + \ MLXSW_SP_PORT_HW_RFC_2863_STATS_LEN + \ MLXSW_SP_PORT_HW_RFC_2819_STATS_LEN + \ @@ -540,7 +593,8 @@ static struct mlxsw_sp_port_hw_stats mlxsw_sp_port_hw_tc_stats[] = { (MLXSW_SP_PORT_HW_PRIO_STATS_LEN * \ IEEE_8021QAZ_MAX_TCS) + \ (MLXSW_SP_PORT_HW_TC_STATS_LEN * \ - TC_MAX_QUEUE)) + TC_MAX_QUEUE) + \ + MLXSW_SP_PORT_HW_TRANSCEIVER_STATS_LEN) static void mlxsw_sp_port_get_prio_strings(u8 **p, int prio) { @@ -616,6 +670,12 @@ static void mlxsw_sp_port_get_strings(struct net_device *dev, mlxsw_sp_port_get_tc_strings(&p, i); mlxsw_sp_port->mlxsw_sp->ptp_ops->get_stats_strings(&p); + + for (i = 0; i < MLXSW_SP_PORT_HW_TRANSCEIVER_STATS_LEN; i++) { + memcpy(p, mlxsw_sp_port_transceiver_stats[i].str, + ETH_GSTRING_LEN); + p += ETH_GSTRING_LEN; + } break; } } @@ -711,6 +771,17 @@ static void __mlxsw_sp_port_get_stats(struct net_device *dev, } } +static void __mlxsw_sp_port_get_env_stats(struct net_device *dev, u64 *data, int data_index, + struct mlxsw_sp_port_stats *port_stats, + int len) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + int i; + + for (i = 0; i < len; i++) + data[data_index + i] = port_stats[i].getter(mlxsw_sp_port); +} + static void mlxsw_sp_port_get_stats(struct net_device *dev, struct ethtool_stats *stats, u64 *data) { @@ -765,6 +836,11 @@ static void mlxsw_sp_port_get_stats(struct net_device *dev, mlxsw_sp_port->mlxsw_sp->ptp_ops->get_stats(mlxsw_sp_port, data, data_index); data_index += mlxsw_sp_port->mlxsw_sp->ptp_ops->get_stats_count(); + + /* Transceiver counters */ + __mlxsw_sp_port_get_env_stats(dev, data, data_index, mlxsw_sp_port_transceiver_stats, + MLXSW_SP_PORT_HW_TRANSCEIVER_STATS_LEN); + data_index += MLXSW_SP_PORT_HW_TRANSCEIVER_STATS_LEN; } static int mlxsw_sp_port_get_sset_count(struct net_device *dev, int sset) @@ -842,6 +918,29 @@ mlxsw_sp_port_connector_port(enum mlxsw_reg_ptys_connector_type connector_type) } } +static int mlxsw_sp_port_ptys_query(struct mlxsw_sp_port *mlxsw_sp_port, + u32 *p_eth_proto_cap, u32 *p_eth_proto_admin, + u32 *p_eth_proto_oper, u8 *p_connector_type) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + const struct mlxsw_sp_port_type_speed_ops *ops; + char ptys_pl[MLXSW_REG_PTYS_LEN]; + int err; + + ops = mlxsw_sp->port_type_speed_ops; + + ops->reg_ptys_eth_pack(mlxsw_sp, ptys_pl, mlxsw_sp_port->local_port, 0, false); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl); + if (err) + return err; + + ops->reg_ptys_eth_unpack(mlxsw_sp, ptys_pl, p_eth_proto_cap, p_eth_proto_admin, + p_eth_proto_oper); + if (p_connector_type) + *p_connector_type = mlxsw_reg_ptys_connector_type_get(ptys_pl); + return 0; +} + static int mlxsw_sp_port_get_link_ksettings(struct net_device *dev, struct ethtool_link_ksettings *cmd) { @@ -849,21 +948,17 @@ static int mlxsw_sp_port_get_link_ksettings(struct net_device *dev, struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; const struct mlxsw_sp_port_type_speed_ops *ops; - char ptys_pl[MLXSW_REG_PTYS_LEN]; u8 connector_type; bool autoneg; int err; - ops = mlxsw_sp->port_type_speed_ops; - - autoneg = mlxsw_sp_port->link.autoneg; - ops->reg_ptys_eth_pack(mlxsw_sp, ptys_pl, mlxsw_sp_port->local_port, - 0, false); - err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl); + err = mlxsw_sp_port_ptys_query(mlxsw_sp_port, ð_proto_cap, ð_proto_admin, + ð_proto_oper, &connector_type); if (err) return err; - ops->reg_ptys_eth_unpack(mlxsw_sp, ptys_pl, ð_proto_cap, - ð_proto_admin, ð_proto_oper); + + ops = mlxsw_sp->port_type_speed_ops; + autoneg = mlxsw_sp_port->link.autoneg; mlxsw_sp_port_get_link_supported(mlxsw_sp, eth_proto_cap, mlxsw_sp_port->mapping.width, cmd); @@ -872,7 +967,6 @@ static int mlxsw_sp_port_get_link_ksettings(struct net_device *dev, mlxsw_sp_port->mapping.width, cmd); cmd->base.autoneg = autoneg ? AUTONEG_ENABLE : AUTONEG_DISABLE; - connector_type = mlxsw_reg_ptys_connector_type_get(ptys_pl); cmd->base.port = mlxsw_sp_port_connector_port(connector_type); ops->from_ptys_speed_duplex(mlxsw_sp, netif_carrier_ok(dev), eth_proto_oper, cmd); @@ -993,22 +1087,12 @@ struct mlxsw_sp1_port_link_mode { static const struct mlxsw_sp1_port_link_mode mlxsw_sp1_port_link_mode[] = { { - .mask = MLXSW_REG_PTYS_ETH_SPEED_100BASE_T, - .mask_ethtool = ETHTOOL_LINK_MODE_100baseT_Full_BIT, - .speed = SPEED_100, - }, - { .mask = MLXSW_REG_PTYS_ETH_SPEED_SGMII | MLXSW_REG_PTYS_ETH_SPEED_1000BASE_KX, .mask_ethtool = ETHTOOL_LINK_MODE_1000baseKX_Full_BIT, .speed = SPEED_1000, }, { - .mask = MLXSW_REG_PTYS_ETH_SPEED_10GBASE_T, - .mask_ethtool = ETHTOOL_LINK_MODE_10000baseT_Full_BIT, - .speed = SPEED_10000, - }, - { .mask = MLXSW_REG_PTYS_ETH_SPEED_10GBASE_CX4 | MLXSW_REG_PTYS_ETH_SPEED_10GBASE_KX4, .mask_ethtool = ETHTOOL_LINK_MODE_10000baseKX4_Full_BIT, @@ -1023,11 +1107,6 @@ static const struct mlxsw_sp1_port_link_mode mlxsw_sp1_port_link_mode[] = { .speed = SPEED_10000, }, { - .mask = MLXSW_REG_PTYS_ETH_SPEED_20GBASE_KR2, - .mask_ethtool = ETHTOOL_LINK_MODE_20000baseKR2_Full_BIT, - .speed = SPEED_20000, - }, - { .mask = MLXSW_REG_PTYS_ETH_SPEED_40GBASE_CR4, .mask_ethtool = ETHTOOL_LINK_MODE_40000baseCR4_Full_BIT, .speed = SPEED_40000, @@ -1092,11 +1171,6 @@ static const struct mlxsw_sp1_port_link_mode mlxsw_sp1_port_link_mode[] = { .mask_ethtool = ETHTOOL_LINK_MODE_100000baseKR4_Full_BIT, .speed = SPEED_100000, }, - { - .mask = MLXSW_REG_PTYS_ETH_SPEED_100GBASE_LR4_ER4, - .mask_ethtool = ETHTOOL_LINK_MODE_100000baseLR4_ER4_Full_BIT, - .speed = SPEED_100000, - }, }; #define MLXSW_SP1_PORT_LINK_MODE_LEN ARRAY_SIZE(mlxsw_sp1_port_link_mode) @@ -1164,6 +1238,27 @@ mlxsw_sp1_from_ptys_speed_duplex(struct mlxsw_sp *mlxsw_sp, bool carrier_ok, cmd->base.duplex = DUPLEX_FULL; } +static int mlxsw_sp1_ptys_max_speed(struct mlxsw_sp_port *mlxsw_sp_port, u32 *p_max_speed) +{ + u32 eth_proto_cap; + u32 max_speed = 0; + int err; + int i; + + err = mlxsw_sp_port_ptys_query(mlxsw_sp_port, ð_proto_cap, NULL, NULL, NULL); + if (err) + return err; + + for (i = 0; i < MLXSW_SP1_PORT_LINK_MODE_LEN; i++) { + if ((eth_proto_cap & mlxsw_sp1_port_link_mode[i].mask) && + mlxsw_sp1_port_link_mode[i].speed > max_speed) + max_speed = mlxsw_sp1_port_link_mode[i].speed; + } + + *p_max_speed = max_speed; + return 0; +} + static u32 mlxsw_sp1_to_ptys_advert_link(struct mlxsw_sp *mlxsw_sp, u8 width, const struct ethtool_link_ksettings *cmd) @@ -1213,6 +1308,7 @@ const struct mlxsw_sp_port_type_speed_ops mlxsw_sp1_port_type_speed_ops = { .from_ptys_link = mlxsw_sp1_from_ptys_link, .from_ptys_speed = mlxsw_sp1_from_ptys_speed, .from_ptys_speed_duplex = mlxsw_sp1_from_ptys_speed_duplex, + .ptys_max_speed = mlxsw_sp1_ptys_max_speed, .to_ptys_advert_link = mlxsw_sp1_to_ptys_advert_link, .to_ptys_speed = mlxsw_sp1_to_ptys_speed, .reg_ptys_eth_pack = mlxsw_sp1_reg_ptys_eth_pack, @@ -1237,14 +1333,6 @@ mlxsw_sp2_mask_ethtool_1000base_x_sgmii[] = { ARRAY_SIZE(mlxsw_sp2_mask_ethtool_1000base_x_sgmii) static const enum ethtool_link_mode_bit_indices -mlxsw_sp2_mask_ethtool_2_5gbase_x_2_5gmii[] = { - ETHTOOL_LINK_MODE_2500baseX_Full_BIT, -}; - -#define MLXSW_SP2_MASK_ETHTOOL_2_5GBASE_X_2_5GMII_LEN \ - ARRAY_SIZE(mlxsw_sp2_mask_ethtool_2_5gbase_x_2_5gmii) - -static const enum ethtool_link_mode_bit_indices mlxsw_sp2_mask_ethtool_5gbase_r[] = { ETHTOOL_LINK_MODE_5000baseT_Full_BIT, }; @@ -1408,16 +1496,6 @@ static const struct mlxsw_sp2_port_link_mode mlxsw_sp2_port_link_mode[] = { .speed = SPEED_1000, }, { - .mask = MLXSW_REG_PTYS_EXT_ETH_SPEED_2_5GBASE_X_2_5GMII, - .mask_ethtool = mlxsw_sp2_mask_ethtool_2_5gbase_x_2_5gmii, - .m_ethtool_len = MLXSW_SP2_MASK_ETHTOOL_2_5GBASE_X_2_5GMII_LEN, - .mask_width = MLXSW_SP_PORT_MASK_WIDTH_1X | - MLXSW_SP_PORT_MASK_WIDTH_2X | - MLXSW_SP_PORT_MASK_WIDTH_4X | - MLXSW_SP_PORT_MASK_WIDTH_8X, - .speed = SPEED_2500, - }, - { .mask = MLXSW_REG_PTYS_EXT_ETH_SPEED_5GBASE_R, .mask_ethtool = mlxsw_sp2_mask_ethtool_5gbase_r, .m_ethtool_len = MLXSW_SP2_MASK_ETHTOOL_5GBASE_R_LEN, @@ -1568,6 +1646,27 @@ mlxsw_sp2_from_ptys_speed_duplex(struct mlxsw_sp *mlxsw_sp, bool carrier_ok, cmd->base.duplex = DUPLEX_FULL; } +static int mlxsw_sp2_ptys_max_speed(struct mlxsw_sp_port *mlxsw_sp_port, u32 *p_max_speed) +{ + u32 eth_proto_cap; + u32 max_speed = 0; + int err; + int i; + + err = mlxsw_sp_port_ptys_query(mlxsw_sp_port, ð_proto_cap, NULL, NULL, NULL); + if (err) + return err; + + for (i = 0; i < MLXSW_SP2_PORT_LINK_MODE_LEN; i++) { + if ((eth_proto_cap & mlxsw_sp2_port_link_mode[i].mask) && + mlxsw_sp2_port_link_mode[i].speed > max_speed) + max_speed = mlxsw_sp2_port_link_mode[i].speed; + } + + *p_max_speed = max_speed; + return 0; +} + static bool mlxsw_sp2_test_bit_ethtool(const struct mlxsw_sp2_port_link_mode *link_mode, const unsigned long *mode) @@ -1637,6 +1736,7 @@ const struct mlxsw_sp_port_type_speed_ops mlxsw_sp2_port_type_speed_ops = { .from_ptys_link = mlxsw_sp2_from_ptys_link, .from_ptys_speed = mlxsw_sp2_from_ptys_speed, .from_ptys_speed_duplex = mlxsw_sp2_from_ptys_speed_duplex, + .ptys_max_speed = mlxsw_sp2_ptys_max_speed, .to_ptys_advert_link = mlxsw_sp2_to_ptys_advert_link, .to_ptys_speed = mlxsw_sp2_to_ptys_speed, .reg_ptys_eth_pack = mlxsw_sp2_reg_ptys_eth_pack, diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c index 9650562fc0ef..ca8090a28dec 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c @@ -314,11 +314,9 @@ static int mlxsw_sp_ptp_parse(struct sk_buff *skb, u8 *p_message_type, u16 *p_sequence_id) { - unsigned int offset = 0; unsigned int ptp_class; - u8 *data; + struct ptp_header *hdr; - data = skb_mac_header(skb); ptp_class = ptp_classify_raw(skb); switch (ptp_class & PTP_CLASS_VMASK) { @@ -329,30 +327,14 @@ static int mlxsw_sp_ptp_parse(struct sk_buff *skb, return -ERANGE; } - if (ptp_class & PTP_CLASS_VLAN) - offset += VLAN_HLEN; - - switch (ptp_class & PTP_CLASS_PMASK) { - case PTP_CLASS_IPV4: - offset += ETH_HLEN + IPV4_HLEN(data + offset) + UDP_HLEN; - break; - case PTP_CLASS_IPV6: - offset += ETH_HLEN + IP6_HLEN + UDP_HLEN; - break; - case PTP_CLASS_L2: - offset += ETH_HLEN; - break; - default: - return -ERANGE; - } - - /* PTP header is 34 bytes. */ - if (skb->len < offset + 34) + hdr = ptp_parse_header(skb, ptp_class); + if (!hdr) return -EINVAL; - *p_message_type = data[offset] & 0x0f; - *p_domain_number = data[offset + 4]; - *p_sequence_id = (u16)(data[offset + 30]) << 8 | data[offset + 31]; + *p_message_type = ptp_get_msgtype(hdr, ptp_class); + *p_domain_number = hdr->domain_number; + *p_sequence_id = be16_to_cpu(hdr->sequence_id); + return 0; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c index 964fd444bb10..fd672c6c9133 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c @@ -140,18 +140,31 @@ static int mlxsw_sp_qdisc_destroy(struct mlxsw_sp_port *mlxsw_sp_port, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc) { + struct mlxsw_sp_qdisc *root_qdisc = &mlxsw_sp_port->qdisc->root_qdisc; + int err_hdroom = 0; int err = 0; if (!mlxsw_sp_qdisc) return 0; + if (root_qdisc == mlxsw_sp_qdisc) { + struct mlxsw_sp_hdroom hdroom = *mlxsw_sp_port->hdroom; + + hdroom.mode = MLXSW_SP_HDROOM_MODE_DCB; + mlxsw_sp_hdroom_prios_reset_buf_idx(&hdroom); + mlxsw_sp_hdroom_bufs_reset_lossiness(&hdroom); + mlxsw_sp_hdroom_bufs_reset_sizes(mlxsw_sp_port, &hdroom); + err_hdroom = mlxsw_sp_hdroom_configure(mlxsw_sp_port, &hdroom); + } + if (mlxsw_sp_qdisc->ops && mlxsw_sp_qdisc->ops->destroy) err = mlxsw_sp_qdisc->ops->destroy(mlxsw_sp_port, mlxsw_sp_qdisc); mlxsw_sp_qdisc->handle = TC_H_UNSPEC; mlxsw_sp_qdisc->ops = NULL; - return err; + + return err_hdroom ?: err; } static int @@ -159,6 +172,8 @@ mlxsw_sp_qdisc_replace(struct mlxsw_sp_port *mlxsw_sp_port, u32 handle, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, struct mlxsw_sp_qdisc_ops *ops, void *params) { + struct mlxsw_sp_qdisc *root_qdisc = &mlxsw_sp_port->qdisc->root_qdisc; + struct mlxsw_sp_hdroom orig_hdroom; int err; if (mlxsw_sp_qdisc->ops && mlxsw_sp_qdisc->ops->type != ops->type) @@ -168,6 +183,21 @@ mlxsw_sp_qdisc_replace(struct mlxsw_sp_port *mlxsw_sp_port, u32 handle, * new one. */ mlxsw_sp_qdisc_destroy(mlxsw_sp_port, mlxsw_sp_qdisc); + + orig_hdroom = *mlxsw_sp_port->hdroom; + if (root_qdisc == mlxsw_sp_qdisc) { + struct mlxsw_sp_hdroom hdroom = orig_hdroom; + + hdroom.mode = MLXSW_SP_HDROOM_MODE_TC; + mlxsw_sp_hdroom_prios_reset_buf_idx(&hdroom); + mlxsw_sp_hdroom_bufs_reset_lossiness(&hdroom); + mlxsw_sp_hdroom_bufs_reset_sizes(mlxsw_sp_port, &hdroom); + + err = mlxsw_sp_hdroom_configure(mlxsw_sp_port, &hdroom); + if (err) + goto err_hdroom_configure; + } + err = ops->check_params(mlxsw_sp_port, mlxsw_sp_qdisc, params); if (err) goto err_bad_param; @@ -191,6 +221,8 @@ mlxsw_sp_qdisc_replace(struct mlxsw_sp_port *mlxsw_sp_port, u32 handle, err_bad_param: err_config: + mlxsw_sp_hdroom_configure(mlxsw_sp_port, &orig_hdroom); +err_hdroom_configure: if (mlxsw_sp_qdisc->handle == handle && ops->unoffload) ops->unoffload(mlxsw_sp_port, mlxsw_sp_qdisc, params); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 460cb523312f..4381f8c6c3fb 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -8038,7 +8038,6 @@ static int __mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) bool usp = net->ipv4.sysctl_ip_fwd_update_priority; char rgcr_pl[MLXSW_REG_RGCR_LEN]; u64 max_rifs; - int err; if (!MLXSW_CORE_RES_VALID(mlxsw_sp->core, MAX_RIFS)) return -EIO; @@ -8047,10 +8046,7 @@ static int __mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) mlxsw_reg_rgcr_pack(rgcr_pl, true, true); mlxsw_reg_rgcr_max_router_interfaces_set(rgcr_pl, max_rifs); mlxsw_reg_rgcr_usp_set(rgcr_pl, usp); - err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(rgcr), rgcr_pl); - if (err) - return err; - return 0; + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(rgcr), rgcr_pl); } static void __mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c index 1d18e41ab255..c6c5826aba41 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c @@ -968,42 +968,26 @@ static int mlxsw_sp_span_entry_put(struct mlxsw_sp *mlxsw_sp, return 0; } -static u32 mlxsw_sp_span_buffsize_get(struct mlxsw_sp *mlxsw_sp, int mtu, - u32 speed) +static int mlxsw_sp_span_port_buffer_update(struct mlxsw_sp_port *mlxsw_sp_port, bool enable) { - u32 buffsize = mlxsw_sp->span_ops->buffsize_get(speed, mtu); + struct mlxsw_sp_hdroom hdroom; - return mlxsw_sp_bytes_cells(mlxsw_sp, buffsize) + 1; + hdroom = *mlxsw_sp_port->hdroom; + hdroom.int_buf.enable = enable; + mlxsw_sp_hdroom_bufs_reset_sizes(mlxsw_sp_port, &hdroom); + + return mlxsw_sp_hdroom_configure(mlxsw_sp_port, &hdroom); } static int -mlxsw_sp_span_port_buffer_update(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu) +mlxsw_sp_span_port_buffer_enable(struct mlxsw_sp_port *mlxsw_sp_port) { - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - char sbib_pl[MLXSW_REG_SBIB_LEN]; - u32 buffsize; - u32 speed; - int err; - - err = mlxsw_sp_port_speed_get(mlxsw_sp_port, &speed); - if (err) - return err; - if (speed == SPEED_UNKNOWN) - speed = 0; - - buffsize = mlxsw_sp_span_buffsize_get(mlxsw_sp, speed, mtu); - buffsize = mlxsw_sp_port_headroom_8x_adjust(mlxsw_sp_port, buffsize); - mlxsw_reg_sbib_pack(sbib_pl, mlxsw_sp_port->local_port, buffsize); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl); + return mlxsw_sp_span_port_buffer_update(mlxsw_sp_port, true); } -static void mlxsw_sp_span_port_buffer_disable(struct mlxsw_sp *mlxsw_sp, - u8 local_port) +static void mlxsw_sp_span_port_buffer_disable(struct mlxsw_sp_port *mlxsw_sp_port) { - char sbib_pl[MLXSW_REG_SBIB_LEN]; - - mlxsw_reg_sbib_pack(sbib_pl, local_port, 0); - mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl); + mlxsw_sp_span_port_buffer_update(mlxsw_sp_port, false); } static struct mlxsw_sp_span_analyzed_port * @@ -1021,48 +1005,6 @@ mlxsw_sp_span_analyzed_port_find(struct mlxsw_sp_span *span, u8 local_port, return NULL; } -int mlxsw_sp_span_port_mtu_update(struct mlxsw_sp_port *port, u16 mtu) -{ - struct mlxsw_sp *mlxsw_sp = port->mlxsw_sp; - int err = 0; - - /* If port is egress mirrored, the shared buffer size should be - * updated according to the mtu value - */ - mutex_lock(&mlxsw_sp->span->analyzed_ports_lock); - - if (mlxsw_sp_span_analyzed_port_find(mlxsw_sp->span, port->local_port, - false)) - err = mlxsw_sp_span_port_buffer_update(port, mtu); - - mutex_unlock(&mlxsw_sp->span->analyzed_ports_lock); - - return err; -} - -void mlxsw_sp_span_speed_update_work(struct work_struct *work) -{ - struct delayed_work *dwork = to_delayed_work(work); - struct mlxsw_sp_port *mlxsw_sp_port; - struct mlxsw_sp *mlxsw_sp; - - mlxsw_sp_port = container_of(dwork, struct mlxsw_sp_port, - span.speed_update_dw); - - /* If port is egress mirrored, the shared buffer size should be - * updated according to the speed value. - */ - mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - mutex_lock(&mlxsw_sp->span->analyzed_ports_lock); - - if (mlxsw_sp_span_analyzed_port_find(mlxsw_sp->span, - mlxsw_sp_port->local_port, false)) - mlxsw_sp_span_port_buffer_update(mlxsw_sp_port, - mlxsw_sp_port->dev->mtu); - - mutex_unlock(&mlxsw_sp->span->analyzed_ports_lock); -} - static const struct mlxsw_sp_span_entry_ops * mlxsw_sp_span_entry_ops(struct mlxsw_sp *mlxsw_sp, const struct net_device *to_dev) @@ -1180,9 +1122,7 @@ mlxsw_sp_span_analyzed_port_create(struct mlxsw_sp_span *span, * does the mirroring. */ if (!ingress) { - u16 mtu = mlxsw_sp_port->dev->mtu; - - err = mlxsw_sp_span_port_buffer_update(mlxsw_sp_port, mtu); + err = mlxsw_sp_span_port_buffer_enable(mlxsw_sp_port); if (err) goto err_buffer_update; } @@ -1196,18 +1136,15 @@ err_buffer_update: } static void -mlxsw_sp_span_analyzed_port_destroy(struct mlxsw_sp_span *span, +mlxsw_sp_span_analyzed_port_destroy(struct mlxsw_sp_port *mlxsw_sp_port, struct mlxsw_sp_span_analyzed_port * analyzed_port) { - struct mlxsw_sp *mlxsw_sp = span->mlxsw_sp; - /* Remove egress mirror buffer now that port is no longer analyzed * at egress. */ if (!analyzed_port->ingress) - mlxsw_sp_span_port_buffer_disable(mlxsw_sp, - analyzed_port->local_port); + mlxsw_sp_span_port_buffer_disable(mlxsw_sp_port); list_del(&analyzed_port->list); kfree(analyzed_port); @@ -1258,7 +1195,7 @@ void mlxsw_sp_span_analyzed_port_put(struct mlxsw_sp_port *mlxsw_sp_port, if (!refcount_dec_and_test(&analyzed_port->ref_count)) goto out_unlock; - mlxsw_sp_span_analyzed_port_destroy(mlxsw_sp->span, analyzed_port); + mlxsw_sp_span_analyzed_port_destroy(mlxsw_sp_port, analyzed_port); out_unlock: mutex_unlock(&mlxsw_sp->span->analyzed_ports_lock); @@ -1712,11 +1649,6 @@ static int mlxsw_sp1_span_init(struct mlxsw_sp *mlxsw_sp) return 0; } -static u32 mlxsw_sp1_span_buffsize_get(int mtu, u32 speed) -{ - return mtu * 5 / 2; -} - static int mlxsw_sp1_span_policer_id_base_set(struct mlxsw_sp *mlxsw_sp, u16 policer_id_base) { @@ -1725,7 +1657,6 @@ static int mlxsw_sp1_span_policer_id_base_set(struct mlxsw_sp *mlxsw_sp, const struct mlxsw_sp_span_ops mlxsw_sp1_span_ops = { .init = mlxsw_sp1_span_init, - .buffsize_get = mlxsw_sp1_span_buffsize_get, .policer_id_base_set = mlxsw_sp1_span_policer_id_base_set, }; @@ -1750,18 +1681,6 @@ static int mlxsw_sp2_span_init(struct mlxsw_sp *mlxsw_sp) #define MLXSW_SP2_SPAN_EG_MIRROR_BUFFER_FACTOR 38 #define MLXSW_SP3_SPAN_EG_MIRROR_BUFFER_FACTOR 50 -static u32 __mlxsw_sp_span_buffsize_get(int mtu, u32 speed, u32 buffer_factor) -{ - return 3 * mtu + buffer_factor * speed / 1000; -} - -static u32 mlxsw_sp2_span_buffsize_get(int mtu, u32 speed) -{ - int factor = MLXSW_SP2_SPAN_EG_MIRROR_BUFFER_FACTOR; - - return __mlxsw_sp_span_buffsize_get(mtu, speed, factor); -} - static int mlxsw_sp2_span_policer_id_base_set(struct mlxsw_sp *mlxsw_sp, u16 policer_id_base) { @@ -1778,19 +1697,10 @@ static int mlxsw_sp2_span_policer_id_base_set(struct mlxsw_sp *mlxsw_sp, const struct mlxsw_sp_span_ops mlxsw_sp2_span_ops = { .init = mlxsw_sp2_span_init, - .buffsize_get = mlxsw_sp2_span_buffsize_get, .policer_id_base_set = mlxsw_sp2_span_policer_id_base_set, }; -static u32 mlxsw_sp3_span_buffsize_get(int mtu, u32 speed) -{ - int factor = MLXSW_SP3_SPAN_EG_MIRROR_BUFFER_FACTOR; - - return __mlxsw_sp_span_buffsize_get(mtu, speed, factor); -} - const struct mlxsw_sp_span_ops mlxsw_sp3_span_ops = { .init = mlxsw_sp2_span_init, - .buffsize_get = mlxsw_sp3_span_buffsize_get, .policer_id_base_set = mlxsw_sp2_span_policer_id_base_set, }; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h index 1c746dd3b1bd..d907718bc8c5 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h @@ -47,7 +47,6 @@ struct mlxsw_sp_span_entry_ops; struct mlxsw_sp_span_ops { int (*init)(struct mlxsw_sp *mlxsw_sp); - u32 (*buffsize_get)(int mtu, u32 speed); int (*policer_id_base_set)(struct mlxsw_sp *mlxsw_sp, u16 policer_id_base); }; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c index 2e41c5519c1b..433f14ade464 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c @@ -291,7 +291,7 @@ static void mlxsw_sp_rx_sample_listener(struct sk_buff *skb, u8 local_port, static const struct mlxsw_sp_trap_policer_item mlxsw_sp_trap_policer_items_arr[] = { { - .policer = MLXSW_SP_TRAP_POLICER(1, 10 * 1024, 128), + .policer = MLXSW_SP_TRAP_POLICER(1, 10 * 1024, 4096), }, { .policer = MLXSW_SP_TRAP_POLICER(2, 128, 128), @@ -303,25 +303,25 @@ mlxsw_sp_trap_policer_items_arr[] = { .policer = MLXSW_SP_TRAP_POLICER(4, 128, 128), }, { - .policer = MLXSW_SP_TRAP_POLICER(5, 16 * 1024, 128), + .policer = MLXSW_SP_TRAP_POLICER(5, 16 * 1024, 8192), }, { .policer = MLXSW_SP_TRAP_POLICER(6, 128, 128), }, { - .policer = MLXSW_SP_TRAP_POLICER(7, 1024, 128), + .policer = MLXSW_SP_TRAP_POLICER(7, 1024, 512), }, { - .policer = MLXSW_SP_TRAP_POLICER(8, 20 * 1024, 1024), + .policer = MLXSW_SP_TRAP_POLICER(8, 20 * 1024, 8192), }, { .policer = MLXSW_SP_TRAP_POLICER(9, 128, 128), }, { - .policer = MLXSW_SP_TRAP_POLICER(10, 1024, 128), + .policer = MLXSW_SP_TRAP_POLICER(10, 1024, 512), }, { - .policer = MLXSW_SP_TRAP_POLICER(11, 360, 128), + .policer = MLXSW_SP_TRAP_POLICER(11, 256, 128), }, { .policer = MLXSW_SP_TRAP_POLICER(12, 128, 128), @@ -330,19 +330,19 @@ mlxsw_sp_trap_policer_items_arr[] = { .policer = MLXSW_SP_TRAP_POLICER(13, 128, 128), }, { - .policer = MLXSW_SP_TRAP_POLICER(14, 1024, 128), + .policer = MLXSW_SP_TRAP_POLICER(14, 1024, 512), }, { - .policer = MLXSW_SP_TRAP_POLICER(15, 1024, 128), + .policer = MLXSW_SP_TRAP_POLICER(15, 1024, 512), }, { - .policer = MLXSW_SP_TRAP_POLICER(16, 24 * 1024, 4096), + .policer = MLXSW_SP_TRAP_POLICER(16, 24 * 1024, 16384), }, { - .policer = MLXSW_SP_TRAP_POLICER(17, 19 * 1024, 4096), + .policer = MLXSW_SP_TRAP_POLICER(17, 19 * 1024, 8192), }, { - .policer = MLXSW_SP_TRAP_POLICER(18, 1024, 128), + .policer = MLXSW_SP_TRAP_POLICER(18, 1024, 512), }, { .policer = MLXSW_SP_TRAP_POLICER(19, 1024, 512), diff --git a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c index 6f9a725662fb..5023d91269f4 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c +++ b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c @@ -551,16 +551,6 @@ struct mlxsw_sx_port_link_mode { static const struct mlxsw_sx_port_link_mode mlxsw_sx_port_link_mode[] = { { - .mask = MLXSW_REG_PTYS_ETH_SPEED_100BASE_T, - .supported = SUPPORTED_100baseT_Full, - .advertised = ADVERTISED_100baseT_Full, - .speed = 100, - }, - { - .mask = MLXSW_REG_PTYS_ETH_SPEED_100BASE_TX, - .speed = 100, - }, - { .mask = MLXSW_REG_PTYS_ETH_SPEED_SGMII | MLXSW_REG_PTYS_ETH_SPEED_1000BASE_KX, .supported = SUPPORTED_1000baseKX_Full, @@ -568,12 +558,6 @@ static const struct mlxsw_sx_port_link_mode mlxsw_sx_port_link_mode[] = { .speed = 1000, }, { - .mask = MLXSW_REG_PTYS_ETH_SPEED_10GBASE_T, - .supported = SUPPORTED_10000baseT_Full, - .advertised = ADVERTISED_10000baseT_Full, - .speed = 10000, - }, - { .mask = MLXSW_REG_PTYS_ETH_SPEED_10GBASE_CX4 | MLXSW_REG_PTYS_ETH_SPEED_10GBASE_KX4, .supported = SUPPORTED_10000baseKX4_Full, @@ -590,12 +574,6 @@ static const struct mlxsw_sx_port_link_mode mlxsw_sx_port_link_mode[] = { .speed = 10000, }, { - .mask = MLXSW_REG_PTYS_ETH_SPEED_20GBASE_KR2, - .supported = SUPPORTED_20000baseKR2_Full, - .advertised = ADVERTISED_20000baseKR2_Full, - .speed = 20000, - }, - { .mask = MLXSW_REG_PTYS_ETH_SPEED_40GBASE_CR4, .supported = SUPPORTED_40000baseCR4_Full, .advertised = ADVERTISED_40000baseCR4_Full, @@ -634,8 +612,7 @@ static const struct mlxsw_sx_port_link_mode mlxsw_sx_port_link_mode[] = { { .mask = MLXSW_REG_PTYS_ETH_SPEED_100GBASE_CR4 | MLXSW_REG_PTYS_ETH_SPEED_100GBASE_SR4 | - MLXSW_REG_PTYS_ETH_SPEED_100GBASE_KR4 | - MLXSW_REG_PTYS_ETH_SPEED_100GBASE_LR4_ER4, + MLXSW_REG_PTYS_ETH_SPEED_100GBASE_KR4, .speed = 100000, }, }; diff --git a/drivers/net/ethernet/mellanox/mlxsw/trap.h b/drivers/net/ethernet/mellanox/mlxsw/trap.h index 33909887d0ac..57f9e24602d0 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/trap.h +++ b/drivers/net/ethernet/mellanox/mlxsw/trap.h @@ -120,8 +120,14 @@ enum { }; enum mlxsw_event_trap_id { + /* Fatal Event generated by FW */ + MLXSW_TRAP_ID_MFDE = 0x3, /* Port Up/Down event generated by hardware */ MLXSW_TRAP_ID_PUDE = 0x8, + /* Port Module Plug/Unplug Event generated by hardware */ + MLXSW_TRAP_ID_PMPE = 0x9, + /* Temperature Warning event generated by hardware */ + MLXSW_TRAP_ID_MTWE = 0xC, /* PTP Ingress FIFO has a new entry */ MLXSW_TRAP_ID_PTP_ING_FIFO = 0x2D, /* PTP Egress FIFO has a new entry */ diff --git a/drivers/net/ethernet/micrel/ks8842.c b/drivers/net/ethernet/micrel/ks8842.c index f3f6dfe3eddc..caa251d0e381 100644 --- a/drivers/net/ethernet/micrel/ks8842.c +++ b/drivers/net/ethernet/micrel/ks8842.c @@ -587,10 +587,10 @@ out: return err; } -static void ks8842_rx_frame_dma_tasklet(unsigned long arg) +static void ks8842_rx_frame_dma_tasklet(struct tasklet_struct *t) { - struct net_device *netdev = (struct net_device *)arg; - struct ks8842_adapter *adapter = netdev_priv(netdev); + struct ks8842_adapter *adapter = from_tasklet(adapter, t, dma_rx.tasklet); + struct net_device *netdev = adapter->netdev; struct ks8842_rx_dma_ctl *ctl = &adapter->dma_rx; struct sk_buff *skb = ctl->skb; dma_addr_t addr = sg_dma_address(&ctl->sg); @@ -720,10 +720,10 @@ static void ks8842_handle_rx_overrun(struct net_device *netdev, netdev->stats.rx_fifo_errors++; } -static void ks8842_tasklet(unsigned long arg) +static void ks8842_tasklet(struct tasklet_struct *t) { - struct net_device *netdev = (struct net_device *)arg; - struct ks8842_adapter *adapter = netdev_priv(netdev); + struct ks8842_adapter *adapter = from_tasklet(adapter, t, tasklet); + struct net_device *netdev = adapter->netdev; u16 isr; unsigned long flags; u16 entry_bank; @@ -953,8 +953,7 @@ static int ks8842_alloc_dma_bufs(struct net_device *netdev) goto err; } - tasklet_init(&rx_ctl->tasklet, ks8842_rx_frame_dma_tasklet, - (unsigned long)netdev); + tasklet_setup(&rx_ctl->tasklet, ks8842_rx_frame_dma_tasklet); return 0; err: @@ -1173,7 +1172,7 @@ static int ks8842_probe(struct platform_device *pdev) adapter->dma_tx.channel = -1; } - tasklet_init(&adapter->tasklet, ks8842_tasklet, (unsigned long)netdev); + tasklet_setup(&adapter->tasklet, ks8842_tasklet); spin_lock_init(&adapter->lock); netdev->netdev_ops = &ks8842_netdev_ops; diff --git a/drivers/net/ethernet/micrel/ksz884x.c b/drivers/net/ethernet/micrel/ksz884x.c index bb646b65cc95..9ed264ed7070 100644 --- a/drivers/net/ethernet/micrel/ksz884x.c +++ b/drivers/net/ethernet/micrel/ksz884x.c @@ -1,5 +1,5 @@ // SPDX-License-Identifier: GPL-2.0-only -/** +/* * drivers/net/ethernet/micrel/ksx884x.c - Micrel KSZ8841/2 PCI Ethernet driver * * Copyright (c) 2009-2010 Micrel, Inc. @@ -959,7 +959,7 @@ struct ksz_sw_desc { * struct ksz_dma_buf - OS dependent DMA buffer data structure * @skb: Associated socket buffer. * @dma: Associated physical DMA address. - * len: Actual len used. + * @len: Actual len used. */ struct ksz_dma_buf { struct sk_buff *skb; @@ -1254,6 +1254,7 @@ struct ksz_port_info { * @multi_list_size: Multicast address list size. * @enabled: Indication of hardware enabled. * @rx_stop: Indication of receive process stop. + * @reserved2: none * @features: Hardware features to enable. * @overrides: Hardware features to override. * @parent: Pointer to parent, network device private structure. @@ -1447,7 +1448,7 @@ struct dev_info { * struct dev_priv - Network device private data structure * @adapter: Adapter device information. * @port: Port information. - * @monitor_time_info: Timer to monitor ports. + * @monitor_timer_info: Timer to monitor ports. * @proc_sem: Semaphore for proc accessing. * @id: Device ID. * @mii_if: MII interface information. @@ -1566,6 +1567,7 @@ static inline void hw_restore_intr(struct ksz_hw *hw, uint interrupt) /** * hw_block_intr - block hardware interrupts + * @hw: The hardware instance. * * This function blocks all interrupts of the hardware and returns the current * interrupt enable mask so that interrupts can be restored later. @@ -1649,8 +1651,7 @@ static inline void set_tx_len(struct ksz_desc *desc, u32 len) #define HW_DELAY(hw, reg) \ do { \ - u16 dummy; \ - dummy = readw(hw->io + reg); \ + readw(hw->io + reg); \ } while (0) /** @@ -1819,6 +1820,7 @@ static void port_r_mib_cnt(struct ksz_hw *hw, int port, u16 addr, u64 *cnt) * port_r_mib_pkt - read dropped packet counts * @hw: The hardware instance. * @port: The port index. + * @last: last one * @cnt: Buffer to store the receive and transmit dropped packet counts. * * This routine reads the dropped packet counts of the port. @@ -1972,7 +1974,7 @@ static void port_cfg(struct ksz_hw *hw, int port, int offset, u16 bits, * port_chk_shift - check port bit * @hw: The hardware instance. * @port: The port index. - * @offset: The offset of the register. + * @addr: The offset of the register. * @shift: Number of bits to shift. * * This function checks whether the specified port is set in the register or @@ -1994,7 +1996,7 @@ static int port_chk_shift(struct ksz_hw *hw, int port, u32 addr, int shift) * port_cfg_shift - set port bit * @hw: The hardware instance. * @port: The port index. - * @offset: The offset of the register. + * @addr: The offset of the register. * @shift: Number of bits to shift. * @set: The flag indicating whether the port is to be set or not. * @@ -4425,6 +4427,8 @@ static int ksz_alloc_desc(struct dev_info *adapter) /** * free_dma_buf - release DMA buffer resources * @adapter: Adapter information structure. + * @dma_buf: pointer to buf + * @direction: to or from device * * This routine is just a helper function to release the DMA buffer resources. */ @@ -4562,6 +4566,7 @@ static void ksz_free_desc(struct dev_info *adapter) * ksz_free_buffers - free buffers used in the descriptors * @adapter: Adapter information structure. * @desc_info: Descriptor information structure. + * @direction: to or from device * * This local routine frees buffers used in the DMA buffers. */ @@ -4721,7 +4726,8 @@ static void send_packet(struct sk_buff *skb, struct net_device *dev) /** * transmit_cleanup - clean up transmit descriptors - * @dev: Network device. + * @hw_priv: Network device. + * @normal: break if owned * * This routine is called to clean up the transmitted buffers. */ @@ -4777,7 +4783,7 @@ static void transmit_cleanup(struct dev_info *hw_priv, int normal) /** * transmit_done - transmit done processing - * @dev: Network device. + * @hw_priv: Network device. * * This routine is called when the transmit interrupt is triggered, indicating * either a packet is sent successfully or there are transmit errors. @@ -4883,6 +4889,7 @@ unlock: /** * netdev_tx_timeout - transmit timeout processing * @dev: Network device. + * @txqueue: index of hanging queue * * This routine is called when the transmit timer expires. That indicates the * hardware is not running correctly because transmit interrupts are not @@ -4978,7 +4985,6 @@ static inline int rx_proc(struct net_device *dev, struct ksz_hw* hw, struct dev_info *hw_priv = priv->adapter; struct ksz_dma_buf *dma_buf; struct sk_buff *skb; - int rx_status; /* Received length includes 4-byte CRC. */ packet_len = status.rx.frame_len - 4; @@ -5014,7 +5020,7 @@ static inline int rx_proc(struct net_device *dev, struct ksz_hw* hw, dev->stats.rx_bytes += packet_len; /* Notify upper layer for received packet. */ - rx_status = netif_rx(skb); + netif_rx(skb); return 0; } @@ -5159,9 +5165,9 @@ release_packet: return received; } -static void rx_proc_task(unsigned long data) +static void rx_proc_task(struct tasklet_struct *t) { - struct dev_info *hw_priv = (struct dev_info *) data; + struct dev_info *hw_priv = from_tasklet(hw_priv, t, rx_tasklet); struct ksz_hw *hw = &hw_priv->hw; if (!hw->enabled) @@ -5181,9 +5187,9 @@ static void rx_proc_task(unsigned long data) } } -static void tx_proc_task(unsigned long data) +static void tx_proc_task(struct tasklet_struct *t) { - struct dev_info *hw_priv = (struct dev_info *) data; + struct dev_info *hw_priv = from_tasklet(hw_priv, t, tx_tasklet); struct ksz_hw *hw = &hw_priv->hw; hw_ack_intr(hw, KS884X_INT_TX_MASK); @@ -5436,10 +5442,8 @@ static int prepare_hardware(struct net_device *dev) rc = request_irq(dev->irq, netdev_intr, IRQF_SHARED, dev->name, dev); if (rc) return rc; - tasklet_init(&hw_priv->rx_tasklet, rx_proc_task, - (unsigned long) hw_priv); - tasklet_init(&hw_priv->tx_tasklet, tx_proc_task, - (unsigned long) hw_priv); + tasklet_setup(&hw_priv->rx_tasklet, rx_proc_task); + tasklet_setup(&hw_priv->tx_tasklet, tx_proc_task); hw->promiscuous = 0; hw->all_multi = 0; @@ -5829,8 +5833,7 @@ static int netdev_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) /* Get address of MII PHY in use. */ case SIOCGMIIPHY: data->phy_id = priv->id; - - /* Fallthrough... */ + fallthrough; /* Read MII PHY register. */ case SIOCGMIIREG: @@ -6078,14 +6081,6 @@ static void netdev_get_drvinfo(struct net_device *dev, sizeof(info->bus_info)); } -/** - * netdev_get_regs_len - get length of register dump - * @dev: Network device. - * - * This function returns the length of the register dump. - * - * Return length of the register dump. - */ static struct hw_regs { int start; int end; @@ -6099,6 +6094,14 @@ static struct hw_regs { { 0, 0 } }; +/** + * netdev_get_regs_len - get length of register dump + * @dev: Network device. + * + * This function returns the length of the register dump. + * + * Return length of the register dump. + */ static int netdev_get_regs_len(struct net_device *dev) { struct hw_regs *range = hw_regs_range; @@ -6240,6 +6243,8 @@ static int netdev_get_eeprom_len(struct net_device *dev) return EEPROM_SIZE * 2; } +#define EEPROM_MAGIC 0x10A18842 + /** * netdev_get_eeprom - get EEPROM data * @dev: Network device. @@ -6250,8 +6255,6 @@ static int netdev_get_eeprom_len(struct net_device *dev) * * Return 0 if successful; otherwise an error code. */ -#define EEPROM_MAGIC 0x10A18842 - static int netdev_get_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom, u8 *data) { @@ -6388,7 +6391,7 @@ static int netdev_set_pauseparam(struct net_device *dev, /** * netdev_get_ringparam - get tx/rx ring parameters * @dev: Network device. - * @pause: Ethtool RING settings data structure. + * @ring: Ethtool RING settings data structure. * * This procedure returns the TX/RX ring settings. */ @@ -6509,7 +6512,6 @@ static void netdev_get_ethtool_stats(struct net_device *dev, int i; int n; int p; - int rc; u64 counter[TOTAL_PORT_COUNTER_NUM]; mutex_lock(&hw_priv->lock); @@ -6530,19 +6532,19 @@ static void netdev_get_ethtool_stats(struct net_device *dev, if (1 == port->mib_port_cnt && n < SWITCH_PORT_NUM) { p = n; - rc = wait_event_interruptible_timeout( + wait_event_interruptible_timeout( hw_priv->counter[p].counter, 2 == hw_priv->counter[p].read, HZ * 1); } else for (i = 0, p = n; i < port->mib_port_cnt - n; i++, p++) { if (0 == i) { - rc = wait_event_interruptible_timeout( + wait_event_interruptible_timeout( hw_priv->counter[p].counter, 2 == hw_priv->counter[p].read, HZ * 2); } else if (hw->port_mib[p].cnt_ptr) { - rc = wait_event_interruptible_timeout( + wait_event_interruptible_timeout( hw_priv->counter[p].counter, 2 == hw_priv->counter[p].read, HZ * 1); @@ -6693,7 +6695,7 @@ static void mib_monitor(struct timer_list *t) /** * dev_monitor - periodic monitoring - * @ptr: Network device pointer. + * @t: timer list containing a network device pointer. * * This routine is run in a kernel timer to monitor the network device. */ diff --git a/drivers/net/ethernet/microchip/encx24j600-regmap.c b/drivers/net/ethernet/microchip/encx24j600-regmap.c index 5bd7fb917b7a..796e46a53926 100644 --- a/drivers/net/ethernet/microchip/encx24j600-regmap.c +++ b/drivers/net/ethernet/microchip/encx24j600-regmap.c @@ -1,5 +1,5 @@ // SPDX-License-Identifier: GPL-2.0-only -/** +/* * Register map access API - ENCX24J600 support * * Copyright 2015 Gridpoint diff --git a/drivers/net/ethernet/microchip/lan743x_main.c b/drivers/net/ethernet/microchip/lan743x_main.c index de93cc6ebc1a..a1938842f828 100644 --- a/drivers/net/ethernet/microchip/lan743x_main.c +++ b/drivers/net/ethernet/microchip/lan743x_main.c @@ -158,9 +158,8 @@ static void lan743x_tx_isr(void *context, u32 int_sts, u32 flags) struct lan743x_tx *tx = context; struct lan743x_adapter *adapter = tx->adapter; bool enable_flag = true; - u32 int_en = 0; - int_en = lan743x_csr_read(adapter, INT_EN_SET); + lan743x_csr_read(adapter, INT_EN_SET); if (flags & LAN743X_VECTOR_FLAG_SOURCE_ENABLE_CLEAR) { lan743x_csr_write(adapter, INT_EN_CLR, INT_BIT_DMA_TX_(tx->channel_number)); @@ -1699,10 +1698,9 @@ static int lan743x_tx_napi_poll(struct napi_struct *napi, int weight) bool start_transmitter = false; unsigned long irq_flags = 0; u32 ioc_bit = 0; - u32 int_sts = 0; ioc_bit = DMAC_INT_BIT_TX_IOC_(tx->channel_number); - int_sts = lan743x_csr_read(adapter, DMAC_INT_STS); + lan743x_csr_read(adapter, DMAC_INT_STS); if (tx->vector_flags & LAN743X_VECTOR_FLAG_SOURCE_STATUS_W2C) lan743x_csr_write(adapter, DMAC_INT_STS, ioc_bit); spin_lock_irqsave(&tx->ring_lock, irq_flags); @@ -3038,7 +3036,6 @@ static int lan743x_pm_suspend(struct device *dev) struct pci_dev *pdev = to_pci_dev(dev); struct net_device *netdev = pci_get_drvdata(pdev); struct lan743x_adapter *adapter = netdev_priv(netdev); - int ret; lan743x_pcidev_shutdown(pdev); @@ -3051,9 +3048,7 @@ static int lan743x_pm_suspend(struct device *dev) lan743x_pm_set_wol(adapter); /* Host sets PME_En, put D3hot */ - ret = pci_prepare_to_sleep(pdev); - - return 0; + return pci_prepare_to_sleep(pdev);; } static int lan743x_pm_resume(struct device *dev) diff --git a/drivers/net/ethernet/mscc/ocelot.c b/drivers/net/ethernet/mscc/ocelot.c index aa002db04250..70bf8c67d7ef 100644 --- a/drivers/net/ethernet/mscc/ocelot.c +++ b/drivers/net/ethernet/mscc/ocelot.c @@ -5,6 +5,7 @@ * Copyright (c) 2017 Microsemi Corporation */ #include <linux/if_bridge.h> +#include <soc/mscc/ocelot_vcap.h> #include "ocelot.h" #include "ocelot_vcap.h" @@ -107,6 +108,13 @@ static void ocelot_vcap_enable(struct ocelot *ocelot, int port) ocelot_write_gix(ocelot, ANA_PORT_VCAP_S2_CFG_S2_ENA | ANA_PORT_VCAP_S2_CFG_S2_IP6_CFG(0xa), ANA_PORT_VCAP_S2_CFG, port); + + ocelot_write_gix(ocelot, ANA_PORT_VCAP_CFG_S1_ENA, + ANA_PORT_VCAP_CFG, port); + + ocelot_rmw_gix(ocelot, REW_PORT_CFG_ES0_EN, + REW_PORT_CFG_ES0_EN, + REW_PORT_CFG, port); } static inline u32 ocelot_vlant_read_vlanaccess(struct ocelot *ocelot) @@ -191,12 +199,28 @@ static int ocelot_port_set_native_vlan(struct ocelot *ocelot, int port, return 0; } -void ocelot_port_vlan_filtering(struct ocelot *ocelot, int port, - bool vlan_aware) +int ocelot_port_vlan_filtering(struct ocelot *ocelot, int port, + bool vlan_aware, struct switchdev_trans *trans) { struct ocelot_port *ocelot_port = ocelot->ports[port]; u32 val; + if (switchdev_trans_ph_prepare(trans)) { + struct ocelot_vcap_block *block = &ocelot->block[VCAP_IS1]; + struct ocelot_vcap_filter *filter; + + list_for_each_entry(filter, &block->rules, list) { + if (filter->ingress_port_mask & BIT(port) && + filter->action.vid_replace_ena) { + dev_err(ocelot->dev, + "Cannot change VLAN state with vlan modify rules active\n"); + return -EBUSY; + } + } + + return 0; + } + ocelot_port->vlan_aware = vlan_aware; if (vlan_aware) @@ -210,6 +234,8 @@ void ocelot_port_vlan_filtering(struct ocelot *ocelot, int port, ANA_PORT_VLAN_CFG, port); ocelot_port_set_native_vlan(ocelot, port, ocelot_port->vid); + + return 0; } EXPORT_SYMBOL(ocelot_port_vlan_filtering); @@ -413,26 +439,20 @@ void ocelot_port_disable(struct ocelot *ocelot, int port) } EXPORT_SYMBOL(ocelot_port_disable); -int ocelot_port_add_txtstamp_skb(struct ocelot_port *ocelot_port, - struct sk_buff *skb) +void ocelot_port_add_txtstamp_skb(struct ocelot *ocelot, int port, + struct sk_buff *clone) { - struct skb_shared_info *shinfo = skb_shinfo(skb); - struct ocelot *ocelot = ocelot_port->ocelot; + struct ocelot_port *ocelot_port = ocelot->ports[port]; - if (ocelot->ptp && shinfo->tx_flags & SKBTX_HW_TSTAMP && - ocelot_port->ptp_cmd == IFH_REW_OP_TWO_STEP_PTP) { - spin_lock(&ocelot_port->ts_id_lock); + spin_lock(&ocelot_port->ts_id_lock); - shinfo->tx_flags |= SKBTX_IN_PROGRESS; - /* Store timestamp ID in cb[0] of sk_buff */ - skb->cb[0] = ocelot_port->ts_id; - ocelot_port->ts_id = (ocelot_port->ts_id + 1) % 4; - skb_queue_tail(&ocelot_port->tx_skbs, skb); + skb_shinfo(clone)->tx_flags |= SKBTX_IN_PROGRESS; + /* Store timestamp ID in cb[0] of sk_buff */ + clone->cb[0] = ocelot_port->ts_id; + ocelot_port->ts_id = (ocelot_port->ts_id + 1) % 4; + skb_queue_tail(&ocelot_port->tx_skbs, clone); - spin_unlock(&ocelot_port->ts_id_lock); - return 0; - } - return -ENODATA; + spin_unlock(&ocelot_port->ts_id_lock); } EXPORT_SYMBOL(ocelot_port_add_txtstamp_skb); @@ -511,9 +531,7 @@ void ocelot_get_txtstamp(struct ocelot *ocelot) /* Set the timestamp into the skb */ memset(&shhwtstamps, 0, sizeof(shhwtstamps)); shhwtstamps.hwtstamp = ktime_set(ts.tv_sec, ts.tv_nsec); - skb_tstamp_tx(skb_match, &shhwtstamps); - - dev_kfree_skb_any(skb_match); + skb_complete_tx_timestamp(skb_match, &shhwtstamps); /* Next ts */ ocelot_write(ocelot, SYS_PTP_NXT_PTP_NXT, SYS_PTP_NXT); @@ -1102,12 +1120,24 @@ EXPORT_SYMBOL(ocelot_port_bridge_join); int ocelot_port_bridge_leave(struct ocelot *ocelot, int port, struct net_device *bridge) { + struct switchdev_trans trans; + int ret; + ocelot->bridge_mask &= ~BIT(port); if (!ocelot->bridge_mask) ocelot->hw_bridge_dev = NULL; - ocelot_port_vlan_filtering(ocelot, port, 0); + trans.ph_prepare = true; + ret = ocelot_port_vlan_filtering(ocelot, port, false, &trans); + if (ret) + return ret; + + trans.ph_prepare = false; + ret = ocelot_port_vlan_filtering(ocelot, port, false, &trans); + if (ret) + return ret; + ocelot_port_set_pvid(ocelot, port, 0); return ocelot_port_set_native_vlan(ocelot, port, 0); } @@ -1354,22 +1384,14 @@ void ocelot_init_port(struct ocelot *ocelot, int port) } EXPORT_SYMBOL(ocelot_init_port); -/* Configure and enable the CPU port module, which is a set of queues. - * If @npi contains a valid port index, the CPU port module is connected - * to the Node Processor Interface (NPI). This is the mode through which - * frames can be injected from and extracted to an external CPU, - * over Ethernet. +/* Configure and enable the CPU port module, which is a set of queues + * accessible through register MMIO, frame DMA or Ethernet (in case + * NPI mode is used). */ -void ocelot_configure_cpu(struct ocelot *ocelot, int npi, - enum ocelot_tag_prefix injection, - enum ocelot_tag_prefix extraction) +static void ocelot_cpu_port_init(struct ocelot *ocelot) { int cpu = ocelot->num_phys_ports; - ocelot->npi = npi; - ocelot->inj_prefix = injection; - ocelot->xtr_prefix = extraction; - /* The unicast destination PGID for the CPU port module is unused */ ocelot_write_rix(ocelot, 0, ANA_PGID_PGID, cpu); /* Instead set up a multicast destination PGID for traffic copied to @@ -1381,31 +1403,13 @@ void ocelot_configure_cpu(struct ocelot *ocelot, int npi, ANA_PORT_PORT_CFG_PORTID_VAL(cpu), ANA_PORT_PORT_CFG, cpu); - if (npi >= 0 && npi < ocelot->num_phys_ports) { - ocelot_write(ocelot, QSYS_EXT_CPU_CFG_EXT_CPUQ_MSK_M | - QSYS_EXT_CPU_CFG_EXT_CPU_PORT(npi), - QSYS_EXT_CPU_CFG); - - /* Enable NPI port */ - ocelot_fields_write(ocelot, npi, - QSYS_SWITCH_PORT_MODE_PORT_ENA, 1); - /* NPI port Injection/Extraction configuration */ - ocelot_fields_write(ocelot, npi, SYS_PORT_MODE_INCL_XTR_HDR, - extraction); - ocelot_fields_write(ocelot, npi, SYS_PORT_MODE_INCL_INJ_HDR, - injection); - - /* Disable transmission of pause frames */ - ocelot_fields_write(ocelot, npi, SYS_PAUSE_CFG_PAUSE_ENA, 0); - } - /* Enable CPU port module */ ocelot_fields_write(ocelot, cpu, QSYS_SWITCH_PORT_MODE_PORT_ENA, 1); /* CPU port Injection/Extraction configuration */ ocelot_fields_write(ocelot, cpu, SYS_PORT_MODE_INCL_XTR_HDR, - extraction); + ocelot->xtr_prefix); ocelot_fields_write(ocelot, cpu, SYS_PORT_MODE_INCL_INJ_HDR, - injection); + ocelot->inj_prefix); /* Configure the CPU port to be VLAN aware */ ocelot_write_gix(ocelot, ANA_PORT_VLAN_CFG_VLAN_VID(0) | @@ -1413,7 +1417,6 @@ void ocelot_configure_cpu(struct ocelot *ocelot, int npi, ANA_PORT_VLAN_CFG_VLAN_POP_CNT(1), ANA_PORT_VLAN_CFG, cpu); } -EXPORT_SYMBOL(ocelot_configure_cpu); int ocelot_init(struct ocelot *ocelot) { @@ -1453,6 +1456,7 @@ int ocelot_init(struct ocelot *ocelot) ocelot_mact_init(ocelot); ocelot_vlan_init(ocelot); ocelot_vcap_init(ocelot); + ocelot_cpu_port_init(ocelot); for (port = 0; port < ocelot->num_phys_ports; port++) { /* Clear all counters (5 groups) */ diff --git a/drivers/net/ethernet/mscc/ocelot.h b/drivers/net/ethernet/mscc/ocelot.h index dc29e05103a1..abb407dff93c 100644 --- a/drivers/net/ethernet/mscc/ocelot.h +++ b/drivers/net/ethernet/mscc/ocelot.h @@ -98,6 +98,8 @@ int ocelot_port_lag_join(struct ocelot *ocelot, int port, struct net_device *bond); void ocelot_port_lag_leave(struct ocelot *ocelot, int port, struct net_device *bond); +struct net_device *ocelot_port_to_netdev(struct ocelot *ocelot, int port); +int ocelot_netdev_to_port(struct net_device *dev); u32 ocelot_port_readl(struct ocelot_port *port, u32 reg); void ocelot_port_writel(struct ocelot_port *port, u32 val, u32 reg); diff --git a/drivers/net/ethernet/mscc/ocelot_flower.c b/drivers/net/ethernet/mscc/ocelot_flower.c index ec1b6e2572ba..729495a1a77e 100644 --- a/drivers/net/ethernet/mscc/ocelot_flower.c +++ b/drivers/net/ethernet/mscc/ocelot_flower.c @@ -5,56 +5,433 @@ #include <net/pkt_cls.h> #include <net/tc_act/tc_gact.h> - +#include <soc/mscc/ocelot_vcap.h> #include "ocelot_vcap.h" -static int ocelot_flower_parse_action(struct flow_cls_offload *f, +/* Arbitrarily chosen constants for encoding the VCAP block and lookup number + * into the chain number. This is UAPI. + */ +#define VCAP_BLOCK 10000 +#define VCAP_LOOKUP 1000 +#define VCAP_IS1_NUM_LOOKUPS 3 +#define VCAP_IS2_NUM_LOOKUPS 2 +#define VCAP_IS2_NUM_PAG 256 +#define VCAP_IS1_CHAIN(lookup) \ + (1 * VCAP_BLOCK + (lookup) * VCAP_LOOKUP) +#define VCAP_IS2_CHAIN(lookup, pag) \ + (2 * VCAP_BLOCK + (lookup) * VCAP_LOOKUP + (pag)) + +static int ocelot_chain_to_block(int chain, bool ingress) +{ + int lookup, pag; + + if (!ingress) { + if (chain == 0) + return VCAP_ES0; + return -EOPNOTSUPP; + } + + /* Backwards compatibility with older, single-chain tc-flower + * offload support in Ocelot + */ + if (chain == 0) + return VCAP_IS2; + + for (lookup = 0; lookup < VCAP_IS1_NUM_LOOKUPS; lookup++) + if (chain == VCAP_IS1_CHAIN(lookup)) + return VCAP_IS1; + + for (lookup = 0; lookup < VCAP_IS2_NUM_LOOKUPS; lookup++) + for (pag = 0; pag < VCAP_IS2_NUM_PAG; pag++) + if (chain == VCAP_IS2_CHAIN(lookup, pag)) + return VCAP_IS2; + + return -EOPNOTSUPP; +} + +/* Caller must ensure this is a valid IS1 or IS2 chain first, + * by calling ocelot_chain_to_block. + */ +static int ocelot_chain_to_lookup(int chain) +{ + return (chain / VCAP_LOOKUP) % 10; +} + +/* Caller must ensure this is a valid IS2 chain first, + * by calling ocelot_chain_to_block. + */ +static int ocelot_chain_to_pag(int chain) +{ + int lookup = ocelot_chain_to_lookup(chain); + + /* calculate PAG value as chain index relative to the first PAG */ + return chain - VCAP_IS2_CHAIN(lookup, 0); +} + +static bool ocelot_is_goto_target_valid(int goto_target, int chain, + bool ingress) +{ + int pag; + + /* Can't offload GOTO in VCAP ES0 */ + if (!ingress) + return (goto_target < 0); + + /* Non-optional GOTOs */ + if (chain == 0) + /* VCAP IS1 can be skipped, either partially or completely */ + return (goto_target == VCAP_IS1_CHAIN(0) || + goto_target == VCAP_IS1_CHAIN(1) || + goto_target == VCAP_IS1_CHAIN(2) || + goto_target == VCAP_IS2_CHAIN(0, 0) || + goto_target == VCAP_IS2_CHAIN(1, 0)); + + if (chain == VCAP_IS1_CHAIN(0)) + return (goto_target == VCAP_IS1_CHAIN(1)); + + if (chain == VCAP_IS1_CHAIN(1)) + return (goto_target == VCAP_IS1_CHAIN(2)); + + /* Lookup 2 of VCAP IS1 can really support non-optional GOTOs, + * using a Policy Association Group (PAG) value, which is an 8-bit + * value encoding a VCAP IS2 target chain. + */ + if (chain == VCAP_IS1_CHAIN(2)) { + for (pag = 0; pag < VCAP_IS2_NUM_PAG; pag++) + if (goto_target == VCAP_IS2_CHAIN(0, pag)) + return true; + + return false; + } + + /* Non-optional GOTO from VCAP IS2 lookup 0 to lookup 1. + * We cannot change the PAG at this point. + */ + for (pag = 0; pag < VCAP_IS2_NUM_PAG; pag++) + if (chain == VCAP_IS2_CHAIN(0, pag)) + return (goto_target == VCAP_IS2_CHAIN(1, pag)); + + /* VCAP IS2 lookup 1 cannot jump anywhere */ + return false; +} + +static struct ocelot_vcap_filter * +ocelot_find_vcap_filter_that_points_at(struct ocelot *ocelot, int chain) +{ + struct ocelot_vcap_filter *filter; + struct ocelot_vcap_block *block; + int block_id; + + block_id = ocelot_chain_to_block(chain, true); + if (block_id < 0) + return NULL; + + if (block_id == VCAP_IS2) { + block = &ocelot->block[VCAP_IS1]; + + list_for_each_entry(filter, &block->rules, list) + if (filter->type == OCELOT_VCAP_FILTER_PAG && + filter->goto_target == chain) + return filter; + } + + list_for_each_entry(filter, &ocelot->dummy_rules, list) + if (filter->goto_target == chain) + return filter; + + return NULL; +} + +static int ocelot_flower_parse_action(struct ocelot *ocelot, int port, + bool ingress, struct flow_cls_offload *f, struct ocelot_vcap_filter *filter) { + struct ocelot_port *ocelot_port = ocelot->ports[port]; + struct netlink_ext_ack *extack = f->common.extack; + bool allow_missing_goto_target = false; const struct flow_action_entry *a; + enum ocelot_tag_tpid_sel tpid; + int i, chain, egress_port; u64 rate; - int i; - - if (!flow_offload_has_one_action(&f->rule->action)) - return -EOPNOTSUPP; if (!flow_action_basic_hw_stats_check(&f->rule->action, f->common.extack)) return -EOPNOTSUPP; + chain = f->common.chain_index; + filter->block_id = ocelot_chain_to_block(chain, ingress); + if (filter->block_id < 0) { + NL_SET_ERR_MSG_MOD(extack, "Cannot offload to this chain"); + return -EOPNOTSUPP; + } + if (filter->block_id == VCAP_IS1 || filter->block_id == VCAP_IS2) + filter->lookup = ocelot_chain_to_lookup(chain); + if (filter->block_id == VCAP_IS2) + filter->pag = ocelot_chain_to_pag(chain); + + filter->goto_target = -1; + filter->type = OCELOT_VCAP_FILTER_DUMMY; + flow_action_for_each(i, a, &f->rule->action) { switch (a->id) { case FLOW_ACTION_DROP: - filter->action = OCELOT_VCAP_ACTION_DROP; + if (filter->block_id != VCAP_IS2) { + NL_SET_ERR_MSG_MOD(extack, + "Drop action can only be offloaded to VCAP IS2"); + return -EOPNOTSUPP; + } + if (filter->goto_target != -1) { + NL_SET_ERR_MSG_MOD(extack, + "Last action must be GOTO"); + return -EOPNOTSUPP; + } + filter->action.mask_mode = OCELOT_MASK_MODE_PERMIT_DENY; + filter->action.port_mask = 0; + filter->action.police_ena = true; + filter->action.pol_ix = OCELOT_POLICER_DISCARD; + filter->type = OCELOT_VCAP_FILTER_OFFLOAD; break; case FLOW_ACTION_TRAP: - filter->action = OCELOT_VCAP_ACTION_TRAP; + if (filter->block_id != VCAP_IS2) { + NL_SET_ERR_MSG_MOD(extack, + "Trap action can only be offloaded to VCAP IS2"); + return -EOPNOTSUPP; + } + if (filter->goto_target != -1) { + NL_SET_ERR_MSG_MOD(extack, + "Last action must be GOTO"); + return -EOPNOTSUPP; + } + filter->action.mask_mode = OCELOT_MASK_MODE_PERMIT_DENY; + filter->action.port_mask = 0; + filter->action.cpu_copy_ena = true; + filter->action.cpu_qu_num = 0; + filter->type = OCELOT_VCAP_FILTER_OFFLOAD; break; case FLOW_ACTION_POLICE: - filter->action = OCELOT_VCAP_ACTION_POLICE; + if (filter->block_id != VCAP_IS2 || + filter->lookup != 0) { + NL_SET_ERR_MSG_MOD(extack, + "Police action can only be offloaded to VCAP IS2 lookup 0"); + return -EOPNOTSUPP; + } + if (filter->goto_target != -1) { + NL_SET_ERR_MSG_MOD(extack, + "Last action must be GOTO"); + return -EOPNOTSUPP; + } + filter->action.police_ena = true; rate = a->police.rate_bytes_ps; - filter->pol.rate = div_u64(rate, 1000) * 8; - filter->pol.burst = a->police.burst; + filter->action.pol.rate = div_u64(rate, 1000) * 8; + filter->action.pol.burst = a->police.burst; + filter->type = OCELOT_VCAP_FILTER_OFFLOAD; + break; + case FLOW_ACTION_REDIRECT: + if (filter->block_id != VCAP_IS2) { + NL_SET_ERR_MSG_MOD(extack, + "Redirect action can only be offloaded to VCAP IS2"); + return -EOPNOTSUPP; + } + if (filter->goto_target != -1) { + NL_SET_ERR_MSG_MOD(extack, + "Last action must be GOTO"); + return -EOPNOTSUPP; + } + egress_port = ocelot->ops->netdev_to_port(a->dev); + if (egress_port < 0) { + NL_SET_ERR_MSG_MOD(extack, + "Destination not an ocelot port"); + return -EOPNOTSUPP; + } + filter->action.mask_mode = OCELOT_MASK_MODE_REDIRECT; + filter->action.port_mask = BIT(egress_port); + filter->type = OCELOT_VCAP_FILTER_OFFLOAD; + break; + case FLOW_ACTION_VLAN_POP: + if (filter->block_id != VCAP_IS1) { + NL_SET_ERR_MSG_MOD(extack, + "VLAN pop action can only be offloaded to VCAP IS1"); + return -EOPNOTSUPP; + } + if (filter->goto_target != -1) { + NL_SET_ERR_MSG_MOD(extack, + "Last action must be GOTO"); + return -EOPNOTSUPP; + } + filter->action.vlan_pop_cnt_ena = true; + filter->action.vlan_pop_cnt++; + if (filter->action.vlan_pop_cnt > 2) { + NL_SET_ERR_MSG_MOD(extack, + "Cannot pop more than 2 VLAN headers"); + return -EOPNOTSUPP; + } + filter->type = OCELOT_VCAP_FILTER_OFFLOAD; + break; + case FLOW_ACTION_VLAN_MANGLE: + if (filter->block_id != VCAP_IS1) { + NL_SET_ERR_MSG_MOD(extack, + "VLAN modify action can only be offloaded to VCAP IS1"); + return -EOPNOTSUPP; + } + if (filter->goto_target != -1) { + NL_SET_ERR_MSG_MOD(extack, + "Last action must be GOTO"); + return -EOPNOTSUPP; + } + if (!ocelot_port->vlan_aware) { + NL_SET_ERR_MSG_MOD(extack, + "Can only modify VLAN under VLAN aware bridge"); + return -EOPNOTSUPP; + } + filter->action.vid_replace_ena = true; + filter->action.pcp_dei_ena = true; + filter->action.vid = a->vlan.vid; + filter->action.pcp = a->vlan.prio; + filter->type = OCELOT_VCAP_FILTER_OFFLOAD; + break; + case FLOW_ACTION_PRIORITY: + if (filter->block_id != VCAP_IS1) { + NL_SET_ERR_MSG_MOD(extack, + "Priority action can only be offloaded to VCAP IS1"); + return -EOPNOTSUPP; + } + if (filter->goto_target != -1) { + NL_SET_ERR_MSG_MOD(extack, + "Last action must be GOTO"); + return -EOPNOTSUPP; + } + filter->action.qos_ena = true; + filter->action.qos_val = a->priority; + filter->type = OCELOT_VCAP_FILTER_OFFLOAD; + break; + case FLOW_ACTION_GOTO: + filter->goto_target = a->chain_index; + + if (filter->block_id == VCAP_IS1 && filter->lookup == 2) { + int pag = ocelot_chain_to_pag(filter->goto_target); + + filter->action.pag_override_mask = 0xff; + filter->action.pag_val = pag; + filter->type = OCELOT_VCAP_FILTER_PAG; + } + break; + case FLOW_ACTION_VLAN_PUSH: + if (filter->block_id != VCAP_ES0) { + NL_SET_ERR_MSG_MOD(extack, + "VLAN push action can only be offloaded to VCAP ES0"); + return -EOPNOTSUPP; + } + switch (ntohs(a->vlan.proto)) { + case ETH_P_8021Q: + tpid = OCELOT_TAG_TPID_SEL_8021Q; + break; + case ETH_P_8021AD: + tpid = OCELOT_TAG_TPID_SEL_8021AD; + break; + default: + NL_SET_ERR_MSG_MOD(extack, + "Cannot push custom TPID"); + return -EOPNOTSUPP; + } + filter->action.tag_a_tpid_sel = tpid; + filter->action.push_outer_tag = OCELOT_ES0_TAG; + filter->action.tag_a_vid_sel = 1; + filter->action.vid_a_val = a->vlan.vid; + filter->action.pcp_a_val = a->vlan.prio; + filter->type = OCELOT_VCAP_FILTER_OFFLOAD; break; default: + NL_SET_ERR_MSG_MOD(extack, "Cannot offload action"); return -EOPNOTSUPP; } } + if (filter->goto_target == -1) { + if ((filter->block_id == VCAP_IS2 && filter->lookup == 1) || + chain == 0) { + allow_missing_goto_target = true; + } else { + NL_SET_ERR_MSG_MOD(extack, "Missing GOTO action"); + return -EOPNOTSUPP; + } + } + + if (!ocelot_is_goto_target_valid(filter->goto_target, chain, ingress) && + !allow_missing_goto_target) { + NL_SET_ERR_MSG_MOD(extack, "Cannot offload this GOTO target"); + return -EOPNOTSUPP; + } + return 0; } -static int ocelot_flower_parse(struct flow_cls_offload *f, - struct ocelot_vcap_filter *filter) +static int ocelot_flower_parse_indev(struct ocelot *ocelot, int port, + struct flow_cls_offload *f, + struct ocelot_vcap_filter *filter) +{ + struct flow_rule *rule = flow_cls_offload_flow_rule(f); + const struct vcap_props *vcap = &ocelot->vcap[VCAP_ES0]; + int key_length = vcap->keys[VCAP_ES0_IGR_PORT].length; + struct netlink_ext_ack *extack = f->common.extack; + struct net_device *dev, *indev; + struct flow_match_meta match; + int ingress_port; + + flow_rule_match_meta(rule, &match); + + if (!match.mask->ingress_ifindex) + return 0; + + if (match.mask->ingress_ifindex != 0xFFFFFFFF) { + NL_SET_ERR_MSG_MOD(extack, "Unsupported ingress ifindex mask"); + return -EOPNOTSUPP; + } + + dev = ocelot->ops->port_to_netdev(ocelot, port); + if (!dev) + return -EINVAL; + + indev = __dev_get_by_index(dev_net(dev), match.key->ingress_ifindex); + if (!indev) { + NL_SET_ERR_MSG_MOD(extack, + "Can't find the ingress port to match on"); + return -ENOENT; + } + + ingress_port = ocelot->ops->netdev_to_port(indev); + if (ingress_port < 0) { + NL_SET_ERR_MSG_MOD(extack, + "Can only offload an ocelot ingress port"); + return -EOPNOTSUPP; + } + if (ingress_port == port) { + NL_SET_ERR_MSG_MOD(extack, + "Ingress port is equal to the egress port"); + return -EINVAL; + } + + filter->ingress_port.value = ingress_port; + filter->ingress_port.mask = GENMASK(key_length - 1, 0); + + return 0; +} + +static int +ocelot_flower_parse_key(struct ocelot *ocelot, int port, bool ingress, + struct flow_cls_offload *f, + struct ocelot_vcap_filter *filter) { struct flow_rule *rule = flow_cls_offload_flow_rule(f); struct flow_dissector *dissector = rule->match.dissector; + struct netlink_ext_ack *extack = f->common.extack; u16 proto = ntohs(f->common.protocol); bool match_protocol = true; + int ret; if (dissector->used_keys & ~(BIT(FLOW_DISSECTOR_KEY_CONTROL) | BIT(FLOW_DISSECTOR_KEY_BASIC) | + BIT(FLOW_DISSECTOR_KEY_META) | BIT(FLOW_DISSECTOR_KEY_PORTS) | BIT(FLOW_DISSECTOR_KEY_VLAN) | BIT(FLOW_DISSECTOR_KEY_IPV4_ADDRS) | @@ -63,6 +440,13 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, return -EOPNOTSUPP; } + /* For VCAP ES0 (egress rewriter) we can match on the ingress port */ + if (!ingress) { + ret = ocelot_flower_parse_indev(ocelot, port, f, filter); + if (ret) + return ret; + } + if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_CONTROL)) { struct flow_match_control match; @@ -72,6 +456,19 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ETH_ADDRS)) { struct flow_match_eth_addrs match; + if (filter->block_id == VCAP_ES0) { + NL_SET_ERR_MSG_MOD(extack, + "VCAP ES0 cannot match on MAC address"); + return -EOPNOTSUPP; + } + + if (filter->block_id == VCAP_IS1 && + !is_zero_ether_addr(match.mask->dst)) { + NL_SET_ERR_MSG_MOD(extack, + "Key type S1_NORMAL cannot match on destination MAC"); + return -EOPNOTSUPP; + } + /* The hw support mac matches only for MAC_ETYPE key, * therefore if other matches(port, tcp flags, etc) are added * then just bail out @@ -103,6 +500,12 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, flow_rule_match_basic(rule, &match); if (ntohs(match.key->n_proto) == ETH_P_IP) { + if (filter->block_id == VCAP_ES0) { + NL_SET_ERR_MSG_MOD(extack, + "VCAP ES0 cannot match on IP protocol"); + return -EOPNOTSUPP; + } + filter->key_type = OCELOT_VCAP_KEY_IPV4; filter->key.ipv4.proto.value[0] = match.key->ip_proto; @@ -111,6 +514,12 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, match_protocol = false; } if (ntohs(match.key->n_proto) == ETH_P_IPV6) { + if (filter->block_id == VCAP_ES0) { + NL_SET_ERR_MSG_MOD(extack, + "VCAP ES0 cannot match on IP protocol"); + return -EOPNOTSUPP; + } + filter->key_type = OCELOT_VCAP_KEY_IPV6; filter->key.ipv6.proto.value[0] = match.key->ip_proto; @@ -125,6 +534,18 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, struct flow_match_ipv4_addrs match; u8 *tmp; + if (filter->block_id == VCAP_ES0) { + NL_SET_ERR_MSG_MOD(extack, + "VCAP ES0 cannot match on IP address"); + return -EOPNOTSUPP; + } + + if (filter->block_id == VCAP_IS1 && *(u32 *)&match.mask->dst) { + NL_SET_ERR_MSG_MOD(extack, + "Key type S1_NORMAL cannot match on destination IP"); + return -EOPNOTSUPP; + } + flow_rule_match_ipv4_addrs(rule, &match); tmp = &filter->key.ipv4.sip.value.addr[0]; memcpy(tmp, &match.key->src, 4); @@ -148,6 +569,12 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_PORTS)) { struct flow_match_ports match; + if (filter->block_id == VCAP_ES0) { + NL_SET_ERR_MSG_MOD(extack, + "VCAP ES0 cannot match on L4 ports"); + return -EOPNOTSUPP; + } + flow_rule_match_ports(rule, &match); filter->key.ipv4.sport.value = ntohs(match.key->src); filter->key.ipv4.sport.mask = ntohs(match.mask->src); @@ -170,6 +597,12 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, finished_key_parsing: if (match_protocol && proto != ETH_P_ALL) { + if (filter->block_id == VCAP_ES0) { + NL_SET_ERR_MSG_MOD(extack, + "VCAP ES0 cannot match on L2 proto"); + return -EOPNOTSUPP; + } + /* TODO: support SNAP, LLC etc */ if (proto < ETH_P_802_3_MIN) return -EOPNOTSUPP; @@ -179,14 +612,28 @@ finished_key_parsing: } /* else, a filter of type OCELOT_VCAP_KEY_ANY is implicitly added */ + return 0; +} + +static int ocelot_flower_parse(struct ocelot *ocelot, int port, bool ingress, + struct flow_cls_offload *f, + struct ocelot_vcap_filter *filter) +{ + int ret; + filter->prio = f->common.prio; filter->id = f->cookie; - return ocelot_flower_parse_action(f, filter); + + ret = ocelot_flower_parse_action(ocelot, port, ingress, f, filter); + if (ret) + return ret; + + return ocelot_flower_parse_key(ocelot, port, ingress, f, filter); } static struct ocelot_vcap_filter -*ocelot_vcap_filter_create(struct ocelot *ocelot, int port, - struct flow_cls_offload *f) +*ocelot_vcap_filter_create(struct ocelot *ocelot, int port, bool ingress, + struct flow_cls_offload *f) { struct ocelot_vcap_filter *filter; @@ -194,26 +641,65 @@ static struct ocelot_vcap_filter if (!filter) return NULL; - filter->ingress_port_mask = BIT(port); + if (ingress) { + filter->ingress_port_mask = BIT(port); + } else { + const struct vcap_props *vcap = &ocelot->vcap[VCAP_ES0]; + int key_length = vcap->keys[VCAP_ES0_EGR_PORT].length; + + filter->egress_port.value = port; + filter->egress_port.mask = GENMASK(key_length - 1, 0); + } + return filter; } +static int ocelot_vcap_dummy_filter_add(struct ocelot *ocelot, + struct ocelot_vcap_filter *filter) +{ + list_add(&filter->list, &ocelot->dummy_rules); + + return 0; +} + +static int ocelot_vcap_dummy_filter_del(struct ocelot *ocelot, + struct ocelot_vcap_filter *filter) +{ + list_del(&filter->list); + kfree(filter); + + return 0; +} + int ocelot_cls_flower_replace(struct ocelot *ocelot, int port, struct flow_cls_offload *f, bool ingress) { + struct netlink_ext_ack *extack = f->common.extack; struct ocelot_vcap_filter *filter; + int chain = f->common.chain_index; int ret; - filter = ocelot_vcap_filter_create(ocelot, port, f); + if (chain && !ocelot_find_vcap_filter_that_points_at(ocelot, chain)) { + NL_SET_ERR_MSG_MOD(extack, "No default GOTO action points to this chain"); + return -EOPNOTSUPP; + } + + filter = ocelot_vcap_filter_create(ocelot, port, ingress, f); if (!filter) return -ENOMEM; - ret = ocelot_flower_parse(f, filter); + ret = ocelot_flower_parse(ocelot, port, ingress, f, filter); if (ret) { kfree(filter); return ret; } + /* The non-optional GOTOs for the TCAM skeleton don't need + * to be actually offloaded. + */ + if (filter->type == OCELOT_VCAP_FILTER_DUMMY) + return ocelot_vcap_dummy_filter_add(ocelot, filter); + return ocelot_vcap_filter_add(ocelot, filter, f->common.extack); } EXPORT_SYMBOL_GPL(ocelot_cls_flower_replace); @@ -221,28 +707,49 @@ EXPORT_SYMBOL_GPL(ocelot_cls_flower_replace); int ocelot_cls_flower_destroy(struct ocelot *ocelot, int port, struct flow_cls_offload *f, bool ingress) { - struct ocelot_vcap_filter filter; + struct ocelot_vcap_filter *filter; + struct ocelot_vcap_block *block; + int block_id; + + block_id = ocelot_chain_to_block(f->common.chain_index, ingress); + if (block_id < 0) + return 0; - filter.prio = f->common.prio; - filter.id = f->cookie; + block = &ocelot->block[block_id]; - return ocelot_vcap_filter_del(ocelot, &filter); + filter = ocelot_vcap_block_find_filter_by_id(block, f->cookie); + if (!filter) + return 0; + + if (filter->type == OCELOT_VCAP_FILTER_DUMMY) + return ocelot_vcap_dummy_filter_del(ocelot, filter); + + return ocelot_vcap_filter_del(ocelot, filter); } EXPORT_SYMBOL_GPL(ocelot_cls_flower_destroy); int ocelot_cls_flower_stats(struct ocelot *ocelot, int port, struct flow_cls_offload *f, bool ingress) { - struct ocelot_vcap_filter filter; - int ret; + struct ocelot_vcap_filter *filter; + struct ocelot_vcap_block *block; + int block_id, ret; + + block_id = ocelot_chain_to_block(f->common.chain_index, ingress); + if (block_id < 0) + return 0; + + block = &ocelot->block[block_id]; + + filter = ocelot_vcap_block_find_filter_by_id(block, f->cookie); + if (!filter || filter->type == OCELOT_VCAP_FILTER_DUMMY) + return 0; - filter.prio = f->common.prio; - filter.id = f->cookie; - ret = ocelot_vcap_filter_stats_update(ocelot, &filter); + ret = ocelot_vcap_filter_stats_update(ocelot, filter); if (ret) return ret; - flow_stats_update(&f->stats, 0x0, filter.stats.pkts, 0, 0x0, + flow_stats_update(&f->stats, 0x0, filter->stats.pkts, 0, 0x0, FLOW_ACTION_HW_STATS_IMMEDIATE); return 0; } diff --git a/drivers/net/ethernet/mscc/ocelot_io.c b/drivers/net/ethernet/mscc/ocelot_io.c index d22711282183..0acb45948418 100644 --- a/drivers/net/ethernet/mscc/ocelot_io.c +++ b/drivers/net/ethernet/mscc/ocelot_io.c @@ -71,6 +71,23 @@ void ocelot_port_writel(struct ocelot_port *port, u32 val, u32 reg) } EXPORT_SYMBOL(ocelot_port_writel); +u32 __ocelot_target_read_ix(struct ocelot *ocelot, enum ocelot_target target, + u32 reg, u32 offset) +{ + u32 val; + + regmap_read(ocelot->targets[target], + ocelot->map[target][reg] + offset, &val); + return val; +} + +void __ocelot_target_write_ix(struct ocelot *ocelot, enum ocelot_target target, + u32 val, u32 reg, u32 offset) +{ + regmap_write(ocelot->targets[target], + ocelot->map[target][reg] + offset, val); +} + int ocelot_regfields_init(struct ocelot *ocelot, const struct reg_field *const regfields) { diff --git a/drivers/net/ethernet/mscc/ocelot_net.c b/drivers/net/ethernet/mscc/ocelot_net.c index 8490e42e9e2d..b34da11acf65 100644 --- a/drivers/net/ethernet/mscc/ocelot_net.c +++ b/drivers/net/ethernet/mscc/ocelot_net.c @@ -330,7 +330,6 @@ static int ocelot_port_xmit(struct sk_buff *skb, struct net_device *dev) u8 grp = 0; /* Send everything on CPU group 0 */ unsigned int i, count, last; int port = priv->chip_port; - bool do_tstamp; val = ocelot_read(ocelot, QS_INJ_STATUS); if (!(val & QS_INJ_STATUS_FIFO_RDY(BIT(grp))) || @@ -345,7 +344,23 @@ static int ocelot_port_xmit(struct sk_buff *skb, struct net_device *dev) info.vid = skb_vlan_tag_get(skb); /* Check if timestamping is needed */ - do_tstamp = (ocelot_port_add_txtstamp_skb(ocelot_port, skb) == 0); + if (ocelot->ptp && (shinfo->tx_flags & SKBTX_HW_TSTAMP)) { + info.rew_op = ocelot_port->ptp_cmd; + + if (ocelot_port->ptp_cmd == IFH_REW_OP_TWO_STEP_PTP) { + struct sk_buff *clone; + + clone = skb_clone_sk(skb); + if (!clone) { + kfree_skb(skb); + return NETDEV_TX_OK; + } + + ocelot_port_add_txtstamp_skb(ocelot, port, clone); + + info.rew_op |= clone->cb[0] << 3; + } + } if (ocelot->ptp && shinfo->tx_flags & SKBTX_HW_TSTAMP) { info.rew_op = ocelot_port->ptp_cmd; @@ -383,8 +398,7 @@ static int ocelot_port_xmit(struct sk_buff *skb, struct net_device *dev) dev->stats.tx_packets++; dev->stats.tx_bytes += skb->len; - if (!do_tstamp) - dev_kfree_skb_any(skb); + kfree_skb(skb); return NETDEV_TX_OK; } @@ -642,6 +656,37 @@ static const struct net_device_ops ocelot_port_netdev_ops = { .ndo_do_ioctl = ocelot_ioctl, }; +struct net_device *ocelot_port_to_netdev(struct ocelot *ocelot, int port) +{ + struct ocelot_port *ocelot_port = ocelot->ports[port]; + struct ocelot_port_private *priv; + + if (!ocelot_port) + return NULL; + + priv = container_of(ocelot_port, struct ocelot_port_private, port); + + return priv->dev; +} + +/* Checks if the net_device instance given to us originates from our driver */ +static bool ocelot_netdevice_dev_check(const struct net_device *dev) +{ + return dev->netdev_ops == &ocelot_port_netdev_ops; +} + +int ocelot_netdev_to_port(struct net_device *dev) +{ + struct ocelot_port_private *priv; + + if (!dev || !ocelot_netdevice_dev_check(dev)) + return -EINVAL; + + priv = netdev_priv(dev); + + return priv->chip_port; +} + static void ocelot_port_get_strings(struct net_device *netdev, u32 sset, u8 *data) { @@ -746,7 +791,7 @@ static int ocelot_port_attr_set(struct net_device *dev, break; case SWITCHDEV_ATTR_ID_BRIDGE_VLAN_FILTERING: ocelot_port_vlan_filtering(ocelot, port, - attr->u.vlan_filtering); + attr->u.vlan_filtering, trans); break; case SWITCHDEV_ATTR_ID_BRIDGE_MC_DISABLED: ocelot_port_attr_mc_set(ocelot, port, !attr->u.mc_disabled); @@ -863,12 +908,6 @@ static int ocelot_port_obj_del(struct net_device *dev, return ret; } -/* Checks if the net_device instance given to us originate from our driver. */ -static bool ocelot_netdevice_dev_check(const struct net_device *dev) -{ - return dev->netdev_ops == &ocelot_port_netdev_ops; -} - static int ocelot_netdevice_port_event(struct net_device *dev, unsigned long event, struct netdev_notifier_changeupper_info *info) diff --git a/drivers/net/ethernet/mscc/ocelot_ptp.c b/drivers/net/ethernet/mscc/ocelot_ptp.c index 1e08fe4daaef..a33ab315cc6b 100644 --- a/drivers/net/ethernet/mscc/ocelot_ptp.c +++ b/drivers/net/ethernet/mscc/ocelot_ptp.c @@ -300,7 +300,8 @@ int ocelot_ptp_enable(struct ptp_clock_info *ptp, } EXPORT_SYMBOL(ocelot_ptp_enable); -int ocelot_init_timestamp(struct ocelot *ocelot, struct ptp_clock_info *info) +int ocelot_init_timestamp(struct ocelot *ocelot, + const struct ptp_clock_info *info) { struct ptp_clock *ptp_clock; int i; diff --git a/drivers/net/ethernet/mscc/ocelot_s2.h b/drivers/net/ethernet/mscc/ocelot_s2.h deleted file mode 100644 index 80107bec2e45..000000000000 --- a/drivers/net/ethernet/mscc/ocelot_s2.h +++ /dev/null @@ -1,64 +0,0 @@ -/* SPDX-License-Identifier: (GPL-2.0 OR MIT) */ -/* Microsemi Ocelot Switch driver - * Copyright (c) 2018 Microsemi Corporation - */ - -#ifndef _OCELOT_S2_CORE_H_ -#define _OCELOT_S2_CORE_H_ - -#define S2_CORE_UPDATE_CTRL_UPDATE_CMD(x) (((x) << 22) & GENMASK(24, 22)) -#define S2_CORE_UPDATE_CTRL_UPDATE_CMD_M GENMASK(24, 22) -#define S2_CORE_UPDATE_CTRL_UPDATE_CMD_X(x) (((x) & GENMASK(24, 22)) >> 22) -#define S2_CORE_UPDATE_CTRL_UPDATE_ENTRY_DIS BIT(21) -#define S2_CORE_UPDATE_CTRL_UPDATE_ACTION_DIS BIT(20) -#define S2_CORE_UPDATE_CTRL_UPDATE_CNT_DIS BIT(19) -#define S2_CORE_UPDATE_CTRL_UPDATE_ADDR(x) (((x) << 3) & GENMASK(18, 3)) -#define S2_CORE_UPDATE_CTRL_UPDATE_ADDR_M GENMASK(18, 3) -#define S2_CORE_UPDATE_CTRL_UPDATE_ADDR_X(x) (((x) & GENMASK(18, 3)) >> 3) -#define S2_CORE_UPDATE_CTRL_UPDATE_SHOT BIT(2) -#define S2_CORE_UPDATE_CTRL_CLEAR_CACHE BIT(1) -#define S2_CORE_UPDATE_CTRL_MV_TRAFFIC_IGN BIT(0) - -#define S2_CORE_MV_CFG_MV_NUM_POS(x) (((x) << 16) & GENMASK(31, 16)) -#define S2_CORE_MV_CFG_MV_NUM_POS_M GENMASK(31, 16) -#define S2_CORE_MV_CFG_MV_NUM_POS_X(x) (((x) & GENMASK(31, 16)) >> 16) -#define S2_CORE_MV_CFG_MV_SIZE(x) ((x) & GENMASK(15, 0)) -#define S2_CORE_MV_CFG_MV_SIZE_M GENMASK(15, 0) - -#define S2_CACHE_ENTRY_DAT_RSZ 0x4 - -#define S2_CACHE_MASK_DAT_RSZ 0x4 - -#define S2_CACHE_ACTION_DAT_RSZ 0x4 - -#define S2_CACHE_CNT_DAT_RSZ 0x4 - -#define S2_STICKY_VCAP_ROW_DELETED_STICKY BIT(0) - -#define S2_BIST_CTRL_TCAM_BIST BIT(1) -#define S2_BIST_CTRL_TCAM_INIT BIT(0) - -#define S2_BIST_CFG_TCAM_BIST_SOE_ENA BIT(8) -#define S2_BIST_CFG_TCAM_HCG_DIS BIT(7) -#define S2_BIST_CFG_TCAM_CG_DIS BIT(6) -#define S2_BIST_CFG_TCAM_BIAS(x) ((x) & GENMASK(5, 0)) -#define S2_BIST_CFG_TCAM_BIAS_M GENMASK(5, 0) - -#define S2_BIST_STAT_BIST_RT_ERR BIT(15) -#define S2_BIST_STAT_BIST_PENC_ERR BIT(14) -#define S2_BIST_STAT_BIST_COMP_ERR BIT(13) -#define S2_BIST_STAT_BIST_ADDR_ERR BIT(12) -#define S2_BIST_STAT_BIST_BL1E_ERR BIT(11) -#define S2_BIST_STAT_BIST_BL1_ERR BIT(10) -#define S2_BIST_STAT_BIST_BL0E_ERR BIT(9) -#define S2_BIST_STAT_BIST_BL0_ERR BIT(8) -#define S2_BIST_STAT_BIST_PH1_ERR BIT(7) -#define S2_BIST_STAT_BIST_PH0_ERR BIT(6) -#define S2_BIST_STAT_BIST_PV1_ERR BIT(5) -#define S2_BIST_STAT_BIST_PV0_ERR BIT(4) -#define S2_BIST_STAT_BIST_RUN BIT(3) -#define S2_BIST_STAT_BIST_ERR BIT(2) -#define S2_BIST_STAT_BIST_BUSY BIT(1) -#define S2_BIST_STAT_TCAM_RDY BIT(0) - -#endif /* _OCELOT_S2_CORE_H_ */ diff --git a/drivers/net/ethernet/mscc/ocelot_vcap.c b/drivers/net/ethernet/mscc/ocelot_vcap.c index 3ef620faf995..d8c778ee6f1b 100644 --- a/drivers/net/ethernet/mscc/ocelot_vcap.c +++ b/drivers/net/ethernet/mscc/ocelot_vcap.c @@ -9,9 +9,7 @@ #include <soc/mscc/ocelot_vcap.h> #include "ocelot_police.h" #include "ocelot_vcap.h" -#include "ocelot_s2.h" -#define OCELOT_POLICER_DISCARD 0x17f #define ENTRY_WIDTH 32 enum vcap_sel { @@ -48,145 +46,174 @@ struct vcap_data { u32 tg_mask; /* Current type-group mask */ }; -static u32 vcap_s2_read_update_ctrl(struct ocelot *ocelot) +static u32 vcap_read_update_ctrl(struct ocelot *ocelot, + const struct vcap_props *vcap) { - return ocelot_read(ocelot, S2_CORE_UPDATE_CTRL); + return ocelot_target_read(ocelot, vcap->target, VCAP_CORE_UPDATE_CTRL); } -static void vcap_cmd(struct ocelot *ocelot, u16 ix, int cmd, int sel) +static void vcap_cmd(struct ocelot *ocelot, const struct vcap_props *vcap, + u16 ix, int cmd, int sel) { - const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; + u32 value = (VCAP_CORE_UPDATE_CTRL_UPDATE_CMD(cmd) | + VCAP_CORE_UPDATE_CTRL_UPDATE_ADDR(ix) | + VCAP_CORE_UPDATE_CTRL_UPDATE_SHOT); - u32 value = (S2_CORE_UPDATE_CTRL_UPDATE_CMD(cmd) | - S2_CORE_UPDATE_CTRL_UPDATE_ADDR(ix) | - S2_CORE_UPDATE_CTRL_UPDATE_SHOT); - - if ((sel & VCAP_SEL_ENTRY) && ix >= vcap_is2->entry_count) + if ((sel & VCAP_SEL_ENTRY) && ix >= vcap->entry_count) return; if (!(sel & VCAP_SEL_ENTRY)) - value |= S2_CORE_UPDATE_CTRL_UPDATE_ENTRY_DIS; + value |= VCAP_CORE_UPDATE_CTRL_UPDATE_ENTRY_DIS; if (!(sel & VCAP_SEL_ACTION)) - value |= S2_CORE_UPDATE_CTRL_UPDATE_ACTION_DIS; + value |= VCAP_CORE_UPDATE_CTRL_UPDATE_ACTION_DIS; if (!(sel & VCAP_SEL_COUNTER)) - value |= S2_CORE_UPDATE_CTRL_UPDATE_CNT_DIS; + value |= VCAP_CORE_UPDATE_CTRL_UPDATE_CNT_DIS; + + ocelot_target_write(ocelot, vcap->target, value, VCAP_CORE_UPDATE_CTRL); - ocelot_write(ocelot, value, S2_CORE_UPDATE_CTRL); - readx_poll_timeout(vcap_s2_read_update_ctrl, ocelot, value, - (value & S2_CORE_UPDATE_CTRL_UPDATE_SHOT) == 0, - 10, 100000); + read_poll_timeout(vcap_read_update_ctrl, value, + (value & VCAP_CORE_UPDATE_CTRL_UPDATE_SHOT) == 0, + 10, 100000, false, ocelot, vcap); } /* Convert from 0-based row to VCAP entry row and run command */ -static void vcap_row_cmd(struct ocelot *ocelot, u32 row, int cmd, int sel) +static void vcap_row_cmd(struct ocelot *ocelot, const struct vcap_props *vcap, + u32 row, int cmd, int sel) { - const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; - - vcap_cmd(ocelot, vcap_is2->entry_count - row - 1, cmd, sel); + vcap_cmd(ocelot, vcap, vcap->entry_count - row - 1, cmd, sel); } -static void vcap_entry2cache(struct ocelot *ocelot, struct vcap_data *data) +static void vcap_entry2cache(struct ocelot *ocelot, + const struct vcap_props *vcap, + struct vcap_data *data) { - const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; u32 entry_words, i; - entry_words = DIV_ROUND_UP(vcap_is2->entry_width, ENTRY_WIDTH); + entry_words = DIV_ROUND_UP(vcap->entry_width, ENTRY_WIDTH); for (i = 0; i < entry_words; i++) { - ocelot_write_rix(ocelot, data->entry[i], S2_CACHE_ENTRY_DAT, i); - ocelot_write_rix(ocelot, ~data->mask[i], S2_CACHE_MASK_DAT, i); + ocelot_target_write_rix(ocelot, vcap->target, data->entry[i], + VCAP_CACHE_ENTRY_DAT, i); + ocelot_target_write_rix(ocelot, vcap->target, ~data->mask[i], + VCAP_CACHE_MASK_DAT, i); } - ocelot_write(ocelot, data->tg, S2_CACHE_TG_DAT); + ocelot_target_write(ocelot, vcap->target, data->tg, VCAP_CACHE_TG_DAT); } -static void vcap_cache2entry(struct ocelot *ocelot, struct vcap_data *data) +static void vcap_cache2entry(struct ocelot *ocelot, + const struct vcap_props *vcap, + struct vcap_data *data) { - const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; u32 entry_words, i; - entry_words = DIV_ROUND_UP(vcap_is2->entry_width, ENTRY_WIDTH); + entry_words = DIV_ROUND_UP(vcap->entry_width, ENTRY_WIDTH); for (i = 0; i < entry_words; i++) { - data->entry[i] = ocelot_read_rix(ocelot, S2_CACHE_ENTRY_DAT, i); + data->entry[i] = ocelot_target_read_rix(ocelot, vcap->target, + VCAP_CACHE_ENTRY_DAT, i); // Invert mask - data->mask[i] = ~ocelot_read_rix(ocelot, S2_CACHE_MASK_DAT, i); + data->mask[i] = ~ocelot_target_read_rix(ocelot, vcap->target, + VCAP_CACHE_MASK_DAT, i); } - data->tg = ocelot_read(ocelot, S2_CACHE_TG_DAT); + data->tg = ocelot_target_read(ocelot, vcap->target, VCAP_CACHE_TG_DAT); } -static void vcap_action2cache(struct ocelot *ocelot, struct vcap_data *data) +static void vcap_action2cache(struct ocelot *ocelot, + const struct vcap_props *vcap, + struct vcap_data *data) { - const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; u32 action_words, mask; int i, width; /* Encode action type */ - width = vcap_is2->action_type_width; + width = vcap->action_type_width; if (width) { mask = GENMASK(width, 0); data->action[0] = ((data->action[0] & ~mask) | data->type); } - action_words = DIV_ROUND_UP(vcap_is2->action_width, ENTRY_WIDTH); + action_words = DIV_ROUND_UP(vcap->action_width, ENTRY_WIDTH); for (i = 0; i < action_words; i++) - ocelot_write_rix(ocelot, data->action[i], S2_CACHE_ACTION_DAT, - i); + ocelot_target_write_rix(ocelot, vcap->target, data->action[i], + VCAP_CACHE_ACTION_DAT, i); - for (i = 0; i < vcap_is2->counter_words; i++) - ocelot_write_rix(ocelot, data->counter[i], S2_CACHE_CNT_DAT, i); + for (i = 0; i < vcap->counter_words; i++) + ocelot_target_write_rix(ocelot, vcap->target, data->counter[i], + VCAP_CACHE_CNT_DAT, i); } -static void vcap_cache2action(struct ocelot *ocelot, struct vcap_data *data) +static void vcap_cache2action(struct ocelot *ocelot, + const struct vcap_props *vcap, + struct vcap_data *data) { - const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; u32 action_words; int i, width; - action_words = DIV_ROUND_UP(vcap_is2->action_width, ENTRY_WIDTH); + action_words = DIV_ROUND_UP(vcap->action_width, ENTRY_WIDTH); for (i = 0; i < action_words; i++) - data->action[i] = ocelot_read_rix(ocelot, S2_CACHE_ACTION_DAT, - i); + data->action[i] = ocelot_target_read_rix(ocelot, vcap->target, + VCAP_CACHE_ACTION_DAT, + i); - for (i = 0; i < vcap_is2->counter_words; i++) - data->counter[i] = ocelot_read_rix(ocelot, S2_CACHE_CNT_DAT, i); + for (i = 0; i < vcap->counter_words; i++) + data->counter[i] = ocelot_target_read_rix(ocelot, vcap->target, + VCAP_CACHE_CNT_DAT, + i); /* Extract action type */ - width = vcap_is2->action_type_width; + width = vcap->action_type_width; data->type = (width ? (data->action[0] & GENMASK(width, 0)) : 0); } /* Calculate offsets for entry */ -static void is2_data_get(struct ocelot *ocelot, struct vcap_data *data, int ix) +static void vcap_data_offset_get(const struct vcap_props *vcap, + struct vcap_data *data, int ix) { - const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; - int i, col, offset, count, cnt, base; - int width = vcap_is2->tg_width; + int num_subwords_per_entry, num_subwords_per_action; + int i, col, offset, num_entries_per_row, base; + u32 width = vcap->tg_width; - count = (data->tg_sw == VCAP_TG_HALF ? 2 : 4); - col = (ix % 2); - cnt = (vcap_is2->sw_count / count); - base = (vcap_is2->sw_count - col * cnt - cnt); + switch (data->tg_sw) { + case VCAP_TG_FULL: + num_entries_per_row = 1; + break; + case VCAP_TG_HALF: + num_entries_per_row = 2; + break; + case VCAP_TG_QUARTER: + num_entries_per_row = 4; + break; + default: + return; + } + + col = (ix % num_entries_per_row); + num_subwords_per_entry = (vcap->sw_count / num_entries_per_row); + base = (vcap->sw_count - col * num_subwords_per_entry - + num_subwords_per_entry); data->tg_value = 0; data->tg_mask = 0; - for (i = 0; i < cnt; i++) { + for (i = 0; i < num_subwords_per_entry; i++) { offset = ((base + i) * width); data->tg_value |= (data->tg_sw << offset); data->tg_mask |= GENMASK(offset + width - 1, offset); } /* Calculate key/action/counter offsets */ - col = (count - col - 1); - data->key_offset = (base * vcap_is2->entry_width) / vcap_is2->sw_count; - data->counter_offset = (cnt * col * vcap_is2->counter_width); + col = (num_entries_per_row - col - 1); + data->key_offset = (base * vcap->entry_width) / vcap->sw_count; + data->counter_offset = (num_subwords_per_entry * col * + vcap->counter_width); i = data->type; - width = vcap_is2->action_table[i].width; - cnt = vcap_is2->action_table[i].count; - data->action_offset = - (((cnt * col * width) / count) + vcap_is2->action_type_width); + width = vcap->action_table[i].width; + num_subwords_per_action = vcap->action_table[i].count; + data->action_offset = ((num_subwords_per_action * col * width) / + num_entries_per_row); + data->action_offset += vcap->action_type_width; } static void vcap_data_set(u32 *data, u32 offset, u32 len, u32 value) @@ -224,22 +251,21 @@ static void vcap_key_field_set(struct vcap_data *data, u32 offset, u32 width, vcap_data_set(data->mask, offset + data->key_offset, width, mask); } -static void vcap_key_set(struct ocelot *ocelot, struct vcap_data *data, - enum vcap_is2_half_key_field field, - u32 value, u32 mask) +static void vcap_key_set(const struct vcap_props *vcap, struct vcap_data *data, + int field, u32 value, u32 mask) { - u32 offset = ocelot->vcap_is2_keys[field].offset; - u32 length = ocelot->vcap_is2_keys[field].length; + u32 offset = vcap->keys[field].offset; + u32 length = vcap->keys[field].length; vcap_key_field_set(data, offset, length, value, mask); } -static void vcap_key_bytes_set(struct ocelot *ocelot, struct vcap_data *data, - enum vcap_is2_half_key_field field, +static void vcap_key_bytes_set(const struct vcap_props *vcap, + struct vcap_data *data, int field, u8 *val, u8 *msk) { - u32 offset = ocelot->vcap_is2_keys[field].offset; - u32 count = ocelot->vcap_is2_keys[field].length; + u32 offset = vcap->keys[field].offset; + u32 count = vcap->keys[field].length; u32 i, j, n = 0, value = 0, mask = 0; WARN_ON(count % 8); @@ -265,37 +291,37 @@ static void vcap_key_bytes_set(struct ocelot *ocelot, struct vcap_data *data, } } -static void vcap_key_l4_port_set(struct ocelot *ocelot, struct vcap_data *data, - enum vcap_is2_half_key_field field, +static void vcap_key_l4_port_set(const struct vcap_props *vcap, + struct vcap_data *data, int field, struct ocelot_vcap_udp_tcp *port) { - u32 offset = ocelot->vcap_is2_keys[field].offset; - u32 length = ocelot->vcap_is2_keys[field].length; + u32 offset = vcap->keys[field].offset; + u32 length = vcap->keys[field].length; WARN_ON(length != 16); vcap_key_field_set(data, offset, length, port->value, port->mask); } -static void vcap_key_bit_set(struct ocelot *ocelot, struct vcap_data *data, - enum vcap_is2_half_key_field field, +static void vcap_key_bit_set(const struct vcap_props *vcap, + struct vcap_data *data, int field, enum ocelot_vcap_bit val) { - u32 offset = ocelot->vcap_is2_keys[field].offset; - u32 length = ocelot->vcap_is2_keys[field].length; u32 value = (val == OCELOT_VCAP_BIT_1 ? 1 : 0); u32 msk = (val == OCELOT_VCAP_BIT_ANY ? 0 : 1); + u32 offset = vcap->keys[field].offset; + u32 length = vcap->keys[field].length; WARN_ON(length != 1); vcap_key_field_set(data, offset, length, value, msk); } -static void vcap_action_set(struct ocelot *ocelot, struct vcap_data *data, - enum vcap_is2_action_field field, u32 value) +static void vcap_action_set(const struct vcap_props *vcap, + struct vcap_data *data, int field, u32 value) { - int offset = ocelot->vcap_is2_actions[field].offset; - int length = ocelot->vcap_is2_actions[field].length; + int offset = vcap->actions[field].offset; + int length = vcap->actions[field].length; vcap_data_set(data->action, offset + data->action_offset, length, value); @@ -304,40 +330,21 @@ static void vcap_action_set(struct ocelot *ocelot, struct vcap_data *data, static void is2_action_set(struct ocelot *ocelot, struct vcap_data *data, struct ocelot_vcap_filter *filter) { - switch (filter->action) { - case OCELOT_VCAP_ACTION_DROP: - vcap_action_set(ocelot, data, VCAP_IS2_ACT_PORT_MASK, 0); - vcap_action_set(ocelot, data, VCAP_IS2_ACT_MASK_MODE, 1); - vcap_action_set(ocelot, data, VCAP_IS2_ACT_POLICE_ENA, 1); - vcap_action_set(ocelot, data, VCAP_IS2_ACT_POLICE_IDX, - OCELOT_POLICER_DISCARD); - vcap_action_set(ocelot, data, VCAP_IS2_ACT_CPU_QU_NUM, 0); - vcap_action_set(ocelot, data, VCAP_IS2_ACT_CPU_COPY_ENA, 0); - break; - case OCELOT_VCAP_ACTION_TRAP: - vcap_action_set(ocelot, data, VCAP_IS2_ACT_PORT_MASK, 0); - vcap_action_set(ocelot, data, VCAP_IS2_ACT_MASK_MODE, 1); - vcap_action_set(ocelot, data, VCAP_IS2_ACT_POLICE_ENA, 0); - vcap_action_set(ocelot, data, VCAP_IS2_ACT_POLICE_IDX, 0); - vcap_action_set(ocelot, data, VCAP_IS2_ACT_CPU_QU_NUM, 0); - vcap_action_set(ocelot, data, VCAP_IS2_ACT_CPU_COPY_ENA, 1); - break; - case OCELOT_VCAP_ACTION_POLICE: - vcap_action_set(ocelot, data, VCAP_IS2_ACT_PORT_MASK, 0); - vcap_action_set(ocelot, data, VCAP_IS2_ACT_MASK_MODE, 0); - vcap_action_set(ocelot, data, VCAP_IS2_ACT_POLICE_ENA, 1); - vcap_action_set(ocelot, data, VCAP_IS2_ACT_POLICE_IDX, - filter->pol_ix); - vcap_action_set(ocelot, data, VCAP_IS2_ACT_CPU_QU_NUM, 0); - vcap_action_set(ocelot, data, VCAP_IS2_ACT_CPU_COPY_ENA, 0); - break; - } + const struct vcap_props *vcap = &ocelot->vcap[VCAP_IS2]; + struct ocelot_vcap_action *a = &filter->action; + + vcap_action_set(vcap, data, VCAP_IS2_ACT_MASK_MODE, a->mask_mode); + vcap_action_set(vcap, data, VCAP_IS2_ACT_PORT_MASK, a->port_mask); + vcap_action_set(vcap, data, VCAP_IS2_ACT_POLICE_ENA, a->police_ena); + vcap_action_set(vcap, data, VCAP_IS2_ACT_POLICE_IDX, a->pol_ix); + vcap_action_set(vcap, data, VCAP_IS2_ACT_CPU_QU_NUM, a->cpu_qu_num); + vcap_action_set(vcap, data, VCAP_IS2_ACT_CPU_COPY_ENA, a->cpu_copy_ena); } static void is2_entry_set(struct ocelot *ocelot, int ix, struct ocelot_vcap_filter *filter) { - const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; + const struct vcap_props *vcap = &ocelot->vcap[VCAP_IS2]; struct ocelot_vcap_key_vlan *tag = &filter->vlan; u32 val, msk, type, type_mask = 0xf, i, count; struct ocelot_vcap_u64 payload; @@ -348,52 +355,55 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, memset(&data, 0, sizeof(data)); /* Read row */ - vcap_row_cmd(ocelot, row, VCAP_CMD_READ, VCAP_SEL_ALL); - vcap_cache2entry(ocelot, &data); - vcap_cache2action(ocelot, &data); + vcap_row_cmd(ocelot, vcap, row, VCAP_CMD_READ, VCAP_SEL_ALL); + vcap_cache2entry(ocelot, vcap, &data); + vcap_cache2action(ocelot, vcap, &data); data.tg_sw = VCAP_TG_HALF; - is2_data_get(ocelot, &data, ix); + vcap_data_offset_get(vcap, &data, ix); data.tg = (data.tg & ~data.tg_mask); if (filter->prio != 0) data.tg |= data.tg_value; data.type = IS2_ACTION_TYPE_NORMAL; - vcap_key_set(ocelot, &data, VCAP_IS2_HK_PAG, 0, 0); - vcap_key_set(ocelot, &data, VCAP_IS2_HK_IGR_PORT_MASK, 0, + vcap_key_set(vcap, &data, VCAP_IS2_HK_PAG, filter->pag, 0xff); + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_FIRST, + (filter->lookup == 0) ? OCELOT_VCAP_BIT_1 : + OCELOT_VCAP_BIT_0); + vcap_key_set(vcap, &data, VCAP_IS2_HK_IGR_PORT_MASK, 0, ~filter->ingress_port_mask); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_FIRST, OCELOT_VCAP_BIT_1); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_HOST_MATCH, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_FIRST, OCELOT_VCAP_BIT_ANY); + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_HOST_MATCH, OCELOT_VCAP_BIT_ANY); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L2_MC, filter->dmac_mc); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L2_BC, filter->dmac_bc); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_VLAN_TAGGED, tag->tagged); - vcap_key_set(ocelot, &data, VCAP_IS2_HK_VID, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_L2_MC, filter->dmac_mc); + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_L2_BC, filter->dmac_bc); + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_VLAN_TAGGED, tag->tagged); + vcap_key_set(vcap, &data, VCAP_IS2_HK_VID, tag->vid.value, tag->vid.mask); - vcap_key_set(ocelot, &data, VCAP_IS2_HK_PCP, + vcap_key_set(vcap, &data, VCAP_IS2_HK_PCP, tag->pcp.value[0], tag->pcp.mask[0]); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_DEI, tag->dei); + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_DEI, tag->dei); switch (filter->key_type) { case OCELOT_VCAP_KEY_ETYPE: { struct ocelot_vcap_key_etype *etype = &filter->key.etype; type = IS2_TYPE_ETYPE; - vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L2_DMAC, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_L2_DMAC, etype->dmac.value, etype->dmac.mask); - vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L2_SMAC, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_L2_SMAC, etype->smac.value, etype->smac.mask); - vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_MAC_ETYPE_ETYPE, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_MAC_ETYPE_ETYPE, etype->etype.value, etype->etype.mask); /* Clear unused bits */ - vcap_key_set(ocelot, &data, VCAP_IS2_HK_MAC_ETYPE_L2_PAYLOAD0, + vcap_key_set(vcap, &data, VCAP_IS2_HK_MAC_ETYPE_L2_PAYLOAD0, 0, 0); - vcap_key_set(ocelot, &data, VCAP_IS2_HK_MAC_ETYPE_L2_PAYLOAD1, + vcap_key_set(vcap, &data, VCAP_IS2_HK_MAC_ETYPE_L2_PAYLOAD1, 0, 0); - vcap_key_set(ocelot, &data, VCAP_IS2_HK_MAC_ETYPE_L2_PAYLOAD2, + vcap_key_set(vcap, &data, VCAP_IS2_HK_MAC_ETYPE_L2_PAYLOAD2, 0, 0); - vcap_key_bytes_set(ocelot, &data, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_MAC_ETYPE_L2_PAYLOAD0, etype->data.value, etype->data.mask); break; @@ -402,15 +412,15 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, struct ocelot_vcap_key_llc *llc = &filter->key.llc; type = IS2_TYPE_LLC; - vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L2_DMAC, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_L2_DMAC, llc->dmac.value, llc->dmac.mask); - vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L2_SMAC, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_L2_SMAC, llc->smac.value, llc->smac.mask); for (i = 0; i < 4; i++) { payload.value[i] = llc->llc.value[i]; payload.mask[i] = llc->llc.mask[i]; } - vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_MAC_LLC_L2_LLC, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_MAC_LLC_L2_LLC, payload.value, payload.mask); break; } @@ -418,11 +428,11 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, struct ocelot_vcap_key_snap *snap = &filter->key.snap; type = IS2_TYPE_SNAP; - vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L2_DMAC, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_L2_DMAC, snap->dmac.value, snap->dmac.mask); - vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L2_SMAC, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_L2_SMAC, snap->smac.value, snap->smac.mask); - vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_MAC_SNAP_L2_SNAP, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_MAC_SNAP_L2_SNAP, filter->key.snap.snap.value, filter->key.snap.snap.mask); break; @@ -431,24 +441,24 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, struct ocelot_vcap_key_arp *arp = &filter->key.arp; type = IS2_TYPE_ARP; - vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_MAC_ARP_SMAC, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_MAC_ARP_SMAC, arp->smac.value, arp->smac.mask); - vcap_key_bit_set(ocelot, &data, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_MAC_ARP_ADDR_SPACE_OK, arp->ethernet); - vcap_key_bit_set(ocelot, &data, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_MAC_ARP_PROTO_SPACE_OK, arp->ip); - vcap_key_bit_set(ocelot, &data, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_MAC_ARP_LEN_OK, arp->length); - vcap_key_bit_set(ocelot, &data, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_MAC_ARP_TARGET_MATCH, arp->dmac_match); - vcap_key_bit_set(ocelot, &data, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_MAC_ARP_SENDER_MATCH, arp->smac_match); - vcap_key_bit_set(ocelot, &data, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_MAC_ARP_OPCODE_UNKNOWN, arp->unknown); @@ -457,15 +467,15 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, (arp->arp == OCELOT_VCAP_BIT_0 ? 2 : 0)); msk = ((arp->req == OCELOT_VCAP_BIT_ANY ? 0 : 1) | (arp->arp == OCELOT_VCAP_BIT_ANY ? 0 : 2)); - vcap_key_set(ocelot, &data, VCAP_IS2_HK_MAC_ARP_OPCODE, + vcap_key_set(vcap, &data, VCAP_IS2_HK_MAC_ARP_OPCODE, val, msk); - vcap_key_bytes_set(ocelot, &data, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_MAC_ARP_L3_IP4_DIP, arp->dip.value.addr, arp->dip.mask.addr); - vcap_key_bytes_set(ocelot, &data, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_MAC_ARP_L3_IP4_SIP, arp->sip.value.addr, arp->sip.mask.addr); - vcap_key_set(ocelot, &data, VCAP_IS2_HK_MAC_ARP_DIP_EQ_SIP, + vcap_key_set(vcap, &data, VCAP_IS2_HK_MAC_ARP_DIP_EQ_SIP, 0, 0); break; } @@ -534,22 +544,22 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, seq_zero = ipv6->seq_zero; } - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_IP4, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_IP4, ipv4 ? OCELOT_VCAP_BIT_1 : OCELOT_VCAP_BIT_0); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L3_FRAGMENT, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_L3_FRAGMENT, fragment); - vcap_key_set(ocelot, &data, VCAP_IS2_HK_L3_FRAG_OFS_GT0, 0, 0); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L3_OPTIONS, + vcap_key_set(vcap, &data, VCAP_IS2_HK_L3_FRAG_OFS_GT0, 0, 0); + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_L3_OPTIONS, options); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_IP4_L3_TTL_GT0, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_IP4_L3_TTL_GT0, ttl); - vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L3_TOS, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_L3_TOS, ds.value, ds.mask); - vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L3_IP4_DIP, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_L3_IP4_DIP, dip.value.addr, dip.mask.addr); - vcap_key_bytes_set(ocelot, &data, VCAP_IS2_HK_L3_IP4_SIP, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_L3_IP4_SIP, sip.value.addr, sip.mask.addr); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_DIP_EQ_SIP, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_DIP_EQ_SIP, sip_eq_dip); val = proto.value[0]; msk = proto.mask[0]; @@ -558,33 +568,33 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, /* UDP/TCP protocol match */ tcp = (val == 6 ? OCELOT_VCAP_BIT_1 : OCELOT_VCAP_BIT_0); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_TCP, tcp); - vcap_key_l4_port_set(ocelot, &data, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_TCP, tcp); + vcap_key_l4_port_set(vcap, &data, VCAP_IS2_HK_L4_DPORT, dport); - vcap_key_l4_port_set(ocelot, &data, + vcap_key_l4_port_set(vcap, &data, VCAP_IS2_HK_L4_SPORT, sport); - vcap_key_set(ocelot, &data, VCAP_IS2_HK_L4_RNG, 0, 0); - vcap_key_bit_set(ocelot, &data, + vcap_key_set(vcap, &data, VCAP_IS2_HK_L4_RNG, 0, 0); + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_L4_SPORT_EQ_DPORT, sport_eq_dport); - vcap_key_bit_set(ocelot, &data, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_L4_SEQUENCE_EQ0, seq_zero); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L4_FIN, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_L4_FIN, tcp_fin); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L4_SYN, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_L4_SYN, tcp_syn); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L4_RST, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_L4_RST, tcp_rst); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L4_PSH, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_L4_PSH, tcp_psh); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L4_ACK, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_L4_ACK, tcp_ack); - vcap_key_bit_set(ocelot, &data, VCAP_IS2_HK_L4_URG, + vcap_key_bit_set(vcap, &data, VCAP_IS2_HK_L4_URG, tcp_urg); - vcap_key_set(ocelot, &data, VCAP_IS2_HK_L4_1588_DOM, + vcap_key_set(vcap, &data, VCAP_IS2_HK_L4_1588_DOM, 0, 0); - vcap_key_set(ocelot, &data, VCAP_IS2_HK_L4_1588_VER, + vcap_key_set(vcap, &data, VCAP_IS2_HK_L4_1588_VER, 0, 0); } else { if (msk == 0) { @@ -598,10 +608,10 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, payload.mask[i] = ip_data->mask[i]; } } - vcap_key_bytes_set(ocelot, &data, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_IP4_L3_PROTO, proto.value, proto.mask); - vcap_key_bytes_set(ocelot, &data, + vcap_key_bytes_set(vcap, &data, VCAP_IS2_HK_L3_PAYLOAD, payload.value, payload.mask); } @@ -611,46 +621,271 @@ static void is2_entry_set(struct ocelot *ocelot, int ix, default: type = 0; type_mask = 0; - count = vcap_is2->entry_width / 2; + count = vcap->entry_width / 2; /* Iterate over the non-common part of the key and * clear entry data */ - for (i = ocelot->vcap_is2_keys[VCAP_IS2_HK_L2_DMAC].offset; + for (i = vcap->keys[VCAP_IS2_HK_L2_DMAC].offset; i < count; i += ENTRY_WIDTH) { vcap_key_field_set(&data, i, min(32u, count - i), 0, 0); } break; } - vcap_key_set(ocelot, &data, VCAP_IS2_TYPE, type, type_mask); + vcap_key_set(vcap, &data, VCAP_IS2_TYPE, type, type_mask); is2_action_set(ocelot, &data, filter); vcap_data_set(data.counter, data.counter_offset, - vcap_is2->counter_width, filter->stats.pkts); + vcap->counter_width, filter->stats.pkts); /* Write row */ - vcap_entry2cache(ocelot, &data); - vcap_action2cache(ocelot, &data); - vcap_row_cmd(ocelot, row, VCAP_CMD_WRITE, VCAP_SEL_ALL); + vcap_entry2cache(ocelot, vcap, &data); + vcap_action2cache(ocelot, vcap, &data); + vcap_row_cmd(ocelot, vcap, row, VCAP_CMD_WRITE, VCAP_SEL_ALL); +} + +static void is1_action_set(struct ocelot *ocelot, struct vcap_data *data, + const struct ocelot_vcap_filter *filter) +{ + const struct vcap_props *vcap = &ocelot->vcap[VCAP_IS1]; + const struct ocelot_vcap_action *a = &filter->action; + + vcap_action_set(vcap, data, VCAP_IS1_ACT_VID_REPLACE_ENA, + a->vid_replace_ena); + vcap_action_set(vcap, data, VCAP_IS1_ACT_VID_ADD_VAL, a->vid); + vcap_action_set(vcap, data, VCAP_IS1_ACT_VLAN_POP_CNT_ENA, + a->vlan_pop_cnt_ena); + vcap_action_set(vcap, data, VCAP_IS1_ACT_VLAN_POP_CNT, + a->vlan_pop_cnt); + vcap_action_set(vcap, data, VCAP_IS1_ACT_PCP_DEI_ENA, a->pcp_dei_ena); + vcap_action_set(vcap, data, VCAP_IS1_ACT_PCP_VAL, a->pcp); + vcap_action_set(vcap, data, VCAP_IS1_ACT_DEI_VAL, a->dei); + vcap_action_set(vcap, data, VCAP_IS1_ACT_QOS_ENA, a->qos_ena); + vcap_action_set(vcap, data, VCAP_IS1_ACT_QOS_VAL, a->qos_val); + vcap_action_set(vcap, data, VCAP_IS1_ACT_PAG_OVERRIDE_MASK, + a->pag_override_mask); + vcap_action_set(vcap, data, VCAP_IS1_ACT_PAG_VAL, a->pag_val); } -static void is2_entry_get(struct ocelot *ocelot, struct ocelot_vcap_filter *filter, - int ix) +static void is1_entry_set(struct ocelot *ocelot, int ix, + struct ocelot_vcap_filter *filter) { - const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; + const struct vcap_props *vcap = &ocelot->vcap[VCAP_IS1]; + struct ocelot_vcap_key_vlan *tag = &filter->vlan; + struct ocelot_vcap_u64 payload; struct vcap_data data; - int row = (ix / 2); - u32 cnt; + int row = ix / 2; + u32 type; + + memset(&payload, 0, sizeof(payload)); + memset(&data, 0, sizeof(data)); + + /* Read row */ + vcap_row_cmd(ocelot, vcap, row, VCAP_CMD_READ, VCAP_SEL_ALL); + vcap_cache2entry(ocelot, vcap, &data); + vcap_cache2action(ocelot, vcap, &data); - vcap_row_cmd(ocelot, row, VCAP_CMD_READ, VCAP_SEL_COUNTER); - vcap_cache2action(ocelot, &data); data.tg_sw = VCAP_TG_HALF; - is2_data_get(ocelot, &data, ix); + data.type = IS1_ACTION_TYPE_NORMAL; + vcap_data_offset_get(vcap, &data, ix); + data.tg = (data.tg & ~data.tg_mask); + if (filter->prio != 0) + data.tg |= data.tg_value; + + vcap_key_set(vcap, &data, VCAP_IS1_HK_LOOKUP, filter->lookup, 0x3); + vcap_key_set(vcap, &data, VCAP_IS1_HK_IGR_PORT_MASK, 0, + ~filter->ingress_port_mask); + vcap_key_bit_set(vcap, &data, VCAP_IS1_HK_L2_MC, filter->dmac_mc); + vcap_key_bit_set(vcap, &data, VCAP_IS1_HK_L2_BC, filter->dmac_bc); + vcap_key_bit_set(vcap, &data, VCAP_IS1_HK_VLAN_TAGGED, tag->tagged); + vcap_key_set(vcap, &data, VCAP_IS1_HK_VID, + tag->vid.value, tag->vid.mask); + vcap_key_set(vcap, &data, VCAP_IS1_HK_PCP, + tag->pcp.value[0], tag->pcp.mask[0]); + type = IS1_TYPE_S1_NORMAL; + + switch (filter->key_type) { + case OCELOT_VCAP_KEY_ETYPE: { + struct ocelot_vcap_key_etype *etype = &filter->key.etype; + + vcap_key_bytes_set(vcap, &data, VCAP_IS1_HK_L2_SMAC, + etype->smac.value, etype->smac.mask); + vcap_key_bytes_set(vcap, &data, VCAP_IS1_HK_ETYPE, + etype->etype.value, etype->etype.mask); + break; + } + case OCELOT_VCAP_KEY_IPV4: { + struct ocelot_vcap_key_ipv4 *ipv4 = &filter->key.ipv4; + struct ocelot_vcap_udp_tcp *sport = &ipv4->sport; + struct ocelot_vcap_udp_tcp *dport = &ipv4->dport; + enum ocelot_vcap_bit tcp_udp = OCELOT_VCAP_BIT_0; + struct ocelot_vcap_u8 proto = ipv4->proto; + struct ocelot_vcap_ipv4 sip = ipv4->sip; + u32 val, msk; + + vcap_key_bit_set(vcap, &data, VCAP_IS1_HK_IP_SNAP, + OCELOT_VCAP_BIT_1); + vcap_key_bit_set(vcap, &data, VCAP_IS1_HK_IP4, + OCELOT_VCAP_BIT_1); + vcap_key_bit_set(vcap, &data, VCAP_IS1_HK_ETYPE_LEN, + OCELOT_VCAP_BIT_1); + vcap_key_bytes_set(vcap, &data, VCAP_IS1_HK_L3_IP4_SIP, + sip.value.addr, sip.mask.addr); + + val = proto.value[0]; + msk = proto.mask[0]; + + if ((val == NEXTHDR_TCP || val == NEXTHDR_UDP) && msk == 0xff) + tcp_udp = OCELOT_VCAP_BIT_1; + vcap_key_bit_set(vcap, &data, VCAP_IS1_HK_TCP_UDP, tcp_udp); + + if (tcp_udp) { + enum ocelot_vcap_bit tcp = OCELOT_VCAP_BIT_0; + + if (val == NEXTHDR_TCP) + tcp = OCELOT_VCAP_BIT_1; + + vcap_key_bit_set(vcap, &data, VCAP_IS1_HK_TCP, tcp); + vcap_key_l4_port_set(vcap, &data, VCAP_IS1_HK_L4_SPORT, + sport); + /* Overloaded field */ + vcap_key_l4_port_set(vcap, &data, VCAP_IS1_HK_ETYPE, + dport); + } else { + /* IPv4 "other" frame */ + struct ocelot_vcap_u16 etype = {0}; + + /* Overloaded field */ + etype.value[0] = proto.value[0]; + etype.mask[0] = proto.mask[0]; + + vcap_key_bytes_set(vcap, &data, VCAP_IS1_HK_ETYPE, + etype.value, etype.mask); + } + } + default: + break; + } + vcap_key_bit_set(vcap, &data, VCAP_IS1_HK_TYPE, + type ? OCELOT_VCAP_BIT_1 : OCELOT_VCAP_BIT_0); + + is1_action_set(ocelot, &data, filter); + vcap_data_set(data.counter, data.counter_offset, + vcap->counter_width, filter->stats.pkts); + + /* Write row */ + vcap_entry2cache(ocelot, vcap, &data); + vcap_action2cache(ocelot, vcap, &data); + vcap_row_cmd(ocelot, vcap, row, VCAP_CMD_WRITE, VCAP_SEL_ALL); +} + +static void es0_action_set(struct ocelot *ocelot, struct vcap_data *data, + const struct ocelot_vcap_filter *filter) +{ + const struct vcap_props *vcap = &ocelot->vcap[VCAP_ES0]; + const struct ocelot_vcap_action *a = &filter->action; + + vcap_action_set(vcap, data, VCAP_ES0_ACT_PUSH_OUTER_TAG, + a->push_outer_tag); + vcap_action_set(vcap, data, VCAP_ES0_ACT_PUSH_INNER_TAG, + a->push_inner_tag); + vcap_action_set(vcap, data, VCAP_ES0_ACT_TAG_A_TPID_SEL, + a->tag_a_tpid_sel); + vcap_action_set(vcap, data, VCAP_ES0_ACT_TAG_A_VID_SEL, + a->tag_a_vid_sel); + vcap_action_set(vcap, data, VCAP_ES0_ACT_TAG_A_PCP_SEL, + a->tag_a_pcp_sel); + vcap_action_set(vcap, data, VCAP_ES0_ACT_VID_A_VAL, a->vid_a_val); + vcap_action_set(vcap, data, VCAP_ES0_ACT_PCP_A_VAL, a->pcp_a_val); + vcap_action_set(vcap, data, VCAP_ES0_ACT_TAG_B_TPID_SEL, + a->tag_b_tpid_sel); + vcap_action_set(vcap, data, VCAP_ES0_ACT_TAG_B_VID_SEL, + a->tag_b_vid_sel); + vcap_action_set(vcap, data, VCAP_ES0_ACT_TAG_B_PCP_SEL, + a->tag_b_pcp_sel); + vcap_action_set(vcap, data, VCAP_ES0_ACT_VID_B_VAL, a->vid_b_val); + vcap_action_set(vcap, data, VCAP_ES0_ACT_PCP_B_VAL, a->pcp_b_val); +} + +static void es0_entry_set(struct ocelot *ocelot, int ix, + struct ocelot_vcap_filter *filter) +{ + const struct vcap_props *vcap = &ocelot->vcap[VCAP_ES0]; + struct ocelot_vcap_key_vlan *tag = &filter->vlan; + struct ocelot_vcap_u64 payload; + struct vcap_data data; + int row = ix; + + memset(&payload, 0, sizeof(payload)); + memset(&data, 0, sizeof(data)); + + /* Read row */ + vcap_row_cmd(ocelot, vcap, row, VCAP_CMD_READ, VCAP_SEL_ALL); + vcap_cache2entry(ocelot, vcap, &data); + vcap_cache2action(ocelot, vcap, &data); + + data.tg_sw = VCAP_TG_FULL; + data.type = ES0_ACTION_TYPE_NORMAL; + vcap_data_offset_get(vcap, &data, ix); + data.tg = (data.tg & ~data.tg_mask); + if (filter->prio != 0) + data.tg |= data.tg_value; + + vcap_key_set(vcap, &data, VCAP_ES0_IGR_PORT, filter->ingress_port.value, + filter->ingress_port.mask); + vcap_key_set(vcap, &data, VCAP_ES0_EGR_PORT, filter->egress_port.value, + filter->egress_port.mask); + vcap_key_bit_set(vcap, &data, VCAP_ES0_L2_MC, filter->dmac_mc); + vcap_key_bit_set(vcap, &data, VCAP_ES0_L2_BC, filter->dmac_bc); + vcap_key_set(vcap, &data, VCAP_ES0_VID, + tag->vid.value, tag->vid.mask); + vcap_key_set(vcap, &data, VCAP_ES0_PCP, + tag->pcp.value[0], tag->pcp.mask[0]); + + es0_action_set(ocelot, &data, filter); + vcap_data_set(data.counter, data.counter_offset, + vcap->counter_width, filter->stats.pkts); + + /* Write row */ + vcap_entry2cache(ocelot, vcap, &data); + vcap_action2cache(ocelot, vcap, &data); + vcap_row_cmd(ocelot, vcap, row, VCAP_CMD_WRITE, VCAP_SEL_ALL); +} + +static void vcap_entry_get(struct ocelot *ocelot, int ix, + struct ocelot_vcap_filter *filter) +{ + const struct vcap_props *vcap = &ocelot->vcap[filter->block_id]; + struct vcap_data data; + int row, count; + u32 cnt; + + if (filter->block_id == VCAP_ES0) + data.tg_sw = VCAP_TG_FULL; + else + data.tg_sw = VCAP_TG_HALF; + + count = (1 << (data.tg_sw - 1)); + row = (ix / count); + vcap_row_cmd(ocelot, vcap, row, VCAP_CMD_READ, VCAP_SEL_COUNTER); + vcap_cache2action(ocelot, vcap, &data); + vcap_data_offset_get(vcap, &data, ix); cnt = vcap_data_get(data.counter, data.counter_offset, - vcap_is2->counter_width); + vcap->counter_width); filter->stats.pkts = cnt; } +static void vcap_entry_set(struct ocelot *ocelot, int ix, + struct ocelot_vcap_filter *filter) +{ + if (filter->block_id == VCAP_IS1) + return is1_entry_set(ocelot, ix, filter); + if (filter->block_id == VCAP_IS2) + return is2_entry_set(ocelot, ix, filter); + if (filter->block_id == VCAP_ES0) + return es0_entry_set(ocelot, ix, filter); +} + static int ocelot_vcap_policer_add(struct ocelot *ocelot, u32 pol_ix, struct ocelot_policer *pol) { @@ -679,11 +914,12 @@ static void ocelot_vcap_policer_del(struct ocelot *ocelot, list_for_each_entry(filter, &block->rules, list) { index++; - if (filter->action == OCELOT_VCAP_ACTION_POLICE && - filter->pol_ix < pol_ix) { - filter->pol_ix += 1; - ocelot_vcap_policer_add(ocelot, filter->pol_ix, - &filter->pol); + if (filter->block_id == VCAP_IS2 && + filter->action.police_ena && + filter->action.pol_ix < pol_ix) { + filter->action.pol_ix += 1; + ocelot_vcap_policer_add(ocelot, filter->action.pol_ix, + &filter->action.pol); is2_entry_set(ocelot, index, filter); } } @@ -701,10 +937,11 @@ static void ocelot_vcap_filter_add_to_block(struct ocelot *ocelot, struct ocelot_vcap_filter *tmp; struct list_head *pos, *n; - if (filter->action == OCELOT_VCAP_ACTION_POLICE) { + if (filter->block_id == VCAP_IS2 && filter->action.police_ena) { block->pol_lpr--; - filter->pol_ix = block->pol_lpr; - ocelot_vcap_policer_add(ocelot, filter->pol_ix, &filter->pol); + filter->action.pol_ix = block->pol_lpr; + ocelot_vcap_policer_add(ocelot, filter->action.pol_ix, + &filter->action.pol); } block->count++; @@ -726,19 +963,20 @@ static int ocelot_vcap_block_get_filter_index(struct ocelot_vcap_block *block, struct ocelot_vcap_filter *filter) { struct ocelot_vcap_filter *tmp; - int index = -1; + int index = 0; list_for_each_entry(tmp, &block->rules, list) { - ++index; if (filter->id == tmp->id) - break; + return index; + index++; } - return index; + + return -ENOENT; } static struct ocelot_vcap_filter* -ocelot_vcap_block_find_filter(struct ocelot_vcap_block *block, - int index) +ocelot_vcap_block_find_filter_by_index(struct ocelot_vcap_block *block, + int index) { struct ocelot_vcap_filter *tmp; int i = 0; @@ -752,6 +990,18 @@ ocelot_vcap_block_find_filter(struct ocelot_vcap_block *block, return NULL; } +struct ocelot_vcap_filter * +ocelot_vcap_block_find_filter_by_id(struct ocelot_vcap_block *block, int id) +{ + struct ocelot_vcap_filter *filter; + + list_for_each_entry(filter, &block->rules, list) + if (filter->id == id) + return filter; + + return NULL; +} + /* If @on=false, then SNAP, ARP, IP and OAM frames will not match on keys based * on destination and source MAC addresses, but only on higher-level protocol * information. The only frame types to match on keys containing MAC addresses @@ -763,23 +1013,23 @@ ocelot_vcap_block_find_filter(struct ocelot_vcap_block *block, * on any _other_ keys than MAC_ETYPE ones. */ static void ocelot_match_all_as_mac_etype(struct ocelot *ocelot, int port, - bool on) + int lookup, bool on) { u32 val = 0; if (on) - val = ANA_PORT_VCAP_S2_CFG_S2_SNAP_DIS(3) | - ANA_PORT_VCAP_S2_CFG_S2_ARP_DIS(3) | - ANA_PORT_VCAP_S2_CFG_S2_IP_TCPUDP_DIS(3) | - ANA_PORT_VCAP_S2_CFG_S2_IP_OTHER_DIS(3) | - ANA_PORT_VCAP_S2_CFG_S2_OAM_DIS(3); + val = ANA_PORT_VCAP_S2_CFG_S2_SNAP_DIS(BIT(lookup)) | + ANA_PORT_VCAP_S2_CFG_S2_ARP_DIS(BIT(lookup)) | + ANA_PORT_VCAP_S2_CFG_S2_IP_TCPUDP_DIS(BIT(lookup)) | + ANA_PORT_VCAP_S2_CFG_S2_IP_OTHER_DIS(BIT(lookup)) | + ANA_PORT_VCAP_S2_CFG_S2_OAM_DIS(BIT(lookup)); ocelot_rmw_gix(ocelot, val, - ANA_PORT_VCAP_S2_CFG_S2_SNAP_DIS_M | - ANA_PORT_VCAP_S2_CFG_S2_ARP_DIS_M | - ANA_PORT_VCAP_S2_CFG_S2_IP_TCPUDP_DIS_M | - ANA_PORT_VCAP_S2_CFG_S2_IP_OTHER_DIS_M | - ANA_PORT_VCAP_S2_CFG_S2_OAM_DIS_M, + ANA_PORT_VCAP_S2_CFG_S2_SNAP_DIS(BIT(lookup)) | + ANA_PORT_VCAP_S2_CFG_S2_ARP_DIS(BIT(lookup)) | + ANA_PORT_VCAP_S2_CFG_S2_IP_TCPUDP_DIS(BIT(lookup)) | + ANA_PORT_VCAP_S2_CFG_S2_IP_OTHER_DIS(BIT(lookup)) | + ANA_PORT_VCAP_S2_CFG_S2_OAM_DIS(BIT(lookup)), ANA_PORT_VCAP_S2_CFG, port); } @@ -825,35 +1075,43 @@ static bool ocelot_exclusive_mac_etype_filter_rules(struct ocelot *ocelot, struct ocelot_vcap_filter *filter) { - struct ocelot_vcap_block *block = &ocelot->block; + struct ocelot_vcap_block *block = &ocelot->block[filter->block_id]; struct ocelot_vcap_filter *tmp; unsigned long port; int i; + /* We only have the S2_IP_TCPUDP_DIS set of knobs for VCAP IS2 */ + if (filter->block_id != VCAP_IS2) + return true; + if (ocelot_vcap_is_problematic_mac_etype(filter)) { /* Search for any non-MAC_ETYPE rules on the port */ for (i = 0; i < block->count; i++) { - tmp = ocelot_vcap_block_find_filter(block, i); + tmp = ocelot_vcap_block_find_filter_by_index(block, i); if (tmp->ingress_port_mask & filter->ingress_port_mask && + tmp->lookup == filter->lookup && ocelot_vcap_is_problematic_non_mac_etype(tmp)) return false; } for_each_set_bit(port, &filter->ingress_port_mask, ocelot->num_phys_ports) - ocelot_match_all_as_mac_etype(ocelot, port, true); + ocelot_match_all_as_mac_etype(ocelot, port, + filter->lookup, true); } else if (ocelot_vcap_is_problematic_non_mac_etype(filter)) { /* Search for any MAC_ETYPE rules on the port */ for (i = 0; i < block->count; i++) { - tmp = ocelot_vcap_block_find_filter(block, i); + tmp = ocelot_vcap_block_find_filter_by_index(block, i); if (tmp->ingress_port_mask & filter->ingress_port_mask && + tmp->lookup == filter->lookup && ocelot_vcap_is_problematic_mac_etype(tmp)) return false; } for_each_set_bit(port, &filter->ingress_port_mask, ocelot->num_phys_ports) - ocelot_match_all_as_mac_etype(ocelot, port, false); + ocelot_match_all_as_mac_etype(ocelot, port, + filter->lookup, false); } return true; @@ -863,12 +1121,12 @@ int ocelot_vcap_filter_add(struct ocelot *ocelot, struct ocelot_vcap_filter *filter, struct netlink_ext_ack *extack) { - struct ocelot_vcap_block *block = &ocelot->block; + struct ocelot_vcap_block *block = &ocelot->block[filter->block_id]; int i, index; if (!ocelot_exclusive_mac_etype_filter_rules(ocelot, filter)) { NL_SET_ERR_MSG_MOD(extack, - "Cannot mix MAC_ETYPE with non-MAC_ETYPE rules"); + "Cannot mix MAC_ETYPE with non-MAC_ETYPE rules, use the other IS2 lookup"); return -EBUSY; } @@ -877,17 +1135,19 @@ int ocelot_vcap_filter_add(struct ocelot *ocelot, /* Get the index of the inserted filter */ index = ocelot_vcap_block_get_filter_index(block, filter); + if (index < 0) + return index; /* Move down the rules to make place for the new filter */ for (i = block->count - 1; i > index; i--) { struct ocelot_vcap_filter *tmp; - tmp = ocelot_vcap_block_find_filter(block, i); - is2_entry_set(ocelot, i, tmp); + tmp = ocelot_vcap_block_find_filter_by_index(block, i); + vcap_entry_set(ocelot, i, tmp); } /* Now insert the new filter */ - is2_entry_set(ocelot, index, filter); + vcap_entry_set(ocelot, index, filter); return 0; } @@ -901,9 +1161,10 @@ static void ocelot_vcap_block_remove_filter(struct ocelot *ocelot, list_for_each_safe(pos, q, &block->rules) { tmp = list_entry(pos, struct ocelot_vcap_filter, list); if (tmp->id == filter->id) { - if (tmp->action == OCELOT_VCAP_ACTION_POLICE) + if (tmp->block_id == VCAP_IS2 && + tmp->action.police_ena) ocelot_vcap_policer_del(ocelot, block, - tmp->pol_ix); + tmp->action.pol_ix); list_del(pos); kfree(tmp); @@ -916,7 +1177,7 @@ static void ocelot_vcap_block_remove_filter(struct ocelot *ocelot, int ocelot_vcap_filter_del(struct ocelot *ocelot, struct ocelot_vcap_filter *filter) { - struct ocelot_vcap_block *block = &ocelot->block; + struct ocelot_vcap_block *block = &ocelot->block[filter->block_id]; struct ocelot_vcap_filter del_filter; int i, index; @@ -924,6 +1185,8 @@ int ocelot_vcap_filter_del(struct ocelot *ocelot, /* Gets index of the filter */ index = ocelot_vcap_block_get_filter_index(block, filter); + if (index < 0) + return index; /* Delete filter */ ocelot_vcap_block_remove_filter(ocelot, block, filter); @@ -932,12 +1195,12 @@ int ocelot_vcap_filter_del(struct ocelot *ocelot, for (i = index; i < block->count; i++) { struct ocelot_vcap_filter *tmp; - tmp = ocelot_vcap_block_find_filter(block, i); - is2_entry_set(ocelot, i, tmp); + tmp = ocelot_vcap_block_find_filter_by_index(block, i); + vcap_entry_set(ocelot, i, tmp); } /* Now delete the last filter, because it is duplicated */ - is2_entry_set(ocelot, block->count, &del_filter); + vcap_entry_set(ocelot, block->count, &del_filter); return 0; } @@ -945,37 +1208,115 @@ int ocelot_vcap_filter_del(struct ocelot *ocelot, int ocelot_vcap_filter_stats_update(struct ocelot *ocelot, struct ocelot_vcap_filter *filter) { - struct ocelot_vcap_block *block = &ocelot->block; - struct ocelot_vcap_filter *tmp; + struct ocelot_vcap_block *block = &ocelot->block[filter->block_id]; + struct ocelot_vcap_filter tmp; int index; index = ocelot_vcap_block_get_filter_index(block, filter); - is2_entry_get(ocelot, filter, index); + if (index < 0) + return index; + + vcap_entry_get(ocelot, index, filter); /* After we get the result we need to clear the counters */ - tmp = ocelot_vcap_block_find_filter(block, index); - tmp->stats.pkts = 0; - is2_entry_set(ocelot, index, tmp); + tmp = *filter; + tmp.stats.pkts = 0; + vcap_entry_set(ocelot, index, &tmp); return 0; } -int ocelot_vcap_init(struct ocelot *ocelot) +static void ocelot_vcap_init_one(struct ocelot *ocelot, + const struct vcap_props *vcap) { - const struct vcap_props *vcap_is2 = &ocelot->vcap[VCAP_IS2]; - struct ocelot_vcap_block *block = &ocelot->block; struct vcap_data data; memset(&data, 0, sizeof(data)); - vcap_entry2cache(ocelot, &data); - ocelot_write(ocelot, vcap_is2->entry_count, S2_CORE_MV_CFG); - vcap_cmd(ocelot, 0, VCAP_CMD_INITIALIZE, VCAP_SEL_ENTRY); + vcap_entry2cache(ocelot, vcap, &data); + ocelot_target_write(ocelot, vcap->target, vcap->entry_count, + VCAP_CORE_MV_CFG); + vcap_cmd(ocelot, vcap, 0, VCAP_CMD_INITIALIZE, VCAP_SEL_ENTRY); - vcap_action2cache(ocelot, &data); - ocelot_write(ocelot, vcap_is2->action_count, S2_CORE_MV_CFG); - vcap_cmd(ocelot, 0, VCAP_CMD_INITIALIZE, + vcap_action2cache(ocelot, vcap, &data); + ocelot_target_write(ocelot, vcap->target, vcap->action_count, + VCAP_CORE_MV_CFG); + vcap_cmd(ocelot, vcap, 0, VCAP_CMD_INITIALIZE, VCAP_SEL_ACTION | VCAP_SEL_COUNTER); +} + +static void ocelot_vcap_detect_constants(struct ocelot *ocelot, + struct vcap_props *vcap) +{ + int counter_memory_width; + int num_default_actions; + int version; + + version = ocelot_target_read(ocelot, vcap->target, + VCAP_CONST_VCAP_VER); + /* Only version 0 VCAP supported for now */ + if (WARN_ON(version != 0)) + return; + + /* Width in bits of type-group field */ + vcap->tg_width = ocelot_target_read(ocelot, vcap->target, + VCAP_CONST_ENTRY_TG_WIDTH); + /* Number of subwords per TCAM row */ + vcap->sw_count = ocelot_target_read(ocelot, vcap->target, + VCAP_CONST_ENTRY_SWCNT); + /* Number of rows in TCAM. There can be this many full keys, or double + * this number half keys, or 4 times this number quarter keys. + */ + vcap->entry_count = ocelot_target_read(ocelot, vcap->target, + VCAP_CONST_ENTRY_CNT); + /* Assuming there are 4 subwords per TCAM row, their layout in the + * actual TCAM (not in the cache) would be: + * + * | SW 3 | TG 3 | SW 2 | TG 2 | SW 1 | TG 1 | SW 0 | TG 0 | + * + * (where SW=subword and TG=Type-Group). + * + * What VCAP_CONST_ENTRY_CNT is giving us is the width of one full TCAM + * row. But when software accesses the TCAM through the cache + * registers, the Type-Group values are written through another set of + * registers VCAP_TG_DAT, and therefore, it appears as though the 4 + * subwords are contiguous in the cache memory. + * Important mention: regardless of the number of key entries per row + * (and therefore of key size: 1 full key or 2 half keys or 4 quarter + * keys), software always has to configure 4 Type-Group values. For + * example, in the case of 1 full key, the driver needs to set all 4 + * Type-Group to be full key. + * + * For this reason, we need to fix up the value that the hardware is + * giving us. We don't actually care about the width of the entry in + * the TCAM. What we care about is the width of the entry in the cache + * registers, which is how we get to interact with it. And since the + * VCAP_ENTRY_DAT cache registers access only the subwords and not the + * Type-Groups, this means we need to subtract the width of the + * Type-Groups when packing and unpacking key entry data in a TCAM row. + */ + vcap->entry_width = ocelot_target_read(ocelot, vcap->target, + VCAP_CONST_ENTRY_WIDTH); + vcap->entry_width -= vcap->tg_width * vcap->sw_count; + num_default_actions = ocelot_target_read(ocelot, vcap->target, + VCAP_CONST_ACTION_DEF_CNT); + vcap->action_count = vcap->entry_count + num_default_actions; + vcap->action_width = ocelot_target_read(ocelot, vcap->target, + VCAP_CONST_ACTION_WIDTH); + /* The width of the counter memory, this is the complete width of all + * counter-fields associated with one full-word entry. There is one + * counter per entry sub-word (see CAP_CORE::ENTRY_SWCNT for number of + * subwords.) + */ + vcap->counter_words = vcap->sw_count; + counter_memory_width = ocelot_target_read(ocelot, vcap->target, + VCAP_CONST_CNT_WIDTH); + vcap->counter_width = counter_memory_width / vcap->counter_words; +} + +int ocelot_vcap_init(struct ocelot *ocelot) +{ + int i; /* Create a policer that will drop the frames for the cpu. * This policer will be used as action in the acl rules to drop @@ -992,9 +1333,18 @@ int ocelot_vcap_init(struct ocelot *ocelot) ocelot_write_gix(ocelot, 0x3fffff, ANA_POL_CIR_STATE, OCELOT_POLICER_DISCARD); - block->pol_lpr = OCELOT_POLICER_DISCARD - 1; + for (i = 0; i < OCELOT_NUM_VCAP_BLOCKS; i++) { + struct ocelot_vcap_block *block = &ocelot->block[i]; + struct vcap_props *vcap = &ocelot->vcap[i]; + + INIT_LIST_HEAD(&block->rules); + block->pol_lpr = OCELOT_POLICER_DISCARD - 1; + + ocelot_vcap_detect_constants(ocelot, vcap); + ocelot_vcap_init_one(ocelot, vcap); + } - INIT_LIST_HEAD(&ocelot->block.rules); + INIT_LIST_HEAD(&ocelot->dummy_rules); return 0; } diff --git a/drivers/net/ethernet/mscc/ocelot_vcap.h b/drivers/net/ethernet/mscc/ocelot_vcap.h index 0dfbfc011b2e..82fd10581a14 100644 --- a/drivers/net/ethernet/mscc/ocelot_vcap.h +++ b/drivers/net/ethernet/mscc/ocelot_vcap.h @@ -11,6 +11,8 @@ #include <net/sch_generic.h> #include <net/pkt_cls.h> +#define OCELOT_POLICER_DISCARD 0x17f + struct ocelot_ipv4 { u8 addr[4]; }; @@ -76,6 +78,11 @@ struct ocelot_vcap_udp_tcp { u16 mask; }; +struct ocelot_vcap_port { + u8 value; + u8 mask; +}; + enum ocelot_vcap_key_type { OCELOT_VCAP_KEY_ANY, OCELOT_VCAP_KEY_ETYPE, @@ -158,6 +165,7 @@ struct ocelot_vcap_key_ipv4 { struct ocelot_vcap_key_ipv6 { struct ocelot_vcap_u8 proto; /* IPv6 protocol */ struct ocelot_vcap_u128 sip; /* IPv6 source (byte 0-7 ignored) */ + struct ocelot_vcap_u128 dip; /* IPv6 destination (byte 0-7 ignored) */ enum ocelot_vcap_bit ttl; /* TTL zero */ struct ocelot_vcap_u8 ds; struct ocelot_vcap_u48 data; /* Not UDP/TCP: IP data */ @@ -174,10 +182,71 @@ struct ocelot_vcap_key_ipv6 { enum ocelot_vcap_bit seq_zero; /* TCP sequence number is zero */ }; -enum ocelot_vcap_action { - OCELOT_VCAP_ACTION_DROP, - OCELOT_VCAP_ACTION_TRAP, - OCELOT_VCAP_ACTION_POLICE, +enum ocelot_mask_mode { + OCELOT_MASK_MODE_NONE, + OCELOT_MASK_MODE_PERMIT_DENY, + OCELOT_MASK_MODE_POLICY, + OCELOT_MASK_MODE_REDIRECT, +}; + +enum ocelot_es0_tag { + OCELOT_NO_ES0_TAG, + OCELOT_ES0_TAG, + OCELOT_FORCE_PORT_TAG, + OCELOT_FORCE_UNTAG, +}; + +enum ocelot_tag_tpid_sel { + OCELOT_TAG_TPID_SEL_8021Q, + OCELOT_TAG_TPID_SEL_8021AD, +}; + +struct ocelot_vcap_action { + union { + /* VCAP ES0 */ + struct { + enum ocelot_es0_tag push_outer_tag; + enum ocelot_es0_tag push_inner_tag; + enum ocelot_tag_tpid_sel tag_a_tpid_sel; + int tag_a_vid_sel; + int tag_a_pcp_sel; + u16 vid_a_val; + u8 pcp_a_val; + u8 dei_a_val; + enum ocelot_tag_tpid_sel tag_b_tpid_sel; + int tag_b_vid_sel; + int tag_b_pcp_sel; + u16 vid_b_val; + u8 pcp_b_val; + u8 dei_b_val; + }; + + /* VCAP IS1 */ + struct { + bool vid_replace_ena; + u16 vid; + bool vlan_pop_cnt_ena; + int vlan_pop_cnt; + bool pcp_dei_ena; + u8 pcp; + u8 dei; + bool qos_ena; + u8 qos_val; + u8 pag_override_mask; + u8 pag_val; + }; + + /* VCAP IS2 */ + struct { + bool cpu_copy_ena; + u8 cpu_qu_num; + enum ocelot_mask_mode mask_mode; + unsigned long port_mask; + bool police_ena; + struct ocelot_policer pol; + u32 pol_ix; + }; + }; }; struct ocelot_vcap_stats { @@ -186,15 +255,30 @@ struct ocelot_vcap_stats { u64 used; }; +enum ocelot_vcap_filter_type { + OCELOT_VCAP_FILTER_DUMMY, + OCELOT_VCAP_FILTER_PAG, + OCELOT_VCAP_FILTER_OFFLOAD, +}; + struct ocelot_vcap_filter { struct list_head list; + enum ocelot_vcap_filter_type type; + int block_id; + int goto_target; + int lookup; + u8 pag; u16 prio; u32 id; - enum ocelot_vcap_action action; + struct ocelot_vcap_action action; struct ocelot_vcap_stats stats; + /* For VCAP IS1 and IS2 */ unsigned long ingress_port_mask; + /* For VCAP ES0 */ + struct ocelot_vcap_port ingress_port; + struct ocelot_vcap_port egress_port; enum ocelot_vcap_bit dmac_mc; enum ocelot_vcap_bit dmac_bc; @@ -210,8 +294,6 @@ struct ocelot_vcap_filter { struct ocelot_vcap_key_ipv4 ipv4; struct ocelot_vcap_key_ipv6 ipv6; } key; - struct ocelot_policer pol; - u32 pol_ix; }; int ocelot_vcap_filter_add(struct ocelot *ocelot, @@ -221,7 +303,10 @@ int ocelot_vcap_filter_del(struct ocelot *ocelot, struct ocelot_vcap_filter *rule); int ocelot_vcap_filter_stats_update(struct ocelot *ocelot, struct ocelot_vcap_filter *rule); +struct ocelot_vcap_filter * +ocelot_vcap_block_find_filter_by_id(struct ocelot_vcap_block *block, int id); +void ocelot_detect_vcap_constants(struct ocelot *ocelot); int ocelot_vcap_init(struct ocelot *ocelot); int ocelot_setup_tc_cls_flower(struct ocelot_port_private *priv, diff --git a/drivers/net/ethernet/mscc/ocelot_vsc7514.c b/drivers/net/ethernet/mscc/ocelot_vsc7514.c index 8a6917691ba6..dc00772950e5 100644 --- a/drivers/net/ethernet/mscc/ocelot_vsc7514.c +++ b/drivers/net/ethernet/mscc/ocelot_vsc7514.c @@ -19,10 +19,6 @@ #include "ocelot.h" #define IFH_EXTRACT_BITFIELD64(x, o, w) (((x) >> (o)) & GENMASK_ULL((w) - 1, 0)) -#define VSC7514_VCAP_IS2_CNT 64 -#define VSC7514_VCAP_IS2_ENTRY_WIDTH 376 -#define VSC7514_VCAP_IS2_ACTION_WIDTH 99 -#define VSC7514_VCAP_PORT_CNT 11 static const u32 ocelot_ana_regmap[] = { REG(ANA_ADVLEARN, 0x009000), @@ -241,14 +237,27 @@ static const u32 ocelot_sys_regmap[] = { REG(SYS_PTP_CFG, 0x0006c4), }; -static const u32 ocelot_s2_regmap[] = { - REG(S2_CORE_UPDATE_CTRL, 0x000000), - REG(S2_CORE_MV_CFG, 0x000004), - REG(S2_CACHE_ENTRY_DAT, 0x000008), - REG(S2_CACHE_MASK_DAT, 0x000108), - REG(S2_CACHE_ACTION_DAT, 0x000208), - REG(S2_CACHE_CNT_DAT, 0x000308), - REG(S2_CACHE_TG_DAT, 0x000388), +static const u32 ocelot_vcap_regmap[] = { + /* VCAP_CORE_CFG */ + REG(VCAP_CORE_UPDATE_CTRL, 0x000000), + REG(VCAP_CORE_MV_CFG, 0x000004), + /* VCAP_CORE_CACHE */ + REG(VCAP_CACHE_ENTRY_DAT, 0x000008), + REG(VCAP_CACHE_MASK_DAT, 0x000108), + REG(VCAP_CACHE_ACTION_DAT, 0x000208), + REG(VCAP_CACHE_CNT_DAT, 0x000308), + REG(VCAP_CACHE_TG_DAT, 0x000388), + /* VCAP_CONST */ + REG(VCAP_CONST_VCAP_VER, 0x000398), + REG(VCAP_CONST_ENTRY_WIDTH, 0x00039c), + REG(VCAP_CONST_ENTRY_CNT, 0x0003a0), + REG(VCAP_CONST_ENTRY_SWCNT, 0x0003a4), + REG(VCAP_CONST_ENTRY_TG_WIDTH, 0x0003a8), + REG(VCAP_CONST_ACTION_DEF_CNT, 0x0003ac), + REG(VCAP_CONST_ACTION_WIDTH, 0x0003b0), + REG(VCAP_CONST_CNT_WIDTH, 0x0003b4), + REG(VCAP_CONST_CORE_CNT, 0x0003b8), + REG(VCAP_CONST_IF_CNT, 0x0003bc), }; static const u32 ocelot_ptp_regmap[] = { @@ -311,7 +320,9 @@ static const u32 *ocelot_regmap[TARGET_MAX] = { [QSYS] = ocelot_qsys_regmap, [REW] = ocelot_rew_regmap, [SYS] = ocelot_sys_regmap, - [S2] = ocelot_s2_regmap, + [S0] = ocelot_vcap_regmap, + [S1] = ocelot_vcap_regmap, + [S2] = ocelot_vcap_regmap, [PTP] = ocelot_ptp_regmap, [DEV_GMII] = ocelot_dev_gmii_regmap, }; @@ -756,6 +767,115 @@ static u16 ocelot_wm_enc(u16 value) static const struct ocelot_ops ocelot_ops = { .reset = ocelot_reset, .wm_enc = ocelot_wm_enc, + .port_to_netdev = ocelot_port_to_netdev, + .netdev_to_port = ocelot_netdev_to_port, +}; + +static const struct vcap_field vsc7514_vcap_es0_keys[] = { + [VCAP_ES0_EGR_PORT] = { 0, 4}, + [VCAP_ES0_IGR_PORT] = { 4, 4}, + [VCAP_ES0_RSV] = { 8, 2}, + [VCAP_ES0_L2_MC] = { 10, 1}, + [VCAP_ES0_L2_BC] = { 11, 1}, + [VCAP_ES0_VID] = { 12, 12}, + [VCAP_ES0_DP] = { 24, 1}, + [VCAP_ES0_PCP] = { 25, 3}, +}; + +static const struct vcap_field vsc7514_vcap_es0_actions[] = { + [VCAP_ES0_ACT_PUSH_OUTER_TAG] = { 0, 2}, + [VCAP_ES0_ACT_PUSH_INNER_TAG] = { 2, 1}, + [VCAP_ES0_ACT_TAG_A_TPID_SEL] = { 3, 2}, + [VCAP_ES0_ACT_TAG_A_VID_SEL] = { 5, 1}, + [VCAP_ES0_ACT_TAG_A_PCP_SEL] = { 6, 2}, + [VCAP_ES0_ACT_TAG_A_DEI_SEL] = { 8, 2}, + [VCAP_ES0_ACT_TAG_B_TPID_SEL] = { 10, 2}, + [VCAP_ES0_ACT_TAG_B_VID_SEL] = { 12, 1}, + [VCAP_ES0_ACT_TAG_B_PCP_SEL] = { 13, 2}, + [VCAP_ES0_ACT_TAG_B_DEI_SEL] = { 15, 2}, + [VCAP_ES0_ACT_VID_A_VAL] = { 17, 12}, + [VCAP_ES0_ACT_PCP_A_VAL] = { 29, 3}, + [VCAP_ES0_ACT_DEI_A_VAL] = { 32, 1}, + [VCAP_ES0_ACT_VID_B_VAL] = { 33, 12}, + [VCAP_ES0_ACT_PCP_B_VAL] = { 45, 3}, + [VCAP_ES0_ACT_DEI_B_VAL] = { 48, 1}, + [VCAP_ES0_ACT_RSV] = { 49, 24}, + [VCAP_ES0_ACT_HIT_STICKY] = { 73, 1}, +}; + +static const struct vcap_field vsc7514_vcap_is1_keys[] = { + [VCAP_IS1_HK_TYPE] = { 0, 1}, + [VCAP_IS1_HK_LOOKUP] = { 1, 2}, + [VCAP_IS1_HK_IGR_PORT_MASK] = { 3, 12}, + [VCAP_IS1_HK_RSV] = { 15, 9}, + [VCAP_IS1_HK_OAM_Y1731] = { 24, 1}, + [VCAP_IS1_HK_L2_MC] = { 25, 1}, + [VCAP_IS1_HK_L2_BC] = { 26, 1}, + [VCAP_IS1_HK_IP_MC] = { 27, 1}, + [VCAP_IS1_HK_VLAN_TAGGED] = { 28, 1}, + [VCAP_IS1_HK_VLAN_DBL_TAGGED] = { 29, 1}, + [VCAP_IS1_HK_TPID] = { 30, 1}, + [VCAP_IS1_HK_VID] = { 31, 12}, + [VCAP_IS1_HK_DEI] = { 43, 1}, + [VCAP_IS1_HK_PCP] = { 44, 3}, + /* Specific Fields for IS1 Half Key S1_NORMAL */ + [VCAP_IS1_HK_L2_SMAC] = { 47, 48}, + [VCAP_IS1_HK_ETYPE_LEN] = { 95, 1}, + [VCAP_IS1_HK_ETYPE] = { 96, 16}, + [VCAP_IS1_HK_IP_SNAP] = {112, 1}, + [VCAP_IS1_HK_IP4] = {113, 1}, + /* Layer-3 Information */ + [VCAP_IS1_HK_L3_FRAGMENT] = {114, 1}, + [VCAP_IS1_HK_L3_FRAG_OFS_GT0] = {115, 1}, + [VCAP_IS1_HK_L3_OPTIONS] = {116, 1}, + [VCAP_IS1_HK_L3_DSCP] = {117, 6}, + [VCAP_IS1_HK_L3_IP4_SIP] = {123, 32}, + /* Layer-4 Information */ + [VCAP_IS1_HK_TCP_UDP] = {155, 1}, + [VCAP_IS1_HK_TCP] = {156, 1}, + [VCAP_IS1_HK_L4_SPORT] = {157, 16}, + [VCAP_IS1_HK_L4_RNG] = {173, 8}, + /* Specific Fields for IS1 Half Key S1_5TUPLE_IP4 */ + [VCAP_IS1_HK_IP4_INNER_TPID] = { 47, 1}, + [VCAP_IS1_HK_IP4_INNER_VID] = { 48, 12}, + [VCAP_IS1_HK_IP4_INNER_DEI] = { 60, 1}, + [VCAP_IS1_HK_IP4_INNER_PCP] = { 61, 3}, + [VCAP_IS1_HK_IP4_IP4] = { 64, 1}, + [VCAP_IS1_HK_IP4_L3_FRAGMENT] = { 65, 1}, + [VCAP_IS1_HK_IP4_L3_FRAG_OFS_GT0] = { 66, 1}, + [VCAP_IS1_HK_IP4_L3_OPTIONS] = { 67, 1}, + [VCAP_IS1_HK_IP4_L3_DSCP] = { 68, 6}, + [VCAP_IS1_HK_IP4_L3_IP4_DIP] = { 74, 32}, + [VCAP_IS1_HK_IP4_L3_IP4_SIP] = {106, 32}, + [VCAP_IS1_HK_IP4_L3_PROTO] = {138, 8}, + [VCAP_IS1_HK_IP4_TCP_UDP] = {146, 1}, + [VCAP_IS1_HK_IP4_TCP] = {147, 1}, + [VCAP_IS1_HK_IP4_L4_RNG] = {148, 8}, + [VCAP_IS1_HK_IP4_IP_PAYLOAD_S1_5TUPLE] = {156, 32}, +}; + +static const struct vcap_field vsc7514_vcap_is1_actions[] = { + [VCAP_IS1_ACT_DSCP_ENA] = { 0, 1}, + [VCAP_IS1_ACT_DSCP_VAL] = { 1, 6}, + [VCAP_IS1_ACT_QOS_ENA] = { 7, 1}, + [VCAP_IS1_ACT_QOS_VAL] = { 8, 3}, + [VCAP_IS1_ACT_DP_ENA] = { 11, 1}, + [VCAP_IS1_ACT_DP_VAL] = { 12, 1}, + [VCAP_IS1_ACT_PAG_OVERRIDE_MASK] = { 13, 8}, + [VCAP_IS1_ACT_PAG_VAL] = { 21, 8}, + [VCAP_IS1_ACT_RSV] = { 29, 9}, + /* The fields below are incorrectly shifted by 2 in the manual */ + [VCAP_IS1_ACT_VID_REPLACE_ENA] = { 38, 1}, + [VCAP_IS1_ACT_VID_ADD_VAL] = { 39, 12}, + [VCAP_IS1_ACT_FID_SEL] = { 51, 2}, + [VCAP_IS1_ACT_FID_VAL] = { 53, 13}, + [VCAP_IS1_ACT_PCP_DEI_ENA] = { 66, 1}, + [VCAP_IS1_ACT_PCP_VAL] = { 67, 3}, + [VCAP_IS1_ACT_DEI_VAL] = { 70, 1}, + [VCAP_IS1_ACT_VLAN_POP_CNT_ENA] = { 71, 1}, + [VCAP_IS1_ACT_VLAN_POP_CNT] = { 72, 2}, + [VCAP_IS1_ACT_CUSTOM_ACE_TYPE_ENA] = { 74, 4}, + [VCAP_IS1_ACT_HIT_STICKY] = { 78, 1}, }; static const struct vcap_field vsc7514_vcap_is2_keys[] = { @@ -856,15 +976,32 @@ static const struct vcap_field vsc7514_vcap_is2_actions[] = { [VCAP_IS2_ACT_HIT_CNT] = { 49, 32}, }; -static const struct vcap_props vsc7514_vcap_props[] = { +static struct vcap_props vsc7514_vcap_props[] = { + [VCAP_ES0] = { + .action_type_width = 0, + .action_table = { + [ES0_ACTION_TYPE_NORMAL] = { + .width = 73, /* HIT_STICKY not included */ + .count = 1, + }, + }, + .target = S0, + .keys = vsc7514_vcap_es0_keys, + .actions = vsc7514_vcap_es0_actions, + }, + [VCAP_IS1] = { + .action_type_width = 0, + .action_table = { + [IS1_ACTION_TYPE_NORMAL] = { + .width = 78, /* HIT_STICKY not included */ + .count = 4, + }, + }, + .target = S1, + .keys = vsc7514_vcap_is1_keys, + .actions = vsc7514_vcap_is1_actions, + }, [VCAP_IS2] = { - .tg_width = 2, - .sw_count = 4, - .entry_count = VSC7514_VCAP_IS2_CNT, - .entry_width = VSC7514_VCAP_IS2_ENTRY_WIDTH, - .action_count = VSC7514_VCAP_IS2_CNT + - VSC7514_VCAP_PORT_CNT + 2, - .action_width = 99, .action_type_width = 1, .action_table = { [IS2_ACTION_TYPE_NORMAL] = { @@ -876,8 +1013,9 @@ static const struct vcap_props vsc7514_vcap_props[] = { .count = 4 }, }, - .counter_words = 4, - .counter_width = 32, + .target = S2, + .keys = vsc7514_vcap_is2_keys, + .actions = vsc7514_vcap_is2_actions, }, }; @@ -932,10 +1070,6 @@ static int mscc_ocelot_init_ports(struct platform_device *pdev, if (!ocelot->ports) return -ENOMEM; - /* No NPI port */ - ocelot_configure_cpu(ocelot, -1, OCELOT_TAG_PREFIX_NONE, - OCELOT_TAG_PREFIX_NONE); - for_each_available_child_of_node(ports, portnp) { struct ocelot_port_private *priv; struct ocelot_port *ocelot_port; @@ -1043,6 +1177,8 @@ static int mscc_ocelot_probe(struct platform_device *pdev) { QSYS, "qsys" }, { ANA, "ana" }, { QS, "qs" }, + { S0, "s0" }, + { S1, "s1" }, { S2, "s2" }, { PTP, "ptp", 1 }, }; @@ -1119,9 +1255,10 @@ static int mscc_ocelot_probe(struct platform_device *pdev) ocelot->num_phys_ports = of_get_child_count(ports); - ocelot->vcap_is2_keys = vsc7514_vcap_is2_keys; - ocelot->vcap_is2_actions = vsc7514_vcap_is2_actions; ocelot->vcap = vsc7514_vcap_props; + ocelot->inj_prefix = OCELOT_TAG_PREFIX_NONE; + ocelot->xtr_prefix = OCELOT_TAG_PREFIX_NONE; + ocelot->npi = -1; err = ocelot_init(ocelot); if (err) diff --git a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c index 4a5beafa0493..1634ca6d4a8f 100644 --- a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c +++ b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c @@ -3543,11 +3543,10 @@ static void myri10ge_free_slices(struct myri10ge_priv *mgp) ss->fw_stats, ss->fw_stats_bus); ss->fw_stats = NULL; } - napi_hash_del(&ss->napi); - netif_napi_del(&ss->napi); + __netif_napi_del(&ss->napi); } /* Wait till napi structs are no longer used, and then free ss. */ - synchronize_rcu(); + synchronize_net(); kfree(mgp->ss); mgp->ss = NULL; } diff --git a/drivers/net/ethernet/natsemi/natsemi.c b/drivers/net/ethernet/natsemi/natsemi.c index 3de8430ee8c5..b81e1487945c 100644 --- a/drivers/net/ethernet/natsemi/natsemi.c +++ b/drivers/net/ethernet/natsemi/natsemi.c @@ -1916,9 +1916,9 @@ static void ns_tx_timeout(struct net_device *dev, unsigned int txqueue) static int alloc_ring(struct net_device *dev) { struct netdev_private *np = netdev_priv(dev); - np->rx_ring = pci_alloc_consistent(np->pci_dev, - sizeof(struct netdev_desc) * (RX_RING_SIZE+TX_RING_SIZE), - &np->ring_dma); + np->rx_ring = dma_alloc_coherent(&np->pci_dev->dev, + sizeof(struct netdev_desc) * (RX_RING_SIZE + TX_RING_SIZE), + &np->ring_dma, GFP_KERNEL); if (!np->rx_ring) return -ENOMEM; np->tx_ring = &np->rx_ring[RX_RING_SIZE]; @@ -1939,10 +1939,10 @@ static void refill_rx(struct net_device *dev) np->rx_skbuff[entry] = skb; if (skb == NULL) break; /* Better luck next round. */ - np->rx_dma[entry] = pci_map_single(np->pci_dev, - skb->data, buflen, PCI_DMA_FROMDEVICE); - if (pci_dma_mapping_error(np->pci_dev, - np->rx_dma[entry])) { + np->rx_dma[entry] = dma_map_single(&np->pci_dev->dev, + skb->data, buflen, + DMA_FROM_DEVICE); + if (dma_mapping_error(&np->pci_dev->dev, np->rx_dma[entry])) { dev_kfree_skb_any(skb); np->rx_skbuff[entry] = NULL; break; /* Better luck next round. */ @@ -2013,9 +2013,8 @@ static void drain_tx(struct net_device *dev) for (i = 0; i < TX_RING_SIZE; i++) { if (np->tx_skbuff[i]) { - pci_unmap_single(np->pci_dev, - np->tx_dma[i], np->tx_skbuff[i]->len, - PCI_DMA_TODEVICE); + dma_unmap_single(&np->pci_dev->dev, np->tx_dma[i], + np->tx_skbuff[i]->len, DMA_TO_DEVICE); dev_kfree_skb(np->tx_skbuff[i]); dev->stats.tx_dropped++; } @@ -2034,9 +2033,9 @@ static void drain_rx(struct net_device *dev) np->rx_ring[i].cmd_status = 0; np->rx_ring[i].addr = cpu_to_le32(0xBADF00D0); /* An invalid address. */ if (np->rx_skbuff[i]) { - pci_unmap_single(np->pci_dev, np->rx_dma[i], - buflen + NATSEMI_PADDING, - PCI_DMA_FROMDEVICE); + dma_unmap_single(&np->pci_dev->dev, np->rx_dma[i], + buflen + NATSEMI_PADDING, + DMA_FROM_DEVICE); dev_kfree_skb(np->rx_skbuff[i]); } np->rx_skbuff[i] = NULL; @@ -2052,9 +2051,9 @@ static void drain_ring(struct net_device *dev) static void free_ring(struct net_device *dev) { struct netdev_private *np = netdev_priv(dev); - pci_free_consistent(np->pci_dev, - sizeof(struct netdev_desc) * (RX_RING_SIZE+TX_RING_SIZE), - np->rx_ring, np->ring_dma); + dma_free_coherent(&np->pci_dev->dev, + sizeof(struct netdev_desc) * (RX_RING_SIZE + TX_RING_SIZE), + np->rx_ring, np->ring_dma); } static void reinit_rx(struct net_device *dev) @@ -2101,9 +2100,9 @@ static netdev_tx_t start_tx(struct sk_buff *skb, struct net_device *dev) entry = np->cur_tx % TX_RING_SIZE; np->tx_skbuff[entry] = skb; - np->tx_dma[entry] = pci_map_single(np->pci_dev, - skb->data,skb->len, PCI_DMA_TODEVICE); - if (pci_dma_mapping_error(np->pci_dev, np->tx_dma[entry])) { + np->tx_dma[entry] = dma_map_single(&np->pci_dev->dev, skb->data, + skb->len, DMA_TO_DEVICE); + if (dma_mapping_error(&np->pci_dev->dev, np->tx_dma[entry])) { np->tx_skbuff[entry] = NULL; dev_kfree_skb_irq(skb); dev->stats.tx_dropped++; @@ -2169,9 +2168,8 @@ static void netdev_tx_done(struct net_device *dev) dev->stats.tx_window_errors++; dev->stats.tx_errors++; } - pci_unmap_single(np->pci_dev,np->tx_dma[entry], - np->tx_skbuff[entry]->len, - PCI_DMA_TODEVICE); + dma_unmap_single(&np->pci_dev->dev, np->tx_dma[entry], + np->tx_skbuff[entry]->len, DMA_TO_DEVICE); /* Free the original skb. */ dev_consume_skb_irq(np->tx_skbuff[entry]); np->tx_skbuff[entry] = NULL; @@ -2359,21 +2357,22 @@ static void netdev_rx(struct net_device *dev, int *work_done, int work_to_do) (skb = netdev_alloc_skb(dev, pkt_len + RX_OFFSET)) != NULL) { /* 16 byte align the IP header */ skb_reserve(skb, RX_OFFSET); - pci_dma_sync_single_for_cpu(np->pci_dev, - np->rx_dma[entry], - buflen, - PCI_DMA_FROMDEVICE); + dma_sync_single_for_cpu(&np->pci_dev->dev, + np->rx_dma[entry], + buflen, + DMA_FROM_DEVICE); skb_copy_to_linear_data(skb, np->rx_skbuff[entry]->data, pkt_len); skb_put(skb, pkt_len); - pci_dma_sync_single_for_device(np->pci_dev, - np->rx_dma[entry], - buflen, - PCI_DMA_FROMDEVICE); + dma_sync_single_for_device(&np->pci_dev->dev, + np->rx_dma[entry], + buflen, + DMA_FROM_DEVICE); } else { - pci_unmap_single(np->pci_dev, np->rx_dma[entry], + dma_unmap_single(&np->pci_dev->dev, + np->rx_dma[entry], buflen + NATSEMI_PADDING, - PCI_DMA_FROMDEVICE); + DMA_FROM_DEVICE); skb_put(skb = np->rx_skbuff[entry], pkt_len); np->rx_skbuff[entry] = NULL; } diff --git a/drivers/net/ethernet/natsemi/ns83820.c b/drivers/net/ethernet/natsemi/ns83820.c index 8e24c7acf79b..72794d158871 100644 --- a/drivers/net/ethernet/natsemi/ns83820.c +++ b/drivers/net/ethernet/natsemi/ns83820.c @@ -526,8 +526,8 @@ static inline int ns83820_add_rx_skb(struct ns83820 *dev, struct sk_buff *skb) dev->rx_info.next_empty = (next_empty + 1) % NR_RX_DESC; cmdsts = REAL_RX_BUF_SIZE | CMDSTS_INTR; - buf = pci_map_single(dev->pci_dev, skb->data, - REAL_RX_BUF_SIZE, PCI_DMA_FROMDEVICE); + buf = dma_map_single(&dev->pci_dev->dev, skb->data, REAL_RX_BUF_SIZE, + DMA_FROM_DEVICE); build_rx_desc(dev, sg, 0, buf, cmdsts, 0); /* update link of previous rx */ if (likely(next_empty != dev->rx_info.next_rx)) @@ -600,12 +600,14 @@ static void phy_intr(struct net_device *ndev) struct ns83820 *dev = PRIV(ndev); static const char *speeds[] = { "10", "100", "1000", "1000(?)", "1000F" }; u32 cfg, new_cfg; - u32 tbisr, tanar, tanlpar; + u32 tanar, tanlpar; int speed, fullduplex, newlinkstate; cfg = readl(dev->base + CFG) ^ SPDSTS_POLARITY; if (dev->CFG_cache & CFG_TBI_EN) { + u32 __maybe_unused tbisr; + /* we have an optical transceiver */ tbisr = readl(dev->base + TBISR); tanar = readl(dev->base + TANAR); @@ -858,8 +860,8 @@ static void rx_irq(struct net_device *ndev) mb(); clear_rx_desc(dev, next_rx); - pci_unmap_single(dev->pci_dev, bufptr, - RX_BUF_SIZE, PCI_DMA_FROMDEVICE); + dma_unmap_single(&dev->pci_dev->dev, bufptr, RX_BUF_SIZE, + DMA_FROM_DEVICE); len = cmdsts & CMDSTS_LEN_MASK; #ifdef NS83820_VLAN_ACCEL_SUPPORT /* NH: As was mentioned below, this chip is kinda @@ -923,10 +925,10 @@ out: spin_unlock_irqrestore(&info->lock, flags); } -static void rx_action(unsigned long _dev) +static void rx_action(struct tasklet_struct *t) { - struct net_device *ndev = (void *)_dev; - struct ns83820 *dev = PRIV(ndev); + struct ns83820 *dev = from_tasklet(dev, t, rx_tasklet); + struct net_device *ndev = dev->ndev; rx_irq(ndev); writel(ihr, dev->base + IHR); @@ -985,17 +987,13 @@ static void do_tx_done(struct net_device *ndev) len = cmdsts & CMDSTS_LEN_MASK; addr = desc_addr_get(desc + DESC_BUFPTR); if (skb) { - pci_unmap_single(dev->pci_dev, - addr, - len, - PCI_DMA_TODEVICE); + dma_unmap_single(&dev->pci_dev->dev, addr, len, + DMA_TO_DEVICE); dev_consume_skb_irq(skb); atomic_dec(&dev->nr_tx_skbs); } else - pci_unmap_page(dev->pci_dev, - addr, - len, - PCI_DMA_TODEVICE); + dma_unmap_page(&dev->pci_dev->dev, addr, len, + DMA_TO_DEVICE); tx_done_idx = (tx_done_idx + 1) % NR_TX_DESC; dev->tx_done_idx = tx_done_idx; @@ -1023,10 +1021,10 @@ static void ns83820_cleanup_tx(struct ns83820 *dev) dev->tx_skbs[i] = NULL; if (skb) { __le32 *desc = dev->tx_descs + (i * DESC_SIZE); - pci_unmap_single(dev->pci_dev, - desc_addr_get(desc + DESC_BUFPTR), - le32_to_cpu(desc[DESC_CMDSTS]) & CMDSTS_LEN_MASK, - PCI_DMA_TODEVICE); + dma_unmap_single(&dev->pci_dev->dev, + desc_addr_get(desc + DESC_BUFPTR), + le32_to_cpu(desc[DESC_CMDSTS]) & CMDSTS_LEN_MASK, + DMA_TO_DEVICE); dev_kfree_skb_irq(skb); atomic_dec(&dev->nr_tx_skbs); } @@ -1121,7 +1119,8 @@ again: len = skb->len; if (nr_frags) len -= skb->data_len; - buf = pci_map_single(dev->pci_dev, skb->data, len, PCI_DMA_TODEVICE); + buf = dma_map_single(&dev->pci_dev->dev, skb->data, len, + DMA_TO_DEVICE); first_desc = dev->tx_descs + (free_idx * DESC_SIZE); @@ -1207,7 +1206,7 @@ static int ns83820_get_link_ksettings(struct net_device *ndev, struct ethtool_link_ksettings *cmd) { struct ns83820 *dev = PRIV(ndev); - u32 cfg, tanar, tbicr; + u32 cfg, tbicr; int fullduplex = 0; u32 supported; @@ -1226,7 +1225,7 @@ static int ns83820_get_link_ksettings(struct net_device *ndev, /* read current configuration */ cfg = readl(dev->base + CFG) ^ SPDSTS_POLARITY; - tanar = readl(dev->base + TANAR); + readl(dev->base + TANAR); tbicr = readl(dev->base + TBICR); fullduplex = (cfg & CFG_DUPSTS) ? 1 : 0; @@ -1902,12 +1901,12 @@ static int ns83820_init_one(struct pci_dev *pci_dev, /* See if we can set the dma mask early on; failure is fatal. */ if (sizeof(dma_addr_t) == 8 && - !pci_set_dma_mask(pci_dev, DMA_BIT_MASK(64))) { + !dma_set_mask(&pci_dev->dev, DMA_BIT_MASK(64))) { using_dac = 1; - } else if (!pci_set_dma_mask(pci_dev, DMA_BIT_MASK(32))) { + } else if (!dma_set_mask(&pci_dev->dev, DMA_BIT_MASK(32))) { using_dac = 0; } else { - dev_warn(&pci_dev->dev, "pci_set_dma_mask failed!\n"); + dev_warn(&pci_dev->dev, "dma_set_mask failed!\n"); return -ENODEV; } @@ -1927,7 +1926,7 @@ static int ns83820_init_one(struct pci_dev *pci_dev, SET_NETDEV_DEV(ndev, &pci_dev->dev); INIT_WORK(&dev->tq_refill, queue_refill); - tasklet_init(&dev->rx_tasklet, rx_action, (unsigned long)ndev); + tasklet_setup(&dev->rx_tasklet, rx_action); err = pci_enable_device(pci_dev); if (err) { @@ -1938,10 +1937,12 @@ static int ns83820_init_one(struct pci_dev *pci_dev, pci_set_master(pci_dev); addr = pci_resource_start(pci_dev, 1); dev->base = ioremap(addr, PAGE_SIZE); - dev->tx_descs = pci_alloc_consistent(pci_dev, - 4 * DESC_SIZE * NR_TX_DESC, &dev->tx_phy_descs); - dev->rx_info.descs = pci_alloc_consistent(pci_dev, - 4 * DESC_SIZE * NR_RX_DESC, &dev->rx_info.phy_descs); + dev->tx_descs = dma_alloc_coherent(&pci_dev->dev, + 4 * DESC_SIZE * NR_TX_DESC, + &dev->tx_phy_descs, GFP_KERNEL); + dev->rx_info.descs = dma_alloc_coherent(&pci_dev->dev, + 4 * DESC_SIZE * NR_RX_DESC, + &dev->rx_info.phy_descs, GFP_KERNEL); err = -ENOMEM; if (!dev->base || !dev->tx_descs || !dev->rx_info.descs) goto out_disable; @@ -2183,8 +2184,10 @@ out_free_irq: out_disable: if (dev->base) iounmap(dev->base); - pci_free_consistent(pci_dev, 4 * DESC_SIZE * NR_TX_DESC, dev->tx_descs, dev->tx_phy_descs); - pci_free_consistent(pci_dev, 4 * DESC_SIZE * NR_RX_DESC, dev->rx_info.descs, dev->rx_info.phy_descs); + dma_free_coherent(&pci_dev->dev, 4 * DESC_SIZE * NR_TX_DESC, + dev->tx_descs, dev->tx_phy_descs); + dma_free_coherent(&pci_dev->dev, 4 * DESC_SIZE * NR_RX_DESC, + dev->rx_info.descs, dev->rx_info.phy_descs); pci_disable_device(pci_dev); out_free: free_netdev(ndev); @@ -2205,10 +2208,10 @@ static void ns83820_remove_one(struct pci_dev *pci_dev) unregister_netdev(ndev); free_irq(dev->pci_dev->irq, ndev); iounmap(dev->base); - pci_free_consistent(dev->pci_dev, 4 * DESC_SIZE * NR_TX_DESC, - dev->tx_descs, dev->tx_phy_descs); - pci_free_consistent(dev->pci_dev, 4 * DESC_SIZE * NR_RX_DESC, - dev->rx_info.descs, dev->rx_info.phy_descs); + dma_free_coherent(&dev->pci_dev->dev, 4 * DESC_SIZE * NR_TX_DESC, + dev->tx_descs, dev->tx_phy_descs); + dma_free_coherent(&dev->pci_dev->dev, 4 * DESC_SIZE * NR_RX_DESC, + dev->rx_info.descs, dev->rx_info.phy_descs); pci_disable_device(dev->pci_dev); free_netdev(ndev); } diff --git a/drivers/net/ethernet/natsemi/sonic.c b/drivers/net/ethernet/natsemi/sonic.c index dd3605aa5f23..d17d1b4f2585 100644 --- a/drivers/net/ethernet/natsemi/sonic.c +++ b/drivers/net/ethernet/natsemi/sonic.c @@ -143,7 +143,7 @@ static int sonic_open(struct net_device *dev) /* * Initialize the SONIC */ - sonic_init(dev); + sonic_init(dev, true); netif_start_queue(dev); @@ -153,7 +153,7 @@ static int sonic_open(struct net_device *dev) } /* Wait for the SONIC to become idle. */ -static void sonic_quiesce(struct net_device *dev, u16 mask) +static void sonic_quiesce(struct net_device *dev, u16 mask, bool may_sleep) { struct sonic_local * __maybe_unused lp = netdev_priv(dev); int i; @@ -163,7 +163,7 @@ static void sonic_quiesce(struct net_device *dev, u16 mask) bits = SONIC_READ(SONIC_CMD) & mask; if (!bits) return; - if (irqs_disabled() || in_interrupt()) + if (!may_sleep) udelay(20); else usleep_range(100, 200); @@ -187,7 +187,7 @@ static int sonic_close(struct net_device *dev) * stop the SONIC, disable interrupts */ SONIC_WRITE(SONIC_CMD, SONIC_CR_RXDIS); - sonic_quiesce(dev, SONIC_CR_ALL); + sonic_quiesce(dev, SONIC_CR_ALL, true); SONIC_WRITE(SONIC_IMR, 0); SONIC_WRITE(SONIC_ISR, 0x7fff); @@ -229,7 +229,7 @@ static void sonic_tx_timeout(struct net_device *dev, unsigned int txqueue) * disable all interrupts before releasing DMA buffers */ SONIC_WRITE(SONIC_CMD, SONIC_CR_RXDIS); - sonic_quiesce(dev, SONIC_CR_ALL); + sonic_quiesce(dev, SONIC_CR_ALL, false); SONIC_WRITE(SONIC_IMR, 0); SONIC_WRITE(SONIC_ISR, 0x7fff); @@ -246,7 +246,7 @@ static void sonic_tx_timeout(struct net_device *dev, unsigned int txqueue) } } /* Try to restart the adaptor. */ - sonic_init(dev); + sonic_init(dev, false); lp->stats.tx_errors++; netif_trans_update(dev); /* prevent tx timeout */ netif_wake_queue(dev); @@ -692,9 +692,9 @@ static void sonic_multicast_list(struct net_device *dev) /* LCAM and TXP commands can't be used simultaneously */ spin_lock_irqsave(&lp->lock, flags); - sonic_quiesce(dev, SONIC_CR_TXP); + sonic_quiesce(dev, SONIC_CR_TXP, false); SONIC_WRITE(SONIC_CMD, SONIC_CR_LCAM); - sonic_quiesce(dev, SONIC_CR_LCAM); + sonic_quiesce(dev, SONIC_CR_LCAM, false); spin_unlock_irqrestore(&lp->lock, flags); } } @@ -708,7 +708,7 @@ static void sonic_multicast_list(struct net_device *dev) /* * Initialize the SONIC ethernet controller. */ -static int sonic_init(struct net_device *dev) +static int sonic_init(struct net_device *dev, bool may_sleep) { struct sonic_local *lp = netdev_priv(dev); int i; @@ -730,7 +730,7 @@ static int sonic_init(struct net_device *dev) */ SONIC_WRITE(SONIC_CMD, 0); SONIC_WRITE(SONIC_CMD, SONIC_CR_RXDIS | SONIC_CR_STP); - sonic_quiesce(dev, SONIC_CR_ALL); + sonic_quiesce(dev, SONIC_CR_ALL, may_sleep); /* * initialize the receive resource area @@ -759,7 +759,7 @@ static int sonic_init(struct net_device *dev) netif_dbg(lp, ifup, dev, "%s: issuing RRRA command\n", __func__); SONIC_WRITE(SONIC_CMD, SONIC_CR_RRRA); - sonic_quiesce(dev, SONIC_CR_RRRA); + sonic_quiesce(dev, SONIC_CR_RRRA, may_sleep); /* * Initialize the receive descriptors so that they @@ -834,7 +834,7 @@ static int sonic_init(struct net_device *dev) * load the CAM */ SONIC_WRITE(SONIC_CMD, SONIC_CR_LCAM); - sonic_quiesce(dev, SONIC_CR_LCAM); + sonic_quiesce(dev, SONIC_CR_LCAM, may_sleep); /* * enable receiver, disable loopback diff --git a/drivers/net/ethernet/natsemi/sonic.h b/drivers/net/ethernet/natsemi/sonic.h index 3cbb62c860c8..a5b803eb8c8a 100644 --- a/drivers/net/ethernet/natsemi/sonic.h +++ b/drivers/net/ethernet/natsemi/sonic.h @@ -338,7 +338,7 @@ static void sonic_rx(struct net_device *dev); static int sonic_close(struct net_device *dev); static struct net_device_stats *sonic_get_stats(struct net_device *dev); static void sonic_multicast_list(struct net_device *dev); -static int sonic_init(struct net_device *dev); +static int sonic_init(struct net_device *dev, bool may_sleep); static void sonic_tx_timeout(struct net_device *dev, unsigned int txqueue); static void sonic_msg_init(struct net_device *dev); static int sonic_alloc_descriptors(struct net_device *dev); diff --git a/drivers/net/ethernet/neterion/s2io.c b/drivers/net/ethernet/neterion/s2io.c index bc94970bea45..d13d92bf7447 100644 --- a/drivers/net/ethernet/neterion/s2io.c +++ b/drivers/net/ethernet/neterion/s2io.c @@ -1000,7 +1000,7 @@ static void free_shared_mem(struct s2io_nic *nic) } } -/** +/* * s2io_verify_pci_mode - */ @@ -1035,7 +1035,7 @@ static int s2io_on_nec_bridge(struct pci_dev *s2io_pdev) } static int bus_speed[8] = {33, 133, 133, 200, 266, 133, 200, 266}; -/** +/* * s2io_print_pci_mode - */ static int s2io_print_pci_mode(struct s2io_nic *nic) @@ -2064,6 +2064,9 @@ static void en_dis_able_nic_intrs(struct s2io_nic *nic, u16 mask, int flag) /** * verify_pcc_quiescent- Checks for PCC quiescent state + * @sp : private member of the device structure, which is a pointer to the + * s2io_nic structure. + * @flag: boolean controlling function path * Return: 1 If PCC is quiescence * 0 If PCC is not quiescence */ @@ -2099,6 +2102,8 @@ static int verify_pcc_quiescent(struct s2io_nic *sp, int flag) } /** * verify_xena_quiescence - Checks whether the H/W is ready + * @sp : private member of the device structure, which is a pointer to the + * s2io_nic structure. * Description: Returns whether the H/W is ready to go or not. Depending * on whether adapter enable bit was written or not the comparison * differs and the calling function passes the input argument flag to @@ -2305,6 +2310,9 @@ static int start_nic(struct s2io_nic *nic) } /** * s2io_txdl_getskb - Get the skb from txdl, unmap and return skb + * @fifo_data: fifo data pointer + * @txdlp: descriptor + * @get_off: unused */ static struct sk_buff *s2io_txdl_getskb(struct fifo_info *fifo_data, struct TxD *txdlp, int get_off) @@ -2391,7 +2399,7 @@ static void free_tx_buffers(struct s2io_nic *nic) /** * stop_nic - To stop the nic - * @nic ; device private variable. + * @nic : device private variable. * Description: * This function does exactly the opposite of what the start_nic() * function does. This function is called to stop the device. @@ -2419,7 +2427,8 @@ static void stop_nic(struct s2io_nic *nic) /** * fill_rx_buffers - Allocates the Rx side skbs - * @ring_info: per ring structure + * @nic : device private variable. + * @ring: per ring structure * @from_card_up: If this is true, we will map the buffer to get * the dma address for buf0 and buf1 to give it to the card. * Else we will sync the already mapped buffer to give it to the card. @@ -2864,7 +2873,7 @@ static void s2io_netpoll(struct net_device *dev) /** * rx_intr_handler - Rx interrupt handler - * @ring_info: per ring structure. + * @ring_data: per ring structure. * @budget: budget for napi processing. * Description: * If the interrupt is because of a received frame or if the @@ -2972,7 +2981,7 @@ static int rx_intr_handler(struct ring_info *ring_data, int budget) /** * tx_intr_handler - Transmit interrupt handler - * @nic : device private variable + * @fifo_data : fifo data pointer * Description: * If an interrupt was raised to indicate DMA complete of the * Tx packet, this function is called. It identifies the last TxD @@ -3153,6 +3162,8 @@ static u64 s2io_mdio_read(u32 mmd_type, u64 addr, struct net_device *dev) /** * s2io_chk_xpak_counter - Function to check the status of the xpak counters * @counter : counter value to be updated + * @regs_stat : registers status + * @index : index * @flag : flag to indicate the status * @type : counter type * Description: @@ -3309,8 +3320,9 @@ static void s2io_updt_xpak_counter(struct net_device *dev) /** * wait_for_cmd_complete - waits for a command to complete. - * @sp : private member of the device structure, which is a pointer to the - * s2io_nic structure. + * @addr: address + * @busy_bit: bit to check for busy + * @bit_state: state to check * Description: Function that waits for a command to Write into RMAC * ADDR DATA registers to be completed and returns either success or * error depending on whether the command was complete or not. @@ -4335,7 +4347,7 @@ static int do_s2io_chk_alarm_bit(u64 value, void __iomem *addr, /** * s2io_handle_errors - Xframe error indication handler - * @nic: device private variable + * @dev_id: opaque handle to dev * Description: Handle alarms such as loss of link, single or * double ECC errors, critical and serious errors. * Return Value: @@ -4739,7 +4751,7 @@ static irqreturn_t s2io_isr(int irq, void *dev_id) return IRQ_HANDLED; } -/** +/* * s2io_updt_stats - */ static void s2io_updt_stats(struct s2io_nic *sp) @@ -5168,7 +5180,7 @@ static u64 do_s2io_read_unicast_mc(struct s2io_nic *sp, int offset) return tmp64 >> 16; } -/** +/* * s2io_set_mac_addr - driver entry point */ @@ -5243,8 +5255,7 @@ static int do_s2io_prog_unicast(struct net_device *dev, u8 *addr) /** * s2io_ethtool_set_link_ksettings - Sets different link parameters. - * @sp : private member of the device structure, which is a pointer to the - * s2io_nic structure. + * @dev : pointer to netdev * @cmd: pointer to the structure with parameters given by ethtool to set * link information. * Description: @@ -5273,8 +5284,7 @@ s2io_ethtool_set_link_ksettings(struct net_device *dev, /** * s2io_ethtol_get_link_ksettings - Return link specific information. - * @sp : private member of the device structure, pointer to the - * s2io_nic structure. + * @dev: pointer to netdev * @cmd : pointer to the structure with parameters given by ethtool * to return link information. * Description: @@ -5313,8 +5323,7 @@ s2io_ethtool_get_link_ksettings(struct net_device *dev, /** * s2io_ethtool_gdrvinfo - Returns driver specific information. - * @sp : private member of the device structure, which is a pointer to the - * s2io_nic structure. + * @dev: pointer to netdev * @info : pointer to the structure with parameters given by ethtool to * return driver information. * Description: @@ -5335,11 +5344,10 @@ static void s2io_ethtool_gdrvinfo(struct net_device *dev, /** * s2io_ethtool_gregs - dumps the entire space of Xfame into the buffer. - * @sp: private member of the device structure, which is a pointer to the - * s2io_nic structure. + * @dev: pointer to netdev * @regs : pointer to the structure with parameters given by ethtool for - * dumping the registers. - * @reg_space: The input argument into which all the registers are dumped. + * dumping the registers. + * @space: The input argument into which all the registers are dumped. * Description: * Dumps the entire register space of xFrame NIC into the user given * buffer area. @@ -5471,8 +5479,7 @@ static void s2io_ethtool_gringparam(struct net_device *dev, /** * s2io_ethtool_getpause_data -Pause frame frame generation and reception. - * @sp : private member of the device structure, which is a pointer to the - * s2io_nic structure. + * @dev: pointer to netdev * @ep : pointer to the structure with pause parameters given by ethtool. * Description: * Returns the Pause frame generation and reception capability of the NIC. @@ -5496,8 +5503,7 @@ static void s2io_ethtool_getpause_data(struct net_device *dev, /** * s2io_ethtool_setpause_data - set/reset pause frame generation. - * @sp : private member of the device structure, which is a pointer to the - * s2io_nic structure. + * @dev: pointer to netdev * @ep : pointer to the structure with pause parameters given by ethtool. * Description: * It can be used to set or reset Pause frame generation or reception @@ -5526,6 +5532,7 @@ static int s2io_ethtool_setpause_data(struct net_device *dev, return 0; } +#define S2IO_DEV_ID 5 /** * read_eeprom - reads 4 bytes of data from user given offset. * @sp : private member of the device structure, which is a pointer to the @@ -5541,8 +5548,6 @@ static int s2io_ethtool_setpause_data(struct net_device *dev, * Return value: * -1 on failure and 0 on success. */ - -#define S2IO_DEV_ID 5 static int read_eeprom(struct s2io_nic *sp, int off, u64 *data) { int ret = -1; @@ -5734,8 +5739,7 @@ static void s2io_vpd_read(struct s2io_nic *nic) /** * s2io_ethtool_geeprom - reads the value stored in the Eeprom. - * @sp : private member of the device structure, which is a pointer to the - * s2io_nic structure. + * @dev: pointer to netdev * @eeprom : pointer to the user level structure provided by ethtool, * containing all relevant information. * @data_buf : user defined value to be written into Eeprom. @@ -5771,11 +5775,10 @@ static int s2io_ethtool_geeprom(struct net_device *dev, /** * s2io_ethtool_seeprom - tries to write the user provided value in Eeprom - * @sp : private member of the device structure, which is a pointer to the - * s2io_nic structure. + * @dev: pointer to netdev * @eeprom : pointer to the user level structure provided by ethtool, * containing all relevant information. - * @data_buf ; user defined value to be written into Eeprom. + * @data_buf : user defined value to be written into Eeprom. * Description: * Tries to write the user provided value in the Eeprom, at the offset * given by the user. @@ -6027,7 +6030,7 @@ static int s2io_bist_test(struct s2io_nic *sp, uint64_t *data) /** * s2io_link_test - verifies the link state of the nic - * @sp ; private member of the device structure, which is a pointer to the + * @sp: private member of the device structure, which is a pointer to the * s2io_nic structure. * @data: variable that returns the result of each of the test conducted by * the driver. @@ -6150,8 +6153,7 @@ static int s2io_rldram_test(struct s2io_nic *sp, uint64_t *data) /** * s2io_ethtool_test - conducts 6 tsets to determine the health of card. - * @sp : private member of the device structure, which is a pointer to the - * s2io_nic structure. + * @dev: pointer to netdev * @ethtest : pointer to a ethtool command specific structure that will be * returned to the user. * @data : variable that returns the result of each of the test @@ -6597,7 +6599,7 @@ static const struct ethtool_ops netdev_ethtool_ops = { /** * s2io_ioctl - Entry point for the Ioctl * @dev : Device pointer. - * @ifr : An IOCTL specefic structure, that can contain a pointer to + * @rq : An IOCTL specefic structure, that can contain a pointer to * a proprietary structure used to pass information to the driver. * @cmd : This is used to distinguish between the different commands that * can be passed to the IOCTL functions. @@ -6650,7 +6652,7 @@ static int s2io_change_mtu(struct net_device *dev, int new_mtu) /** * s2io_set_link - Set the LInk status - * @data: long pointer to device private structue + * @work: work struct containing a pointer to device private structue * Description: Sets the link status for the adapter */ @@ -7187,7 +7189,7 @@ static int s2io_card_up(struct s2io_nic *sp) /** * s2io_restart_nic - Resets the NIC. - * @data : long pointer to the device private structure + * @work : work struct containing a pointer to the device private structure * Description: * This function is scheduled to be run by the s2io_tx_watchdog * function after 0.5 secs to reset the NIC. The idea is to reduce @@ -7218,6 +7220,7 @@ out_unlock: /** * s2io_tx_watchdog - Watchdog for transmit side. * @dev : Pointer to net device structure + * @txqueue: index of the hanging queue * Description: * This function is triggered if the Tx Queue is stopped * for a pre-defined amount of time when the Interface is still up. @@ -7242,11 +7245,8 @@ static void s2io_tx_watchdog(struct net_device *dev, unsigned int txqueue) /** * rx_osm_handler - To perform some OS related operations on SKB. - * @sp: private member of the device structure,pointer to s2io_nic structure. - * @skb : the socket buffer pointer. - * @len : length of the packet - * @cksum : FCS checksum of the frame. - * @ring_no : the ring from which this RxD was extracted. + * @ring_data : the ring from which this RxD was extracted. + * @rxdp: descriptor * Description: * This function is called by the Rx interrupt serivce routine to perform * some OS related operations on the SKB before passing it to the upper @@ -7576,9 +7576,10 @@ static int s2io_verify_parm(struct pci_dev *pdev, u8 *dev_intr_type, } /** - * rts_ds_steer - Receive traffic steering based on IPv4 or IPv6 TOS - * or Traffic class respectively. + * rts_ds_steer - Receive traffic steering based on IPv4 or IPv6 TOS or Traffic class respectively. * @nic: device private variable + * @ds_codepoint: data + * @ring: ring index * Description: The function configures the receive steering to * desired receive ring. * Return Value: SUCCESS on success and diff --git a/drivers/net/ethernet/neterion/vxge/vxge-config.c b/drivers/net/ethernet/neterion/vxge/vxge-config.c index 78eba10300ae..f5d48d7c4ce2 100644 --- a/drivers/net/ethernet/neterion/vxge/vxge-config.c +++ b/drivers/net/ethernet/neterion/vxge/vxge-config.c @@ -988,6 +988,9 @@ exit: /** * vxge_hw_device_hw_info_get - Get the hw information + * @bar0: the bar + * @hw_info: the hw_info struct + * * Returns the vpath mask that has the bits set for each vpath allocated * for the driver, FW version information, and the first mac address for * each vpath @@ -2303,16 +2306,9 @@ exit: static inline void vxge_os_dma_malloc_async(struct pci_dev *pdev, void *devh, unsigned long size) { - gfp_t flags; void *vaddr; - if (in_interrupt()) - flags = GFP_ATOMIC | GFP_DMA; - else - flags = GFP_KERNEL | GFP_DMA; - - vaddr = kmalloc((size), flags); - + vaddr = kmalloc(size, GFP_KERNEL | GFP_DMA); vxge_hw_blockpool_block_add(devh, vaddr, size, pdev, pdev); } @@ -3926,7 +3922,7 @@ exit: /** * vxge_hw_vpath_check_leak - Check for memory leak - * @ringh: Handle to the ring object used for receive + * @ring: Handle to the ring object used for receive * * If PRC_RXD_DOORBELL_VPn.NEW_QW_CNT is larger or equal to * PRC_CFG6_VPn.RXD_SPAT then a leak has occurred. diff --git a/drivers/net/ethernet/neterion/vxge/vxge-config.h b/drivers/net/ethernet/neterion/vxge/vxge-config.h index 373165119850..0cd0750484ae 100644 --- a/drivers/net/ethernet/neterion/vxge/vxge-config.h +++ b/drivers/net/ethernet/neterion/vxge/vxge-config.h @@ -1899,18 +1899,13 @@ static inline void *vxge_os_dma_malloc(struct pci_dev *pdev, struct pci_dev **p_dmah, struct pci_dev **p_dma_acch) { - gfp_t flags; void *vaddr; unsigned long misaligned = 0; int realloc_flag = 0; *p_dma_acch = *p_dmah = NULL; - if (in_interrupt()) - flags = GFP_ATOMIC | GFP_DMA; - else - flags = GFP_KERNEL | GFP_DMA; realloc: - vaddr = kmalloc((size), flags); + vaddr = kmalloc(size, GFP_KERNEL | GFP_DMA); if (vaddr == NULL) return vaddr; misaligned = (unsigned long)VXGE_ALIGN((unsigned long)vaddr, diff --git a/drivers/net/ethernet/neterion/vxge/vxge-ethtool.c b/drivers/net/ethernet/neterion/vxge/vxge-ethtool.c index 03c3d1230c17..4d91026485ae 100644 --- a/drivers/net/ethernet/neterion/vxge/vxge-ethtool.c +++ b/drivers/net/ethernet/neterion/vxge/vxge-ethtool.c @@ -119,7 +119,7 @@ static void vxge_ethtool_gdrvinfo(struct net_device *dev, * @dev: device pointer. * @regs: pointer to the structure with parameters given by ethtool for * dumping the registers. - * @reg_space: The input argument into which all the registers are dumped. + * @space: The input argument into which all the registers are dumped. * * Dumps the vpath register space of Titan NIC into the user given * buffer area. diff --git a/drivers/net/ethernet/neterion/vxge/vxge-main.c b/drivers/net/ethernet/neterion/vxge/vxge-main.c index 1ded4e275086..87892bd992b1 100644 --- a/drivers/net/ethernet/neterion/vxge/vxge-main.c +++ b/drivers/net/ethernet/neterion/vxge/vxge-main.c @@ -1275,6 +1275,7 @@ _set_all_mcast: /** * vxge_set_mac_addr * @dev: pointer to the device structure + * @p: socket info * * Update entry "0" (default MAC addr) */ @@ -1799,7 +1800,7 @@ static void vxge_reset(struct work_struct *work) /** * vxge_poll - Receive handler when Receive Polling is used. - * @dev: pointer to the device structure. + * @napi: pointer to the napi structure. * @budget: Number of packets budgeted to be processed in this iteration. * * This function comes into picture only if Receive side is being handled @@ -3096,7 +3097,7 @@ static int vxge_change_mtu(struct net_device *dev, int new_mtu) /** * vxge_get_stats64 * @dev: pointer to the device structure - * @stats: pointer to struct rtnl_link_stats64 + * @net_stats: pointer to struct rtnl_link_stats64 * */ static void @@ -3245,7 +3246,7 @@ static int vxge_hwtstamp_get(struct vxgedev *vdev, void __user *data) /** * vxge_ioctl * @dev: Device pointer. - * @ifr: An IOCTL specific structure, that can contain a pointer to + * @rq: An IOCTL specific structure, that can contain a pointer to * a proprietary structure used to pass information to the driver. * @cmd: This is used to distinguish between the different commands that * can be passed to the IOCTL functions. @@ -3269,6 +3270,7 @@ static int vxge_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) /** * vxge_tx_watchdog * @dev: pointer to net device structure + * @txqueue: index of the hanging queue * * Watchdog for transmit side. * This function is triggered if the Tx Queue is stopped @@ -4002,6 +4004,7 @@ static void vxge_print_parm(struct vxgedev *vdev, u64 vpath_mask) /** * vxge_pm_suspend - vxge power management suspend entry point + * @dev_d: device pointer * */ static int __maybe_unused vxge_pm_suspend(struct device *dev_d) @@ -4010,6 +4013,7 @@ static int __maybe_unused vxge_pm_suspend(struct device *dev_d) } /** * vxge_pm_resume - vxge power management resume entry point + * @dev_d: device pointer * */ static int __maybe_unused vxge_pm_resume(struct device *dev_d) @@ -4539,7 +4543,7 @@ vxge_probe(struct pci_dev *pdev, const struct pci_device_id *pre) * due to the fact that HWTS is using the FCS as the location of the * timestamp. The HW FCS checking will still correctly determine if * there is a valid checksum, and the FCS is being removed by the driver - * anyway. So no fucntionality is being lost. Since it is always + * anyway. So no functionality is being lost. Since it is always * enabled, we now simply use the ioctl call to set whether or not the * driver should be paying attention to the HWTS. */ diff --git a/drivers/net/ethernet/neterion/vxge/vxge-traffic.c b/drivers/net/ethernet/neterion/vxge/vxge-traffic.c index 709d20d9938f..ee164970b267 100644 --- a/drivers/net/ethernet/neterion/vxge/vxge-traffic.c +++ b/drivers/net/ethernet/neterion/vxge/vxge-traffic.c @@ -30,8 +30,6 @@ */ enum vxge_hw_status vxge_hw_vpath_intr_enable(struct __vxge_hw_vpath_handle *vp) { - u64 val64; - struct __vxge_hw_virtualpath *vpath; struct vxge_hw_vpath_reg __iomem *vp_reg; enum vxge_hw_status status = VXGE_HW_OK; @@ -84,7 +82,7 @@ enum vxge_hw_status vxge_hw_vpath_intr_enable(struct __vxge_hw_vpath_handle *vp) __vxge_hw_pio_mem_write32_upper((u32)VXGE_HW_INTR_MASK_ALL, &vp_reg->xgmac_vp_int_status); - val64 = readq(&vp_reg->vpath_general_int_status); + readq(&vp_reg->vpath_general_int_status); /* Mask unwanted interrupts */ @@ -157,8 +155,6 @@ exit: enum vxge_hw_status vxge_hw_vpath_intr_disable( struct __vxge_hw_vpath_handle *vp) { - u64 val64; - struct __vxge_hw_virtualpath *vpath; enum vxge_hw_status status = VXGE_HW_OK; struct vxge_hw_vpath_reg __iomem *vp_reg; @@ -179,8 +175,6 @@ enum vxge_hw_status vxge_hw_vpath_intr_disable( (u32)VXGE_HW_INTR_MASK_ALL, &vp_reg->vpath_general_int_mask); - val64 = VXGE_HW_TIM_CLR_INT_EN_VP(1 << (16 - vpath->vp_id)); - writeq(VXGE_HW_INTR_MASK_ALL, &vp_reg->kdfcctl_errors_mask); __vxge_hw_pio_mem_write32_upper((u32)VXGE_HW_INTR_MASK_ALL, @@ -284,7 +278,7 @@ void vxge_hw_vpath_dynamic_rti_rtimer_set(struct __vxge_hw_ring *ring) /** * vxge_hw_channel_msix_mask - Mask MSIX Vector. - * @channeh: Channel for rx or tx handle + * @channel: Channel for rx or tx handle * @msix_id: MSIX ID * * The function masks the msix interrupt for the given msix_id @@ -301,7 +295,7 @@ void vxge_hw_channel_msix_mask(struct __vxge_hw_channel *channel, int msix_id) /** * vxge_hw_channel_msix_unmask - Unmask the MSIX Vector. - * @channeh: Channel for rx or tx handle + * @channel: Channel for rx or tx handle * @msix_id: MSI ID * * The function unmasks the msix interrupt for the given msix_id @@ -356,8 +350,6 @@ u32 vxge_hw_device_set_intr_type(struct __vxge_hw_device *hldev, u32 intr_mode) /** * vxge_hw_device_intr_enable - Enable interrupts. * @hldev: HW device handle. - * @op: One of the enum vxge_hw_device_intr enumerated values specifying - * the type(s) of interrupts to enable. * * Enable Titan interrupts. The function is to be executed the last in * Titan initialization sequence. @@ -411,8 +403,6 @@ void vxge_hw_device_intr_enable(struct __vxge_hw_device *hldev) /** * vxge_hw_device_intr_disable - Disable Titan interrupts. * @hldev: HW device handle. - * @op: One of the enum vxge_hw_device_intr enumerated values specifying - * the type(s) of interrupts to disable. * * Disable Titan interrupts. * @@ -487,9 +477,7 @@ void vxge_hw_device_unmask_all(struct __vxge_hw_device *hldev) */ void vxge_hw_device_flush_io(struct __vxge_hw_device *hldev) { - u32 val32; - - val32 = readl(&hldev->common_reg->titan_general_int_status); + readl(&hldev->common_reg->titan_general_int_status); } /** @@ -1414,7 +1402,7 @@ u32 vxge_hw_fifo_free_txdl_count_get(struct __vxge_hw_fifo *fifoh) /** * vxge_hw_fifo_txdl_reserve - Reserve fifo descriptor. - * @fifoh: Handle to the fifo object used for non offload send + * @fifo: Handle to the fifo object used for non offload send * @txdlh: Reserved descriptor. On success HW fills this "out" parameter * with a valid handle. * @txdl_priv: Buffer to return the pointer to per txdl space @@ -1525,8 +1513,6 @@ void vxge_hw_fifo_txdl_buffer_set(struct __vxge_hw_fifo *fifo, * vxge_hw_fifo_txdl_post - Post descriptor on the fifo channel. * @fifo: Handle to the fifo object used for non offload send * @txdlh: Descriptor obtained via vxge_hw_fifo_txdl_reserve() - * @frags: Number of contiguous buffers that are part of a single - * transmit operation. * * Post descriptor on the 'fifo' type channel for transmission. * Prior to posting the descriptor should be filled in accordance with @@ -1699,8 +1685,7 @@ void vxge_hw_fifo_txdl_free(struct __vxge_hw_fifo *fifo, void *txdlh) } /** - * vxge_hw_vpath_mac_addr_add - Add the mac address entry for this vpath - * to MAC address table. + * vxge_hw_vpath_mac_addr_add - Add the mac address entry for this vpath to MAC address table. * @vp: Vpath handle. * @macaddr: MAC address to be added for this vpath into the list * @macaddr_mask: MAC address mask for macaddr @@ -1716,8 +1701,8 @@ void vxge_hw_fifo_txdl_free(struct __vxge_hw_fifo *fifo, void *txdlh) enum vxge_hw_status vxge_hw_vpath_mac_addr_add( struct __vxge_hw_vpath_handle *vp, - u8 (macaddr)[ETH_ALEN], - u8 (macaddr_mask)[ETH_ALEN], + u8 *macaddr, + u8 *macaddr_mask, enum vxge_hw_vpath_mac_addr_add_mode duplicate_mode) { u32 i; @@ -1765,13 +1750,13 @@ exit: } /** - * vxge_hw_vpath_mac_addr_get - Get the first mac address entry for this vpath - * from MAC address table. + * vxge_hw_vpath_mac_addr_get - Get the first mac address entry * @vp: Vpath handle. * @macaddr: First MAC address entry for this vpath in the list * @macaddr_mask: MAC address mask for macaddr * - * Returns the first mac address and mac address mask in the list for this + * Get the first mac address entry for this vpath from MAC address table. + * Return: the first mac address and mac address mask in the list for this * vpath. * see also: vxge_hw_vpath_mac_addr_get_next * @@ -1779,8 +1764,8 @@ exit: enum vxge_hw_status vxge_hw_vpath_mac_addr_get( struct __vxge_hw_vpath_handle *vp, - u8 (macaddr)[ETH_ALEN], - u8 (macaddr_mask)[ETH_ALEN]) + u8 *macaddr, + u8 *macaddr_mask) { u32 i; u64 data1 = 0ULL; @@ -1816,14 +1801,13 @@ exit: } /** - * vxge_hw_vpath_mac_addr_get_next - Get the next mac address entry for this - * vpath - * from MAC address table. + * vxge_hw_vpath_mac_addr_get_next - Get the next mac address entry * @vp: Vpath handle. * @macaddr: Next MAC address entry for this vpath in the list * @macaddr_mask: MAC address mask for macaddr * - * Returns the next mac address and mac address mask in the list for this + * Get the next mac address entry for this vpath from MAC address table. + * Return: the next mac address and mac address mask in the list for this * vpath. * see also: vxge_hw_vpath_mac_addr_get * @@ -1831,8 +1815,8 @@ exit: enum vxge_hw_status vxge_hw_vpath_mac_addr_get_next( struct __vxge_hw_vpath_handle *vp, - u8 (macaddr)[ETH_ALEN], - u8 (macaddr_mask)[ETH_ALEN]) + u8 *macaddr, + u8 *macaddr_mask) { u32 i; u64 data1 = 0ULL; @@ -1869,8 +1853,7 @@ exit: } /** - * vxge_hw_vpath_mac_addr_delete - Delete the mac address entry for this vpath - * to MAC address table. + * vxge_hw_vpath_mac_addr_delete - Delete the mac address entry for this vpath to MAC address table. * @vp: Vpath handle. * @macaddr: MAC address to be added for this vpath into the list * @macaddr_mask: MAC address mask for macaddr @@ -1884,8 +1867,8 @@ exit: enum vxge_hw_status vxge_hw_vpath_mac_addr_delete( struct __vxge_hw_vpath_handle *vp, - u8 (macaddr)[ETH_ALEN], - u8 (macaddr_mask)[ETH_ALEN]) + u8 *macaddr, + u8 *macaddr_mask) { u32 i; u64 data1 = 0ULL; @@ -1916,8 +1899,7 @@ exit: } /** - * vxge_hw_vpath_vid_add - Add the vlan id entry for this vpath - * to vlan id table. + * vxge_hw_vpath_vid_add - Add the vlan id entry for this vpath to vlan id table. * @vp: Vpath handle. * @vid: vlan id to be added for this vpath into the list * @@ -2375,7 +2357,6 @@ enum vxge_hw_status vxge_hw_vpath_poll_rx(struct __vxge_hw_ring *ring) u8 t_code; enum vxge_hw_status status = VXGE_HW_OK; void *first_rxdh; - u64 val64 = 0; int new_count = 0; ring->cmpl_cnt = 0; @@ -2403,8 +2384,7 @@ enum vxge_hw_status vxge_hw_vpath_poll_rx(struct __vxge_hw_ring *ring) } writeq(VXGE_HW_PRC_RXD_DOORBELL_NEW_QW_CNT(new_count), &ring->vp_reg->prc_rxd_doorbell); - val64 = - readl(&ring->common_reg->titan_general_int_status); + readl(&ring->common_reg->titan_general_int_status); ring->doorbell_cnt = 0; } } @@ -2413,9 +2393,11 @@ enum vxge_hw_status vxge_hw_vpath_poll_rx(struct __vxge_hw_ring *ring) } /** - * vxge_hw_vpath_poll_tx - Poll Tx for completed descriptors and process - * the same. + * vxge_hw_vpath_poll_tx - Poll Tx for completed descriptors and process the same. * @fifo: Handle to the fifo object used for non offload send + * @skb_ptr: pointer to skb + * @nr_skb: number of skbs + * @more: more is coming * * The function polls the Tx for the completed descriptors and calls * the driver via supplied completion callback. diff --git a/drivers/net/ethernet/netronome/nfp/bpf/offload.c b/drivers/net/ethernet/netronome/nfp/bpf/offload.c index ac02369174a9..53851853562c 100644 --- a/drivers/net/ethernet/netronome/nfp/bpf/offload.c +++ b/drivers/net/ethernet/netronome/nfp/bpf/offload.c @@ -111,7 +111,9 @@ static int nfp_map_ptrs_record(struct nfp_app_bpf *bpf, struct nfp_prog *nfp_prog, struct bpf_prog *prog) { - int i, cnt, err; + int i, cnt, err = 0; + + mutex_lock(&prog->aux->used_maps_mutex); /* Quickly count the maps we will have to remember */ cnt = 0; @@ -119,13 +121,15 @@ nfp_map_ptrs_record(struct nfp_app_bpf *bpf, struct nfp_prog *nfp_prog, if (bpf_map_offload_neutral(prog->aux->used_maps[i])) cnt++; if (!cnt) - return 0; + goto out; nfp_prog->map_records = kmalloc_array(cnt, sizeof(nfp_prog->map_records[0]), GFP_KERNEL); - if (!nfp_prog->map_records) - return -ENOMEM; + if (!nfp_prog->map_records) { + err = -ENOMEM; + goto out; + } for (i = 0; i < prog->aux->used_map_cnt; i++) if (bpf_map_offload_neutral(prog->aux->used_maps[i])) { @@ -133,12 +137,14 @@ nfp_map_ptrs_record(struct nfp_app_bpf *bpf, struct nfp_prog *nfp_prog, prog->aux->used_maps[i]); if (err) { nfp_map_ptrs_forget(bpf, nfp_prog); - return err; + goto out; } } WARN_ON(cnt != nfp_prog->map_records_cnt); - return 0; +out: + mutex_unlock(&prog->aux->used_maps_mutex); + return err; } static int diff --git a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h index bf516285510f..a2926b1b3cff 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/cmsg.h +++ b/drivers/net/ethernet/netronome/nfp/flower/cmsg.h @@ -24,6 +24,7 @@ #define NFP_FLOWER_LAYER_VXLAN BIT(7) #define NFP_FLOWER_LAYER2_GRE BIT(0) +#define NFP_FLOWER_LAYER2_QINQ BIT(4) #define NFP_FLOWER_LAYER2_GENEVE BIT(5) #define NFP_FLOWER_LAYER2_GENEVE_OP BIT(6) #define NFP_FLOWER_LAYER2_TUN_IPV6 BIT(7) @@ -319,6 +320,22 @@ struct nfp_flower_mac_mpls { __be32 mpls_lse; }; +/* VLAN details (2W/8B) + * 3 2 1 + * 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | outer_tpid | outer_tci | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | inner_tpid | inner_tci | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + */ +struct nfp_flower_vlan { + __be16 outer_tpid; + __be16 outer_tci; + __be16 inner_tpid; + __be16 inner_tci; +}; + /* L4 ports (for UDP, TCP, SCTP) (1W/4B) * 3 2 1 * 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 diff --git a/drivers/net/ethernet/netronome/nfp/flower/main.h b/drivers/net/ethernet/netronome/nfp/flower/main.h index 3bf9c1afa45e..caf12eec9945 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/main.h +++ b/drivers/net/ethernet/netronome/nfp/flower/main.h @@ -30,6 +30,8 @@ struct nfp_app; #define NFP_FLOWER_MASK_ELEMENT_RS 1 #define NFP_FLOWER_MASK_HASH_BITS 10 +#define NFP_FLOWER_KEY_MAX_LW 32 + #define NFP_FL_META_FLAG_MANAGE_MASK BIT(7) #define NFP_FL_MASK_REUSE_TIME_NS 40000 @@ -44,6 +46,7 @@ struct nfp_app; #define NFP_FL_FEATS_FLOW_MOD BIT(5) #define NFP_FL_FEATS_PRE_TUN_RULES BIT(6) #define NFP_FL_FEATS_IPV6_TUN BIT(7) +#define NFP_FL_FEATS_VLAN_QINQ BIT(8) #define NFP_FL_FEATS_HOST_ACK BIT(31) #define NFP_FL_ENABLE_FLOW_MERGE BIT(0) @@ -57,7 +60,8 @@ struct nfp_app; NFP_FL_FEATS_VF_RLIM | \ NFP_FL_FEATS_FLOW_MOD | \ NFP_FL_FEATS_PRE_TUN_RULES | \ - NFP_FL_FEATS_IPV6_TUN) + NFP_FL_FEATS_IPV6_TUN | \ + NFP_FL_FEATS_VLAN_QINQ) struct nfp_fl_mask_id { struct circ_buf mask_id_free_list; diff --git a/drivers/net/ethernet/netronome/nfp/flower/match.c b/drivers/net/ethernet/netronome/nfp/flower/match.c index f7f01e2e3dce..255a4dff6288 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/match.c +++ b/drivers/net/ethernet/netronome/nfp/flower/match.c @@ -10,7 +10,7 @@ static void nfp_flower_compile_meta_tci(struct nfp_flower_meta_tci *ext, struct nfp_flower_meta_tci *msk, - struct flow_rule *rule, u8 key_type) + struct flow_rule *rule, u8 key_type, bool qinq_sup) { u16 tmp_tci; @@ -24,7 +24,7 @@ nfp_flower_compile_meta_tci(struct nfp_flower_meta_tci *ext, msk->nfp_flow_key_layer = key_type; msk->mask_id = ~0; - if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_VLAN)) { + if (!qinq_sup && flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_VLAN)) { struct flow_match_vlan match; flow_rule_match_vlan(rule, &match); @@ -231,6 +231,50 @@ nfp_flower_compile_ip_ext(struct nfp_flower_ip_ext *ext, } static void +nfp_flower_fill_vlan(struct flow_dissector_key_vlan *key, + struct nfp_flower_vlan *frame, + bool outer_vlan) +{ + u16 tci; + + tci = NFP_FLOWER_MASK_VLAN_PRESENT; + tci |= FIELD_PREP(NFP_FLOWER_MASK_VLAN_PRIO, + key->vlan_priority) | + FIELD_PREP(NFP_FLOWER_MASK_VLAN_VID, + key->vlan_id); + + if (outer_vlan) { + frame->outer_tci = cpu_to_be16(tci); + frame->outer_tpid = key->vlan_tpid; + } else { + frame->inner_tci = cpu_to_be16(tci); + frame->inner_tpid = key->vlan_tpid; + } +} + +static void +nfp_flower_compile_vlan(struct nfp_flower_vlan *ext, + struct nfp_flower_vlan *msk, + struct flow_rule *rule) +{ + struct flow_match_vlan match; + + memset(ext, 0, sizeof(struct nfp_flower_vlan)); + memset(msk, 0, sizeof(struct nfp_flower_vlan)); + + if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_VLAN)) { + flow_rule_match_vlan(rule, &match); + nfp_flower_fill_vlan(match.key, ext, true); + nfp_flower_fill_vlan(match.mask, msk, true); + } + if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_CVLAN)) { + flow_rule_match_cvlan(rule, &match); + nfp_flower_fill_vlan(match.key, ext, false); + nfp_flower_fill_vlan(match.mask, msk, false); + } +} + +static void nfp_flower_compile_ipv4(struct nfp_flower_ipv4 *ext, struct nfp_flower_ipv4 *msk, struct flow_rule *rule) { @@ -433,7 +477,10 @@ int nfp_flower_compile_flow_match(struct nfp_app *app, struct netlink_ext_ack *extack) { struct flow_rule *rule = flow_cls_offload_flow_rule(flow); + struct nfp_flower_priv *priv = app->priv; + bool qinq_sup; u32 port_id; + int ext_len; int err; u8 *ext; u8 *msk; @@ -446,9 +493,11 @@ int nfp_flower_compile_flow_match(struct nfp_app *app, ext = nfp_flow->unmasked_data; msk = nfp_flow->mask_data; + qinq_sup = !!(priv->flower_ext_feats & NFP_FL_FEATS_VLAN_QINQ); + nfp_flower_compile_meta_tci((struct nfp_flower_meta_tci *)ext, (struct nfp_flower_meta_tci *)msk, - rule, key_ls->key_layer); + rule, key_ls->key_layer, qinq_sup); ext += sizeof(struct nfp_flower_meta_tci); msk += sizeof(struct nfp_flower_meta_tci); @@ -547,6 +596,14 @@ int nfp_flower_compile_flow_match(struct nfp_app *app, } } + if (NFP_FLOWER_LAYER2_QINQ & key_ls->key_layer_two) { + nfp_flower_compile_vlan((struct nfp_flower_vlan *)ext, + (struct nfp_flower_vlan *)msk, + rule); + ext += sizeof(struct nfp_flower_vlan); + msk += sizeof(struct nfp_flower_vlan); + } + if (key_ls->key_layer & NFP_FLOWER_LAYER_VXLAN || key_ls->key_layer_two & NFP_FLOWER_LAYER2_GENEVE) { if (key_ls->key_layer_two & NFP_FLOWER_LAYER2_TUN_IPV6) { @@ -589,5 +646,15 @@ int nfp_flower_compile_flow_match(struct nfp_app *app, } } + /* Check that the flow key does not exceed the maximum limit. + * All structures in the key is multiples of 4 bytes, so use u32. + */ + ext_len = (u32 *)ext - (u32 *)nfp_flow->unmasked_data; + if (ext_len > NFP_FLOWER_KEY_MAX_LW) { + NL_SET_ERR_MSG_MOD(extack, + "unsupported offload: flow key too long"); + return -EOPNOTSUPP; + } + return 0; } diff --git a/drivers/net/ethernet/netronome/nfp/flower/offload.c b/drivers/net/ethernet/netronome/nfp/flower/offload.c index 36356f96661d..1c59aff2163c 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/offload.c +++ b/drivers/net/ethernet/netronome/nfp/flower/offload.c @@ -31,6 +31,7 @@ BIT(FLOW_DISSECTOR_KEY_PORTS) | \ BIT(FLOW_DISSECTOR_KEY_ETH_ADDRS) | \ BIT(FLOW_DISSECTOR_KEY_VLAN) | \ + BIT(FLOW_DISSECTOR_KEY_CVLAN) | \ BIT(FLOW_DISSECTOR_KEY_ENC_KEYID) | \ BIT(FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS) | \ BIT(FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS) | \ @@ -66,7 +67,8 @@ NFP_FLOWER_LAYER_IPV6) #define NFP_FLOWER_PRE_TUN_RULE_FIELDS \ - (NFP_FLOWER_LAYER_PORT | \ + (NFP_FLOWER_LAYER_EXT_META | \ + NFP_FLOWER_LAYER_PORT | \ NFP_FLOWER_LAYER_MAC | \ NFP_FLOWER_LAYER_IPV4 | \ NFP_FLOWER_LAYER_IPV6) @@ -285,6 +287,30 @@ nfp_flower_calculate_key_layers(struct nfp_app *app, NL_SET_ERR_MSG_MOD(extack, "unsupported offload: loaded firmware does not support VLAN PCP offload"); return -EOPNOTSUPP; } + if (priv->flower_ext_feats & NFP_FL_FEATS_VLAN_QINQ && + !(key_layer_two & NFP_FLOWER_LAYER2_QINQ)) { + key_layer |= NFP_FLOWER_LAYER_EXT_META; + key_size += sizeof(struct nfp_flower_ext_meta); + key_size += sizeof(struct nfp_flower_vlan); + key_layer_two |= NFP_FLOWER_LAYER2_QINQ; + } + } + + if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_CVLAN)) { + struct flow_match_vlan cvlan; + + if (!(priv->flower_ext_feats & NFP_FL_FEATS_VLAN_QINQ)) { + NL_SET_ERR_MSG_MOD(extack, "unsupported offload: loaded firmware does not support VLAN QinQ offload"); + return -EOPNOTSUPP; + } + + flow_rule_match_vlan(rule, &cvlan); + if (!(key_layer_two & NFP_FLOWER_LAYER2_QINQ)) { + key_layer |= NFP_FLOWER_LAYER_EXT_META; + key_size += sizeof(struct nfp_flower_ext_meta); + key_size += sizeof(struct nfp_flower_vlan); + key_layer_two |= NFP_FLOWER_LAYER2_QINQ; + } } if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_CONTROL)) { @@ -1066,6 +1092,7 @@ err_destroy_merge_flow: * nfp_flower_validate_pre_tun_rule() * @app: Pointer to the APP handle * @flow: Pointer to NFP flow representation of rule + * @key_ls: Pointer to NFP key layers structure * @extack: Netlink extended ACK report * * Verifies the flow as a pre-tunnel rule. @@ -1075,10 +1102,13 @@ err_destroy_merge_flow: static int nfp_flower_validate_pre_tun_rule(struct nfp_app *app, struct nfp_fl_payload *flow, + struct nfp_fl_key_ls *key_ls, struct netlink_ext_ack *extack) { + struct nfp_flower_priv *priv = app->priv; struct nfp_flower_meta_tci *meta_tci; struct nfp_flower_mac_mpls *mac; + u8 *ext = flow->unmasked_data; struct nfp_fl_act_head *act; u8 *mask = flow->mask_data; bool vlan = false; @@ -1086,20 +1116,25 @@ nfp_flower_validate_pre_tun_rule(struct nfp_app *app, u8 key_layer; meta_tci = (struct nfp_flower_meta_tci *)flow->unmasked_data; - if (meta_tci->tci & cpu_to_be16(NFP_FLOWER_MASK_VLAN_PRESENT)) { - u16 vlan_tci = be16_to_cpu(meta_tci->tci); - - vlan_tci &= ~NFP_FLOWER_MASK_VLAN_PRESENT; - flow->pre_tun_rule.vlan_tci = cpu_to_be16(vlan_tci); - vlan = true; - } else { - flow->pre_tun_rule.vlan_tci = cpu_to_be16(0xffff); + key_layer = key_ls->key_layer; + if (!(priv->flower_ext_feats & NFP_FL_FEATS_VLAN_QINQ)) { + if (meta_tci->tci & cpu_to_be16(NFP_FLOWER_MASK_VLAN_PRESENT)) { + u16 vlan_tci = be16_to_cpu(meta_tci->tci); + + vlan_tci &= ~NFP_FLOWER_MASK_VLAN_PRESENT; + flow->pre_tun_rule.vlan_tci = cpu_to_be16(vlan_tci); + vlan = true; + } else { + flow->pre_tun_rule.vlan_tci = cpu_to_be16(0xffff); + } } - key_layer = meta_tci->nfp_flow_key_layer; if (key_layer & ~NFP_FLOWER_PRE_TUN_RULE_FIELDS) { NL_SET_ERR_MSG_MOD(extack, "unsupported pre-tunnel rule: too many match fields"); return -EOPNOTSUPP; + } else if (key_ls->key_layer_two & ~NFP_FLOWER_LAYER2_QINQ) { + NL_SET_ERR_MSG_MOD(extack, "unsupported pre-tunnel rule: non-vlan in extended match fields"); + return -EOPNOTSUPP; } if (!(key_layer & NFP_FLOWER_LAYER_MAC)) { @@ -1109,7 +1144,13 @@ nfp_flower_validate_pre_tun_rule(struct nfp_app *app, /* Skip fields known to exist. */ mask += sizeof(struct nfp_flower_meta_tci); + ext += sizeof(struct nfp_flower_meta_tci); + if (key_ls->key_layer_two) { + mask += sizeof(struct nfp_flower_ext_meta); + ext += sizeof(struct nfp_flower_ext_meta); + } mask += sizeof(struct nfp_flower_in_port); + ext += sizeof(struct nfp_flower_in_port); /* Ensure destination MAC address is fully matched. */ mac = (struct nfp_flower_mac_mpls *)mask; @@ -1118,6 +1159,8 @@ nfp_flower_validate_pre_tun_rule(struct nfp_app *app, return -EOPNOTSUPP; } + mask += sizeof(struct nfp_flower_mac_mpls); + ext += sizeof(struct nfp_flower_mac_mpls); if (key_layer & NFP_FLOWER_LAYER_IPV4 || key_layer & NFP_FLOWER_LAYER_IPV6) { /* Flags and proto fields have same offset in IPv4 and IPv6. */ @@ -1130,7 +1173,6 @@ nfp_flower_validate_pre_tun_rule(struct nfp_app *app, sizeof(struct nfp_flower_ipv4) : sizeof(struct nfp_flower_ipv6); - mask += sizeof(struct nfp_flower_mac_mpls); /* Ensure proto and flags are the only IP layer fields. */ for (i = 0; i < size; i++) @@ -1138,6 +1180,25 @@ nfp_flower_validate_pre_tun_rule(struct nfp_app *app, NL_SET_ERR_MSG_MOD(extack, "unsupported pre-tunnel rule: only flags and proto can be matched in ip header"); return -EOPNOTSUPP; } + ext += size; + mask += size; + } + + if ((priv->flower_ext_feats & NFP_FL_FEATS_VLAN_QINQ)) { + if (key_ls->key_layer_two & NFP_FLOWER_LAYER2_QINQ) { + struct nfp_flower_vlan *vlan_tags; + u16 vlan_tci; + + vlan_tags = (struct nfp_flower_vlan *)ext; + + vlan_tci = be16_to_cpu(vlan_tags->outer_tci); + + vlan_tci &= ~NFP_FLOWER_MASK_VLAN_PRESENT; + flow->pre_tun_rule.vlan_tci = cpu_to_be16(vlan_tci); + vlan = true; + } else { + flow->pre_tun_rule.vlan_tci = cpu_to_be16(0xffff); + } } /* Action must be a single egress or pop_vlan and egress. */ @@ -1220,7 +1281,7 @@ nfp_flower_add_offload(struct nfp_app *app, struct net_device *netdev, goto err_destroy_flow; if (flow_pay->pre_tun_rule.dev) { - err = nfp_flower_validate_pre_tun_rule(app, flow_pay, extack); + err = nfp_flower_validate_pre_tun_rule(app, flow_pay, key_layer, extack); if (err) goto err_destroy_flow; } diff --git a/drivers/net/ethernet/netronome/nfp/nfp_devlink.c b/drivers/net/ethernet/netronome/nfp/nfp_devlink.c index be52510d446b..97d2b03208de 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_devlink.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_devlink.c @@ -329,12 +329,11 @@ err_close_nsp: } static int -nfp_devlink_flash_update(struct devlink *devlink, const char *path, - const char *component, struct netlink_ext_ack *extack) +nfp_devlink_flash_update(struct devlink *devlink, + struct devlink_flash_update_params *params, + struct netlink_ext_ack *extack) { - if (component) - return -EOPNOTSUPP; - return nfp_flash_update_common(devlink_priv(devlink), path, extack); + return nfp_flash_update_common(devlink_priv(devlink), params->file_name, extack); } const struct devlink_ops nfp_devlink_ops = { diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index 21ea22694e47..b150da43adb2 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -2287,9 +2287,9 @@ static bool nfp_ctrl_rx(struct nfp_net_r_vector *r_vec) return budget; } -static void nfp_ctrl_poll(unsigned long arg) +static void nfp_ctrl_poll(struct tasklet_struct *t) { - struct nfp_net_r_vector *r_vec = (void *)arg; + struct nfp_net_r_vector *r_vec = from_tasklet(r_vec, t, tasklet); spin_lock(&r_vec->lock); nfp_net_tx_complete(r_vec->tx_ring, 0); @@ -2337,8 +2337,7 @@ static void nfp_net_vecs_init(struct nfp_net *nn) __skb_queue_head_init(&r_vec->queue); spin_lock_init(&r_vec->lock); - tasklet_init(&r_vec->tasklet, nfp_ctrl_poll, - (unsigned long)r_vec); + tasklet_setup(&r_vec->tasklet, nfp_ctrl_poll); tasklet_disable(&r_vec->tasklet); } diff --git a/drivers/net/ethernet/ni/nixge.c b/drivers/net/ethernet/ni/nixge.c index 4075f5e59955..a6861df9904f 100644 --- a/drivers/net/ethernet/ni/nixge.c +++ b/drivers/net/ethernet/ni/nixge.c @@ -787,9 +787,9 @@ out: return IRQ_HANDLED; } -static void nixge_dma_err_handler(unsigned long data) +static void nixge_dma_err_handler(struct tasklet_struct *t) { - struct nixge_priv *lp = (struct nixge_priv *)data; + struct nixge_priv *lp = from_tasklet(lp, t, dma_err_tasklet); struct nixge_hw_dma_bd *cur_p; struct nixge_tx_skb *tx_skb; u32 cr, i; @@ -879,8 +879,7 @@ static int nixge_open(struct net_device *ndev) phy_start(phy); /* Enable tasklets for Axi DMA error handling */ - tasklet_init(&priv->dma_err_tasklet, nixge_dma_err_handler, - (unsigned long)priv); + tasklet_setup(&priv->dma_err_tasklet, nixge_dma_err_handler); napi_enable(&priv->napi); diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_ethtool.c b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_ethtool.c index b36aa5bf3c5f..a58f14aca10c 100644 --- a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_ethtool.c +++ b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_ethtool.c @@ -8,7 +8,7 @@ #include "pch_gbe.h" #include "pch_gbe_phy.h" -/** +/* * pch_gbe_stats - Stats item information */ struct pch_gbe_stats { @@ -24,7 +24,7 @@ struct pch_gbe_stats { .offset = offsetof(struct pch_gbe_hw_stats, m), \ } -/** +/* * pch_gbe_gstrings_stats - ethtool information status name list */ static const struct pch_gbe_stats pch_gbe_gstrings_stats[] = { diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c index 23f7c76737c9..ade8c44c01cd 100644 --- a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c +++ b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c @@ -295,7 +295,7 @@ static s32 pch_gbe_mac_read_mac_addr(struct pch_gbe_hw *hw) /** * pch_gbe_wait_clr_bit - Wait to clear a bit * @reg: Pointer of register - * @busy: Busy bit + * @bit: Busy bit */ static void pch_gbe_wait_clr_bit(void *reg, u32 bit) { @@ -1034,7 +1034,7 @@ static void pch_gbe_set_mode(struct pch_gbe_adapter *adapter, u16 speed, /** * pch_gbe_watchdog - Watchdog process - * @data: Board private structure + * @t: timer list containing a Board private structure */ static void pch_gbe_watchdog(struct timer_list *t) { @@ -2270,6 +2270,7 @@ static int pch_gbe_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) /** * pch_gbe_tx_timeout - Respond to a Tx Hang * @netdev: Network interface device structure + * @txqueue: index of hanging queue */ static void pch_gbe_tx_timeout(struct net_device *netdev, unsigned int txqueue) { diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_param.c b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_param.c index dceec80fd642..81fc5a6e3221 100644 --- a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_param.c +++ b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_param.c @@ -13,7 +13,7 @@ #define OPTION_DISABLED 0 #define OPTION_ENABLED 1 -/** +/* * TxDescriptors - Transmit Descriptor Count * @Valid Range: PCH_GBE_MIN_TXD - PCH_GBE_MAX_TXD * @Default Value: PCH_GBE_DEFAULT_TXD @@ -22,7 +22,7 @@ static int TxDescriptors = OPTION_UNSET; module_param(TxDescriptors, int, 0); MODULE_PARM_DESC(TxDescriptors, "Number of transmit descriptors"); -/** +/* * RxDescriptors -Receive Descriptor Count * @Valid Range: PCH_GBE_MIN_RXD - PCH_GBE_MAX_RXD * @Default Value: PCH_GBE_DEFAULT_RXD @@ -31,7 +31,7 @@ static int RxDescriptors = OPTION_UNSET; module_param(RxDescriptors, int, 0); MODULE_PARM_DESC(RxDescriptors, "Number of receive descriptors"); -/** +/* * Speed - User Specified Speed Override * @Valid Range: 0, 10, 100, 1000 * - 0: auto-negotiate at all supported speeds @@ -44,7 +44,7 @@ static int Speed = OPTION_UNSET; module_param(Speed, int, 0); MODULE_PARM_DESC(Speed, "Speed setting"); -/** +/* * Duplex - User Specified Duplex Override * @Valid Range: 0-2 * - 0: auto-negotiate for duplex @@ -59,7 +59,7 @@ MODULE_PARM_DESC(Duplex, "Duplex setting"); #define HALF_DUPLEX 1 #define FULL_DUPLEX 2 -/** +/* * AutoNeg - Auto-negotiation Advertisement Override * @Valid Range: 0x01-0x0F, 0x20-0x2F * @@ -85,7 +85,7 @@ MODULE_PARM_DESC(AutoNeg, "Advertised auto-negotiation setting"); #define PHY_ADVERTISE_1000_FULL 0x0020 #define PCH_AUTONEG_ADVERTISE_DEFAULT 0x2F -/** +/* * FlowControl - User Specified Flow Control Override * @Valid Range: 0-3 * - 0: No Flow Control @@ -124,7 +124,7 @@ MODULE_PARM_DESC(XsumTX, "Disable or enable Transmit Checksum offload"); #define PCH_GBE_DEFAULT_TX_CSUM true /* trueorfalse */ -/** +/* * pch_gbe_option - Force the MAC's flow control settings * @hw: Pointer to the HW structure * Returns: diff --git a/drivers/net/ethernet/packetengines/yellowfin.c b/drivers/net/ethernet/packetengines/yellowfin.c index 3da075307178..d1dd9bc1bc7f 100644 --- a/drivers/net/ethernet/packetengines/yellowfin.c +++ b/drivers/net/ethernet/packetengines/yellowfin.c @@ -1060,7 +1060,7 @@ static int yellowfin_rx(struct net_device *dev) struct sk_buff *rx_skb = yp->rx_skbuff[entry]; s16 frame_status; u16 desc_status; - int data_size, yf_size; + int data_size, __maybe_unused yf_size; u8 *buf_addr; if(!desc->result_status) diff --git a/drivers/net/ethernet/pensando/Kconfig b/drivers/net/ethernet/pensando/Kconfig index 76f8cc502bf9..5f8b0bb3af6e 100644 --- a/drivers/net/ethernet/pensando/Kconfig +++ b/drivers/net/ethernet/pensando/Kconfig @@ -21,6 +21,7 @@ config IONIC tristate "Pensando Ethernet IONIC Support" depends on 64BIT && PCI select NET_DEVLINK + select DIMLIB help This enables the support for the Pensando family of Ethernet adapters. More specific information on this driver can be diff --git a/drivers/net/ethernet/pensando/ionic/Makefile b/drivers/net/ethernet/pensando/ionic/Makefile index 29f304d75261..8d3c2d3cb10d 100644 --- a/drivers/net/ethernet/pensando/ionic/Makefile +++ b/drivers/net/ethernet/pensando/ionic/Makefile @@ -5,4 +5,4 @@ obj-$(CONFIG_IONIC) := ionic.o ionic-y := ionic_main.o ionic_bus_pci.o ionic_devlink.o ionic_dev.o \ ionic_debugfs.o ionic_lif.o ionic_rx_filter.o ionic_ethtool.o \ - ionic_txrx.o ionic_stats.o + ionic_txrx.o ionic_stats.o ionic_fw.o diff --git a/drivers/net/ethernet/pensando/ionic/ionic.h b/drivers/net/ethernet/pensando/ionic/ionic.h index f5a910c458ba..084a924431d5 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic.h +++ b/drivers/net/ethernet/pensando/ionic/ionic.h @@ -42,13 +42,11 @@ struct ionic { struct ionic_dev_bar bars[IONIC_BARS_MAX]; unsigned int num_bars; struct ionic_identity ident; - struct list_head lifs; - struct ionic_lif *master_lif; + struct ionic_lif *lif; unsigned int nnqs_per_lif; unsigned int neqs_per_lif; unsigned int ntxqs_per_lif; unsigned int nrxqs_per_lif; - DECLARE_BITMAP(lifbits, IONIC_LIFS_MAX); unsigned int nintrs; DECLARE_BITMAP(intrs, IONIC_INTR_CTRL_REGS_MAX); struct work_struct nb_work; @@ -66,9 +64,6 @@ struct ionic_admin_ctx { union ionic_adminq_comp comp; }; -int ionic_napi(struct napi_struct *napi, int budget, ionic_cq_cb cb, - ionic_cq_done_cb done_cb, void *done_arg); - int ionic_adminq_post_wait(struct ionic_lif *lif, struct ionic_admin_ctx *ctx); int ionic_dev_cmd_wait(struct ionic *ionic, unsigned long max_wait); int ionic_set_dma_mask(struct ionic *ionic); diff --git a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c index 85c686c16741..b0d8499d373b 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c @@ -266,6 +266,7 @@ static int ionic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) dev_err(dev, "Cannot identify device: %d, aborting\n", err); goto err_out_teardown; } + ionic_debugfs_add_ident(ionic); err = ionic_init(ionic); if (err) { @@ -286,29 +287,22 @@ static int ionic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto err_out_reset; } - /* Configure LIFs */ - err = ionic_lif_identify(ionic, IONIC_LIF_TYPE_CLASSIC, - &ionic->ident.lif); + /* Allocate and init the LIF */ + err = ionic_lif_size(ionic); if (err) { - dev_err(dev, "Cannot identify LIFs: %d, aborting\n", err); + dev_err(dev, "Cannot size LIF: %d, aborting\n", err); goto err_out_port_reset; } - err = ionic_lifs_size(ionic); + err = ionic_lif_alloc(ionic); if (err) { - dev_err(dev, "Cannot size LIFs: %d, aborting\n", err); - goto err_out_port_reset; - } - - err = ionic_lifs_alloc(ionic); - if (err) { - dev_err(dev, "Cannot allocate LIFs: %d, aborting\n", err); + dev_err(dev, "Cannot allocate LIF: %d, aborting\n", err); goto err_out_free_irqs; } - err = ionic_lifs_init(ionic); + err = ionic_lif_init(ionic->lif); if (err) { - dev_err(dev, "Cannot init LIFs: %d, aborting\n", err); + dev_err(dev, "Cannot init LIF: %d, aborting\n", err); goto err_out_free_lifs; } @@ -321,9 +315,9 @@ static int ionic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) dev_err(dev, "Cannot enable existing VFs: %d\n", err); } - err = ionic_lifs_register(ionic); + err = ionic_lif_register(ionic->lif); if (err) { - dev_err(dev, "Cannot register LIFs: %d, aborting\n", err); + dev_err(dev, "Cannot register LIF: %d, aborting\n", err); goto err_out_deinit_lifs; } @@ -336,12 +330,13 @@ static int ionic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) return 0; err_out_deregister_lifs: - ionic_lifs_unregister(ionic); + ionic_lif_unregister(ionic->lif); err_out_deinit_lifs: ionic_vf_dealloc(ionic); - ionic_lifs_deinit(ionic); + ionic_lif_deinit(ionic->lif); err_out_free_lifs: - ionic_lifs_free(ionic); + ionic_lif_free(ionic->lif); + ionic->lif = NULL; err_out_free_irqs: ionic_bus_free_irq_vectors(ionic); err_out_port_reset: @@ -349,7 +344,7 @@ err_out_port_reset: err_out_reset: ionic_reset(ionic); err_out_teardown: - ionic_dev_teardown(ionic); + del_timer_sync(&ionic->watchdog_timer); pci_clear_master(pdev); /* Don't fail the probe for these errors, keep * the hw interface around for inspection @@ -377,17 +372,19 @@ static void ionic_remove(struct pci_dev *pdev) if (!ionic) return; - if (ionic->master_lif) { + del_timer_sync(&ionic->watchdog_timer); + + if (ionic->lif) { ionic_devlink_unregister(ionic); - ionic_lifs_unregister(ionic); - ionic_lifs_deinit(ionic); - ionic_lifs_free(ionic); + ionic_lif_unregister(ionic->lif); + ionic_lif_deinit(ionic->lif); + ionic_lif_free(ionic->lif); + ionic->lif = NULL; ionic_bus_free_irq_vectors(ionic); } ionic_port_reset(ionic); ionic_reset(ionic); - ionic_dev_teardown(ionic); pci_clear_master(pdev); ionic_unmap_bars(ionic); pci_release_regions(pdev); diff --git a/drivers/net/ethernet/pensando/ionic/ionic_debugfs.c b/drivers/net/ethernet/pensando/ionic/ionic_debugfs.c index 11621ccc1faf..39f59849720d 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_debugfs.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_debugfs.c @@ -76,7 +76,7 @@ static int q_tail_show(struct seq_file *seq, void *v) { struct ionic_queue *q = seq->private; - seq_printf(seq, "%d\n", q->tail->index); + seq_printf(seq, "%d\n", q->tail_idx); return 0; } @@ -86,7 +86,7 @@ static int q_head_show(struct seq_file *seq, void *v) { struct ionic_queue *q = seq->private; - seq_printf(seq, "%d\n", q->head->index); + seq_printf(seq, "%d\n", q->head_idx); return 0; } @@ -96,7 +96,7 @@ static int cq_tail_show(struct seq_file *seq, void *v) { struct ionic_cq *cq = seq->private; - seq_printf(seq, "%d\n", cq->tail->index); + seq_printf(seq, "%d\n", cq->tail_idx); return 0; } @@ -112,7 +112,8 @@ static const struct debugfs_reg32 intr_ctrl_regs[] = { void ionic_debugfs_add_qcq(struct ionic_lif *lif, struct ionic_qcq *qcq) { - struct dentry *q_dentry, *cq_dentry, *intr_dentry, *stats_dentry; + struct dentry *qcq_dentry, *q_dentry, *cq_dentry; + struct dentry *intr_dentry, *stats_dentry; struct ionic_dev *idev = &lif->ionic->idev; struct debugfs_regset32 *intr_ctrl_regset; struct ionic_intr_info *intr = &qcq->intr; @@ -121,21 +122,21 @@ void ionic_debugfs_add_qcq(struct ionic_lif *lif, struct ionic_qcq *qcq) struct ionic_queue *q = &qcq->q; struct ionic_cq *cq = &qcq->cq; - qcq->dentry = debugfs_create_dir(q->name, lif->dentry); + qcq_dentry = debugfs_create_dir(q->name, lif->dentry); + if (IS_ERR_OR_NULL(qcq_dentry)) + return; + qcq->dentry = qcq_dentry; - debugfs_create_x32("total_size", 0400, qcq->dentry, &qcq->total_size); - debugfs_create_x64("base_pa", 0400, qcq->dentry, &qcq->base_pa); + debugfs_create_x64("q_base_pa", 0400, qcq_dentry, &qcq->q_base_pa); + debugfs_create_x32("q_size", 0400, qcq_dentry, &qcq->q_size); + debugfs_create_x64("cq_base_pa", 0400, qcq_dentry, &qcq->cq_base_pa); + debugfs_create_x32("cq_size", 0400, qcq_dentry, &qcq->cq_size); + debugfs_create_x64("sg_base_pa", 0400, qcq_dentry, &qcq->sg_base_pa); + debugfs_create_x32("sg_size", 0400, qcq_dentry, &qcq->sg_size); q_dentry = debugfs_create_dir("q", qcq->dentry); debugfs_create_u32("index", 0400, q_dentry, &q->index); - debugfs_create_x64("base_pa", 0400, q_dentry, &q->base_pa); - if (qcq->flags & IONIC_QCQ_F_SG) { - debugfs_create_x64("sg_base_pa", 0400, q_dentry, - &q->sg_base_pa); - debugfs_create_u32("sg_desc_size", 0400, q_dentry, - &q->sg_desc_size); - } debugfs_create_u32("num_descs", 0400, q_dentry, &q->num_descs); debugfs_create_u32("desc_size", 0400, q_dentry, &q->desc_size); debugfs_create_u32("pid", 0400, q_dentry, &q->pid); @@ -188,6 +189,8 @@ void ionic_debugfs_add_qcq(struct ionic_lif *lif, struct ionic_qcq *qcq) &intr->index); debugfs_create_u32("vector", 0400, intr_dentry, &intr->vector); + debugfs_create_u32("dim_coal_hw", 0400, intr_dentry, + &intr->dim_coal_hw); intr_ctrl_regset = devm_kzalloc(dev, sizeof(*intr_ctrl_regset), GFP_KERNEL); diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.c b/drivers/net/ethernet/pensando/ionic/ionic_dev.c index d83eff0ae0ac..545c99b15df8 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_dev.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.c @@ -19,10 +19,13 @@ static void ionic_watchdog_cb(struct timer_list *t) mod_timer(&ionic->watchdog_timer, round_jiffies(jiffies + ionic->watchdog_period)); + if (!ionic->lif) + return; + hb = ionic_heartbeat_check(ionic); - if (hb >= 0 && ionic->master_lif) - ionic_link_status_check_request(ionic->master_lif); + if (hb >= 0) + ionic_link_status_check_request(ionic->lif, false); } void ionic_init_devinfo(struct ionic *ionic) @@ -98,11 +101,6 @@ int ionic_dev_setup(struct ionic *ionic) return 0; } -void ionic_dev_teardown(struct ionic *ionic) -{ - del_timer_sync(&ionic->watchdog_timer); -} - /* Devcmd Interface */ int ionic_heartbeat_check(struct ionic *ionic) { @@ -126,7 +124,7 @@ int ionic_heartbeat_check(struct ionic *ionic) /* is this a transition? */ if (fw_status != idev->last_fw_status && idev->last_fw_status != 0xff) { - struct ionic_lif *lif = ionic->master_lif; + struct ionic_lif *lif = ionic->lif; bool trigger = false; if (!fw_status || fw_status == 0xff) { @@ -467,9 +465,7 @@ int ionic_cq_init(struct ionic_lif *lif, struct ionic_cq *cq, struct ionic_intr_info *intr, unsigned int num_descs, size_t desc_size) { - struct ionic_cq_info *cur; unsigned int ring_size; - unsigned int i; if (desc_size == 0 || !is_power_of_2(num_descs)) return -EINVAL; @@ -482,22 +478,9 @@ int ionic_cq_init(struct ionic_lif *lif, struct ionic_cq *cq, cq->bound_intr = intr; cq->num_descs = num_descs; cq->desc_size = desc_size; - cq->tail = cq->info; + cq->tail_idx = 0; cq->done_color = 1; - cur = cq->info; - - for (i = 0; i < num_descs; i++) { - if (i + 1 == num_descs) { - cur->next = cq->info; - cur->last = true; - } else { - cur->next = cur + 1; - } - cur->index = i; - cur++; - } - return 0; } @@ -522,15 +505,18 @@ unsigned int ionic_cq_service(struct ionic_cq *cq, unsigned int work_to_do, ionic_cq_cb cb, ionic_cq_done_cb done_cb, void *done_arg) { + struct ionic_cq_info *cq_info; unsigned int work_done = 0; if (work_to_do == 0) return 0; - while (cb(cq, cq->tail)) { - if (cq->tail->last) + cq_info = &cq->info[cq->tail_idx]; + while (cb(cq, cq_info)) { + if (cq->tail_idx == cq->num_descs - 1) cq->done_color = !cq->done_color; - cq->tail = cq->tail->next; + cq->tail_idx = (cq->tail_idx + 1) & (cq->num_descs - 1); + cq_info = &cq->info[cq->tail_idx]; DEBUG_STATS_CQE_CNT(cq); if (++work_done >= work_to_do) @@ -548,9 +534,7 @@ int ionic_q_init(struct ionic_lif *lif, struct ionic_dev *idev, unsigned int num_descs, size_t desc_size, size_t sg_desc_size, unsigned int pid) { - struct ionic_desc_info *cur; unsigned int ring_size; - unsigned int i; if (desc_size == 0 || !is_power_of_2(num_descs)) return -EINVAL; @@ -565,24 +549,12 @@ int ionic_q_init(struct ionic_lif *lif, struct ionic_dev *idev, q->num_descs = num_descs; q->desc_size = desc_size; q->sg_desc_size = sg_desc_size; - q->tail = q->info; - q->head = q->tail; + q->tail_idx = 0; + q->head_idx = 0; q->pid = pid; snprintf(q->name, sizeof(q->name), "L%d-%s%u", lif->index, name, index); - cur = q->info; - - for (i = 0; i < num_descs; i++) { - if (i + 1 == num_descs) - cur->next = q->info; - else - cur->next = cur + 1; - cur->index = i; - cur->left = num_descs - i; - cur++; - } - return 0; } @@ -614,19 +586,22 @@ void ionic_q_post(struct ionic_queue *q, bool ring_doorbell, ionic_desc_cb cb, void *cb_arg) { struct device *dev = q->lif->ionic->dev; + struct ionic_desc_info *desc_info; struct ionic_lif *lif = q->lif; - q->head->cb = cb; - q->head->cb_arg = cb_arg; - q->head = q->head->next; + desc_info = &q->info[q->head_idx]; + desc_info->cb = cb; + desc_info->cb_arg = cb_arg; + + q->head_idx = (q->head_idx + 1) & (q->num_descs - 1); dev_dbg(dev, "lif=%d qname=%s qid=%d qtype=%d p_index=%d ringdb=%d\n", q->lif->index, q->name, q->hw_type, q->hw_index, - q->head->index, ring_doorbell); + q->head_idx, ring_doorbell); if (ring_doorbell) ionic_dbell_ring(lif->kern_dbpage, q->hw_type, - q->dbval | q->head->index); + q->dbval | q->head_idx); } static bool ionic_q_is_posted(struct ionic_queue *q, unsigned int pos) @@ -634,8 +609,8 @@ static bool ionic_q_is_posted(struct ionic_queue *q, unsigned int pos) unsigned int mask, tail, head; mask = q->num_descs - 1; - tail = q->tail->index; - head = q->head->index; + tail = q->tail_idx; + head = q->head_idx; return ((pos - tail) & mask) < ((head - tail) & mask); } @@ -646,20 +621,22 @@ void ionic_q_service(struct ionic_queue *q, struct ionic_cq_info *cq_info, struct ionic_desc_info *desc_info; ionic_desc_cb cb; void *cb_arg; + u16 index; /* check for empty queue */ - if (q->tail->index == q->head->index) + if (q->tail_idx == q->head_idx) return; /* stop index must be for a descriptor that is not yet completed */ if (unlikely(!ionic_q_is_posted(q, stop_index))) dev_err(q->lif->ionic->dev, "ionic stop is not posted %s stop %u tail %u head %u\n", - q->name, stop_index, q->tail->index, q->head->index); + q->name, stop_index, q->tail_idx, q->head_idx); do { - desc_info = q->tail; - q->tail = desc_info->next; + desc_info = &q->info[q->tail_idx]; + index = q->tail_idx; + q->tail_idx = (q->tail_idx + 1) & (q->num_descs - 1); cb = desc_info->cb; cb_arg = desc_info->cb_arg; @@ -669,5 +646,5 @@ void ionic_q_service(struct ionic_queue *q, struct ionic_cq_info *cq_info, if (cb) cb(q, desc_info, cq_info, cb_arg); - } while (desc_info->index != stop_index); + } while (index != stop_index); } diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.h b/drivers/net/ethernet/pensando/ionic/ionic_dev.h index d5cba502abca..c109cd5a0471 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_dev.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.h @@ -149,10 +149,13 @@ struct ionic_dev { }; struct ionic_cq_info { - void *cq_desc; - struct ionic_cq_info *next; - unsigned int index; - bool last; + union { + void *cq_desc; + struct ionic_txq_comp *txcq; + struct ionic_rxq_comp *rxcq; + struct ionic_admin_comp *admincq; + struct ionic_notifyq_event *notifyq; + }; }; struct ionic_queue; @@ -169,11 +172,17 @@ struct ionic_page_info { }; struct ionic_desc_info { - void *desc; - void *sg_desc; - struct ionic_desc_info *next; - unsigned int index; - unsigned int left; + union { + void *desc; + struct ionic_txq_desc *txq_desc; + struct ionic_rxq_desc *rxq_desc; + struct ionic_admin_cmd *adminq_desc; + }; + union { + void *sg_desc; + struct ionic_txq_sg_desc *txq_sg_desc; + struct ionic_rxq_sg_desc *rxq_sgl_desc; + }; unsigned int npages; struct ionic_page_info pages[IONIC_RX_MAX_SG_ELEMS + 1]; ionic_desc_cb cb; @@ -183,25 +192,35 @@ struct ionic_desc_info { #define IONIC_QUEUE_NAME_MAX_SZ 32 struct ionic_queue { + struct device *dev; + struct ionic_lif *lif; + struct ionic_desc_info *info; + u16 head_idx; + u16 tail_idx; + unsigned int index; + unsigned int num_descs; u64 dbell_count; - u64 drop; u64 stop; u64 wake; - struct ionic_lif *lif; - struct ionic_desc_info *info; - struct ionic_desc_info *tail; - struct ionic_desc_info *head; + u64 drop; struct ionic_dev *idev; - unsigned int index; unsigned int type; unsigned int hw_index; unsigned int hw_type; u64 dbval; - void *base; - void *sg_base; + union { + void *base; + struct ionic_txq_desc *txq; + struct ionic_rxq_desc *rxq; + struct ionic_admin_cmd *adminq; + }; + union { + void *sg_base; + struct ionic_txq_sg_desc *txq_sgl; + struct ionic_rxq_sg_desc *rxq_sgl; + }; dma_addr_t base_pa; dma_addr_t sg_base_pa; - unsigned int num_descs; unsigned int desc_size; unsigned int sg_desc_size; unsigned int pid; @@ -218,20 +237,21 @@ struct ionic_intr_info { u64 rearm_count; unsigned int cpu; cpumask_t affinity_mask; + u32 dim_coal_hw; }; struct ionic_cq { - void *base; - dma_addr_t base_pa; struct ionic_lif *lif; struct ionic_cq_info *info; - struct ionic_cq_info *tail; struct ionic_queue *bound_q; struct ionic_intr_info *bound_intr; + u16 tail_idx; bool done_color; unsigned int num_descs; - u64 compl_count; unsigned int desc_size; + u64 compl_count; + void *base; + dma_addr_t base_pa; }; struct ionic; @@ -246,12 +266,12 @@ static inline void ionic_intr_init(struct ionic_dev *idev, static inline unsigned int ionic_q_space_avail(struct ionic_queue *q) { - unsigned int avail = q->tail->index; + unsigned int avail = q->tail_idx; - if (q->head->index >= avail) - avail += q->head->left - 1; + if (q->head_idx >= avail) + avail += q->num_descs - q->head_idx - 1; else - avail -= q->head->index + 1; + avail -= q->head_idx + 1; return avail; } @@ -263,7 +283,6 @@ static inline bool ionic_q_has_space(struct ionic_queue *q, unsigned int want) void ionic_init_devinfo(struct ionic *ionic); int ionic_dev_setup(struct ionic *ionic); -void ionic_dev_teardown(struct ionic *ionic); void ionic_dev_cmd_go(struct ionic_dev *idev, union ionic_dev_cmd *cmd); u8 ionic_dev_cmd_status(struct ionic_dev *idev); diff --git a/drivers/net/ethernet/pensando/ionic/ionic_devlink.c b/drivers/net/ethernet/pensando/ionic/ionic_devlink.c index c4f4fd469fe3..51d64718ed9f 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_devlink.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_devlink.c @@ -9,6 +9,15 @@ #include "ionic_lif.h" #include "ionic_devlink.h" +static int ionic_dl_flash_update(struct devlink *dl, + struct devlink_flash_update_params *params, + struct netlink_ext_ack *extack) +{ + struct ionic *ionic = devlink_priv(dl); + + return ionic_firmware_update(ionic->lif, params->file_name, extack); +} + static int ionic_dl_info_get(struct devlink *dl, struct devlink_info_req *req, struct netlink_ext_ack *extack) { @@ -48,6 +57,7 @@ static int ionic_dl_info_get(struct devlink *dl, struct devlink_info_req *req, static const struct devlink_ops ionic_dl_ops = { .info_get = ionic_dl_info_get, + .flash_update = ionic_dl_flash_update, }; struct ionic *ionic_devlink_alloc(struct device *dev) @@ -85,7 +95,7 @@ int ionic_devlink_register(struct ionic *ionic) dev_err(ionic->dev, "devlink_port_register failed: %d\n", err); else devlink_port_type_eth_set(&ionic->dl_port, - ionic->master_lif->netdev); + ionic->lif->netdev); return err; } diff --git a/drivers/net/ethernet/pensando/ionic/ionic_devlink.h b/drivers/net/ethernet/pensando/ionic/ionic_devlink.h index 0690172fc57a..5c01a9e306d8 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_devlink.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_devlink.h @@ -6,6 +6,9 @@ #include <net/devlink.h> +int ionic_firmware_update(struct ionic_lif *lif, const char *fw_name, + struct netlink_ext_ack *extack); + struct ionic *ionic_devlink_alloc(struct device *dev); void ionic_devlink_free(struct ionic *ionic); int ionic_devlink_register(struct ionic *ionic); diff --git a/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c b/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c index 3c57c331729f..ed9808fc743b 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c @@ -298,8 +298,8 @@ static void ionic_get_pauseparam(struct net_device *netdev, pause_type = lif->ionic->idev.port_info->config.pause_type; if (pause_type) { - pause->rx_pause = pause_type & IONIC_PAUSE_F_RX ? 1 : 0; - pause->tx_pause = pause_type & IONIC_PAUSE_F_TX ? 1 : 0; + pause->rx_pause = (pause_type & IONIC_PAUSE_F_RX) ? 1 : 0; + pause->tx_pause = (pause_type & IONIC_PAUSE_F_TX) ? 1 : 0; } } @@ -406,6 +406,13 @@ static int ionic_get_coalesce(struct net_device *netdev, coalesce->tx_coalesce_usecs = lif->tx_coalesce_usecs; coalesce->rx_coalesce_usecs = lif->rx_coalesce_usecs; + if (test_bit(IONIC_LIF_F_SPLIT_INTR, lif->state)) + coalesce->use_adaptive_tx_coalesce = test_bit(IONIC_LIF_F_TX_DIM_INTR, lif->state); + else + coalesce->use_adaptive_tx_coalesce = 0; + + coalesce->use_adaptive_rx_coalesce = test_bit(IONIC_LIF_F_RX_DIM_INTR, lif->state); + return 0; } @@ -414,10 +421,9 @@ static int ionic_set_coalesce(struct net_device *netdev, { struct ionic_lif *lif = netdev_priv(netdev); struct ionic_identity *ident; - struct ionic_qcq *qcq; + u32 rx_coal, rx_dim; + u32 tx_coal, tx_dim; unsigned int i; - u32 rx_coal; - u32 tx_coal; ident = &lif->ionic->ident; if (ident->dev.intr_coal_div == 0) { @@ -426,10 +432,11 @@ static int ionic_set_coalesce(struct net_device *netdev, return -EIO; } - /* Tx normally shares Rx interrupt, so only change Rx */ + /* Tx normally shares Rx interrupt, so only change Rx if not split */ if (!test_bit(IONIC_LIF_F_SPLIT_INTR, lif->state) && - coalesce->tx_coalesce_usecs != lif->rx_coalesce_usecs) { - netdev_warn(netdev, "only the rx-usecs can be changed\n"); + (coalesce->tx_coalesce_usecs != lif->rx_coalesce_usecs || + coalesce->use_adaptive_tx_coalesce)) { + netdev_warn(netdev, "only rx parameters can be changed\n"); return -EINVAL; } @@ -449,32 +456,44 @@ static int ionic_set_coalesce(struct net_device *netdev, /* Save the new values */ lif->rx_coalesce_usecs = coalesce->rx_coalesce_usecs; - if (rx_coal != lif->rx_coalesce_hw) { - lif->rx_coalesce_hw = rx_coal; - - if (test_bit(IONIC_LIF_F_UP, lif->state)) { - for (i = 0; i < lif->nxqs; i++) { - qcq = lif->rxqcqs[i].qcq; - ionic_intr_coal_init(lif->ionic->idev.intr_ctrl, - qcq->intr.index, - lif->rx_coalesce_hw); - } - } - } + lif->rx_coalesce_hw = rx_coal; if (test_bit(IONIC_LIF_F_SPLIT_INTR, lif->state)) lif->tx_coalesce_usecs = coalesce->tx_coalesce_usecs; else lif->tx_coalesce_usecs = coalesce->rx_coalesce_usecs; - if (tx_coal != lif->tx_coalesce_hw) { - lif->tx_coalesce_hw = tx_coal; + lif->tx_coalesce_hw = tx_coal; - if (test_bit(IONIC_LIF_F_UP, lif->state)) { - for (i = 0; i < lif->nxqs; i++) { - qcq = lif->txqcqs[i].qcq; + if (coalesce->use_adaptive_rx_coalesce) { + set_bit(IONIC_LIF_F_RX_DIM_INTR, lif->state); + rx_dim = rx_coal; + } else { + clear_bit(IONIC_LIF_F_RX_DIM_INTR, lif->state); + rx_dim = 0; + } + + if (coalesce->use_adaptive_tx_coalesce) { + set_bit(IONIC_LIF_F_TX_DIM_INTR, lif->state); + tx_dim = tx_coal; + } else { + clear_bit(IONIC_LIF_F_TX_DIM_INTR, lif->state); + tx_dim = 0; + } + + if (test_bit(IONIC_LIF_F_UP, lif->state)) { + for (i = 0; i < lif->nxqs; i++) { + if (lif->rxqcqs[i]->flags & IONIC_QCQ_F_INTR) { ionic_intr_coal_init(lif->ionic->idev.intr_ctrl, - qcq->intr.index, + lif->rxqcqs[i]->intr.index, + lif->rx_coalesce_hw); + lif->rxqcqs[i]->intr.dim_coal_hw = rx_dim; + } + + if (lif->txqcqs[i]->flags & IONIC_QCQ_F_INTR) { + ionic_intr_coal_init(lif->ionic->idev.intr_ctrl, + lif->txqcqs[i]->intr.index, lif->tx_coalesce_hw); + lif->txqcqs[i]->intr.dim_coal_hw = tx_dim; } } } @@ -493,18 +512,14 @@ static void ionic_get_ringparam(struct net_device *netdev, ring->rx_pending = lif->nrxq_descs; } -static void ionic_set_ringsize(struct ionic_lif *lif, void *arg) -{ - struct ethtool_ringparam *ring = arg; - - lif->ntxq_descs = ring->tx_pending; - lif->nrxq_descs = ring->rx_pending; -} - static int ionic_set_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring) { struct ionic_lif *lif = netdev_priv(netdev); + struct ionic_queue_params qparam; + int err; + + ionic_init_queue_params(lif, &qparam); if (ring->rx_mini_pending || ring->rx_jumbo_pending) { netdev_info(netdev, "Changing jumbo or mini descriptors not supported\n"); @@ -522,7 +537,28 @@ static int ionic_set_ringparam(struct net_device *netdev, ring->rx_pending == lif->nrxq_descs) return 0; - return ionic_reset_queues(lif, ionic_set_ringsize, ring); + if (ring->tx_pending != lif->ntxq_descs) + netdev_info(netdev, "Changing Tx ring size from %d to %d\n", + lif->ntxq_descs, ring->tx_pending); + + if (ring->rx_pending != lif->nrxq_descs) + netdev_info(netdev, "Changing Rx ring size from %d to %d\n", + lif->nrxq_descs, ring->rx_pending); + + /* if we're not running, just set the values and return */ + if (!netif_running(lif->netdev)) { + lif->ntxq_descs = ring->tx_pending; + lif->nrxq_descs = ring->rx_pending; + return 0; + } + + qparam.ntxq_descs = ring->tx_pending; + qparam.nrxq_descs = ring->rx_pending; + err = ionic_reconfigure_queues(lif, &qparam); + if (err) + netdev_info(netdev, "Ring reconfiguration failed, changes canceled: %d\n", err); + + return err; } static void ionic_get_channels(struct net_device *netdev, @@ -544,32 +580,15 @@ static void ionic_get_channels(struct net_device *netdev, } } -static void ionic_set_queuecount(struct ionic_lif *lif, void *arg) -{ - struct ethtool_channels *ch = arg; - - if (ch->combined_count) { - lif->nxqs = ch->combined_count; - if (test_bit(IONIC_LIF_F_SPLIT_INTR, lif->state)) { - clear_bit(IONIC_LIF_F_SPLIT_INTR, lif->state); - lif->tx_coalesce_usecs = lif->rx_coalesce_usecs; - lif->tx_coalesce_hw = lif->rx_coalesce_hw; - netdev_info(lif->netdev, "Sharing queue interrupts\n"); - } - } else { - lif->nxqs = ch->rx_count; - if (!test_bit(IONIC_LIF_F_SPLIT_INTR, lif->state)) { - set_bit(IONIC_LIF_F_SPLIT_INTR, lif->state); - netdev_info(lif->netdev, "Splitting queue interrupts\n"); - } - } -} - static int ionic_set_channels(struct net_device *netdev, struct ethtool_channels *ch) { struct ionic_lif *lif = netdev_priv(netdev); - int new_cnt; + struct ionic_queue_params qparam; + int max_cnt; + int err; + + ionic_init_queue_params(lif, &qparam); if (ch->rx_count != ch->tx_count) { netdev_info(netdev, "The rx and tx count must be equal\n"); @@ -577,20 +596,63 @@ static int ionic_set_channels(struct net_device *netdev, } if (ch->combined_count && ch->rx_count) { - netdev_info(netdev, "Use either combined_count or rx/tx_count, not both\n"); + netdev_info(netdev, "Use either combined or rx and tx, not both\n"); return -EINVAL; } - if (ch->combined_count) - new_cnt = ch->combined_count; - else - new_cnt = ch->rx_count; + max_cnt = lif->ionic->ntxqs_per_lif; + if (ch->combined_count) { + if (ch->combined_count > max_cnt) + return -EINVAL; + + if (test_bit(IONIC_LIF_F_SPLIT_INTR, lif->state)) + netdev_info(lif->netdev, "Sharing queue interrupts\n"); + else if (ch->combined_count == lif->nxqs) + return 0; - if (lif->nxqs != new_cnt) - netdev_info(netdev, "Changing queue count from %d to %d\n", - lif->nxqs, new_cnt); + if (lif->nxqs != ch->combined_count) + netdev_info(netdev, "Changing queue count from %d to %d\n", + lif->nxqs, ch->combined_count); - return ionic_reset_queues(lif, ionic_set_queuecount, ch); + qparam.nxqs = ch->combined_count; + qparam.intr_split = 0; + } else { + max_cnt /= 2; + if (ch->rx_count > max_cnt) + return -EINVAL; + + if (!test_bit(IONIC_LIF_F_SPLIT_INTR, lif->state)) + netdev_info(lif->netdev, "Splitting queue interrupts\n"); + else if (ch->rx_count == lif->nxqs) + return 0; + + if (lif->nxqs != ch->rx_count) + netdev_info(netdev, "Changing queue count from %d to %d\n", + lif->nxqs, ch->rx_count); + + qparam.nxqs = ch->rx_count; + qparam.intr_split = 1; + } + + /* if we're not running, just set the values and return */ + if (!netif_running(lif->netdev)) { + lif->nxqs = qparam.nxqs; + + if (qparam.intr_split) { + set_bit(IONIC_LIF_F_SPLIT_INTR, lif->state); + } else { + clear_bit(IONIC_LIF_F_SPLIT_INTR, lif->state); + lif->tx_coalesce_usecs = lif->rx_coalesce_usecs; + lif->tx_coalesce_hw = lif->rx_coalesce_hw; + } + return 0; + } + + err = ionic_reconfigure_queues(lif, &qparam); + if (err) + netdev_info(netdev, "Queue reconfiguration failed, changes canceled: %d\n", err); + + return err; } static u32 ionic_get_priv_flags(struct net_device *netdev) @@ -807,7 +869,9 @@ static int ionic_nway_reset(struct net_device *netdev) } static const struct ethtool_ops ionic_ethtool_ops = { - .supported_coalesce_params = ETHTOOL_COALESCE_USECS, + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_USE_ADAPTIVE_RX | + ETHTOOL_COALESCE_USE_ADAPTIVE_TX, .get_drvinfo = ionic_get_drvinfo, .get_regs_len = ionic_get_regs_len, .get_regs = ionic_get_regs, diff --git a/drivers/net/ethernet/pensando/ionic/ionic_fw.c b/drivers/net/ethernet/pensando/ionic/ionic_fw.c new file mode 100644 index 000000000000..f492ae406a60 --- /dev/null +++ b/drivers/net/ethernet/pensando/ionic/ionic_fw.c @@ -0,0 +1,206 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2020 Pensando Systems, Inc */ + +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/errno.h> +#include <linux/firmware.h> + +#include "ionic.h" +#include "ionic_dev.h" +#include "ionic_lif.h" +#include "ionic_devlink.h" + +/* The worst case wait for the install activity is about 25 minutes when + * installing a new CPLD, which is very seldom. Normal is about 30-35 + * seconds. Since the driver can't tell if a CPLD update will happen we + * set the timeout for the ugly case. + */ +#define IONIC_FW_INSTALL_TIMEOUT (25 * 60) +#define IONIC_FW_SELECT_TIMEOUT 30 + +/* Number of periodic log updates during fw file download */ +#define IONIC_FW_INTERVAL_FRACTION 32 + +static void ionic_dev_cmd_firmware_download(struct ionic_dev *idev, u64 addr, + u32 offset, u32 length) +{ + union ionic_dev_cmd cmd = { + .fw_download.opcode = IONIC_CMD_FW_DOWNLOAD, + .fw_download.offset = offset, + .fw_download.addr = addr, + .fw_download.length = length + }; + + ionic_dev_cmd_go(idev, &cmd); +} + +static void ionic_dev_cmd_firmware_install(struct ionic_dev *idev) +{ + union ionic_dev_cmd cmd = { + .fw_control.opcode = IONIC_CMD_FW_CONTROL, + .fw_control.oper = IONIC_FW_INSTALL_ASYNC + }; + + ionic_dev_cmd_go(idev, &cmd); +} + +static void ionic_dev_cmd_firmware_activate(struct ionic_dev *idev, u8 slot) +{ + union ionic_dev_cmd cmd = { + .fw_control.opcode = IONIC_CMD_FW_CONTROL, + .fw_control.oper = IONIC_FW_ACTIVATE_ASYNC, + .fw_control.slot = slot + }; + + ionic_dev_cmd_go(idev, &cmd); +} + +static int ionic_fw_status_long_wait(struct ionic *ionic, + const char *label, + unsigned long timeout, + u8 fw_cmd, + struct netlink_ext_ack *extack) +{ + union ionic_dev_cmd cmd = { + .fw_control.opcode = IONIC_CMD_FW_CONTROL, + .fw_control.oper = fw_cmd, + }; + unsigned long start_time; + unsigned long end_time; + int err; + + start_time = jiffies; + end_time = start_time + (timeout * HZ); + do { + mutex_lock(&ionic->dev_cmd_lock); + ionic_dev_cmd_go(&ionic->idev, &cmd); + err = ionic_dev_cmd_wait(ionic, DEVCMD_TIMEOUT); + mutex_unlock(&ionic->dev_cmd_lock); + + msleep(20); + } while (time_before(jiffies, end_time) && (err == -EAGAIN || err == -ETIMEDOUT)); + + if (err == -EAGAIN || err == -ETIMEDOUT) { + NL_SET_ERR_MSG_MOD(extack, "Firmware wait timed out"); + dev_err(ionic->dev, "DEV_CMD firmware wait %s timed out\n", label); + } else if (err) { + NL_SET_ERR_MSG_MOD(extack, "Firmware wait failed"); + } + + return err; +} + +int ionic_firmware_update(struct ionic_lif *lif, const char *fw_name, + struct netlink_ext_ack *extack) +{ + struct ionic_dev *idev = &lif->ionic->idev; + struct net_device *netdev = lif->netdev; + struct ionic *ionic = lif->ionic; + union ionic_dev_cmd_comp comp; + u32 buf_sz, copy_sz, offset; + const struct firmware *fw; + struct devlink *dl; + int next_interval; + int err = 0; + u8 fw_slot; + + netdev_info(netdev, "Installing firmware %s\n", fw_name); + + dl = priv_to_devlink(ionic); + devlink_flash_update_begin_notify(dl); + devlink_flash_update_status_notify(dl, "Preparing to flash", NULL, 0, 0); + + err = request_firmware(&fw, fw_name, ionic->dev); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Unable to find firmware file"); + goto err_out; + } + + buf_sz = sizeof(idev->dev_cmd_regs->data); + + netdev_dbg(netdev, + "downloading firmware - size %d part_sz %d nparts %lu\n", + (int)fw->size, buf_sz, DIV_ROUND_UP(fw->size, buf_sz)); + + offset = 0; + next_interval = 0; + while (offset < fw->size) { + if (offset >= next_interval) { + devlink_flash_update_status_notify(dl, "Downloading", NULL, + offset, fw->size); + next_interval = offset + (fw->size / IONIC_FW_INTERVAL_FRACTION); + } + + copy_sz = min_t(unsigned int, buf_sz, fw->size - offset); + mutex_lock(&ionic->dev_cmd_lock); + memcpy_toio(&idev->dev_cmd_regs->data, fw->data + offset, copy_sz); + ionic_dev_cmd_firmware_download(idev, + offsetof(union ionic_dev_cmd_regs, data), + offset, copy_sz); + err = ionic_dev_cmd_wait(ionic, DEVCMD_TIMEOUT); + mutex_unlock(&ionic->dev_cmd_lock); + if (err) { + netdev_err(netdev, + "download failed offset 0x%x addr 0x%lx len 0x%x\n", + offset, offsetof(union ionic_dev_cmd_regs, data), + copy_sz); + NL_SET_ERR_MSG_MOD(extack, "Segment download failed"); + goto err_out; + } + offset += copy_sz; + } + devlink_flash_update_status_notify(dl, "Downloading", NULL, + fw->size, fw->size); + + devlink_flash_update_timeout_notify(dl, "Installing", NULL, + IONIC_FW_INSTALL_TIMEOUT); + + mutex_lock(&ionic->dev_cmd_lock); + ionic_dev_cmd_firmware_install(idev); + err = ionic_dev_cmd_wait(ionic, DEVCMD_TIMEOUT); + ionic_dev_cmd_comp(idev, (union ionic_dev_cmd_comp *)&comp); + fw_slot = comp.fw_control.slot; + mutex_unlock(&ionic->dev_cmd_lock); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Failed to start firmware install"); + goto err_out; + } + + err = ionic_fw_status_long_wait(ionic, "Installing", + IONIC_FW_INSTALL_TIMEOUT, + IONIC_FW_INSTALL_STATUS, + extack); + if (err) + goto err_out; + + devlink_flash_update_timeout_notify(dl, "Selecting", NULL, + IONIC_FW_SELECT_TIMEOUT); + + mutex_lock(&ionic->dev_cmd_lock); + ionic_dev_cmd_firmware_activate(idev, fw_slot); + err = ionic_dev_cmd_wait(ionic, DEVCMD_TIMEOUT); + mutex_unlock(&ionic->dev_cmd_lock); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Failed to start firmware select"); + goto err_out; + } + + err = ionic_fw_status_long_wait(ionic, "Selecting", + IONIC_FW_SELECT_TIMEOUT, + IONIC_FW_ACTIVATE_STATUS, + extack); + if (err) + goto err_out; + + netdev_info(netdev, "Firmware update completed\n"); + +err_out: + if (err) + devlink_flash_update_status_notify(dl, "Flash failed", NULL, 0, 0); + else + devlink_flash_update_status_notify(dl, "Flash done", NULL, 0, 0); + release_firmware(fw); + devlink_flash_update_end_notify(dl); + return err; +} diff --git a/drivers/net/ethernet/pensando/ionic/ionic_if.h b/drivers/net/ethernet/pensando/ionic/ionic_if.h index acc94b244cf3..31ccfcdc2b0a 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_if.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_if.h @@ -63,8 +63,10 @@ enum ionic_cmd_opcode { IONIC_CMD_QOS_RESET = 245, /* Firmware commands */ - IONIC_CMD_FW_DOWNLOAD = 254, - IONIC_CMD_FW_CONTROL = 255, + IONIC_CMD_FW_DOWNLOAD = 252, + IONIC_CMD_FW_CONTROL = 253, + IONIC_CMD_FW_DOWNLOAD_V1 = 254, + IONIC_CMD_FW_CONTROL_V1 = 255, }; /** @@ -94,6 +96,7 @@ enum ionic_status_code { IONIC_RC_ERROR = 29, /* Generic error */ IONIC_RC_ERDMA = 30, /* Generic RDMA error */ IONIC_RC_EVFID = 31, /* VF ID does not exist */ + IONIC_RC_EBAD_FW = 32, /* FW file is invalid or corrupted */ }; enum ionic_notifyq_opcode { @@ -2069,14 +2072,23 @@ typedef struct ionic_admin_comp ionic_fw_download_comp; /** * enum ionic_fw_control_oper - FW control operations - * @IONIC_FW_RESET: Reset firmware - * @IONIC_FW_INSTALL: Install firmware - * @IONIC_FW_ACTIVATE: Activate firmware + * @IONIC_FW_RESET: Reset firmware + * @IONIC_FW_INSTALL: Install firmware + * @IONIC_FW_ACTIVATE: Activate firmware + * @IONIC_FW_INSTALL_ASYNC: Install firmware asynchronously + * @IONIC_FW_INSTALL_STATUS: Firmware installation status + * @IONIC_FW_ACTIVATE_ASYNC: Activate firmware asynchronously + * @IONIC_FW_ACTIVATE_STATUS: Firmware activate status */ enum ionic_fw_control_oper { - IONIC_FW_RESET = 0, - IONIC_FW_INSTALL = 1, - IONIC_FW_ACTIVATE = 2, + IONIC_FW_RESET = 0, + IONIC_FW_INSTALL = 1, + IONIC_FW_ACTIVATE = 2, + IONIC_FW_INSTALL_ASYNC = 3, + IONIC_FW_INSTALL_STATUS = 4, + IONIC_FW_ACTIVATE_ASYNC = 5, + IONIC_FW_ACTIVATE_STATUS = 6, + IONIC_FW_UPDATE_CLEANUP = 7, }; /** @@ -2689,6 +2701,9 @@ union ionic_dev_cmd { struct ionic_q_identify_cmd q_identify; struct ionic_q_init_cmd q_init; struct ionic_q_control_cmd q_control; + + struct ionic_fw_download_cmd fw_download; + struct ionic_fw_control_cmd fw_control; }; union ionic_dev_cmd_comp { @@ -2722,6 +2737,9 @@ union ionic_dev_cmd_comp { struct ionic_q_identify_comp q_identify; struct ionic_q_init_comp q_init; + + ionic_fw_download_comp fw_download; + struct ionic_fw_control_comp fw_control; }; /** diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c index 26988ad7ec97..d655a7ae3058 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c @@ -36,25 +36,44 @@ static void ionic_lif_handle_fw_down(struct ionic_lif *lif); static void ionic_lif_handle_fw_up(struct ionic_lif *lif); static void ionic_lif_set_netdev_info(struct ionic_lif *lif); +static void ionic_txrx_deinit(struct ionic_lif *lif); +static int ionic_txrx_init(struct ionic_lif *lif); static int ionic_start_queues(struct ionic_lif *lif); static void ionic_stop_queues(struct ionic_lif *lif); static void ionic_lif_queue_identify(struct ionic_lif *lif); +static void ionic_dim_work(struct work_struct *work) +{ + struct dim *dim = container_of(work, struct dim, work); + struct dim_cq_moder cur_moder; + struct ionic_qcq *qcq; + u32 new_coal; + + cur_moder = net_dim_get_rx_moderation(dim->mode, dim->profile_ix); + qcq = container_of(dim, struct ionic_qcq, dim); + new_coal = ionic_coal_usec_to_hw(qcq->q.lif->ionic, cur_moder.usec); + qcq->intr.dim_coal_hw = new_coal ? new_coal : 1; + dim->state = DIM_START_MEASURE; +} + static void ionic_lif_deferred_work(struct work_struct *work) { struct ionic_lif *lif = container_of(work, struct ionic_lif, deferred.work); struct ionic_deferred *def = &lif->deferred; struct ionic_deferred_work *w = NULL; - spin_lock_bh(&def->lock); - if (!list_empty(&def->list)) { - w = list_first_entry(&def->list, - struct ionic_deferred_work, list); - list_del(&w->list); - } - spin_unlock_bh(&def->lock); + do { + spin_lock_bh(&def->lock); + if (!list_empty(&def->list)) { + w = list_first_entry(&def->list, + struct ionic_deferred_work, list); + list_del(&w->list); + } + spin_unlock_bh(&def->lock); + + if (!w) + break; - if (w) { switch (w->type) { case IONIC_DW_TYPE_RX_MODE: ionic_lif_rx_mode(lif, w->rx_mode); @@ -78,8 +97,8 @@ static void ionic_lif_deferred_work(struct work_struct *work) break; } kfree(w); - schedule_work(&def->work); - } + w = NULL; + } while (true); } void ionic_lif_deferred_enqueue(struct ionic_deferred *def, @@ -135,7 +154,7 @@ static void ionic_link_status_check(struct ionic_lif *lif) clear_bit(IONIC_LIF_F_LINK_CHECK_REQUESTED, lif->state); } -void ionic_link_status_check_request(struct ionic_lif *lif) +void ionic_link_status_check_request(struct ionic_lif *lif, bool can_sleep) { struct ionic_deferred_work *work; @@ -143,10 +162,12 @@ void ionic_link_status_check_request(struct ionic_lif *lif) if (test_and_set_bit(IONIC_LIF_F_LINK_CHECK_REQUESTED, lif->state)) return; - if (in_interrupt()) { + if (!can_sleep) { work = kzalloc(sizeof(*work), GFP_ATOMIC); - if (!work) + if (!work) { + clear_bit(IONIC_LIF_F_LINK_CHECK_REQUESTED, lif->state); return; + } work->type = IONIC_DW_TYPE_LINK_STATUS; ionic_lif_deferred_enqueue(&lif->deferred, work); @@ -243,31 +264,30 @@ static int ionic_qcq_enable(struct ionic_qcq *qcq) return ionic_adminq_post_wait(lif, &ctx); } -static int ionic_qcq_disable(struct ionic_qcq *qcq) +static int ionic_qcq_disable(struct ionic_qcq *qcq, bool send_to_hw) { - struct ionic_queue *q = &qcq->q; - struct ionic_lif *lif = q->lif; - struct ionic_dev *idev; - struct device *dev; + struct ionic_queue *q; + struct ionic_lif *lif; + int err = 0; struct ionic_admin_ctx ctx = { .work = COMPLETION_INITIALIZER_ONSTACK(ctx.work), .cmd.q_control = { .opcode = IONIC_CMD_Q_CONTROL, - .lif_index = cpu_to_le16(lif->index), - .type = q->type, - .index = cpu_to_le32(q->index), .oper = IONIC_Q_DISABLE, }, }; - idev = &lif->ionic->idev; - dev = lif->ionic->dev; + if (!qcq) + return -ENXIO; - dev_dbg(dev, "q_disable.index %d q_disable.qtype %d\n", - ctx.cmd.q_control.index, ctx.cmd.q_control.type); + q = &qcq->q; + lif = q->lif; if (qcq->flags & IONIC_QCQ_F_INTR) { + struct ionic_dev *idev = &lif->ionic->idev; + + cancel_work_sync(&qcq->dim.work); ionic_intr_mask(idev->intr_ctrl, qcq->intr.index, IONIC_INTR_MASK_SET); synchronize_irq(qcq->intr.vector); @@ -275,7 +295,17 @@ static int ionic_qcq_disable(struct ionic_qcq *qcq) napi_disable(&qcq->napi); } - return ionic_adminq_post_wait(lif, &ctx); + if (send_to_hw) { + ctx.cmd.q_control.lif_index = cpu_to_le16(lif->index); + ctx.cmd.q_control.type = q->type; + ctx.cmd.q_control.index = cpu_to_le32(q->index); + dev_dbg(lif->ionic->dev, "q_disable.index %d q_disable.qtype %d\n", + ctx.cmd.q_control.index, ctx.cmd.q_control.type); + + err = ionic_adminq_post_wait(lif, &ctx); + } + + return err; } static void ionic_lif_qcq_deinit(struct ionic_lif *lif, struct ionic_qcq *qcq) @@ -297,6 +327,18 @@ static void ionic_lif_qcq_deinit(struct ionic_lif *lif, struct ionic_qcq *qcq) qcq->flags &= ~IONIC_QCQ_F_INITED; } +static void ionic_qcq_intr_free(struct ionic_lif *lif, struct ionic_qcq *qcq) +{ + if (!(qcq->flags & IONIC_QCQ_F_INTR) || qcq->intr.vector == 0) + return; + + irq_set_affinity_hint(qcq->intr.vector, NULL); + devm_free_irq(lif->ionic->dev, qcq->intr.vector, &qcq->napi); + qcq->intr.vector = 0; + ionic_intr_free(lif->ionic, qcq->intr.index); + qcq->intr.index = IONIC_INTR_INDEX_NOT_ASSIGNED; +} + static void ionic_qcq_free(struct ionic_lif *lif, struct ionic_qcq *qcq) { struct device *dev = lif->ionic->dev; @@ -306,51 +348,62 @@ static void ionic_qcq_free(struct ionic_lif *lif, struct ionic_qcq *qcq) ionic_debugfs_del_qcq(qcq); - dma_free_coherent(dev, qcq->total_size, qcq->base, qcq->base_pa); - qcq->base = NULL; - qcq->base_pa = 0; + if (qcq->q_base) { + dma_free_coherent(dev, qcq->q_size, qcq->q_base, qcq->q_base_pa); + qcq->q_base = NULL; + qcq->q_base_pa = 0; + } - if (qcq->flags & IONIC_QCQ_F_INTR) { - irq_set_affinity_hint(qcq->intr.vector, NULL); - devm_free_irq(dev, qcq->intr.vector, &qcq->napi); - qcq->intr.vector = 0; - ionic_intr_free(lif->ionic, qcq->intr.index); + if (qcq->cq_base) { + dma_free_coherent(dev, qcq->cq_size, qcq->cq_base, qcq->cq_base_pa); + qcq->cq_base = NULL; + qcq->cq_base_pa = 0; + } + + if (qcq->sg_base) { + dma_free_coherent(dev, qcq->sg_size, qcq->sg_base, qcq->sg_base_pa); + qcq->sg_base = NULL; + qcq->sg_base_pa = 0; } - devm_kfree(dev, qcq->cq.info); - qcq->cq.info = NULL; - devm_kfree(dev, qcq->q.info); - qcq->q.info = NULL; - devm_kfree(dev, qcq); + ionic_qcq_intr_free(lif, qcq); + + if (qcq->cq.info) { + devm_kfree(dev, qcq->cq.info); + qcq->cq.info = NULL; + } + if (qcq->q.info) { + devm_kfree(dev, qcq->q.info); + qcq->q.info = NULL; + } } static void ionic_qcqs_free(struct ionic_lif *lif) { struct device *dev = lif->ionic->dev; - unsigned int i; if (lif->notifyqcq) { ionic_qcq_free(lif, lif->notifyqcq); + devm_kfree(dev, lif->notifyqcq); lif->notifyqcq = NULL; } if (lif->adminqcq) { ionic_qcq_free(lif, lif->adminqcq); + devm_kfree(dev, lif->adminqcq); lif->adminqcq = NULL; } if (lif->rxqcqs) { - for (i = 0; i < lif->nxqs; i++) - if (lif->rxqcqs[i].stats) - devm_kfree(dev, lif->rxqcqs[i].stats); + devm_kfree(dev, lif->rxqstats); + lif->rxqstats = NULL; devm_kfree(dev, lif->rxqcqs); lif->rxqcqs = NULL; } if (lif->txqcqs) { - for (i = 0; i < lif->nxqs; i++) - if (lif->txqcqs[i].stats) - devm_kfree(dev, lif->txqcqs[i].stats); + devm_kfree(dev, lif->txqstats); + lif->txqstats = NULL; devm_kfree(dev, lif->txqcqs); lif->txqcqs = NULL; } @@ -368,6 +421,53 @@ static void ionic_link_qcq_interrupts(struct ionic_qcq *src_qcq, n_qcq->intr.index = src_qcq->intr.index; } +static int ionic_alloc_qcq_interrupt(struct ionic_lif *lif, struct ionic_qcq *qcq) +{ + int err; + + if (!(qcq->flags & IONIC_QCQ_F_INTR)) { + qcq->intr.index = IONIC_INTR_INDEX_NOT_ASSIGNED; + return 0; + } + + err = ionic_intr_alloc(lif, &qcq->intr); + if (err) { + netdev_warn(lif->netdev, "no intr for %s: %d\n", + qcq->q.name, err); + goto err_out; + } + + err = ionic_bus_get_irq(lif->ionic, qcq->intr.index); + if (err < 0) { + netdev_warn(lif->netdev, "no vector for %s: %d\n", + qcq->q.name, err); + goto err_out_free_intr; + } + qcq->intr.vector = err; + ionic_intr_mask_assert(lif->ionic->idev.intr_ctrl, qcq->intr.index, + IONIC_INTR_MASK_SET); + + err = ionic_request_irq(lif, qcq); + if (err) { + netdev_warn(lif->netdev, "irq request failed %d\n", err); + goto err_out_free_intr; + } + + /* try to get the irq on the local numa node first */ + qcq->intr.cpu = cpumask_local_spread(qcq->intr.index, + dev_to_node(lif->ionic->dev)); + if (qcq->intr.cpu != -1) + cpumask_set_cpu(qcq->intr.cpu, &qcq->intr.affinity_mask); + + netdev_dbg(lif->netdev, "%s: Interrupt index %d\n", qcq->q.name, qcq->intr.index); + return 0; + +err_out_free_intr: + ionic_intr_free(lif->ionic, qcq->intr.index); +err_out: + return err; +} + static int ionic_qcq_alloc(struct ionic_lif *lif, unsigned int type, unsigned int index, const char *name, unsigned int flags, @@ -377,7 +477,6 @@ static int ionic_qcq_alloc(struct ionic_lif *lif, unsigned int type, unsigned int pid, struct ionic_qcq **qcq) { struct ionic_dev *idev = &lif->ionic->idev; - u32 q_size, cq_size, sg_size, total_size; struct device *dev = lif->ionic->dev; void *q_base, *cq_base, *sg_base; dma_addr_t cq_base_pa = 0; @@ -388,21 +487,6 @@ static int ionic_qcq_alloc(struct ionic_lif *lif, unsigned int type, *qcq = NULL; - q_size = num_descs * desc_size; - cq_size = num_descs * cq_desc_size; - sg_size = num_descs * sg_desc_size; - - total_size = ALIGN(q_size, PAGE_SIZE) + ALIGN(cq_size, PAGE_SIZE); - /* Note: aligning q_size/cq_size is not enough due to cq_base - * address aligning as q_base could be not aligned to the page. - * Adding PAGE_SIZE. - */ - total_size += PAGE_SIZE; - if (flags & IONIC_QCQ_F_SG) { - total_size += ALIGN(sg_size, PAGE_SIZE); - total_size += PAGE_SIZE; - } - new = devm_kzalloc(dev, sizeof(*new), GFP_KERNEL); if (!new) { netdev_err(lif->netdev, "Cannot allocate queue structure\n"); @@ -417,7 +501,7 @@ static int ionic_qcq_alloc(struct ionic_lif *lif, unsigned int type, if (!new->q.info) { netdev_err(lif->netdev, "Cannot allocate queue info\n"); err = -ENOMEM; - goto err_out; + goto err_out_free_qcq; } new->q.type = type; @@ -426,41 +510,12 @@ static int ionic_qcq_alloc(struct ionic_lif *lif, unsigned int type, desc_size, sg_desc_size, pid); if (err) { netdev_err(lif->netdev, "Cannot initialize queue\n"); - goto err_out; + goto err_out_free_q_info; } - if (flags & IONIC_QCQ_F_INTR) { - err = ionic_intr_alloc(lif, &new->intr); - if (err) { - netdev_warn(lif->netdev, "no intr for %s: %d\n", - name, err); - goto err_out; - } - - err = ionic_bus_get_irq(lif->ionic, new->intr.index); - if (err < 0) { - netdev_warn(lif->netdev, "no vector for %s: %d\n", - name, err); - goto err_out_free_intr; - } - new->intr.vector = err; - ionic_intr_mask_assert(idev->intr_ctrl, new->intr.index, - IONIC_INTR_MASK_SET); - - err = ionic_request_irq(lif, new); - if (err) { - netdev_warn(lif->netdev, "irq request failed %d\n", err); - goto err_out_free_intr; - } - - new->intr.cpu = cpumask_local_spread(new->intr.index, - dev_to_node(dev)); - if (new->intr.cpu != -1) - cpumask_set_cpu(new->intr.cpu, - &new->intr.affinity_mask); - } else { - new->intr.index = IONIC_INTR_INDEX_NOT_ASSIGNED; - } + err = ionic_alloc_qcq_interrupt(lif, new); + if (err) + goto err_out; new->cq.info = devm_kcalloc(dev, num_descs, sizeof(*new->cq.info), GFP_KERNEL); @@ -473,46 +528,95 @@ static int ionic_qcq_alloc(struct ionic_lif *lif, unsigned int type, err = ionic_cq_init(lif, &new->cq, &new->intr, num_descs, cq_desc_size); if (err) { netdev_err(lif->netdev, "Cannot initialize completion queue\n"); - goto err_out_free_irq; + goto err_out_free_cq_info; } - new->base = dma_alloc_coherent(dev, total_size, &new->base_pa, - GFP_KERNEL); - if (!new->base) { - netdev_err(lif->netdev, "Cannot allocate queue DMA memory\n"); - err = -ENOMEM; - goto err_out_free_irq; - } - - new->total_size = total_size; + if (flags & IONIC_QCQ_F_NOTIFYQ) { + int q_size, cq_size; - q_base = new->base; - q_base_pa = new->base_pa; + /* q & cq need to be contiguous in case of notifyq */ + q_size = ALIGN(num_descs * desc_size, PAGE_SIZE); + cq_size = ALIGN(num_descs * cq_desc_size, PAGE_SIZE); - cq_base = (void *)ALIGN((uintptr_t)q_base + q_size, PAGE_SIZE); - cq_base_pa = ALIGN(q_base_pa + q_size, PAGE_SIZE); + new->q_size = PAGE_SIZE + q_size + cq_size; + new->q_base = dma_alloc_coherent(dev, new->q_size, + &new->q_base_pa, GFP_KERNEL); + if (!new->q_base) { + netdev_err(lif->netdev, "Cannot allocate qcq DMA memory\n"); + err = -ENOMEM; + goto err_out_free_cq_info; + } + q_base = PTR_ALIGN(new->q_base, PAGE_SIZE); + q_base_pa = ALIGN(new->q_base_pa, PAGE_SIZE); + ionic_q_map(&new->q, q_base, q_base_pa); + + cq_base = PTR_ALIGN(q_base + q_size, PAGE_SIZE); + cq_base_pa = ALIGN(new->q_base_pa + q_size, PAGE_SIZE); + ionic_cq_map(&new->cq, cq_base, cq_base_pa); + ionic_cq_bind(&new->cq, &new->q); + } else { + new->q_size = PAGE_SIZE + (num_descs * desc_size); + new->q_base = dma_alloc_coherent(dev, new->q_size, &new->q_base_pa, + GFP_KERNEL); + if (!new->q_base) { + netdev_err(lif->netdev, "Cannot allocate queue DMA memory\n"); + err = -ENOMEM; + goto err_out_free_cq_info; + } + q_base = PTR_ALIGN(new->q_base, PAGE_SIZE); + q_base_pa = ALIGN(new->q_base_pa, PAGE_SIZE); + ionic_q_map(&new->q, q_base, q_base_pa); + + new->cq_size = PAGE_SIZE + (num_descs * cq_desc_size); + new->cq_base = dma_alloc_coherent(dev, new->cq_size, &new->cq_base_pa, + GFP_KERNEL); + if (!new->cq_base) { + netdev_err(lif->netdev, "Cannot allocate cq DMA memory\n"); + err = -ENOMEM; + goto err_out_free_q; + } + cq_base = PTR_ALIGN(new->cq_base, PAGE_SIZE); + cq_base_pa = ALIGN(new->cq_base_pa, PAGE_SIZE); + ionic_cq_map(&new->cq, cq_base, cq_base_pa); + ionic_cq_bind(&new->cq, &new->q); + } if (flags & IONIC_QCQ_F_SG) { - sg_base = (void *)ALIGN((uintptr_t)cq_base + cq_size, - PAGE_SIZE); - sg_base_pa = ALIGN(cq_base_pa + cq_size, PAGE_SIZE); + new->sg_size = PAGE_SIZE + (num_descs * sg_desc_size); + new->sg_base = dma_alloc_coherent(dev, new->sg_size, &new->sg_base_pa, + GFP_KERNEL); + if (!new->sg_base) { + netdev_err(lif->netdev, "Cannot allocate sg DMA memory\n"); + err = -ENOMEM; + goto err_out_free_cq; + } + sg_base = PTR_ALIGN(new->sg_base, PAGE_SIZE); + sg_base_pa = ALIGN(new->sg_base_pa, PAGE_SIZE); ionic_q_sg_map(&new->q, sg_base, sg_base_pa); } - ionic_q_map(&new->q, q_base, q_base_pa); - ionic_cq_map(&new->cq, cq_base, cq_base_pa); - ionic_cq_bind(&new->cq, &new->q); + INIT_WORK(&new->dim.work, ionic_dim_work); + new->dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE; *qcq = new; return 0; +err_out_free_cq: + dma_free_coherent(dev, new->cq_size, new->cq_base, new->cq_base_pa); +err_out_free_q: + dma_free_coherent(dev, new->q_size, new->q_base, new->q_base_pa); +err_out_free_cq_info: + devm_kfree(dev, new->cq.info); err_out_free_irq: - if (flags & IONIC_QCQ_F_INTR) + if (flags & IONIC_QCQ_F_INTR) { devm_free_irq(dev, new->intr.vector, &new->napi); -err_out_free_intr: - if (flags & IONIC_QCQ_F_INTR) ionic_intr_free(lif->ionic, new->intr.index); + } +err_out_free_q_info: + devm_kfree(dev, new->q.info); +err_out_free_qcq: + devm_kfree(dev, new); err_out: dev_err(dev, "qcq alloc of %s%d failed %d\n", name, index, err); return err; @@ -521,10 +625,8 @@ err_out: static int ionic_qcqs_alloc(struct ionic_lif *lif) { struct device *dev = lif->ionic->dev; - unsigned int q_list_size; unsigned int flags; int err; - int i; flags = IONIC_QCQ_F_INTR; err = ionic_qcq_alloc(lif, IONIC_QTYPE_ADMINQ, 0, "admin", flags, @@ -544,63 +646,50 @@ static int ionic_qcqs_alloc(struct ionic_lif *lif) sizeof(union ionic_notifyq_comp), 0, lif->kern_pid, &lif->notifyqcq); if (err) - goto err_out_free_adminqcq; + goto err_out; ionic_debugfs_add_qcq(lif, lif->notifyqcq); /* Let the notifyq ride on the adminq interrupt */ ionic_link_qcq_interrupts(lif->adminqcq, lif->notifyqcq); } - q_list_size = sizeof(*lif->txqcqs) * lif->nxqs; err = -ENOMEM; - lif->txqcqs = devm_kzalloc(dev, q_list_size, GFP_KERNEL); + lif->txqcqs = devm_kcalloc(dev, lif->ionic->ntxqs_per_lif, + sizeof(struct ionic_qcq *), GFP_KERNEL); if (!lif->txqcqs) - goto err_out_free_notifyqcq; - for (i = 0; i < lif->nxqs; i++) { - lif->txqcqs[i].stats = devm_kzalloc(dev, - sizeof(struct ionic_q_stats), - GFP_KERNEL); - if (!lif->txqcqs[i].stats) - goto err_out_free_tx_stats; - } - - lif->rxqcqs = devm_kzalloc(dev, q_list_size, GFP_KERNEL); + goto err_out; + lif->rxqcqs = devm_kcalloc(dev, lif->ionic->nrxqs_per_lif, + sizeof(struct ionic_qcq *), GFP_KERNEL); if (!lif->rxqcqs) - goto err_out_free_tx_stats; - for (i = 0; i < lif->nxqs; i++) { - lif->rxqcqs[i].stats = devm_kzalloc(dev, - sizeof(struct ionic_q_stats), - GFP_KERNEL); - if (!lif->rxqcqs[i].stats) - goto err_out_free_rx_stats; - } + goto err_out; - return 0; + lif->txqstats = devm_kcalloc(dev, lif->ionic->ntxqs_per_lif, + sizeof(struct ionic_tx_stats), GFP_KERNEL); + if (!lif->txqstats) + goto err_out; + lif->rxqstats = devm_kcalloc(dev, lif->ionic->nrxqs_per_lif, + sizeof(struct ionic_rx_stats), GFP_KERNEL); + if (!lif->rxqstats) + goto err_out; -err_out_free_rx_stats: - for (i = 0; i < lif->nxqs; i++) - if (lif->rxqcqs[i].stats) - devm_kfree(dev, lif->rxqcqs[i].stats); - devm_kfree(dev, lif->rxqcqs); - lif->rxqcqs = NULL; -err_out_free_tx_stats: - for (i = 0; i < lif->nxqs; i++) - if (lif->txqcqs[i].stats) - devm_kfree(dev, lif->txqcqs[i].stats); - devm_kfree(dev, lif->txqcqs); - lif->txqcqs = NULL; -err_out_free_notifyqcq: - if (lif->notifyqcq) { - ionic_qcq_free(lif, lif->notifyqcq); - lif->notifyqcq = NULL; - } -err_out_free_adminqcq: - ionic_qcq_free(lif, lif->adminqcq); - lif->adminqcq = NULL; + return 0; +err_out: + ionic_qcqs_free(lif); return err; } +static void ionic_qcq_sanitize(struct ionic_qcq *qcq) +{ + qcq->q.tail_idx = 0; + qcq->q.head_idx = 0; + qcq->cq.tail_idx = 0; + qcq->cq.done_color = 1; + memset(qcq->q_base, 0, qcq->q_size); + memset(qcq->cq_base, 0, qcq->cq_size); + memset(qcq->sg_base, 0, qcq->sg_size); +} + static int ionic_lif_txq_init(struct ionic_lif *lif, struct ionic_qcq *qcq) { struct device *dev = lif->ionic->dev; @@ -626,10 +715,10 @@ static int ionic_lif_txq_init(struct ionic_lif *lif, struct ionic_qcq *qcq) unsigned int intr_index; int err; - if (test_bit(IONIC_LIF_F_SPLIT_INTR, lif->state)) + if (qcq->flags & IONIC_QCQ_F_INTR) intr_index = qcq->intr.index; else - intr_index = lif->rxqcqs[q->index].qcq->intr.index; + intr_index = lif->rxqcqs[q->index]->intr.index; ctx.cmd.q_init.intr_index = cpu_to_le16(intr_index); dev_dbg(dev, "txq_init.pid %d\n", ctx.cmd.q_init.pid); @@ -640,9 +729,7 @@ static int ionic_lif_txq_init(struct ionic_lif *lif, struct ionic_qcq *qcq) dev_dbg(dev, "txq_init.ver %d\n", ctx.cmd.q_init.ver); dev_dbg(dev, "txq_init.intr_index %d\n", ctx.cmd.q_init.intr_index); - q->tail = q->info; - q->head = q->tail; - cq->tail = cq->info; + ionic_qcq_sanitize(qcq); err = ionic_adminq_post_wait(lif, &ctx); if (err) @@ -697,9 +784,7 @@ static int ionic_lif_rxq_init(struct ionic_lif *lif, struct ionic_qcq *qcq) dev_dbg(dev, "rxq_init.ver %d\n", ctx.cmd.q_init.ver); dev_dbg(dev, "rxq_init.intr_index %d\n", ctx.cmd.q_init.intr_index); - q->tail = q->info; - q->head = q->tail; - cq->tail = cq->info; + ionic_qcq_sanitize(qcq); err = ionic_adminq_post_wait(lif, &ctx); if (err) @@ -751,7 +836,7 @@ static bool ionic_notifyq_service(struct ionic_cq *cq, switch (le16_to_cpu(comp->event.ecode)) { case IONIC_EVENT_LINK_CHANGE: - ionic_link_status_check_request(lif); + ionic_link_status_check_request(lif, false); break; case IONIC_EVENT_RESET: work = kzalloc(sizeof(*work), GFP_ATOMIC); @@ -771,21 +856,6 @@ static bool ionic_notifyq_service(struct ionic_cq *cq, return true; } -static int ionic_notifyq_clean(struct ionic_lif *lif, int budget) -{ - struct ionic_dev *idev = &lif->ionic->idev; - struct ionic_cq *cq = &lif->notifyqcq->cq; - u32 work_done; - - work_done = ionic_cq_service(cq, budget, ionic_notifyq_service, - NULL, NULL); - if (work_done) - ionic_intr_credits(idev->intr_ctrl, cq->bound_intr->index, - work_done, IONIC_INTR_CRED_RESET_COALESCE); - - return work_done; -} - static bool ionic_adminq_service(struct ionic_cq *cq, struct ionic_cq_info *cq_info) { @@ -801,15 +871,36 @@ static bool ionic_adminq_service(struct ionic_cq *cq, static int ionic_adminq_napi(struct napi_struct *napi, int budget) { + struct ionic_intr_info *intr = napi_to_cq(napi)->bound_intr; struct ionic_lif *lif = napi_to_cq(napi)->lif; + struct ionic_dev *idev = &lif->ionic->idev; + unsigned int flags = 0; int n_work = 0; int a_work = 0; + int work_done; - if (likely(lif->notifyqcq && lif->notifyqcq->flags & IONIC_QCQ_F_INITED)) - n_work = ionic_notifyq_clean(lif, budget); - a_work = ionic_napi(napi, budget, ionic_adminq_service, NULL, NULL); + if (lif->notifyqcq && lif->notifyqcq->flags & IONIC_QCQ_F_INITED) + n_work = ionic_cq_service(&lif->notifyqcq->cq, budget, + ionic_notifyq_service, NULL, NULL); + + if (lif->adminqcq && lif->adminqcq->flags & IONIC_QCQ_F_INITED) + a_work = ionic_cq_service(&lif->adminqcq->cq, budget, + ionic_adminq_service, NULL, NULL); + + work_done = max(n_work, a_work); + if (work_done < budget && napi_complete_done(napi, work_done)) { + flags |= IONIC_INTR_CRED_UNMASK; + lif->adminqcq->cq.bound_intr->rearm_count++; + } + + if (work_done || flags) { + flags |= IONIC_INTR_CRED_RESET_COALESCE; + ionic_intr_credits(idev->intr_ctrl, + intr->index, + n_work + a_work, flags); + } - return max(n_work, a_work); + return work_done; } void ionic_get_stats64(struct net_device *netdev, @@ -928,9 +1019,9 @@ static int ionic_lif_addr_del(struct ionic_lif *lif, const u8 *addr) return 0; } -static int ionic_lif_addr(struct ionic_lif *lif, const u8 *addr, bool add) +static int ionic_lif_addr(struct ionic_lif *lif, const u8 *addr, bool add, + bool can_sleep) { - struct ionic *ionic = lif->ionic; struct ionic_deferred_work *work; unsigned int nmfilters; unsigned int nufilters; @@ -940,8 +1031,8 @@ static int ionic_lif_addr(struct ionic_lif *lif, const u8 *addr, bool add) * here before checking the need for deferral so that we * can return an overflow error to the stack. */ - nmfilters = le32_to_cpu(ionic->ident.lif.eth.max_mcast_filters); - nufilters = le32_to_cpu(ionic->ident.lif.eth.max_ucast_filters); + nmfilters = le32_to_cpu(lif->identity->eth.max_mcast_filters); + nufilters = le32_to_cpu(lif->identity->eth.max_ucast_filters); if ((is_multicast_ether_addr(addr) && lif->nmcast < nmfilters)) lif->nmcast++; @@ -957,7 +1048,7 @@ static int ionic_lif_addr(struct ionic_lif *lif, const u8 *addr, bool add) lif->nucast--; } - if (in_interrupt()) { + if (!can_sleep) { work = kzalloc(sizeof(*work), GFP_ATOMIC); if (!work) { netdev_err(lif->netdev, "%s OOM\n", __func__); @@ -983,12 +1074,22 @@ static int ionic_lif_addr(struct ionic_lif *lif, const u8 *addr, bool add) static int ionic_addr_add(struct net_device *netdev, const u8 *addr) { - return ionic_lif_addr(netdev_priv(netdev), addr, true); + return ionic_lif_addr(netdev_priv(netdev), addr, true, true); +} + +static int ionic_ndo_addr_add(struct net_device *netdev, const u8 *addr) +{ + return ionic_lif_addr(netdev_priv(netdev), addr, true, false); } static int ionic_addr_del(struct net_device *netdev, const u8 *addr) { - return ionic_lif_addr(netdev_priv(netdev), addr, false); + return ionic_lif_addr(netdev_priv(netdev), addr, false, true); +} + +static int ionic_ndo_addr_del(struct net_device *netdev, const u8 *addr) +{ + return ionic_lif_addr(netdev_priv(netdev), addr, false, false); } static void ionic_lif_rx_mode(struct ionic_lif *lif, unsigned int rx_mode) @@ -1028,11 +1129,12 @@ static void ionic_lif_rx_mode(struct ionic_lif *lif, unsigned int rx_mode) lif->rx_mode = rx_mode; } -static void _ionic_lif_rx_mode(struct ionic_lif *lif, unsigned int rx_mode) +static void _ionic_lif_rx_mode(struct ionic_lif *lif, unsigned int rx_mode, + bool from_ndo) { struct ionic_deferred_work *work; - if (in_interrupt()) { + if (from_ndo) { work = kzalloc(sizeof(*work), GFP_ATOMIC); if (!work) { netdev_err(lif->netdev, "%s OOM\n", __func__); @@ -1047,15 +1149,21 @@ static void _ionic_lif_rx_mode(struct ionic_lif *lif, unsigned int rx_mode) } } -static void ionic_set_rx_mode(struct net_device *netdev) +static void ionic_dev_uc_sync(struct net_device *netdev, bool from_ndo) +{ + if (from_ndo) + __dev_uc_sync(netdev, ionic_ndo_addr_add, ionic_ndo_addr_del); + else + __dev_uc_sync(netdev, ionic_addr_add, ionic_addr_del); + +} + +static void ionic_set_rx_mode(struct net_device *netdev, bool from_ndo) { struct ionic_lif *lif = netdev_priv(netdev); - struct ionic_identity *ident; unsigned int nfilters; unsigned int rx_mode; - ident = &lif->ionic->ident; - rx_mode = IONIC_RX_MODE_F_UNICAST; rx_mode |= (netdev->flags & IFF_MULTICAST) ? IONIC_RX_MODE_F_MULTICAST : 0; rx_mode |= (netdev->flags & IFF_BROADCAST) ? IONIC_RX_MODE_F_BROADCAST : 0; @@ -1069,8 +1177,8 @@ static void ionic_set_rx_mode(struct net_device *netdev) * we remove our overflow flag and check the netdev flags * to see if we can disable NIC PROMISC */ - __dev_uc_sync(netdev, ionic_addr_add, ionic_addr_del); - nfilters = le32_to_cpu(ident->lif.eth.max_ucast_filters); + ionic_dev_uc_sync(netdev, from_ndo); + nfilters = le32_to_cpu(lif->identity->eth.max_ucast_filters); if (netdev_uc_count(netdev) + 1 > nfilters) { rx_mode |= IONIC_RX_MODE_F_PROMISC; lif->uc_overflow = true; @@ -1081,8 +1189,8 @@ static void ionic_set_rx_mode(struct net_device *netdev) } /* same for multicast */ - __dev_mc_sync(netdev, ionic_addr_add, ionic_addr_del); - nfilters = le32_to_cpu(ident->lif.eth.max_mcast_filters); + ionic_dev_uc_sync(netdev, from_ndo); + nfilters = le32_to_cpu(lif->identity->eth.max_mcast_filters); if (netdev_mc_count(netdev) > nfilters) { rx_mode |= IONIC_RX_MODE_F_ALLMULTI; lif->mc_overflow = true; @@ -1093,7 +1201,12 @@ static void ionic_set_rx_mode(struct net_device *netdev) } if (lif->rx_mode != rx_mode) - _ionic_lif_rx_mode(lif, rx_mode); + _ionic_lif_rx_mode(lif, rx_mode, from_ndo); +} + +static void ionic_ndo_set_rx_mode(struct net_device *netdev) +{ + ionic_set_rx_mode(netdev, true); } static __le64 ionic_netdev_features_to_nic(netdev_features_t features) @@ -1315,6 +1428,35 @@ static int ionic_set_mac_address(struct net_device *netdev, void *sa) return ionic_addr_add(netdev, mac); } +static void ionic_stop_queues_reconfig(struct ionic_lif *lif) +{ + /* Stop and clean the queues before reconfiguration */ + mutex_lock(&lif->queue_lock); + netif_device_detach(lif->netdev); + ionic_stop_queues(lif); + ionic_txrx_deinit(lif); +} + +static int ionic_start_queues_reconfig(struct ionic_lif *lif) +{ + int err; + + /* Re-init the queues after reconfiguration */ + + /* The only way txrx_init can fail here is if communication + * with FW is suddenly broken. There's not much we can do + * at this point - error messages have already been printed, + * so we can continue on and the user can eventually do a + * DOWN and UP to try to reset and clear the issue. + */ + err = ionic_txrx_init(lif); + mutex_unlock(&lif->queue_lock); + ionic_link_status_check_request(lif, true); + netif_device_attach(lif->netdev); + + return err; +} + static int ionic_change_mtu(struct net_device *netdev, int new_mtu) { struct ionic_lif *lif = netdev_priv(netdev); @@ -1334,9 +1476,12 @@ static int ionic_change_mtu(struct net_device *netdev, int new_mtu) return err; netdev->mtu = new_mtu; - err = ionic_reset_queues(lif, NULL, NULL); + /* if we're not running, nothing more to do */ + if (!netif_running(netdev)) + return 0; - return err; + ionic_stop_queues_reconfig(lif); + return ionic_start_queues_reconfig(lif); } static void ionic_tx_timeout_work(struct work_struct *ws) @@ -1345,9 +1490,14 @@ static void ionic_tx_timeout_work(struct work_struct *ws) netdev_info(lif->netdev, "Tx Timeout recovery\n"); - rtnl_lock(); - ionic_reset_queues(lif, NULL, NULL); - rtnl_unlock(); + /* if we were stopped before this scheduled job was launched, + * don't bother the queues as they are already stopped. + */ + if (!netif_running(lif->netdev)) + return; + + ionic_stop_queues_reconfig(lif); + ionic_start_queues_reconfig(lif); } static void ionic_tx_timeout(struct net_device *netdev, unsigned int txqueue) @@ -1478,22 +1628,16 @@ static void ionic_lif_rss_deinit(struct ionic_lif *lif) static void ionic_txrx_disable(struct ionic_lif *lif) { unsigned int i; - int err; + int err = 0; if (lif->txqcqs) { - for (i = 0; i < lif->nxqs; i++) { - err = ionic_qcq_disable(lif->txqcqs[i].qcq); - if (err == -ETIMEDOUT) - break; - } + for (i = 0; i < lif->nxqs; i++) + err = ionic_qcq_disable(lif->txqcqs[i], (err != -ETIMEDOUT)); } if (lif->rxqcqs) { - for (i = 0; i < lif->nxqs; i++) { - err = ionic_qcq_disable(lif->rxqcqs[i].qcq); - if (err == -ETIMEDOUT) - break; - } + for (i = 0; i < lif->nxqs; i++) + err = ionic_qcq_disable(lif->rxqcqs[i], (err != -ETIMEDOUT)); } } @@ -1502,18 +1646,18 @@ static void ionic_txrx_deinit(struct ionic_lif *lif) unsigned int i; if (lif->txqcqs) { - for (i = 0; i < lif->nxqs; i++) { - ionic_lif_qcq_deinit(lif, lif->txqcqs[i].qcq); - ionic_tx_flush(&lif->txqcqs[i].qcq->cq); - ionic_tx_empty(&lif->txqcqs[i].qcq->q); + for (i = 0; i < lif->nxqs && lif->txqcqs[i]; i++) { + ionic_lif_qcq_deinit(lif, lif->txqcqs[i]); + ionic_tx_flush(&lif->txqcqs[i]->cq); + ionic_tx_empty(&lif->txqcqs[i]->q); } } if (lif->rxqcqs) { - for (i = 0; i < lif->nxqs; i++) { - ionic_lif_qcq_deinit(lif, lif->rxqcqs[i].qcq); - ionic_rx_flush(&lif->rxqcqs[i].qcq->cq); - ionic_rx_empty(&lif->rxqcqs[i].qcq->q); + for (i = 0; i < lif->nxqs && lif->rxqcqs[i]; i++) { + ionic_lif_qcq_deinit(lif, lif->rxqcqs[i]); + ionic_rx_flush(&lif->rxqcqs[i]->cq); + ionic_rx_empty(&lif->rxqcqs[i]->q); } } lif->rx_mode = 0; @@ -1524,16 +1668,18 @@ static void ionic_txrx_free(struct ionic_lif *lif) unsigned int i; if (lif->txqcqs) { - for (i = 0; i < lif->nxqs; i++) { - ionic_qcq_free(lif, lif->txqcqs[i].qcq); - lif->txqcqs[i].qcq = NULL; + for (i = 0; i < lif->ionic->ntxqs_per_lif && lif->txqcqs[i]; i++) { + ionic_qcq_free(lif, lif->txqcqs[i]); + devm_kfree(lif->ionic->dev, lif->txqcqs[i]); + lif->txqcqs[i] = NULL; } } if (lif->rxqcqs) { - for (i = 0; i < lif->nxqs; i++) { - ionic_qcq_free(lif, lif->rxqcqs[i].qcq); - lif->rxqcqs[i].qcq = NULL; + for (i = 0; i < lif->ionic->nrxqs_per_lif && lif->rxqcqs[i]; i++) { + ionic_qcq_free(lif, lif->rxqcqs[i]); + devm_kfree(lif->ionic->dev, lif->rxqcqs[i]); + lif->rxqcqs[i] = NULL; } } } @@ -1561,17 +1707,19 @@ static int ionic_txrx_alloc(struct ionic_lif *lif) sizeof(struct ionic_txq_desc), sizeof(struct ionic_txq_comp), sg_desc_sz, - lif->kern_pid, &lif->txqcqs[i].qcq); + lif->kern_pid, &lif->txqcqs[i]); if (err) goto err_out; - if (flags & IONIC_QCQ_F_INTR) + if (flags & IONIC_QCQ_F_INTR) { ionic_intr_coal_init(lif->ionic->idev.intr_ctrl, - lif->txqcqs[i].qcq->intr.index, + lif->txqcqs[i]->intr.index, lif->tx_coalesce_hw); + if (test_bit(IONIC_LIF_F_TX_DIM_INTR, lif->state)) + lif->txqcqs[i]->intr.dim_coal_hw = lif->tx_coalesce_hw; + } - lif->txqcqs[i].qcq->stats = lif->txqcqs[i].stats; - ionic_debugfs_add_qcq(lif, lif->txqcqs[i].qcq); + ionic_debugfs_add_qcq(lif, lif->txqcqs[i]); } flags = IONIC_QCQ_F_RX_STATS | IONIC_QCQ_F_SG | IONIC_QCQ_F_INTR; @@ -1581,20 +1729,21 @@ static int ionic_txrx_alloc(struct ionic_lif *lif) sizeof(struct ionic_rxq_desc), sizeof(struct ionic_rxq_comp), sizeof(struct ionic_rxq_sg_desc), - lif->kern_pid, &lif->rxqcqs[i].qcq); + lif->kern_pid, &lif->rxqcqs[i]); if (err) goto err_out; ionic_intr_coal_init(lif->ionic->idev.intr_ctrl, - lif->rxqcqs[i].qcq->intr.index, + lif->rxqcqs[i]->intr.index, lif->rx_coalesce_hw); + if (test_bit(IONIC_LIF_F_RX_DIM_INTR, lif->state)) + lif->rxqcqs[i]->intr.dim_coal_hw = lif->rx_coalesce_hw; if (!test_bit(IONIC_LIF_F_SPLIT_INTR, lif->state)) - ionic_link_qcq_interrupts(lif->rxqcqs[i].qcq, - lif->txqcqs[i].qcq); + ionic_link_qcq_interrupts(lif->rxqcqs[i], + lif->txqcqs[i]); - lif->rxqcqs[i].qcq->stats = lif->rxqcqs[i].stats; - ionic_debugfs_add_qcq(lif, lif->rxqcqs[i].qcq); + ionic_debugfs_add_qcq(lif, lif->rxqcqs[i]); } return 0; @@ -1611,13 +1760,13 @@ static int ionic_txrx_init(struct ionic_lif *lif) int err; for (i = 0; i < lif->nxqs; i++) { - err = ionic_lif_txq_init(lif, lif->txqcqs[i].qcq); + err = ionic_lif_txq_init(lif, lif->txqcqs[i]); if (err) goto err_out; - err = ionic_lif_rxq_init(lif, lif->rxqcqs[i].qcq); + err = ionic_lif_rxq_init(lif, lif->rxqcqs[i]); if (err) { - ionic_lif_qcq_deinit(lif, lif->txqcqs[i].qcq); + ionic_lif_qcq_deinit(lif, lif->txqcqs[i]); goto err_out; } } @@ -1625,14 +1774,14 @@ static int ionic_txrx_init(struct ionic_lif *lif) if (lif->netdev->features & NETIF_F_RXHASH) ionic_lif_rss_init(lif); - ionic_set_rx_mode(lif->netdev); + ionic_set_rx_mode(lif->netdev, false); return 0; err_out: while (i--) { - ionic_lif_qcq_deinit(lif, lif->txqcqs[i].qcq); - ionic_lif_qcq_deinit(lif, lif->rxqcqs[i].qcq); + ionic_lif_qcq_deinit(lif, lif->txqcqs[i]); + ionic_lif_qcq_deinit(lif, lif->rxqcqs[i]); } return err; @@ -1640,18 +1789,24 @@ err_out: static int ionic_txrx_enable(struct ionic_lif *lif) { + int derr = 0; int i, err; for (i = 0; i < lif->nxqs; i++) { - ionic_rx_fill(&lif->rxqcqs[i].qcq->q); - err = ionic_qcq_enable(lif->rxqcqs[i].qcq); + if (!(lif->rxqcqs[i] && lif->txqcqs[i])) { + dev_err(lif->ionic->dev, "%s: bad qcq %d\n", __func__, i); + err = -ENXIO; + goto err_out; + } + + ionic_rx_fill(&lif->rxqcqs[i]->q); + err = ionic_qcq_enable(lif->rxqcqs[i]); if (err) goto err_out; - err = ionic_qcq_enable(lif->txqcqs[i].qcq); + err = ionic_qcq_enable(lif->txqcqs[i]); if (err) { - if (err != -ETIMEDOUT) - ionic_qcq_disable(lif->rxqcqs[i].qcq); + derr = ionic_qcq_disable(lif->rxqcqs[i], (err != -ETIMEDOUT)); goto err_out; } } @@ -1660,12 +1815,8 @@ static int ionic_txrx_enable(struct ionic_lif *lif) err_out: while (i--) { - err = ionic_qcq_disable(lif->txqcqs[i].qcq); - if (err == -ETIMEDOUT) - break; - err = ionic_qcq_disable(lif->rxqcqs[i].qcq); - if (err == -ETIMEDOUT) - break; + derr = ionic_qcq_disable(lif->txqcqs[i], (derr != -ETIMEDOUT)); + derr = ionic_qcq_disable(lif->rxqcqs[i], (derr != -ETIMEDOUT)); } return err; @@ -1688,7 +1839,7 @@ static int ionic_start_queues(struct ionic_lif *lif) return 0; } -int ionic_open(struct net_device *netdev) +static int ionic_open(struct net_device *netdev) { struct ionic_lif *lif = netdev_priv(netdev); int err; @@ -1734,7 +1885,7 @@ static void ionic_stop_queues(struct ionic_lif *lif) ionic_txrx_disable(lif); } -int ionic_stop(struct net_device *netdev) +static int ionic_stop(struct net_device *netdev) { struct ionic_lif *lif = netdev_priv(netdev); @@ -1998,7 +2149,7 @@ static const struct net_device_ops ionic_netdev_ops = { .ndo_stop = ionic_stop, .ndo_start_xmit = ionic_start_xmit, .ndo_get_stats64 = ionic_get_stats64, - .ndo_set_rx_mode = ionic_set_rx_mode, + .ndo_set_rx_mode = ionic_ndo_set_rx_mode, .ndo_set_features = ionic_set_features, .ndo_set_mac_address = ionic_set_mac_address, .ndo_validate_addr = eth_validate_addr, @@ -2016,35 +2167,227 @@ static const struct net_device_ops ionic_netdev_ops = { .ndo_get_vf_stats = ionic_get_vf_stats, }; -int ionic_reset_queues(struct ionic_lif *lif, ionic_reset_cb cb, void *arg) -{ - bool running; - int err = 0; +static void ionic_swap_queues(struct ionic_qcq *a, struct ionic_qcq *b) +{ + /* only swapping the queues, not the napi, flags, or other stuff */ + swap(a->q.num_descs, b->q.num_descs); + swap(a->q.base, b->q.base); + swap(a->q.base_pa, b->q.base_pa); + swap(a->q.info, b->q.info); + swap(a->q_base, b->q_base); + swap(a->q_base_pa, b->q_base_pa); + swap(a->q_size, b->q_size); + + swap(a->q.sg_base, b->q.sg_base); + swap(a->q.sg_base_pa, b->q.sg_base_pa); + swap(a->sg_base, b->sg_base); + swap(a->sg_base_pa, b->sg_base_pa); + swap(a->sg_size, b->sg_size); + + swap(a->cq.num_descs, b->cq.num_descs); + swap(a->cq.base, b->cq.base); + swap(a->cq.base_pa, b->cq.base_pa); + swap(a->cq.info, b->cq.info); + swap(a->cq_base, b->cq_base); + swap(a->cq_base_pa, b->cq_base_pa); + swap(a->cq_size, b->cq_size); +} + +int ionic_reconfigure_queues(struct ionic_lif *lif, + struct ionic_queue_params *qparam) +{ + struct ionic_qcq **tx_qcqs = NULL; + struct ionic_qcq **rx_qcqs = NULL; + unsigned int sg_desc_sz; + unsigned int flags; + int err = -ENOMEM; + unsigned int i; - mutex_lock(&lif->queue_lock); - running = netif_running(lif->netdev); - if (running) { - netif_device_detach(lif->netdev); - err = ionic_stop(lif->netdev); + /* allocate temporary qcq arrays to hold new queue structs */ + if (qparam->nxqs != lif->nxqs || qparam->ntxq_descs != lif->ntxq_descs) { + tx_qcqs = devm_kcalloc(lif->ionic->dev, lif->ionic->ntxqs_per_lif, + sizeof(struct ionic_qcq *), GFP_KERNEL); + if (!tx_qcqs) + goto err_out; + } + if (qparam->nxqs != lif->nxqs || qparam->nrxq_descs != lif->nrxq_descs) { + rx_qcqs = devm_kcalloc(lif->ionic->dev, lif->ionic->nrxqs_per_lif, + sizeof(struct ionic_qcq *), GFP_KERNEL); + if (!rx_qcqs) + goto err_out; + } + + /* allocate new desc_info and rings, but leave the interrupt setup + * until later so as to not mess with the still-running queues + */ + if (lif->qtype_info[IONIC_QTYPE_TXQ].version >= 1 && + lif->qtype_info[IONIC_QTYPE_TXQ].sg_desc_sz == + sizeof(struct ionic_txq_sg_desc_v1)) + sg_desc_sz = sizeof(struct ionic_txq_sg_desc_v1); + else + sg_desc_sz = sizeof(struct ionic_txq_sg_desc); + + if (tx_qcqs) { + for (i = 0; i < qparam->nxqs; i++) { + flags = lif->txqcqs[i]->flags & ~IONIC_QCQ_F_INTR; + err = ionic_qcq_alloc(lif, IONIC_QTYPE_TXQ, i, "tx", flags, + qparam->ntxq_descs, + sizeof(struct ionic_txq_desc), + sizeof(struct ionic_txq_comp), + sg_desc_sz, + lif->kern_pid, &tx_qcqs[i]); + if (err) + goto err_out; + } + } + + if (rx_qcqs) { + for (i = 0; i < qparam->nxqs; i++) { + flags = lif->rxqcqs[i]->flags & ~IONIC_QCQ_F_INTR; + err = ionic_qcq_alloc(lif, IONIC_QTYPE_RXQ, i, "rx", flags, + qparam->nrxq_descs, + sizeof(struct ionic_rxq_desc), + sizeof(struct ionic_rxq_comp), + sizeof(struct ionic_rxq_sg_desc), + lif->kern_pid, &rx_qcqs[i]); + if (err) + goto err_out; + } + } + + /* stop and clean the queues */ + ionic_stop_queues_reconfig(lif); + + if (qparam->nxqs != lif->nxqs) { + err = netif_set_real_num_tx_queues(lif->netdev, qparam->nxqs); if (err) - goto reset_out; + goto err_out_reinit_unlock; + err = netif_set_real_num_rx_queues(lif->netdev, qparam->nxqs); + if (err) { + netif_set_real_num_tx_queues(lif->netdev, lif->nxqs); + goto err_out_reinit_unlock; + } } - if (cb) - cb(lif, arg); + /* swap new desc_info and rings, keeping existing interrupt config */ + if (tx_qcqs) { + lif->ntxq_descs = qparam->ntxq_descs; + for (i = 0; i < qparam->nxqs; i++) + ionic_swap_queues(lif->txqcqs[i], tx_qcqs[i]); + } - if (running) { - err = ionic_open(lif->netdev); - netif_device_attach(lif->netdev); + if (rx_qcqs) { + lif->nrxq_descs = qparam->nrxq_descs; + for (i = 0; i < qparam->nxqs; i++) + ionic_swap_queues(lif->rxqcqs[i], rx_qcqs[i]); } -reset_out: - mutex_unlock(&lif->queue_lock); + /* if we need to change the interrupt layout, this is the time */ + if (qparam->intr_split != test_bit(IONIC_LIF_F_SPLIT_INTR, lif->state) || + qparam->nxqs != lif->nxqs) { + if (qparam->intr_split) { + set_bit(IONIC_LIF_F_SPLIT_INTR, lif->state); + } else { + clear_bit(IONIC_LIF_F_SPLIT_INTR, lif->state); + lif->tx_coalesce_usecs = lif->rx_coalesce_usecs; + lif->tx_coalesce_hw = lif->rx_coalesce_hw; + } + + /* clear existing interrupt assignments */ + for (i = 0; i < lif->ionic->ntxqs_per_lif; i++) { + ionic_qcq_intr_free(lif, lif->txqcqs[i]); + ionic_qcq_intr_free(lif, lif->rxqcqs[i]); + } + + /* re-assign the interrupts */ + for (i = 0; i < qparam->nxqs; i++) { + lif->rxqcqs[i]->flags |= IONIC_QCQ_F_INTR; + err = ionic_alloc_qcq_interrupt(lif, lif->rxqcqs[i]); + ionic_intr_coal_init(lif->ionic->idev.intr_ctrl, + lif->rxqcqs[i]->intr.index, + lif->rx_coalesce_hw); + + if (qparam->intr_split) { + lif->txqcqs[i]->flags |= IONIC_QCQ_F_INTR; + err = ionic_alloc_qcq_interrupt(lif, lif->txqcqs[i]); + ionic_intr_coal_init(lif->ionic->idev.intr_ctrl, + lif->txqcqs[i]->intr.index, + lif->tx_coalesce_hw); + if (test_bit(IONIC_LIF_F_TX_DIM_INTR, lif->state)) + lif->txqcqs[i]->intr.dim_coal_hw = lif->tx_coalesce_hw; + } else { + lif->txqcqs[i]->flags &= ~IONIC_QCQ_F_INTR; + ionic_link_qcq_interrupts(lif->rxqcqs[i], lif->txqcqs[i]); + } + } + } + + /* now we can rework the debugfs mappings */ + if (tx_qcqs) { + for (i = 0; i < qparam->nxqs; i++) { + ionic_debugfs_del_qcq(lif->txqcqs[i]); + ionic_debugfs_add_qcq(lif, lif->txqcqs[i]); + } + } + + if (rx_qcqs) { + for (i = 0; i < qparam->nxqs; i++) { + ionic_debugfs_del_qcq(lif->rxqcqs[i]); + ionic_debugfs_add_qcq(lif, lif->rxqcqs[i]); + } + } + + swap(lif->nxqs, qparam->nxqs); + +err_out_reinit_unlock: + /* re-init the queues, but don't loose an error code */ + if (err) + ionic_start_queues_reconfig(lif); + else + err = ionic_start_queues_reconfig(lif); + +err_out: + /* free old allocs without cleaning intr */ + for (i = 0; i < qparam->nxqs; i++) { + if (tx_qcqs && tx_qcqs[i]) { + tx_qcqs[i]->flags &= ~IONIC_QCQ_F_INTR; + ionic_qcq_free(lif, tx_qcqs[i]); + devm_kfree(lif->ionic->dev, tx_qcqs[i]); + tx_qcqs[i] = NULL; + } + if (rx_qcqs && rx_qcqs[i]) { + rx_qcqs[i]->flags &= ~IONIC_QCQ_F_INTR; + ionic_qcq_free(lif, rx_qcqs[i]); + devm_kfree(lif->ionic->dev, rx_qcqs[i]); + rx_qcqs[i] = NULL; + } + } + + /* free q array */ + if (rx_qcqs) { + devm_kfree(lif->ionic->dev, rx_qcqs); + rx_qcqs = NULL; + } + if (tx_qcqs) { + devm_kfree(lif->ionic->dev, tx_qcqs); + tx_qcqs = NULL; + } + + /* clean the unused dma and info allocations when new set is smaller + * than the full array, but leave the qcq shells in place + */ + for (i = lif->nxqs; i < lif->ionic->ntxqs_per_lif; i++) { + lif->txqcqs[i]->flags &= ~IONIC_QCQ_F_INTR; + ionic_qcq_free(lif, lif->txqcqs[i]); + + lif->rxqcqs[i]->flags &= ~IONIC_QCQ_F_INTR; + ionic_qcq_free(lif, lif->rxqcqs[i]); + } return err; } -static struct ionic_lif *ionic_lif_alloc(struct ionic *ionic, unsigned int index) +int ionic_lif_alloc(struct ionic *ionic) { struct device *dev = ionic->dev; union ionic_lif_identity *lid; @@ -2055,7 +2398,7 @@ static struct ionic_lif *ionic_lif_alloc(struct ionic *ionic, unsigned int index lid = kzalloc(sizeof(*lid), GFP_KERNEL); if (!lid) - return ERR_PTR(-ENOMEM); + return -ENOMEM; netdev = alloc_etherdev_mqs(sizeof(*lif), ionic->ntxqs_per_lif, ionic->ntxqs_per_lif); @@ -2069,7 +2412,7 @@ static struct ionic_lif *ionic_lif_alloc(struct ionic *ionic, unsigned int index lif = netdev_priv(netdev); lif->netdev = netdev; - ionic->master_lif = lif; + ionic->lif = lif; netdev->netdev_ops = &ionic_netdev_ops; ionic_ethtool_set_ops(netdev); @@ -2078,8 +2421,14 @@ static struct ionic_lif *ionic_lif_alloc(struct ionic *ionic, unsigned int index lif->identity = lid; lif->lif_type = IONIC_LIF_TYPE_CLASSIC; - ionic_lif_identify(ionic, lif->lif_type, lif->identity); - lif->netdev->min_mtu = le32_to_cpu(lif->identity->eth.min_frame_size); + err = ionic_lif_identify(ionic, lif->lif_type, lif->identity); + if (err) { + dev_err(ionic->dev, "Cannot identify type %d: %d\n", + lif->lif_type, err); + goto err_out_free_netdev; + } + lif->netdev->min_mtu = max_t(unsigned int, ETH_MIN_MTU, + le32_to_cpu(lif->identity->eth.min_frame_size)); lif->netdev->max_mtu = le32_to_cpu(lif->identity->eth.max_frame_size) - ETH_HLEN - VLAN_HLEN; @@ -2087,7 +2436,7 @@ static struct ionic_lif *ionic_lif_alloc(struct ionic *ionic, unsigned int index lif->nxqs = ionic->ntxqs_per_lif; lif->ionic = ionic; - lif->index = index; + lif->index = 0; lif->ntxq_descs = IONIC_DEF_TXRX_DESC; lif->nrxq_descs = IONIC_DEF_TXRX_DESC; lif->tx_budget = IONIC_TX_BUDGET_DEFAULT; @@ -2098,8 +2447,10 @@ static struct ionic_lif *ionic_lif_alloc(struct ionic *ionic, unsigned int index lif->rx_coalesce_usecs); lif->tx_coalesce_usecs = lif->rx_coalesce_usecs; lif->tx_coalesce_hw = lif->rx_coalesce_hw; + set_bit(IONIC_LIF_F_RX_DIM_INTR, lif->state); + set_bit(IONIC_LIF_F_TX_DIM_INTR, lif->state); - snprintf(lif->name, sizeof(lif->name), "lif%u", index); + snprintf(lif->name, sizeof(lif->name), "lif%u", lif->index); spin_lock_init(&lif->adminq_lock); @@ -2119,7 +2470,8 @@ static struct ionic_lif *ionic_lif_alloc(struct ionic *ionic, unsigned int index ionic_debugfs_add_lif(lif); - /* allocate queues */ + /* allocate control queues and txrx queue arrays */ + ionic_lif_queue_identify(lif); err = ionic_qcqs_alloc(lif); if (err) goto err_out_free_lif_info; @@ -2138,9 +2490,7 @@ static struct ionic_lif *ionic_lif_alloc(struct ionic *ionic, unsigned int index } netdev_rss_key_fill(lif->rss_hash_key, IONIC_RSS_HASH_KEY_SIZE); - list_add_tail(&lif->list, &ionic->lifs); - - return lif; + return 0; err_out_free_qcqs: ionic_qcqs_free(lif); @@ -2154,27 +2504,7 @@ err_out_free_netdev: err_out_free_lid: kfree(lid); - return ERR_PTR(err); -} - -int ionic_lifs_alloc(struct ionic *ionic) -{ - struct ionic_lif *lif; - - INIT_LIST_HEAD(&ionic->lifs); - - /* only build the first lif, others are for later features */ - set_bit(0, ionic->lifbits); - - lif = ionic_lif_alloc(ionic, 0); - if (IS_ERR_OR_NULL(lif)) { - clear_bit(0, ionic->lifbits); - return -ENOMEM; - } - - ionic_lif_queue_identify(lif); - - return 0; + return err; } static void ionic_lif_reset(struct ionic_lif *lif) @@ -2209,7 +2539,7 @@ static void ionic_lif_handle_fw_down(struct ionic_lif *lif) ionic_txrx_deinit(lif); ionic_txrx_free(lif); } - ionic_lifs_deinit(ionic); + ionic_lif_deinit(lif); ionic_reset(ionic); ionic_qcqs_free(lif); @@ -2227,12 +2557,20 @@ static void ionic_lif_handle_fw_up(struct ionic_lif *lif) dev_info(ionic->dev, "FW Up: restarting LIFs\n"); ionic_init_devinfo(ionic); - ionic_port_init(ionic); + err = ionic_identify(ionic); + if (err) + goto err_out; + err = ionic_port_identify(ionic); + if (err) + goto err_out; + err = ionic_port_init(ionic); + if (err) + goto err_out; err = ionic_qcqs_alloc(lif); if (err) goto err_out; - err = ionic_lifs_init(ionic); + err = ionic_lif_init(lif); if (err) goto err_qcqs_free; @@ -2252,7 +2590,7 @@ static void ionic_lif_handle_fw_up(struct ionic_lif *lif) } clear_bit(IONIC_LIF_F_FW_RESET, lif->state); - ionic_link_status_check_request(lif); + ionic_link_status_check_request(lif, true); netif_device_attach(lif->netdev); dev_info(ionic->dev, "FW Up: LIFs restarted\n"); @@ -2261,14 +2599,14 @@ static void ionic_lif_handle_fw_up(struct ionic_lif *lif) err_txrx_free: ionic_txrx_free(lif); err_lifs_deinit: - ionic_lifs_deinit(ionic); + ionic_lif_deinit(lif); err_qcqs_free: ionic_qcqs_free(lif); err_out: dev_err(ionic->dev, "FW Up: LIFs restart failed - err %d\n", err); } -static void ionic_lif_free(struct ionic_lif *lif) +void ionic_lif_free(struct ionic_lif *lif) { struct device *dev = lif->ionic->dev; @@ -2297,23 +2635,10 @@ static void ionic_lif_free(struct ionic_lif *lif) /* free netdev & lif */ ionic_debugfs_del_lif(lif); - list_del(&lif->list); free_netdev(lif->netdev); } -void ionic_lifs_free(struct ionic *ionic) -{ - struct list_head *cur, *tmp; - struct ionic_lif *lif; - - list_for_each_safe(cur, tmp, &ionic->lifs) { - lif = list_entry(cur, struct ionic_lif, list); - - ionic_lif_free(lif); - } -} - -static void ionic_lif_deinit(struct ionic_lif *lif) +void ionic_lif_deinit(struct ionic_lif *lif) { if (!test_and_clear_bit(IONIC_LIF_F_INITED, lif->state)) return; @@ -2334,17 +2659,6 @@ static void ionic_lif_deinit(struct ionic_lif *lif) ionic_lif_reset(lif); } -void ionic_lifs_deinit(struct ionic *ionic) -{ - struct list_head *cur, *tmp; - struct ionic_lif *lif; - - list_for_each_safe(cur, tmp, &ionic->lifs) { - lif = list_entry(cur, struct ionic_lif, list); - ionic_lif_deinit(lif); - } -} - static int ionic_lif_adminq_init(struct ionic_lif *lif) { struct device *dev = lif->ionic->dev; @@ -2468,7 +2782,7 @@ static int ionic_station_set(struct ionic_lif *lif) */ if (!ether_addr_equal(ctx.comp.lif_getattr.mac, netdev->dev_addr)) - ionic_lif_addr(lif, netdev->dev_addr, true); + ionic_lif_addr(lif, netdev->dev_addr, true, true); } else { /* Update the netdev mac with the device's mac */ memcpy(addr.sa_data, ctx.comp.lif_getattr.mac, netdev->addr_len); @@ -2485,12 +2799,12 @@ static int ionic_station_set(struct ionic_lif *lif) netdev_dbg(lif->netdev, "adding station MAC addr %pM\n", netdev->dev_addr); - ionic_lif_addr(lif, netdev->dev_addr, true); + ionic_lif_addr(lif, netdev->dev_addr, true, true); return 0; } -static int ionic_lif_init(struct ionic_lif *lif) +int ionic_lif_init(struct ionic_lif *lif) { struct ionic_dev *idev = &lif->ionic->idev; struct device *dev = lif->ionic->dev; @@ -2580,22 +2894,6 @@ err_out_free_dbid: return err; } -int ionic_lifs_init(struct ionic *ionic) -{ - struct list_head *cur, *tmp; - struct ionic_lif *lif; - int err; - - list_for_each_safe(cur, tmp, &ionic->lifs) { - lif = list_entry(cur, struct ionic_lif, list); - err = ionic_lif_init(lif); - if (err) - return err; - } - - return 0; -} - static void ionic_lif_notify_work(struct work_struct *ws) { } @@ -2644,45 +2942,41 @@ static int ionic_lif_notify(struct notifier_block *nb, return NOTIFY_DONE; } -int ionic_lifs_register(struct ionic *ionic) +int ionic_lif_register(struct ionic_lif *lif) { int err; - INIT_WORK(&ionic->nb_work, ionic_lif_notify_work); + INIT_WORK(&lif->ionic->nb_work, ionic_lif_notify_work); - ionic->nb.notifier_call = ionic_lif_notify; + lif->ionic->nb.notifier_call = ionic_lif_notify; - err = register_netdevice_notifier(&ionic->nb); + err = register_netdevice_notifier(&lif->ionic->nb); if (err) - ionic->nb.notifier_call = NULL; + lif->ionic->nb.notifier_call = NULL; /* only register LIF0 for now */ - err = register_netdev(ionic->master_lif->netdev); + err = register_netdev(lif->netdev); if (err) { - dev_err(ionic->dev, "Cannot register net device, aborting\n"); + dev_err(lif->ionic->dev, "Cannot register net device, aborting\n"); return err; } - ionic->master_lif->registered = true; - ionic_lif_set_netdev_info(ionic->master_lif); + lif->registered = true; + ionic_lif_set_netdev_info(lif); return 0; } -void ionic_lifs_unregister(struct ionic *ionic) +void ionic_lif_unregister(struct ionic_lif *lif) { - if (ionic->nb.notifier_call) { - unregister_netdevice_notifier(&ionic->nb); - cancel_work_sync(&ionic->nb_work); - ionic->nb.notifier_call = NULL; + if (lif->ionic->nb.notifier_call) { + unregister_netdevice_notifier(&lif->ionic->nb); + cancel_work_sync(&lif->ionic->nb_work); + lif->ionic->nb.notifier_call = NULL; } - /* There is only one lif ever registered in the - * current model, so don't bother searching the - * ionic->lif for candidates to unregister - */ - if (ionic->master_lif && - ionic->master_lif->netdev->reg_state == NETREG_REGISTERED) - unregister_netdev(ionic->master_lif->netdev); + if (lif->netdev->reg_state == NETREG_REGISTERED) + unregister_netdev(lif->netdev); + lif->registered = false; } static void ionic_lif_queue_identify(struct ionic_lif *lif) @@ -2801,7 +3095,7 @@ int ionic_lif_identify(struct ionic *ionic, u8 lif_type, return 0; } -int ionic_lifs_size(struct ionic *ionic) +int ionic_lif_size(struct ionic *ionic) { struct ionic_identity *ident = &ionic->ident; unsigned int nintrs, dev_nintrs; diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.h b/drivers/net/ethernet/pensando/ionic/ionic_lif.h index 1ee3b14c8d50..0224dfd24b8a 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.h @@ -4,6 +4,7 @@ #ifndef _IONIC_LIF_H_ #define _IONIC_LIF_H_ +#include <linux/dim.h> #include <linux/pci.h> #include "ionic_rx_filter.h" @@ -16,32 +17,32 @@ #define IONIC_TX_BUDGET_DEFAULT 256 struct ionic_tx_stats { - u64 dma_map_err; u64 pkts; u64 bytes; - u64 clean; - u64 linearize; u64 csum_none; u64 csum; - u64 crc32_csum; u64 tso; u64 tso_bytes; u64 frags; u64 vlan_inserted; + u64 clean; + u64 linearize; + u64 crc32_csum; u64 sg_cntr[IONIC_MAX_NUM_SG_CNTR]; + u64 dma_map_err; }; struct ionic_rx_stats { - u64 dma_map_err; - u64 alloc_err; u64 pkts; u64 bytes; u64 csum_none; u64 csum_complete; - u64 csum_error; u64 buffers_posted; u64 dropped; u64 vlan_stripped; + u64 csum_error; + u64 dma_map_err; + u64 alloc_err; }; #define IONIC_QCQ_F_INITED BIT(0) @@ -56,35 +57,29 @@ struct ionic_napi_stats { u64 work_done_cntr[IONIC_MAX_NUM_NAPI_CNTR]; }; -struct ionic_q_stats { - union { - struct ionic_tx_stats tx; - struct ionic_rx_stats rx; - }; -}; - struct ionic_qcq { - void *base; - dma_addr_t base_pa; - unsigned int total_size; + void *q_base; + dma_addr_t q_base_pa; + u32 q_size; + void *cq_base; + dma_addr_t cq_base_pa; + u32 cq_size; + void *sg_base; + dma_addr_t sg_base_pa; + u32 sg_size; + struct dim dim; struct ionic_queue q; struct ionic_cq cq; struct ionic_intr_info intr; struct napi_struct napi; struct ionic_napi_stats napi_stats; - struct ionic_q_stats *stats; unsigned int flags; struct dentry *dentry; }; -struct ionic_qcqst { - struct ionic_qcq *qcq; - struct ionic_q_stats *stats; -}; - #define q_to_qcq(q) container_of(q, struct ionic_qcq, q) -#define q_to_tx_stats(q) (&q_to_qcq(q)->stats->tx) -#define q_to_rx_stats(q) (&q_to_qcq(q)->stats->rx) +#define q_to_tx_stats(q) (&(q)->lif->txqstats[(q)->index]) +#define q_to_rx_stats(q) (&(q)->lif->rxqstats[(q)->index]) #define napi_to_qcq(napi) container_of(napi, struct ionic_qcq, napi) #define napi_to_cq(napi) (&napi_to_qcq(napi)->cq) @@ -138,6 +133,8 @@ enum ionic_lif_state_flags { IONIC_LIF_F_LINK_CHECK_REQUESTED, IONIC_LIF_F_FW_RESET, IONIC_LIF_F_SPLIT_INTR, + IONIC_LIF_F_TX_DIM_INTR, + IONIC_LIF_F_RX_DIM_INTR, /* leave this as last */ IONIC_LIF_F_STATE_SIZE @@ -170,8 +167,10 @@ struct ionic_lif { spinlock_t adminq_lock; /* lock for AdminQ operations */ struct ionic_qcq *adminqcq; struct ionic_qcq *notifyqcq; - struct ionic_qcqst *txqcqs; - struct ionic_qcqst *rxqcqs; + struct ionic_qcq **txqcqs; + struct ionic_tx_stats *txqstats; + struct ionic_qcq **rxqcqs; + struct ionic_rx_stats *rxqstats; u64 last_eid; unsigned int neqs; unsigned int nxqs; @@ -212,12 +211,21 @@ struct ionic_lif { struct work_struct tx_timeout_work; }; -#define lif_to_txqcq(lif, i) ((lif)->txqcqs[i].qcq) -#define lif_to_rxqcq(lif, i) ((lif)->rxqcqs[i].qcq) -#define lif_to_txstats(lif, i) ((lif)->txqcqs[i].stats->tx) -#define lif_to_rxstats(lif, i) ((lif)->rxqcqs[i].stats->rx) -#define lif_to_txq(lif, i) (&lif_to_txqcq((lif), i)->q) -#define lif_to_rxq(lif, i) (&lif_to_txqcq((lif), i)->q) +struct ionic_queue_params { + unsigned int nxqs; + unsigned int ntxq_descs; + unsigned int nrxq_descs; + unsigned int intr_split; +}; + +static inline void ionic_init_queue_params(struct ionic_lif *lif, + struct ionic_queue_params *qparam) +{ + qparam->nxqs = lif->nxqs; + qparam->ntxq_descs = lif->ntxq_descs; + qparam->nrxq_descs = lif->nrxq_descs; + qparam->intr_split = test_bit(IONIC_LIF_F_SPLIT_INTR, lif->state); +} static inline u32 ionic_coal_usec_to_hw(struct ionic *ionic, u32 usecs) { @@ -237,39 +245,38 @@ static inline u32 ionic_coal_usec_to_hw(struct ionic *ionic, u32 usecs) typedef void (*ionic_reset_cb)(struct ionic_lif *lif, void *arg); -void ionic_link_status_check_request(struct ionic_lif *lif); +void ionic_link_status_check_request(struct ionic_lif *lif, bool can_sleep); void ionic_get_stats64(struct net_device *netdev, struct rtnl_link_stats64 *ns); void ionic_lif_deferred_enqueue(struct ionic_deferred *def, struct ionic_deferred_work *work); -int ionic_lifs_alloc(struct ionic *ionic); -void ionic_lifs_free(struct ionic *ionic); -void ionic_lifs_deinit(struct ionic *ionic); -int ionic_lifs_init(struct ionic *ionic); -int ionic_lifs_register(struct ionic *ionic); -void ionic_lifs_unregister(struct ionic *ionic); +int ionic_lif_alloc(struct ionic *ionic); +int ionic_lif_init(struct ionic_lif *lif); +void ionic_lif_free(struct ionic_lif *lif); +void ionic_lif_deinit(struct ionic_lif *lif); +int ionic_lif_register(struct ionic_lif *lif); +void ionic_lif_unregister(struct ionic_lif *lif); int ionic_lif_identify(struct ionic *ionic, u8 lif_type, union ionic_lif_identity *lif_ident); -int ionic_lifs_size(struct ionic *ionic); +int ionic_lif_size(struct ionic *ionic); int ionic_lif_rss_config(struct ionic_lif *lif, u16 types, const u8 *key, const u32 *indir); +int ionic_reconfigure_queues(struct ionic_lif *lif, + struct ionic_queue_params *qparam); -int ionic_open(struct net_device *netdev); -int ionic_stop(struct net_device *netdev); -int ionic_reset_queues(struct ionic_lif *lif, ionic_reset_cb cb, void *arg); - -static inline void debug_stats_txq_post(struct ionic_qcq *qcq, - struct ionic_txq_desc *desc, bool dbell) +static inline void debug_stats_txq_post(struct ionic_queue *q, bool dbell) { - u8 num_sg_elems = ((le64_to_cpu(desc->cmd) >> IONIC_TXQ_DESC_NSGE_SHIFT) - & IONIC_TXQ_DESC_NSGE_MASK); + struct ionic_txq_desc *desc = &q->txq[q->head_idx]; + u8 num_sg_elems; - qcq->q.dbell_count += dbell; + q->dbell_count += dbell; + num_sg_elems = ((le64_to_cpu(desc->cmd) >> IONIC_TXQ_DESC_NSGE_SHIFT) + & IONIC_TXQ_DESC_NSGE_MASK); if (num_sg_elems > (IONIC_MAX_NUM_SG_CNTR - 1)) num_sg_elems = IONIC_MAX_NUM_SG_CNTR - 1; - qcq->stats->tx.sg_cntr[num_sg_elems]++; + q->lif->txqstats[q->index].sg_cntr[num_sg_elems]++; } static inline void debug_stats_napi_poll(struct ionic_qcq *qcq, @@ -284,10 +291,8 @@ static inline void debug_stats_napi_poll(struct ionic_qcq *qcq, } #define DEBUG_STATS_CQE_CNT(cq) ((cq)->compl_count++) -#define DEBUG_STATS_RX_BUFF_CNT(qcq) ((qcq)->stats->rx.buffers_posted++) -#define DEBUG_STATS_INTR_REARM(intr) ((intr)->rearm_count++) -#define DEBUG_STATS_TXQ_POST(qcq, txdesc, dbell) \ - debug_stats_txq_post(qcq, txdesc, dbell) +#define DEBUG_STATS_RX_BUFF_CNT(q) ((q)->lif->rxqstats[q->index].buffers_posted++) +#define DEBUG_STATS_TXQ_POST(q, dbell) debug_stats_txq_post(q, dbell) #define DEBUG_STATS_NAPI_POLL(qcq, work_done) \ debug_stats_napi_poll(qcq, work_done) diff --git a/drivers/net/ethernet/pensando/ionic/ionic_main.c b/drivers/net/ethernet/pensando/ionic/ionic_main.c index df5b9bcc3aba..ee0740881af3 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_main.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_main.c @@ -64,6 +64,8 @@ static const char *ionic_error_to_str(enum ionic_status_code code) return "IONIC_RC_ERROR"; case IONIC_RC_ERDMA: return "IONIC_RC_ERDMA"; + case IONIC_RC_EBAD_FW: + return "IONIC_RC_EBAD_FW"; default: return "IONIC_RC_UNKNOWN"; } @@ -170,6 +172,10 @@ static const char *ionic_opcode_to_str(enum ionic_cmd_opcode opcode) return "IONIC_CMD_FW_DOWNLOAD"; case IONIC_CMD_FW_CONTROL: return "IONIC_CMD_FW_CONTROL"; + case IONIC_CMD_FW_DOWNLOAD_V1: + return "IONIC_CMD_FW_DOWNLOAD_V1"; + case IONIC_CMD_FW_CONTROL_V1: + return "IONIC_CMD_FW_CONTROL_V1"; case IONIC_CMD_VF_GETATTR: return "IONIC_CMD_VF_GETATTR"; case IONIC_CMD_VF_SETATTR: @@ -181,15 +187,17 @@ static const char *ionic_opcode_to_str(enum ionic_cmd_opcode opcode) static void ionic_adminq_flush(struct ionic_lif *lif) { - struct ionic_queue *adminq = &lif->adminqcq->q; + struct ionic_queue *q = &lif->adminqcq->q; + struct ionic_desc_info *desc_info; spin_lock(&lif->adminq_lock); - while (adminq->tail != adminq->head) { - memset(adminq->tail->desc, 0, sizeof(union ionic_adminq_cmd)); - adminq->tail->cb = NULL; - adminq->tail->cb_arg = NULL; - adminq->tail = adminq->tail->next; + while (q->tail_idx != q->head_idx) { + desc_info = &q->info[q->tail_idx]; + memset(desc_info->desc, 0, sizeof(union ionic_adminq_cmd)); + desc_info->cb = NULL; + desc_info->cb_arg = NULL; + q->tail_idx = (q->tail_idx + 1) & (q->num_descs - 1); } spin_unlock(&lif->adminq_lock); } @@ -245,18 +253,17 @@ static void ionic_adminq_cb(struct ionic_queue *q, static int ionic_adminq_post(struct ionic_lif *lif, struct ionic_admin_ctx *ctx) { - struct ionic_queue *adminq; + struct ionic_desc_info *desc_info; + struct ionic_queue *q; int err = 0; - WARN_ON(in_interrupt()); - if (!lif->adminqcq) return -EIO; - adminq = &lif->adminqcq->q; + q = &lif->adminqcq->q; spin_lock(&lif->adminq_lock); - if (!ionic_q_has_space(adminq, 1)) { + if (!ionic_q_has_space(q, 1)) { err = -ENOSPC; goto err_out; } @@ -265,13 +272,14 @@ static int ionic_adminq_post(struct ionic_lif *lif, struct ionic_admin_ctx *ctx) if (err) goto err_out; - memcpy(adminq->head->desc, &ctx->cmd, sizeof(ctx->cmd)); + desc_info = &q->info[q->head_idx]; + memcpy(desc_info->desc, &ctx->cmd, sizeof(ctx->cmd)); dev_dbg(&lif->netdev->dev, "post admin queue command:\n"); dynamic_hex_dump("cmd ", DUMP_PREFIX_OFFSET, 16, 1, &ctx->cmd, sizeof(ctx->cmd), true); - ionic_q_post(adminq, true, ionic_adminq_cb, ctx); + ionic_q_post(q, true, ionic_adminq_cb, ctx); err_out: spin_unlock(&lif->adminq_lock); @@ -301,32 +309,6 @@ int ionic_adminq_post_wait(struct ionic_lif *lif, struct ionic_admin_ctx *ctx) return ionic_adminq_check_err(lif, ctx, (remaining == 0)); } -int ionic_napi(struct napi_struct *napi, int budget, ionic_cq_cb cb, - ionic_cq_done_cb done_cb, void *done_arg) -{ - struct ionic_qcq *qcq = napi_to_qcq(napi); - struct ionic_cq *cq = &qcq->cq; - u32 work_done, flags = 0; - - work_done = ionic_cq_service(cq, budget, cb, done_cb, done_arg); - - if (work_done < budget && napi_complete_done(napi, work_done)) { - flags |= IONIC_INTR_CRED_UNMASK; - DEBUG_STATS_INTR_REARM(cq->bound_intr); - } - - if (work_done || flags) { - flags |= IONIC_INTR_CRED_RESET_COALESCE; - ionic_intr_credits(cq->lif->ionic->idev.intr_ctrl, - cq->bound_intr->index, - work_done, flags); - } - - DEBUG_STATS_NAPI_POLL(qcq, work_done); - - return work_done; -} - static void ionic_dev_cmd_clean(struct ionic *ionic) { union ionic_dev_cmd_regs *regs = ionic->idev.dev_cmd_regs; @@ -346,24 +328,27 @@ int ionic_dev_cmd_wait(struct ionic *ionic, unsigned long max_seconds) int done; int err; - WARN_ON(in_interrupt()); - /* Wait for dev cmd to complete, retrying if we get EAGAIN, * but don't wait any longer than max_seconds. */ max_wait = jiffies + (max_seconds * HZ); try_again: + opcode = idev->dev_cmd_regs->cmd.cmd.opcode; start_time = jiffies; do { done = ionic_dev_cmd_done(idev); if (done) break; - msleep(5); - hb = ionic_heartbeat_check(ionic); + usleep_range(100, 200); + + /* Don't check the heartbeat on FW_CONTROL commands as they are + * notorious for interrupting the firmware's heartbeat update. + */ + if (opcode != IONIC_CMD_FW_CONTROL) + hb = ionic_heartbeat_check(ionic); } while (!done && !hb && time_before(jiffies, max_wait)); duration = jiffies - start_time; - opcode = idev->dev_cmd_regs->cmd.cmd.opcode; dev_dbg(ionic->dev, "DEVCMD %s (%d) done=%d took %ld secs (%ld jiffies)\n", ionic_opcode_to_str(opcode), opcode, done, duration / HZ, duration); @@ -387,8 +372,9 @@ try_again: err = ionic_dev_cmd_status(&ionic->idev); if (err) { - if (err == IONIC_RC_EAGAIN && !time_after(jiffies, max_wait)) { - dev_err(ionic->dev, "DEV_CMD %s (%d) error, %s (%d) retrying...\n", + if (err == IONIC_RC_EAGAIN && + time_before(jiffies, (max_wait - HZ))) { + dev_dbg(ionic->dev, "DEV_CMD %s (%d), %s (%d) retrying...\n", ionic_opcode_to_str(opcode), opcode, ionic_error_to_str(err), err); @@ -398,9 +384,10 @@ try_again: goto try_again; } - dev_err(ionic->dev, "DEV_CMD %s (%d) error, %s (%d) failed\n", - ionic_opcode_to_str(opcode), opcode, - ionic_error_to_str(err), err); + if (!(opcode == IONIC_CMD_FW_CONTROL && err == IONIC_RC_EAGAIN)) + dev_err(ionic->dev, "DEV_CMD %s (%d) error, %s (%d) failed\n", + ionic_opcode_to_str(opcode), opcode, + ionic_error_to_str(err), err); return ionic_error_to_errno(err); } @@ -444,17 +431,23 @@ int ionic_identify(struct ionic *ionic) sz = min(sizeof(ident->dev), sizeof(idev->dev_cmd_regs->data)); memcpy_fromio(&ident->dev, &idev->dev_cmd_regs->data, sz); } - mutex_unlock(&ionic->dev_cmd_lock); - if (err) - goto err_out_unmap; + if (err) { + dev_err(ionic->dev, "Cannot identify ionic: %dn", err); + goto err_out; + } - ionic_debugfs_add_ident(ionic); + err = ionic_lif_identify(ionic, IONIC_LIF_TYPE_CLASSIC, + &ionic->ident.lif); + if (err) { + dev_err(ionic->dev, "Cannot identify LIFs: %d\n", err); + goto err_out; + } return 0; -err_out_unmap: +err_out: return err; } diff --git a/drivers/net/ethernet/pensando/ionic/ionic_stats.c b/drivers/net/ethernet/pensando/ionic/ionic_stats.c index 2a1885da58a6..ff20a2ac4c2f 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_stats.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_stats.c @@ -179,36 +179,28 @@ static const struct ionic_stat_desc ionic_dbg_napi_stats_desc[] = { static void ionic_get_lif_stats(struct ionic_lif *lif, struct ionic_lif_sw_stats *stats) { - struct ionic_tx_stats *tstats; - struct ionic_rx_stats *rstats; + struct ionic_tx_stats *txstats; + struct ionic_rx_stats *rxstats; struct rtnl_link_stats64 ns; - struct ionic_qcq *txqcq; - struct ionic_qcq *rxqcq; int q_num; memset(stats, 0, sizeof(*stats)); for (q_num = 0; q_num < MAX_Q(lif); q_num++) { - txqcq = lif_to_txqcq(lif, q_num); - if (txqcq && txqcq->stats) { - tstats = &txqcq->stats->tx; - stats->tx_packets += tstats->pkts; - stats->tx_bytes += tstats->bytes; - stats->tx_tso += tstats->tso; - stats->tx_tso_bytes += tstats->tso_bytes; - stats->tx_csum_none += tstats->csum_none; - stats->tx_csum += tstats->csum; - } - - rxqcq = lif_to_rxqcq(lif, q_num); - if (rxqcq && rxqcq->stats) { - rstats = &rxqcq->stats->rx; - stats->rx_packets += rstats->pkts; - stats->rx_bytes += rstats->bytes; - stats->rx_csum_none += rstats->csum_none; - stats->rx_csum_complete += rstats->csum_complete; - stats->rx_csum_error += rstats->csum_error; - } + txstats = &lif->txqstats[q_num]; + stats->tx_packets += txstats->pkts; + stats->tx_bytes += txstats->bytes; + stats->tx_tso += txstats->tso; + stats->tx_tso_bytes += txstats->tso_bytes; + stats->tx_csum_none += txstats->csum_none; + stats->tx_csum += txstats->csum; + + rxstats = &lif->rxqstats[q_num]; + stats->rx_packets += rxstats->pkts; + stats->rx_bytes += rxstats->bytes; + stats->rx_csum_none += rxstats->csum_none; + stats->rx_csum_complete += rxstats->csum_complete; + stats->rx_csum_error += rxstats->csum_error; } ionic_get_stats64(lif->netdev, &ns); @@ -371,7 +363,7 @@ static void ionic_sw_stats_get_values(struct ionic_lif *lif, u64 **buf) } for (q_num = 0; q_num < MAX_Q(lif); q_num++) { - txstats = &lif_to_txstats(lif, q_num); + txstats = &lif->txqstats[q_num]; for (i = 0; i < IONIC_NUM_TX_STATS; i++) { **buf = IONIC_READ_STAT64(txstats, @@ -381,7 +373,7 @@ static void ionic_sw_stats_get_values(struct ionic_lif *lif, u64 **buf) if (test_bit(IONIC_LIF_F_UP, lif->state) && test_bit(IONIC_LIF_F_SW_DEBUG_STATS, lif->state)) { - txqcq = lif_to_txqcq(lif, q_num); + txqcq = lif->txqcqs[q_num]; for (i = 0; i < IONIC_NUM_TX_Q_STATS; i++) { **buf = IONIC_READ_STAT64(&txqcq->q, &ionic_txq_stats_desc[i]); @@ -405,7 +397,7 @@ static void ionic_sw_stats_get_values(struct ionic_lif *lif, u64 **buf) } for (q_num = 0; q_num < MAX_Q(lif); q_num++) { - rxstats = &lif_to_rxstats(lif, q_num); + rxstats = &lif->rxqstats[q_num]; for (i = 0; i < IONIC_NUM_RX_STATS; i++) { **buf = IONIC_READ_STAT64(rxstats, @@ -415,7 +407,7 @@ static void ionic_sw_stats_get_values(struct ionic_lif *lif, u64 **buf) if (test_bit(IONIC_LIF_F_UP, lif->state) && test_bit(IONIC_LIF_F_SW_DEBUG_STATS, lif->state)) { - rxqcq = lif_to_rxqcq(lif, q_num); + rxqcq = lif->rxqcqs[q_num]; for (i = 0; i < IONIC_NUM_DBG_CQ_STATS; i++) { **buf = IONIC_READ_STAT64(&rxqcq->cq, &ionic_dbg_cq_stats_desc[i]); diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c index def65fee27b5..169ac4f54640 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c @@ -22,7 +22,7 @@ static bool ionic_tx_service(struct ionic_cq *cq, struct ionic_cq_info *cq_info) static inline void ionic_txq_post(struct ionic_queue *q, bool ring_dbell, ionic_desc_cb cb_func, void *cb_arg) { - DEBUG_STATS_TXQ_POST(q_to_qcq(q), q->head->desc, ring_dbell); + DEBUG_STATS_TXQ_POST(q, ring_dbell); ionic_q_post(q, ring_dbell, cb_func, cb_arg); } @@ -32,7 +32,7 @@ static inline void ionic_rxq_post(struct ionic_queue *q, bool ring_dbell, { ionic_q_post(q, ring_dbell, cb_func, cb_arg); - DEBUG_STATS_RX_BUFF_CNT(q_to_qcq(q)); + DEBUG_STATS_RX_BUFF_CNT(q); } static inline struct netdev_queue *q_to_ndq(struct ionic_queue *q) @@ -49,7 +49,7 @@ static struct sk_buff *ionic_rx_skb_alloc(struct ionic_queue *q, struct sk_buff *skb; netdev = lif->netdev; - stats = q_to_rx_stats(q); + stats = &q->lif->rxqstats[q->index]; if (frags) skb = napi_get_frags(&q_to_qcq(q)->napi); @@ -235,14 +235,14 @@ static bool ionic_rx_service(struct ionic_cq *cq, struct ionic_cq_info *cq_info) return false; /* check for empty queue */ - if (q->tail->index == q->head->index) + if (q->tail_idx == q->head_idx) return false; - desc_info = q->tail; - if (desc_info->index != le16_to_cpu(comp->comp_index)) + if (q->tail_idx != le16_to_cpu(comp->comp_index)) return false; - q->tail = desc_info->next; + desc_info = &q->info[q->tail_idx]; + q->tail_idx = (q->tail_idx + 1) & (q->num_descs - 1); /* clean the related q entry, only one per qc completion */ ionic_rx_clean(q, desc_info, cq_info, desc_info->cb_arg); @@ -266,40 +266,49 @@ void ionic_rx_flush(struct ionic_cq *cq) work_done, IONIC_INTR_CRED_RESET_COALESCE); } -static struct page *ionic_rx_page_alloc(struct ionic_queue *q, - dma_addr_t *dma_addr) +static int ionic_rx_page_alloc(struct ionic_queue *q, + struct ionic_page_info *page_info) { struct ionic_lif *lif = q->lif; struct ionic_rx_stats *stats; struct net_device *netdev; struct device *dev; - struct page *page; netdev = lif->netdev; dev = lif->ionic->dev; stats = q_to_rx_stats(q); - page = alloc_page(GFP_ATOMIC); - if (unlikely(!page)) { - net_err_ratelimited("%s: Page alloc failed on %s!\n", + + if (unlikely(!page_info)) { + net_err_ratelimited("%s: %s invalid page_info in alloc\n", + netdev->name, q->name); + return -EINVAL; + } + + page_info->page = dev_alloc_page(); + if (unlikely(!page_info->page)) { + net_err_ratelimited("%s: %s page alloc failed\n", netdev->name, q->name); stats->alloc_err++; - return NULL; + return -ENOMEM; } - *dma_addr = dma_map_page(dev, page, 0, PAGE_SIZE, DMA_FROM_DEVICE); - if (unlikely(dma_mapping_error(dev, *dma_addr))) { - __free_page(page); - net_err_ratelimited("%s: DMA single map failed on %s!\n", + page_info->dma_addr = dma_map_page(dev, page_info->page, 0, PAGE_SIZE, + DMA_FROM_DEVICE); + if (unlikely(dma_mapping_error(dev, page_info->dma_addr))) { + put_page(page_info->page); + page_info->dma_addr = 0; + page_info->page = NULL; + net_err_ratelimited("%s: %s dma map failed\n", netdev->name, q->name); stats->dma_map_err++; - return NULL; + return -EIO; } - return page; + return 0; } -static void ionic_rx_page_free(struct ionic_queue *q, struct page *page, - dma_addr_t dma_addr) +static void ionic_rx_page_free(struct ionic_queue *q, + struct ionic_page_info *page_info) { struct ionic_lif *lif = q->lif; struct net_device *netdev; @@ -308,15 +317,23 @@ static void ionic_rx_page_free(struct ionic_queue *q, struct page *page, netdev = lif->netdev; dev = lif->ionic->dev; - if (unlikely(!page)) { - net_err_ratelimited("%s: Trying to free unallocated buffer on %s!\n", + if (unlikely(!page_info)) { + net_err_ratelimited("%s: %s invalid page_info in free\n", + netdev->name, q->name); + return; + } + + if (unlikely(!page_info->page)) { + net_err_ratelimited("%s: %s invalid page in free\n", netdev->name, q->name); return; } - dma_unmap_page(dev, dma_addr, PAGE_SIZE, DMA_FROM_DEVICE); + dma_unmap_page(dev, page_info->dma_addr, PAGE_SIZE, DMA_FROM_DEVICE); - __free_page(page); + put_page(page_info->page); + page_info->dma_addr = 0; + page_info->page = NULL; } void ionic_rx_fill(struct ionic_queue *q) @@ -338,7 +355,7 @@ void ionic_rx_fill(struct ionic_queue *q) for (i = ionic_q_space_avail(q); i; i--) { remain_len = len; - desc_info = q->head; + desc_info = &q->info[q->head_idx]; desc = desc_info->desc; sg_desc = desc_info->sg_desc; page_info = &desc_info->pages[0]; @@ -352,8 +369,7 @@ void ionic_rx_fill(struct ionic_queue *q) desc->opcode = (nfrags > 1) ? IONIC_RXQ_DESC_OPCODE_SG : IONIC_RXQ_DESC_OPCODE_SIMPLE; desc_info->npages = nfrags; - page_info->page = ionic_rx_page_alloc(q, &page_info->dma_addr); - if (unlikely(!page_info->page)) { + if (unlikely(ionic_rx_page_alloc(q, page_info))) { desc->addr = 0; desc->len = 0; return; @@ -370,8 +386,7 @@ void ionic_rx_fill(struct ionic_queue *q) continue; sg_elem = &sg_desc->elems[j]; - page_info->page = ionic_rx_page_alloc(q, &page_info->dma_addr); - if (unlikely(!page_info->page)) { + if (unlikely(ionic_rx_page_alloc(q, page_info))) { sg_elem->addr = 0; sg_elem->len = 0; return; @@ -387,7 +402,7 @@ void ionic_rx_fill(struct ionic_queue *q) } ionic_dbell_ring(q->lif->kern_dbpage, q->hw_type, - q->dbval | q->head->index); + q->dbval | q->head_idx); } static void ionic_rx_fill_cb(void *arg) @@ -397,28 +412,50 @@ static void ionic_rx_fill_cb(void *arg) void ionic_rx_empty(struct ionic_queue *q) { - struct ionic_desc_info *cur; + struct ionic_desc_info *desc_info; struct ionic_rxq_desc *desc; unsigned int i; + u16 idx; - for (cur = q->tail; cur != q->head; cur = cur->next) { - desc = cur->desc; + idx = q->tail_idx; + while (idx != q->head_idx) { + desc_info = &q->info[idx]; + desc = desc_info->desc; desc->addr = 0; desc->len = 0; - for (i = 0; i < cur->npages; i++) { - if (likely(cur->pages[i].page)) { - ionic_rx_page_free(q, cur->pages[i].page, - cur->pages[i].dma_addr); - cur->pages[i].page = NULL; - cur->pages[i].dma_addr = 0; - } - } + for (i = 0; i < desc_info->npages; i++) + ionic_rx_page_free(q, &desc_info->pages[i]); - cur->cb_arg = NULL; + desc_info->cb_arg = NULL; + idx = (idx + 1) & (q->num_descs - 1); } } +static void ionic_dim_update(struct ionic_qcq *qcq) +{ + struct dim_sample dim_sample; + struct ionic_lif *lif; + unsigned int qi; + + if (!qcq->intr.dim_coal_hw) + return; + + lif = qcq->q.lif; + qi = qcq->cq.bound_q->index; + + ionic_intr_coal_init(lif->ionic->idev.intr_ctrl, + lif->rxqcqs[qi]->intr.index, + qcq->intr.dim_coal_hw); + + dim_update_sample(qcq->cq.bound_intr->rearm_count, + lif->txqstats[qi].pkts, + lif->txqstats[qi].bytes, + &dim_sample); + + net_dim(&qcq->dim, dim_sample); +} + int ionic_tx_napi(struct napi_struct *napi, int budget) { struct ionic_qcq *qcq = napi_to_qcq(napi); @@ -435,8 +472,9 @@ int ionic_tx_napi(struct napi_struct *napi, int budget) ionic_tx_service, NULL, NULL); if (work_done < budget && napi_complete_done(napi, work_done)) { + ionic_dim_update(qcq); flags |= IONIC_INTR_CRED_UNMASK; - DEBUG_STATS_INTR_REARM(cq->bound_intr); + cq->bound_intr->rearm_count++; } if (work_done || flags) { @@ -470,8 +508,9 @@ int ionic_rx_napi(struct napi_struct *napi, int budget) ionic_rx_fill(cq->bound_q); if (work_done < budget && napi_complete_done(napi, work_done)) { + ionic_dim_update(qcq); flags |= IONIC_INTR_CRED_UNMASK; - DEBUG_STATS_INTR_REARM(cq->bound_intr); + cq->bound_intr->rearm_count++; } if (work_done || flags) { @@ -500,7 +539,7 @@ int ionic_txrx_napi(struct napi_struct *napi, int budget) lif = rxcq->bound_q->lif; idev = &lif->ionic->idev; - txcq = &lif->txqcqs[qi].qcq->cq; + txcq = &lif->txqcqs[qi]->cq; tx_work_done = ionic_cq_service(txcq, lif->tx_budget, ionic_tx_service, NULL, NULL); @@ -511,8 +550,9 @@ int ionic_txrx_napi(struct napi_struct *napi, int budget) ionic_rx_fill_cb(rxcq->bound_q); if (rx_work_done < budget && napi_complete_done(napi, rx_work_done)) { + ionic_dim_update(qcq); flags |= IONIC_INTR_CRED_UNMASK; - DEBUG_STATS_INTR_REARM(rxcq->bound_intr); + rxcq->bound_intr->rearm_count++; } if (rx_work_done || flags) { @@ -615,6 +655,7 @@ static bool ionic_tx_service(struct ionic_cq *cq, struct ionic_cq_info *cq_info) struct ionic_txq_comp *comp = cq_info->cq_desc; struct ionic_queue *q = cq->bound_q; struct ionic_desc_info *desc_info; + u16 index; if (!color_match(comp->color, cq->done_color)) return false; @@ -623,12 +664,13 @@ static bool ionic_tx_service(struct ionic_cq *cq, struct ionic_cq_info *cq_info) * several q entries completed for each cq completion */ do { - desc_info = q->tail; - q->tail = desc_info->next; - ionic_tx_clean(q, desc_info, cq->tail, desc_info->cb_arg); + desc_info = &q->info[q->tail_idx]; + index = q->tail_idx; + q->tail_idx = (q->tail_idx + 1) & (q->num_descs - 1); + ionic_tx_clean(q, desc_info, cq_info, desc_info->cb_arg); desc_info->cb = NULL; desc_info->cb_arg = NULL; - } while (desc_info->index != le16_to_cpu(comp->comp_index)); + } while (index != le16_to_cpu(comp->comp_index)); return true; } @@ -648,16 +690,14 @@ void ionic_tx_flush(struct ionic_cq *cq) void ionic_tx_empty(struct ionic_queue *q) { struct ionic_desc_info *desc_info; - int done = 0; /* walk the not completed tx entries, if any */ - while (q->head != q->tail) { - desc_info = q->tail; - q->tail = desc_info->next; + while (q->head_idx != q->tail_idx) { + desc_info = &q->info[q->tail_idx]; + q->tail_idx = (q->tail_idx + 1) & (q->num_descs - 1); ionic_tx_clean(q, desc_info, NULL, desc_info->cb_arg); desc_info->cb = NULL; desc_info->cb_arg = NULL; - done++; } } @@ -741,8 +781,8 @@ static void ionic_tx_tso_post(struct ionic_queue *q, struct ionic_txq_desc *desc static struct ionic_txq_desc *ionic_tx_tso_next(struct ionic_queue *q, struct ionic_txq_sg_elem **elem) { - struct ionic_txq_sg_desc *sg_desc = q->head->sg_desc; - struct ionic_txq_desc *desc = q->head->desc; + struct ionic_txq_sg_desc *sg_desc = q->info[q->head_idx].txq_sg_desc; + struct ionic_txq_desc *desc = q->info[q->head_idx].txq_desc; *elem = sg_desc->elems; return desc; @@ -751,13 +791,13 @@ static struct ionic_txq_desc *ionic_tx_tso_next(struct ionic_queue *q, static int ionic_tx_tso(struct ionic_queue *q, struct sk_buff *skb) { struct ionic_tx_stats *stats = q_to_tx_stats(q); - struct ionic_desc_info *abort = q->head; + struct ionic_desc_info *rewind_desc_info; struct device *dev = q->lif->ionic->dev; - struct ionic_desc_info *rewind = abort; struct ionic_txq_sg_elem *elem; struct ionic_txq_desc *desc; unsigned int frag_left = 0; unsigned int offset = 0; + u16 abort = q->head_idx; unsigned int len_left; dma_addr_t desc_addr; unsigned int hdrlen; @@ -765,6 +805,7 @@ static int ionic_tx_tso(struct ionic_queue *q, struct sk_buff *skb) unsigned int seglen; u64 total_bytes = 0; u64 total_pkts = 0; + u16 rewind = abort; unsigned int left; unsigned int len; unsigned int mss; @@ -909,19 +950,20 @@ static int ionic_tx_tso(struct ionic_queue *q, struct sk_buff *skb) return 0; err_out_abort: - while (rewind->desc != q->head->desc) { - ionic_tx_clean(q, rewind, NULL, NULL); - rewind = rewind->next; + while (rewind != q->head_idx) { + rewind_desc_info = &q->info[rewind]; + ionic_tx_clean(q, rewind_desc_info, NULL, NULL); + rewind = (rewind + 1) & (q->num_descs - 1); } - q->head = abort; + q->head_idx = abort; return -ENOMEM; } static int ionic_tx_calc_csum(struct ionic_queue *q, struct sk_buff *skb) { + struct ionic_txq_desc *desc = q->info[q->head_idx].txq_desc; struct ionic_tx_stats *stats = q_to_tx_stats(q); - struct ionic_txq_desc *desc = q->head->desc; struct device *dev = q->lif->ionic->dev; dma_addr_t dma_addr; bool has_vlan; @@ -960,8 +1002,8 @@ static int ionic_tx_calc_csum(struct ionic_queue *q, struct sk_buff *skb) static int ionic_tx_calc_no_csum(struct ionic_queue *q, struct sk_buff *skb) { + struct ionic_txq_desc *desc = q->info[q->head_idx].txq_desc; struct ionic_tx_stats *stats = q_to_tx_stats(q); - struct ionic_txq_desc *desc = q->head->desc; struct device *dev = q->lif->ionic->dev; dma_addr_t dma_addr; bool has_vlan; @@ -995,7 +1037,7 @@ static int ionic_tx_calc_no_csum(struct ionic_queue *q, struct sk_buff *skb) static int ionic_tx_skb_frags(struct ionic_queue *q, struct sk_buff *skb) { - struct ionic_txq_sg_desc *sg_desc = q->head->sg_desc; + struct ionic_txq_sg_desc *sg_desc = q->info[q->head_idx].txq_sg_desc; unsigned int len_left = skb->len - skb_headlen(skb); struct ionic_txq_sg_elem *elem = sg_desc->elems; struct ionic_tx_stats *stats = q_to_tx_stats(q); @@ -1104,9 +1146,9 @@ netdev_tx_t ionic_start_xmit(struct sk_buff *skb, struct net_device *netdev) return NETDEV_TX_OK; } - if (unlikely(!lif_to_txqcq(lif, queue_index))) + if (unlikely(queue_index >= lif->nxqs)) queue_index = 0; - q = lif_to_txq(lif, queue_index); + q = &lif->txqcqs[queue_index]->q; ndescs = ionic_tx_descs_needed(q, skb); if (ndescs < 0) diff --git a/drivers/net/ethernet/qlogic/Kconfig b/drivers/net/ethernet/qlogic/Kconfig index 8f743d80760b..4366c7a8de95 100644 --- a/drivers/net/ethernet/qlogic/Kconfig +++ b/drivers/net/ethernet/qlogic/Kconfig @@ -80,7 +80,7 @@ config QED select CRC8 select NET_DEVLINK help - This enables the support for ... + This enables the support for Marvell FastLinQ adapters family. config QED_LL2 bool @@ -100,7 +100,8 @@ config QEDE depends on QED imply PTP_1588_CLOCK help - This enables the support for ... + This enables the support for Marvell FastLinQ adapters family, + ethernet driver. config QED_RDMA bool diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic.h b/drivers/net/ethernet/qlogic/netxen/netxen_nic.h index 86153660d245..e5c51256243a 100644 --- a/drivers/net/ethernet/qlogic/netxen/netxen_nic.h +++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic.h @@ -1189,9 +1189,6 @@ typedef struct { #define NX_FORCE_FW_RESET 0xdeaddead -/* Fw dump levels */ -static const u32 FW_DUMP_LEVELS[] = { 0x3, 0x7, 0xf, 0x1f, 0x3f, 0x7f, 0xff }; - /* Flash read/write address */ #define NX_FW_DUMP_REG1 0x00130060 #define NX_FW_DUMP_REG2 0x001e0000 diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool.c index c3f50ddbe824..dd22cb056d03 100644 --- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool.c +++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool.c @@ -814,6 +814,9 @@ netxen_get_dump_flag(struct net_device *netdev, struct ethtool_dump *dump) return 0; } +/* Fw dump levels */ +static const u32 FW_DUMP_LEVELS[] = { 0x3, 0x7, 0xf, 0x1f, 0x3f, 0x7f, 0xff }; + static int netxen_set_dump(struct net_device *netdev, struct ethtool_dump *val) { diff --git a/drivers/net/ethernet/qlogic/qed/Makefile b/drivers/net/ethernet/qlogic/qed/Makefile index f947b105cf14..8251755ec18c 100644 --- a/drivers/net/ethernet/qlogic/qed/Makefile +++ b/drivers/net/ethernet/qlogic/qed/Makefile @@ -9,6 +9,7 @@ qed-y := \ qed_dcbx.o \ qed_debug.o \ qed_dev.o \ + qed_devlink.o \ qed_hw.o \ qed_init_fw_funcs.o \ qed_init_ops.o \ diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h index b2a7b53ee760..a20cb8a0c377 100644 --- a/drivers/net/ethernet/qlogic/qed/qed.h +++ b/drivers/net/ethernet/qlogic/qed/qed.h @@ -572,7 +572,7 @@ struct qed_hwfn { struct qed_consq *p_consq; /* Slow-Path definitions */ - struct tasklet_struct *sp_dpc; + struct tasklet_struct sp_dpc; bool b_sp_dpc_enabled; struct qed_ptt *p_main_ptt; @@ -807,6 +807,7 @@ struct qed_dev { struct qed_llh_info *p_llh_info; /* Linux specific here */ + struct qed_dev_info common_dev_info; struct qede_dev *edev; struct pci_dev *pdev; u32 flags; @@ -849,7 +850,6 @@ struct qed_dev { u32 rdma_max_srq_sge; u16 tunn_feature_mask; - struct devlink *dl; bool iwarp_cmt; }; @@ -981,6 +981,7 @@ void qed_bw_update(struct qed_hwfn *hwfn, struct qed_ptt *ptt); u32 qed_unzip_data(struct qed_hwfn *p_hwfn, u32 input_len, u8 *input_buf, u32 max_size, u8 *unzip_buf); +int qed_recovery_process(struct qed_dev *cdev); void qed_schedule_recovery_handler(struct qed_hwfn *p_hwfn); void qed_hw_error_occurred(struct qed_hwfn *p_hwfn, enum qed_hw_err_type err_type); diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index 3db181f3617a..d2f5855b2ea7 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -3973,6 +3973,7 @@ static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) struct qed_mcp_link_speed_params *ext_speed; struct qed_mcp_link_capabilities *p_caps; struct qed_mcp_link_params *link; + int i; /* Read global nvm_cfg address */ nvm_cfg_addr = qed_rd(p_hwfn, p_ptt, MISC_REG_GEN_PURP_CR0); @@ -4299,6 +4300,14 @@ static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) __set_bit(QED_DEV_CAP_ROCE, &p_hwfn->hw_info.device_capabilities); + /* Read device serial number information from shmem */ + addr = MCP_REG_SCRATCH + nvm_cfg1_offset + + offsetof(struct nvm_cfg1, glob) + + offsetof(struct nvm_cfg1_glob, serial_number); + + for (i = 0; i < 4; i++) + p_hwfn->hw_info.part_num[i] = qed_rd(p_hwfn, p_ptt, addr + i * 4); + return qed_mcp_fill_shmem_func_info(p_hwfn, p_ptt); } diff --git a/drivers/net/ethernet/qlogic/qed/qed_devlink.c b/drivers/net/ethernet/qlogic/qed/qed_devlink.c new file mode 100644 index 000000000000..cf7f4da68e69 --- /dev/null +++ b/drivers/net/ethernet/qlogic/qed/qed_devlink.c @@ -0,0 +1,259 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* Marvell/Qlogic FastLinQ NIC driver + * + * Copyright (C) 2020 Marvell International Ltd. + */ + +#include <linux/kernel.h> +#include <linux/qed/qed_if.h> +#include <linux/vmalloc.h> +#include "qed.h" +#include "qed_devlink.h" + +enum qed_devlink_param_id { + QED_DEVLINK_PARAM_ID_BASE = DEVLINK_PARAM_GENERIC_ID_MAX, + QED_DEVLINK_PARAM_ID_IWARP_CMT, +}; + +struct qed_fw_fatal_ctx { + enum qed_hw_err_type err_type; +}; + +int qed_report_fatal_error(struct devlink *devlink, enum qed_hw_err_type err_type) +{ + struct qed_devlink *qdl = devlink_priv(devlink); + struct qed_fw_fatal_ctx fw_fatal_ctx = { + .err_type = err_type, + }; + + if (qdl->fw_reporter) + devlink_health_report(qdl->fw_reporter, + "Fatal error occurred", &fw_fatal_ctx); + + return 0; +} + +static int +qed_fw_fatal_reporter_dump(struct devlink_health_reporter *reporter, + struct devlink_fmsg *fmsg, void *priv_ctx, + struct netlink_ext_ack *extack) +{ + struct qed_devlink *qdl = devlink_health_reporter_priv(reporter); + struct qed_fw_fatal_ctx *fw_fatal_ctx = priv_ctx; + struct qed_dev *cdev = qdl->cdev; + u32 dbg_data_buf_size; + u8 *p_dbg_data_buf; + int err; + + /* Having context means that was a dump request after fatal, + * so we enable extra debugging while gathering the dump, + * just in case + */ + cdev->print_dbg_data = fw_fatal_ctx ? true : false; + + dbg_data_buf_size = qed_dbg_all_data_size(cdev); + p_dbg_data_buf = vzalloc(dbg_data_buf_size); + if (!p_dbg_data_buf) { + DP_NOTICE(cdev, + "Failed to allocate memory for a debug data buffer\n"); + return -ENOMEM; + } + + err = qed_dbg_all_data(cdev, p_dbg_data_buf); + if (err) { + DP_NOTICE(cdev, "Failed to obtain debug data\n"); + vfree(p_dbg_data_buf); + return err; + } + + err = devlink_fmsg_binary_pair_put(fmsg, "dump_data", + p_dbg_data_buf, dbg_data_buf_size); + + vfree(p_dbg_data_buf); + + return err; +} + +static int +qed_fw_fatal_reporter_recover(struct devlink_health_reporter *reporter, + void *priv_ctx, + struct netlink_ext_ack *extack) +{ + struct qed_devlink *qdl = devlink_health_reporter_priv(reporter); + struct qed_dev *cdev = qdl->cdev; + + qed_recovery_process(cdev); + + return 0; +} + +static const struct devlink_health_reporter_ops qed_fw_fatal_reporter_ops = { + .name = "fw_fatal", + .recover = qed_fw_fatal_reporter_recover, + .dump = qed_fw_fatal_reporter_dump, +}; + +#define QED_REPORTER_FW_GRACEFUL_PERIOD 1200000 + +void qed_fw_reporters_create(struct devlink *devlink) +{ + struct qed_devlink *dl = devlink_priv(devlink); + + dl->fw_reporter = devlink_health_reporter_create(devlink, &qed_fw_fatal_reporter_ops, + QED_REPORTER_FW_GRACEFUL_PERIOD, dl); + if (IS_ERR(dl->fw_reporter)) { + DP_NOTICE(dl->cdev, "Failed to create fw reporter, err = %ld\n", + PTR_ERR(dl->fw_reporter)); + dl->fw_reporter = NULL; + } +} + +void qed_fw_reporters_destroy(struct devlink *devlink) +{ + struct qed_devlink *dl = devlink_priv(devlink); + struct devlink_health_reporter *rep; + + rep = dl->fw_reporter; + + if (!IS_ERR_OR_NULL(rep)) + devlink_health_reporter_destroy(rep); +} + +static int qed_dl_param_get(struct devlink *dl, u32 id, + struct devlink_param_gset_ctx *ctx) +{ + struct qed_devlink *qed_dl = devlink_priv(dl); + struct qed_dev *cdev; + + cdev = qed_dl->cdev; + ctx->val.vbool = cdev->iwarp_cmt; + + return 0; +} + +static int qed_dl_param_set(struct devlink *dl, u32 id, + struct devlink_param_gset_ctx *ctx) +{ + struct qed_devlink *qed_dl = devlink_priv(dl); + struct qed_dev *cdev; + + cdev = qed_dl->cdev; + cdev->iwarp_cmt = ctx->val.vbool; + + return 0; +} + +static const struct devlink_param qed_devlink_params[] = { + DEVLINK_PARAM_DRIVER(QED_DEVLINK_PARAM_ID_IWARP_CMT, + "iwarp_cmt", DEVLINK_PARAM_TYPE_BOOL, + BIT(DEVLINK_PARAM_CMODE_RUNTIME), + qed_dl_param_get, qed_dl_param_set, NULL), +}; + +static int qed_devlink_info_get(struct devlink *devlink, + struct devlink_info_req *req, + struct netlink_ext_ack *extack) +{ + struct qed_devlink *qed_dl = devlink_priv(devlink); + struct qed_dev *cdev = qed_dl->cdev; + struct qed_dev_info *dev_info; + char buf[100]; + int err; + + dev_info = &cdev->common_dev_info; + + err = devlink_info_driver_name_put(req, KBUILD_MODNAME); + if (err) + return err; + + memcpy(buf, cdev->hwfns[0].hw_info.part_num, sizeof(cdev->hwfns[0].hw_info.part_num)); + buf[sizeof(cdev->hwfns[0].hw_info.part_num)] = 0; + + if (buf[0]) { + err = devlink_info_board_serial_number_put(req, buf); + if (err) + return err; + } + + snprintf(buf, sizeof(buf), "%d.%d.%d.%d", + GET_MFW_FIELD(dev_info->mfw_rev, QED_MFW_VERSION_3), + GET_MFW_FIELD(dev_info->mfw_rev, QED_MFW_VERSION_2), + GET_MFW_FIELD(dev_info->mfw_rev, QED_MFW_VERSION_1), + GET_MFW_FIELD(dev_info->mfw_rev, QED_MFW_VERSION_0)); + + err = devlink_info_version_stored_put(req, + DEVLINK_INFO_VERSION_GENERIC_FW_MGMT, buf); + if (err) + return err; + + snprintf(buf, sizeof(buf), "%d.%d.%d.%d", + dev_info->fw_major, + dev_info->fw_minor, + dev_info->fw_rev, + dev_info->fw_eng); + + return devlink_info_version_running_put(req, + DEVLINK_INFO_VERSION_GENERIC_FW_APP, buf); +} + +static const struct devlink_ops qed_dl_ops = { + .info_get = qed_devlink_info_get, +}; + +struct devlink *qed_devlink_register(struct qed_dev *cdev) +{ + union devlink_param_value value; + struct qed_devlink *qdevlink; + struct devlink *dl; + int rc; + + dl = devlink_alloc(&qed_dl_ops, sizeof(struct qed_devlink)); + if (!dl) + return ERR_PTR(-ENOMEM); + + qdevlink = devlink_priv(dl); + qdevlink->cdev = cdev; + + rc = devlink_register(dl, &cdev->pdev->dev); + if (rc) + goto err_free; + + rc = devlink_params_register(dl, qed_devlink_params, + ARRAY_SIZE(qed_devlink_params)); + if (rc) + goto err_unregister; + + value.vbool = false; + devlink_param_driverinit_value_set(dl, + QED_DEVLINK_PARAM_ID_IWARP_CMT, + value); + + devlink_params_publish(dl); + cdev->iwarp_cmt = false; + + qed_fw_reporters_create(dl); + + return dl; + +err_unregister: + devlink_unregister(dl); + +err_free: + devlink_free(dl); + + return ERR_PTR(rc); +} + +void qed_devlink_unregister(struct devlink *devlink) +{ + if (!devlink) + return; + + qed_fw_reporters_destroy(devlink); + + devlink_params_unregister(devlink, qed_devlink_params, + ARRAY_SIZE(qed_devlink_params)); + + devlink_unregister(devlink); + devlink_free(devlink); +} diff --git a/drivers/net/ethernet/qlogic/qed/qed_devlink.h b/drivers/net/ethernet/qlogic/qed/qed_devlink.h new file mode 100644 index 000000000000..ccc7d1d1bfd4 --- /dev/null +++ b/drivers/net/ethernet/qlogic/qed/qed_devlink.h @@ -0,0 +1,20 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* Marvell/Qlogic FastLinQ NIC driver + * + * Copyright (C) 2020 Marvell International Ltd. + */ +#ifndef _QED_DEVLINK_H +#define _QED_DEVLINK_H + +#include <linux/qed/qed_if.h> +#include <net/devlink.h> + +struct devlink *qed_devlink_register(struct qed_dev *cdev); +void qed_devlink_unregister(struct devlink *devlink); + +void qed_fw_reporters_create(struct devlink *devlink); +void qed_fw_reporters_destroy(struct devlink *devlink); + +int qed_report_fatal_error(struct devlink *dl, enum qed_hw_err_type err_type); + +#endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.c b/drivers/net/ethernet/qlogic/qed/qed_int.c index f8c5a864812d..578935f643b8 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_int.c +++ b/drivers/net/ethernet/qlogic/qed/qed_int.c @@ -1216,9 +1216,9 @@ static void qed_sb_ack_attn(struct qed_hwfn *p_hwfn, barrier(); } -void qed_int_sp_dpc(unsigned long hwfn_cookie) +void qed_int_sp_dpc(struct tasklet_struct *t) { - struct qed_hwfn *p_hwfn = (struct qed_hwfn *)hwfn_cookie; + struct qed_hwfn *p_hwfn = from_tasklet(p_hwfn, t, sp_dpc); struct qed_pi_info *pi_info = NULL; struct qed_sb_attn_info *sb_attn; struct qed_sb_info *sb_info; @@ -2285,34 +2285,14 @@ u64 qed_int_igu_read_sisr_reg(struct qed_hwfn *p_hwfn) static void qed_int_sp_dpc_setup(struct qed_hwfn *p_hwfn) { - tasklet_init(p_hwfn->sp_dpc, - qed_int_sp_dpc, (unsigned long)p_hwfn); + tasklet_setup(&p_hwfn->sp_dpc, qed_int_sp_dpc); p_hwfn->b_sp_dpc_enabled = true; } -static int qed_int_sp_dpc_alloc(struct qed_hwfn *p_hwfn) -{ - p_hwfn->sp_dpc = kmalloc(sizeof(*p_hwfn->sp_dpc), GFP_KERNEL); - if (!p_hwfn->sp_dpc) - return -ENOMEM; - - return 0; -} - -static void qed_int_sp_dpc_free(struct qed_hwfn *p_hwfn) -{ - kfree(p_hwfn->sp_dpc); - p_hwfn->sp_dpc = NULL; -} - int qed_int_alloc(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { int rc = 0; - rc = qed_int_sp_dpc_alloc(p_hwfn); - if (rc) - return rc; - rc = qed_int_sp_sb_alloc(p_hwfn, p_ptt); if (rc) return rc; @@ -2326,7 +2306,6 @@ void qed_int_free(struct qed_hwfn *p_hwfn) { qed_int_sp_sb_free(p_hwfn); qed_int_sb_attn_free(p_hwfn); - qed_int_sp_dpc_free(p_hwfn); } void qed_int_setup(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.h b/drivers/net/ethernet/qlogic/qed/qed_int.h index 86809d7bc2de..c5550e96bbe1 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_int.h +++ b/drivers/net/ethernet/qlogic/qed/qed_int.h @@ -140,7 +140,7 @@ int qed_int_sb_release(struct qed_hwfn *p_hwfn, * @param p_hwfn - pointer to hwfn * */ -void qed_int_sp_dpc(unsigned long hwfn_cookie); +void qed_int_sp_dpc(struct tasklet_struct *t); /** * @brief qed_int_get_num_sbs - get the number of status diff --git a/drivers/net/ethernet/qlogic/qed/qed_ll2.c b/drivers/net/ethernet/qlogic/qed/qed_ll2.c index 0452b728c527..49783f365079 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_ll2.c +++ b/drivers/net/ethernet/qlogic/qed/qed_ll2.c @@ -1185,7 +1185,7 @@ static int qed_ll2_acquire_connection_tx(struct qed_hwfn *p_hwfn, .elem_size = sizeof(struct core_tx_bd), }; struct qed_ll2_tx_packet *p_descq; - u32 desc_size; + size_t desc_size; u32 capacity; int rc = 0; @@ -1198,10 +1198,9 @@ static int qed_ll2_acquire_connection_tx(struct qed_hwfn *p_hwfn, goto out; capacity = qed_chain_get_capacity(&p_ll2_info->tx_queue.txq_chain); - /* First element is part of the packet, rest are flexibly added */ - desc_size = (sizeof(*p_descq) + - (p_ll2_info->input.tx_max_bds_per_packet - 1) * - sizeof(p_descq->bds_set)); + /* All bds_set elements are flexibily added. */ + desc_size = struct_size(p_descq, bds_set, + p_ll2_info->input.tx_max_bds_per_packet); p_descq = kcalloc(capacity, desc_size, GFP_KERNEL); if (!p_descq) { @@ -1524,7 +1523,7 @@ int qed_ll2_establish_connection(void *cxt, u8 connection_handle) struct qed_ptt *p_ptt; int rc = -EINVAL; u32 i, capacity; - u32 desc_size; + size_t desc_size; u8 qid; p_ptt = qed_ptt_acquire(p_hwfn); @@ -1558,10 +1557,9 @@ int qed_ll2_establish_connection(void *cxt, u8 connection_handle) INIT_LIST_HEAD(&p_tx->sending_descq); spin_lock_init(&p_tx->lock); capacity = qed_chain_get_capacity(&p_tx->txq_chain); - /* First element is part of the packet, rest are flexibly added */ - desc_size = (sizeof(*p_pkt) + - (p_ll2_conn->input.tx_max_bds_per_packet - 1) * - sizeof(p_pkt->bds_set)); + /* All bds_set elements are flexibily added. */ + desc_size = struct_size(p_pkt, bds_set, + p_ll2_conn->input.tx_max_bds_per_packet); for (i = 0; i < capacity; i++) { p_pkt = p_tx->descq_mem + desc_size * i; diff --git a/drivers/net/ethernet/qlogic/qed/qed_ll2.h b/drivers/net/ethernet/qlogic/qed/qed_ll2.h index 500d0c4f8077..df88d00053a2 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_ll2.h +++ b/drivers/net/ethernet/qlogic/qed/qed_ll2.h @@ -56,7 +56,7 @@ struct qed_ll2_tx_packet { struct core_tx_bd *txq_bd; dma_addr_t tx_frag; u16 frag_len; - } bds_set[1]; + } bds_set[]; }; struct qed_ll2_rx_queue { @@ -86,9 +86,6 @@ struct qed_ll2_tx_queue { struct list_head active_descq; struct list_head free_descq; struct list_head sending_descq; - void *descq_mem; /* memory for variable sized qed_ll2_tx_packet*/ - struct qed_ll2_tx_packet *cur_send_packet; - struct qed_ll2_tx_packet cur_completing_packet; u16 cur_completing_bd_idx; void __iomem *doorbell_addr; struct core_db_data db_msg; @@ -96,6 +93,9 @@ struct qed_ll2_tx_queue { u16 cur_send_frag_num; u16 cur_completing_frag_num; bool b_completing_packet; + void *descq_mem; /* memory for variable sized qed_ll2_tx_packet*/ + struct qed_ll2_tx_packet *cur_send_packet; + struct qed_ll2_tx_packet cur_completing_packet; }; struct qed_ll2_info { diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index 50e5eb22e60a..5bd58c65e163 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -39,6 +39,7 @@ #include "qed_hw.h" #include "qed_selftest.h" #include "qed_debug.h" +#include "qed_devlink.h" #define QED_ROCE_QPS (8192) #define QED_ROCE_DPIS (8) @@ -480,6 +481,7 @@ int qed_fill_dev_info(struct qed_dev *cdev, } dev_info->mtu = hw_info->mtu; + cdev->common_dev_info = *dev_info; return 0; } @@ -512,107 +514,6 @@ static int qed_set_power_state(struct qed_dev *cdev, pci_power_t state) return 0; } -struct qed_devlink { - struct qed_dev *cdev; -}; - -enum qed_devlink_param_id { - QED_DEVLINK_PARAM_ID_BASE = DEVLINK_PARAM_GENERIC_ID_MAX, - QED_DEVLINK_PARAM_ID_IWARP_CMT, -}; - -static int qed_dl_param_get(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) -{ - struct qed_devlink *qed_dl; - struct qed_dev *cdev; - - qed_dl = devlink_priv(dl); - cdev = qed_dl->cdev; - ctx->val.vbool = cdev->iwarp_cmt; - - return 0; -} - -static int qed_dl_param_set(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) -{ - struct qed_devlink *qed_dl; - struct qed_dev *cdev; - - qed_dl = devlink_priv(dl); - cdev = qed_dl->cdev; - cdev->iwarp_cmt = ctx->val.vbool; - - return 0; -} - -static const struct devlink_param qed_devlink_params[] = { - DEVLINK_PARAM_DRIVER(QED_DEVLINK_PARAM_ID_IWARP_CMT, - "iwarp_cmt", DEVLINK_PARAM_TYPE_BOOL, - BIT(DEVLINK_PARAM_CMODE_RUNTIME), - qed_dl_param_get, qed_dl_param_set, NULL), -}; - -static const struct devlink_ops qed_dl_ops; - -static int qed_devlink_register(struct qed_dev *cdev) -{ - union devlink_param_value value; - struct qed_devlink *qed_dl; - struct devlink *dl; - int rc; - - dl = devlink_alloc(&qed_dl_ops, sizeof(*qed_dl)); - if (!dl) - return -ENOMEM; - - qed_dl = devlink_priv(dl); - - cdev->dl = dl; - qed_dl->cdev = cdev; - - rc = devlink_register(dl, &cdev->pdev->dev); - if (rc) - goto err_free; - - rc = devlink_params_register(dl, qed_devlink_params, - ARRAY_SIZE(qed_devlink_params)); - if (rc) - goto err_unregister; - - value.vbool = false; - devlink_param_driverinit_value_set(dl, - QED_DEVLINK_PARAM_ID_IWARP_CMT, - value); - - devlink_params_publish(dl); - cdev->iwarp_cmt = false; - - return 0; - -err_unregister: - devlink_unregister(dl); - -err_free: - cdev->dl = NULL; - devlink_free(dl); - - return rc; -} - -static void qed_devlink_unregister(struct qed_dev *cdev) -{ - if (!cdev->dl) - return; - - devlink_params_unregister(cdev->dl, qed_devlink_params, - ARRAY_SIZE(qed_devlink_params)); - - devlink_unregister(cdev->dl); - devlink_free(cdev->dl); -} - /* probing */ static struct qed_dev *qed_probe(struct pci_dev *pdev, struct qed_probe_params *params) @@ -641,12 +542,6 @@ static struct qed_dev *qed_probe(struct pci_dev *pdev, } DP_INFO(cdev, "PCI init completed successfully\n"); - rc = qed_devlink_register(cdev); - if (rc) { - DP_INFO(cdev, "Failed to register devlink.\n"); - goto err2; - } - rc = qed_hw_prepare(cdev, QED_PCI_DEFAULT); if (rc) { DP_ERR(cdev, "hw prepare failed\n"); @@ -676,8 +571,6 @@ static void qed_remove(struct qed_dev *cdev) qed_set_power_state(cdev, PCI_D3hot); - qed_devlink_unregister(cdev); - qed_free_cdev(cdev); } @@ -843,7 +736,7 @@ static irqreturn_t qed_single_int(int irq, void *dev_instance) /* Slowpath interrupt */ if (unlikely(status & 0x1)) { - tasklet_schedule(hwfn->sp_dpc); + tasklet_schedule(&hwfn->sp_dpc); status &= ~0x1; rc = IRQ_HANDLED; } @@ -889,7 +782,7 @@ int qed_slowpath_irq_req(struct qed_hwfn *hwfn) id, cdev->pdev->bus->number, PCI_SLOT(cdev->pdev->devfn), hwfn->abs_pf_id); rc = request_irq(cdev->int_params.msix_table[id].vector, - qed_msix_sp_int, 0, hwfn->name, hwfn->sp_dpc); + qed_msix_sp_int, 0, hwfn->name, &hwfn->sp_dpc); } else { unsigned long flags = 0; @@ -921,8 +814,8 @@ static void qed_slowpath_tasklet_flush(struct qed_hwfn *p_hwfn) * enable function makes this sequence a flush-like operation. */ if (p_hwfn->b_sp_dpc_enabled) { - tasklet_disable(p_hwfn->sp_dpc); - tasklet_enable(p_hwfn->sp_dpc); + tasklet_disable(&p_hwfn->sp_dpc); + tasklet_enable(&p_hwfn->sp_dpc); } } @@ -951,7 +844,7 @@ static void qed_slowpath_irq_free(struct qed_dev *cdev) break; synchronize_irq(cdev->int_params.msix_table[i].vector); free_irq(cdev->int_params.msix_table[i].vector, - cdev->hwfns[i].sp_dpc); + &cdev->hwfns[i].sp_dpc); } } else { if (QED_LEADING_HWFN(cdev)->b_int_requested) @@ -970,11 +863,11 @@ static int qed_nic_stop(struct qed_dev *cdev) struct qed_hwfn *p_hwfn = &cdev->hwfns[i]; if (p_hwfn->b_sp_dpc_enabled) { - tasklet_disable(p_hwfn->sp_dpc); + tasklet_disable(&p_hwfn->sp_dpc); p_hwfn->b_sp_dpc_enabled = false; DP_VERBOSE(cdev, NETIF_MSG_IFDOWN, "Disabled sp tasklet [hwfn %d] at %p\n", - i, p_hwfn->sp_dpc); + i, &p_hwfn->sp_dpc); } } @@ -2926,7 +2819,7 @@ static int qed_set_led(struct qed_dev *cdev, enum qed_led_mode mode) return status; } -static int qed_recovery_process(struct qed_dev *cdev) +int qed_recovery_process(struct qed_dev *cdev) { struct qed_hwfn *p_hwfn = QED_LEADING_HWFN(cdev); struct qed_ptt *p_ptt; @@ -3114,6 +3007,9 @@ const struct qed_common_ops qed_common_ops_pass = { .get_link = &qed_get_current_link, .drain = &qed_drain, .update_msglvl = &qed_init_dp, + .devlink_register = qed_devlink_register, + .devlink_unregister = qed_devlink_unregister, + .report_fatal_error = qed_report_fatal_error, .dbg_all_data = &qed_dbg_all_data, .dbg_all_data_size = &qed_dbg_all_data_size, .chain_alloc = &qed_chain_alloc, diff --git a/drivers/net/ethernet/qlogic/qed/qed_rdma.c b/drivers/net/ethernet/qlogic/qed/qed_rdma.c index a4bcde522cdf..d3136556a1e9 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_rdma.c +++ b/drivers/net/ethernet/qlogic/qed/qed_rdma.c @@ -1151,7 +1151,6 @@ qed_rdma_destroy_cq(void *rdma_cxt, DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "icid = %08x\n", in_params->icid); p_ramrod_res = - (struct rdma_destroy_cq_output_params *) dma_alloc_coherent(&p_hwfn->cdev->pdev->dev, sizeof(struct rdma_destroy_cq_output_params), &ramrod_res_phys, GFP_KERNEL); @@ -1463,14 +1462,14 @@ static int qed_rdma_modify_qp(void *rdma_cxt, switch (qp->qp_type) { case QED_RDMA_QP_TYPE_XRC_INI: - qp->has_req = 1; + qp->has_req = true; break; case QED_RDMA_QP_TYPE_XRC_TGT: - qp->has_resp = 1; + qp->has_resp = true; break; default: - qp->has_req = 1; - qp->has_resp = 1; + qp->has_req = true; + qp->has_resp = true; } if (QED_IS_IWARP_PERSONALITY(p_hwfn)) { diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h index 803c1fcca8ad..3efc5899f656 100644 --- a/drivers/net/ethernet/qlogic/qede/qede.h +++ b/drivers/net/ethernet/qlogic/qede/qede.h @@ -172,6 +172,7 @@ struct qede_dev { struct qed_dev *cdev; struct net_device *ndev; struct pci_dev *pdev; + struct devlink *devlink; u32 dp_module; u8 dp_level; @@ -263,6 +264,7 @@ struct qede_dev { struct bpf_prog *xdp_prog; + enum qed_hw_err_type last_err_type; unsigned long err_flags; #define QEDE_ERR_IS_HANDLED 31 #define QEDE_ERR_ATTN_CLR_EN 0 diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index 9e1f41ba766c..05e3a3b60269 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -1170,10 +1170,23 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, rc = -ENOMEM; goto err2; } + + edev->devlink = qed_ops->common->devlink_register(cdev); + if (IS_ERR(edev->devlink)) { + DP_NOTICE(edev, "Cannot register devlink\n"); + edev->devlink = NULL; + /* Go on, we can live without devlink */ + } } else { struct net_device *ndev = pci_get_drvdata(pdev); edev = netdev_priv(ndev); + + if (edev->devlink) { + struct qed_devlink *qdl = devlink_priv(edev->devlink); + + qdl->cdev = cdev; + } edev->cdev = cdev; memset(&edev->stats, 0, sizeof(edev->stats)); memcpy(&edev->dev_info, &dev_info, sizeof(dev_info)); @@ -1225,7 +1238,10 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, err4: qede_rdma_dev_remove(edev, (mode == QEDE_PROBE_RECOVERY)); err3: - free_netdev(edev->ndev); + if (mode != QEDE_PROBE_RECOVERY) + free_netdev(edev->ndev); + else + edev->cdev = NULL; err2: qed_ops->common->slowpath_stop(cdev); err1: @@ -1296,6 +1312,11 @@ static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode) qed_ops->common->slowpath_stop(cdev); if (system_state == SYSTEM_POWER_OFF) return; + + if (mode != QEDE_REMOVE_RECOVERY && edev->devlink) { + qed_ops->common->devlink_unregister(edev->devlink); + edev->devlink = NULL; + } qed_ops->common->remove(cdev); edev->cdev = NULL; @@ -2454,7 +2475,8 @@ static int qede_close(struct net_device *ndev) qede_unload(edev, QEDE_UNLOAD_NORMAL, false); - edev->ops->common->update_drv_state(edev->cdev, false); + if (edev->cdev) + edev->ops->common->update_drv_state(edev->cdev, false); return 0; } @@ -2576,19 +2598,12 @@ static void qede_atomic_hw_err_handler(struct qede_dev *edev) static void qede_generic_hw_err_handler(struct qede_dev *edev) { - struct qed_dev *cdev = edev->cdev; - DP_NOTICE(edev, "Generic sleepable HW error handling started - err_flags 0x%lx\n", edev->err_flags); - /* Trigger a recovery process. - * This is placed in the sleep requiring section just to make - * sure it is the last one, and that all the other operations - * were completed. - */ - if (test_bit(QEDE_ERR_IS_RECOVERABLE, &edev->err_flags)) - edev->ops->common->recovery_process(cdev); + if (edev->devlink) + edev->ops->common->report_fatal_error(edev->devlink, edev->last_err_type); clear_bit(QEDE_ERR_IS_HANDLED, &edev->err_flags); @@ -2642,6 +2657,7 @@ static void qede_schedule_hw_err_handler(void *dev, return; } + edev->last_err_type = err_type; qede_set_hw_err_flags(edev, err_type); qede_atomic_hw_err_handler(edev); set_bit(QEDE_SP_HW_ERR, &edev->sp_flags); diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c index 31ad3a5cd128..d8882d0b6b49 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c @@ -657,11 +657,10 @@ int qlcnic_83xx_cam_lock(struct qlcnic_adapter *adapter) void qlcnic_83xx_cam_unlock(struct qlcnic_adapter *adapter) { void __iomem *addr; - u32 val; struct qlcnic_hardware_context *ahw = adapter->ahw; addr = ahw->pci_base0 + QLC_83XX_SEM_UNLOCK_FUNC(ahw->pci_func); - val = readl(addr); + readl(addr); } void qlcnic_83xx_read_crb(struct qlcnic_adapter *adapter, char *buf, @@ -3812,7 +3811,6 @@ static int qlcnic_83xx_shutdown(struct pci_dev *pdev) { struct qlcnic_adapter *adapter = pci_get_drvdata(pdev); struct net_device *netdev = adapter->netdev; - int retval; netif_device_detach(netdev); qlcnic_cancel_idc_work(adapter); @@ -3823,11 +3821,7 @@ static int qlcnic_83xx_shutdown(struct pci_dev *pdev) qlcnic_83xx_disable_mbx_intr(adapter); cancel_delayed_work_sync(&adapter->idc_aen_work); - retval = pci_save_state(pdev); - if (retval) - return retval; - - return 0; + return pci_save_state(pdev); } static int qlcnic_83xx_resume(struct qlcnic_adapter *adapter) diff --git a/drivers/net/ethernet/qualcomm/emac/emac.c b/drivers/net/ethernet/qualcomm/emac/emac.c index 1166b98d8bb2..8543bf3c3484 100644 --- a/drivers/net/ethernet/qualcomm/emac/emac.c +++ b/drivers/net/ethernet/qualcomm/emac/emac.c @@ -292,6 +292,7 @@ static void emac_tx_timeout(struct net_device *netdev, unsigned int txqueue) /** * emac_update_hw_stats - read the EMAC stat registers + * @adpt: pointer to adapter struct * * Reads the stats registers and write the values to adpt->stats. * diff --git a/drivers/net/ethernet/qualcomm/qca_uart.c b/drivers/net/ethernet/qualcomm/qca_uart.c index 375a844cd27c..362b4f5c162c 100644 --- a/drivers/net/ethernet/qualcomm/qca_uart.c +++ b/drivers/net/ethernet/qualcomm/qca_uart.c @@ -167,7 +167,7 @@ static void qca_tty_wakeup(struct serdev_device *serdev) schedule_work(&qca->tx_work); } -static struct serdev_device_ops qca_serdev_ops = { +static const struct serdev_device_ops qca_serdev_ops = { .receive_buf = qca_tty_receive, .write_wakeup = qca_tty_wakeup, }; diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c index e291e6ac40cb..4e44313b7651 100644 --- a/drivers/net/ethernet/realtek/8139cp.c +++ b/drivers/net/ethernet/realtek/8139cp.c @@ -1239,7 +1239,7 @@ static void cp_tx_timeout(struct net_device *dev, unsigned int txqueue) { struct cp_private *cp = netdev_priv(dev); unsigned long flags; - int rc, i; + int i; netdev_warn(dev, "Transmit timeout, status %2x %4x %4x %4x\n", cpr8(Cmd), cpr16(CpCmd), @@ -1260,7 +1260,7 @@ static void cp_tx_timeout(struct net_device *dev, unsigned int txqueue) cp_stop_hw(cp); cp_clean_rings(cp); - rc = cp_init_rings(cp); + cp_init_rings(cp); cp_start_hw(cp); __cp_set_rx_mode(dev); cpw16_f(IntrMask, cp_norx_intr_mask); diff --git a/drivers/net/ethernet/realtek/8139too.c b/drivers/net/ethernet/realtek/8139too.c index 227139d42227..1e5a453dea14 100644 --- a/drivers/net/ethernet/realtek/8139too.c +++ b/drivers/net/ethernet/realtek/8139too.c @@ -978,7 +978,7 @@ static int rtl8139_init_one(struct pci_dev *pdev, pdev->subsystem_vendor == PCI_VENDOR_ID_ATHEROS && pdev->subsystem_device == PCI_DEVICE_ID_REALTEK_8139) { pr_info("OQO Model 2 detected. Forcing PIO\n"); - use_io = 1; + use_io = true; } dev = rtl8139_init_board (pdev); diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index 11e6962a18e4..7d366b0362cb 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -617,7 +617,6 @@ struct rtl8169_private { struct work_struct work; } wk; - unsigned irq_enabled:1; unsigned supports_gmii:1; unsigned aspm_manageable:1; dma_addr_t counters_phys_addr; @@ -701,6 +700,27 @@ static bool rtl_supports_eee(struct rtl8169_private *tp) tp->mac_version != RTL_GIGA_MAC_VER_39; } +static void rtl_get_priv_stats(struct rtl8169_stats *stats, + u64 *pkts, u64 *bytes) +{ + unsigned int start; + + do { + start = u64_stats_fetch_begin_irq(&stats->syncp); + *pkts = stats->packets; + *bytes = stats->bytes; + } while (u64_stats_fetch_retry_irq(&stats->syncp, start)); +} + +static void rtl_inc_priv_stats(struct rtl8169_stats *stats, + u64 pkts, u64 bytes) +{ + u64_stats_update_begin(&stats->syncp); + stats->packets += pkts; + stats->bytes += bytes; + u64_stats_update_end(&stats->syncp); +} + static void rtl_read_mac_from_reg(struct rtl8169_private *tp, u8 *mac, int reg) { int i; @@ -1280,12 +1300,10 @@ static void rtl_irq_disable(struct rtl8169_private *tp) RTL_W32(tp, IntrMask_8125, 0); else RTL_W16(tp, IntrMask, 0); - tp->irq_enabled = 0; } static void rtl_irq_enable(struct rtl8169_private *tp) { - tp->irq_enabled = 1; if (rtl_is_8125(tp)) RTL_W32(tp, IntrMask_8125, tp->irq_mask); else @@ -4399,10 +4417,7 @@ static void rtl_tx(struct net_device *dev, struct rtl8169_private *tp, if (tp->dirty_tx != dirty_tx) { netdev_completed_queue(dev, pkts_compl, bytes_compl); - u64_stats_update_begin(&tp->tx_stats.syncp); - tp->tx_stats.packets += pkts_compl; - tp->tx_stats.bytes += bytes_compl; - u64_stats_update_end(&tp->tx_stats.syncp); + rtl_inc_priv_stats(&tp->tx_stats, pkts_compl, bytes_compl); tp->dirty_tx = dirty_tx; /* Sync with rtl8169_start_xmit: @@ -4524,11 +4539,7 @@ static int rtl_rx(struct net_device *dev, struct rtl8169_private *tp, u32 budget napi_gro_receive(&tp->napi, skb); - u64_stats_update_begin(&tp->rx_stats.syncp); - tp->rx_stats.packets++; - tp->rx_stats.bytes += pkt_size; - u64_stats_update_end(&tp->rx_stats.syncp); - + rtl_inc_priv_stats(&tp->rx_stats, 1, pkt_size); release_descriptor: rtl8169_mark_to_asic(desc); } @@ -4544,8 +4555,7 @@ static irqreturn_t rtl8169_interrupt(int irq, void *dev_instance) struct rtl8169_private *tp = dev_instance; u32 status = rtl_get_events(tp); - if (!tp->irq_enabled || (status & 0xffff) == 0xffff || - !(status & tp->irq_mask)) + if ((status & 0xffff) == 0xffff || !(status & tp->irq_mask)) return IRQ_NONE; if (unlikely(status & SYSErr)) { @@ -4599,10 +4609,8 @@ static int rtl8169_poll(struct napi_struct *napi, int budget) rtl_tx(dev, tp, budget); - if (work_done < budget) { - napi_complete_done(napi, work_done); + if (work_done < budget && napi_complete_done(napi, work_done)) rtl_irq_enable(tp); - } return work_done; } @@ -4778,23 +4786,13 @@ rtl8169_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats) struct rtl8169_private *tp = netdev_priv(dev); struct pci_dev *pdev = tp->pci_dev; struct rtl8169_counters *counters = tp->counters; - unsigned int start; pm_runtime_get_noresume(&pdev->dev); netdev_stats_to_stats64(stats, &dev->stats); - do { - start = u64_stats_fetch_begin_irq(&tp->rx_stats.syncp); - stats->rx_packets = tp->rx_stats.packets; - stats->rx_bytes = tp->rx_stats.bytes; - } while (u64_stats_fetch_retry_irq(&tp->rx_stats.syncp, start)); - - do { - start = u64_stats_fetch_begin_irq(&tp->tx_stats.syncp); - stats->tx_packets = tp->tx_stats.packets; - stats->tx_bytes = tp->tx_stats.bytes; - } while (u64_stats_fetch_retry_irq(&tp->tx_stats.syncp, start)); + rtl_get_priv_stats(&tp->rx_stats, &stats->rx_packets, &stats->rx_bytes); + rtl_get_priv_stats(&tp->tx_stats, &stats->tx_packets, &stats->tx_bytes); /* * Fetch additional counter values missing in stats collected by driver diff --git a/drivers/net/ethernet/renesas/ravb.h b/drivers/net/ethernet/renesas/ravb.h index 9f88b5db4f89..7453b17a37a2 100644 --- a/drivers/net/ethernet/renesas/ravb.h +++ b/drivers/net/ethernet/renesas/ravb.h @@ -1036,7 +1036,10 @@ struct ravb_private { unsigned no_avb_link:1; unsigned avb_link_active_low:1; unsigned wol_enabled:1; - int num_tx_desc; /* TX descriptors per packet */ + unsigned rxcidm:1; /* RX Clock Internal Delay Mode */ + unsigned txcidm:1; /* TX Clock Internal Delay Mode */ + unsigned rgmii_override:1; /* Deprecated rgmii-*id behavior */ + int num_tx_desc; /* TX descriptors per packet */ }; static inline u32 ravb_read(struct net_device *ndev, enum ravb_reg reg) diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 99f7aae102ce..9c4df4ede011 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -162,7 +162,7 @@ static int ravb_get_mdio_data(struct mdiobb_ctrl *ctrl) } /* MDIO bus control struct */ -static struct mdiobb_ops bb_ops = { +static const struct mdiobb_ops bb_ops = { .owner = THIS_MODULE, .set_mdc = ravb_set_mdc, .set_mdio_dir = ravb_set_mdio_dir, @@ -1034,11 +1034,8 @@ static int ravb_phy_init(struct net_device *ndev) pn = of_node_get(np); } - iface = priv->phy_interface; - if (priv->chip_id != RCAR_GEN2 && phy_interface_mode_is_rgmii(iface)) { - /* ravb_set_delay_mode() takes care of internal delay mode */ - iface = PHY_INTERFACE_MODE_RGMII; - } + iface = priv->rgmii_override ? PHY_INTERFACE_MODE_RGMII + : priv->phy_interface; phydev = of_phy_connect(ndev, pn, ravb_adjust_link, 0, iface); of_node_put(pn); if (!phydev) { @@ -1989,23 +1986,53 @@ static const struct soc_device_attribute ravb_delay_mode_quirk_match[] = { }; /* Set tx and rx clock internal delay modes */ -static void ravb_set_delay_mode(struct net_device *ndev) +static void ravb_parse_delay_mode(struct device_node *np, struct net_device *ndev) { struct ravb_private *priv = netdev_priv(ndev); - int set = 0; + bool explicit_delay = false; + u32 delay; + + if (!of_property_read_u32(np, "rx-internal-delay-ps", &delay)) { + /* Valid values are 0 and 1800, according to DT bindings */ + priv->rxcidm = !!delay; + explicit_delay = true; + } + if (!of_property_read_u32(np, "tx-internal-delay-ps", &delay)) { + /* Valid values are 0 and 2000, according to DT bindings */ + priv->txcidm = !!delay; + explicit_delay = true; + } + if (explicit_delay) + return; + + /* Fall back to legacy rgmii-*id behavior */ if (priv->phy_interface == PHY_INTERFACE_MODE_RGMII_ID || - priv->phy_interface == PHY_INTERFACE_MODE_RGMII_RXID) - set |= APSR_DM_RDM; + priv->phy_interface == PHY_INTERFACE_MODE_RGMII_RXID) { + priv->rxcidm = 1; + priv->rgmii_override = 1; + } if (priv->phy_interface == PHY_INTERFACE_MODE_RGMII_ID || priv->phy_interface == PHY_INTERFACE_MODE_RGMII_TXID) { if (!WARN(soc_device_match(ravb_delay_mode_quirk_match), "phy-mode %s requires TX clock internal delay mode which is not supported by this hardware revision. Please update device tree", - phy_modes(priv->phy_interface))) - set |= APSR_DM_TDM; + phy_modes(priv->phy_interface))) { + priv->txcidm = 1; + priv->rgmii_override = 1; + } } +} + +static void ravb_set_delay_mode(struct net_device *ndev) +{ + struct ravb_private *priv = netdev_priv(ndev); + u32 set = 0; + if (priv->rxcidm) + set |= APSR_DM_RDM; + if (priv->txcidm) + set |= APSR_DM_TDM; ravb_modify(ndev, APSR, APSR_DM, set); } @@ -2138,8 +2165,10 @@ static int ravb_probe(struct platform_device *pdev) /* Request GTI loading */ ravb_modify(ndev, GCCR, GCCR_LTI, GCCR_LTI); - if (priv->chip_id != RCAR_GEN2) + if (priv->chip_id != RCAR_GEN2) { + ravb_parse_delay_mode(np, ndev); ravb_set_delay_mode(ndev); + } /* Allocate descriptor base address table */ priv->desc_bat_size = sizeof(struct ravb_desc) * DBAT_ENTRY_NUM; diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c index f45331ed90b0..c63304632935 100644 --- a/drivers/net/ethernet/renesas/sh_eth.c +++ b/drivers/net/ethernet/renesas/sh_eth.c @@ -45,6 +45,15 @@ #define SH_ETH_OFFSET_DEFAULTS \ [0 ... SH_ETH_MAX_REGISTER_OFFSET - 1] = SH_ETH_OFFSET_INVALID +/* use some intentionally tricky logic here to initialize the whole struct to + * 0xffff, but then override certain fields, requiring us to indicate that we + * "know" that there are overrides in this structure, and we'll need to disable + * that warning from W=1 builds. GCC has supported this option since 4.2.X, but + * the macros available to do this only define GCC 8. + */ +__diag_push(); +__diag_ignore(GCC, 8, "-Woverride-init", + "logic to initialize all and then override some is OK"); static const u16 sh_eth_offset_gigabit[SH_ETH_MAX_REGISTER_OFFSET] = { SH_ETH_OFFSET_DEFAULTS, @@ -332,6 +341,7 @@ static const u16 sh_eth_offset_fast_sh3_sh2[SH_ETH_MAX_REGISTER_OFFSET] = { [TSU_ADRH0] = 0x0100, }; +__diag_pop(); static void sh_eth_rcv_snd_disable(struct net_device *ndev); static struct net_device_stats *sh_eth_get_stats(struct net_device *ndev); @@ -1202,7 +1212,7 @@ static void sh_mdc_ctrl(struct mdiobb_ctrl *ctrl, int bit) } /* mdio bus control struct */ -static struct mdiobb_ops bb_ops = { +static const struct mdiobb_ops bb_ops = { .owner = THIS_MODULE, .set_mdc = sh_mdc_ctrl, .set_mdio_dir = sh_mmd_ctrl, diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c index 9cc31f7e0df1..dd0bc7f0aaee 100644 --- a/drivers/net/ethernet/rocker/rocker_main.c +++ b/drivers/net/ethernet/rocker/rocker_main.c @@ -200,9 +200,9 @@ static int rocker_dma_test_offset(const struct rocker *rocker, buf = alloc + offset; expect = buf + ROCKER_TEST_DMA_BUF_SIZE; - dma_handle = pci_map_single(pdev, buf, ROCKER_TEST_DMA_BUF_SIZE, - PCI_DMA_BIDIRECTIONAL); - if (pci_dma_mapping_error(pdev, dma_handle)) { + dma_handle = dma_map_single(&pdev->dev, buf, ROCKER_TEST_DMA_BUF_SIZE, + DMA_BIDIRECTIONAL); + if (dma_mapping_error(&pdev->dev, dma_handle)) { err = -EIO; goto free_alloc; } @@ -234,8 +234,8 @@ static int rocker_dma_test_offset(const struct rocker *rocker, goto unmap; unmap: - pci_unmap_single(pdev, dma_handle, ROCKER_TEST_DMA_BUF_SIZE, - PCI_DMA_BIDIRECTIONAL); + dma_unmap_single(&pdev->dev, dma_handle, ROCKER_TEST_DMA_BUF_SIZE, + DMA_BIDIRECTIONAL); free_alloc: kfree(alloc); @@ -441,9 +441,9 @@ static int rocker_dma_ring_create(const struct rocker *rocker, if (!info->desc_info) return -ENOMEM; - info->desc = pci_alloc_consistent(rocker->pdev, - info->size * sizeof(*info->desc), - &info->mapaddr); + info->desc = dma_alloc_coherent(&rocker->pdev->dev, + info->size * sizeof(*info->desc), + &info->mapaddr, GFP_KERNEL); if (!info->desc) { kfree(info->desc_info); return -ENOMEM; @@ -465,9 +465,9 @@ static void rocker_dma_ring_destroy(const struct rocker *rocker, { rocker_write64(rocker, DMA_DESC_ADDR(info->type), 0); - pci_free_consistent(rocker->pdev, - info->size * sizeof(struct rocker_desc), - info->desc, info->mapaddr); + dma_free_coherent(&rocker->pdev->dev, + info->size * sizeof(struct rocker_desc), info->desc, + info->mapaddr); kfree(info->desc_info); } @@ -506,8 +506,9 @@ static int rocker_dma_ring_bufs_alloc(const struct rocker *rocker, goto rollback; } - dma_handle = pci_map_single(pdev, buf, buf_size, direction); - if (pci_dma_mapping_error(pdev, dma_handle)) { + dma_handle = dma_map_single(&pdev->dev, buf, buf_size, + direction); + if (dma_mapping_error(&pdev->dev, dma_handle)) { kfree(buf); err = -EIO; goto rollback; @@ -526,7 +527,8 @@ rollback: for (i--; i >= 0; i--) { const struct rocker_desc_info *desc_info = &info->desc_info[i]; - pci_unmap_single(pdev, dma_unmap_addr(desc_info, mapaddr), + dma_unmap_single(&pdev->dev, + dma_unmap_addr(desc_info, mapaddr), desc_info->data_size, direction); kfree(desc_info->data); } @@ -546,7 +548,8 @@ static void rocker_dma_ring_bufs_free(const struct rocker *rocker, desc->buf_addr = 0; desc->buf_size = 0; - pci_unmap_single(pdev, dma_unmap_addr(desc_info, mapaddr), + dma_unmap_single(&pdev->dev, + dma_unmap_addr(desc_info, mapaddr), desc_info->data_size, direction); kfree(desc_info->data); } @@ -615,7 +618,7 @@ static int rocker_dma_rings_init(struct rocker *rocker) spin_lock_init(&rocker->cmd_ring_lock); err = rocker_dma_ring_bufs_alloc(rocker, &rocker->cmd_ring, - PCI_DMA_BIDIRECTIONAL, PAGE_SIZE); + DMA_BIDIRECTIONAL, PAGE_SIZE); if (err) { dev_err(&pdev->dev, "failed to alloc command dma ring buffers\n"); goto err_dma_cmd_ring_bufs_alloc; @@ -636,7 +639,7 @@ static int rocker_dma_rings_init(struct rocker *rocker) } err = rocker_dma_ring_bufs_alloc(rocker, &rocker->event_ring, - PCI_DMA_FROMDEVICE, PAGE_SIZE); + DMA_FROM_DEVICE, PAGE_SIZE); if (err) { dev_err(&pdev->dev, "failed to alloc event dma ring buffers\n"); goto err_dma_event_ring_bufs_alloc; @@ -650,7 +653,7 @@ err_dma_event_ring_create: rocker_dma_cmd_ring_waits_free(rocker); err_dma_cmd_ring_waits_alloc: rocker_dma_ring_bufs_free(rocker, &rocker->cmd_ring, - PCI_DMA_BIDIRECTIONAL); + DMA_BIDIRECTIONAL); err_dma_cmd_ring_bufs_alloc: rocker_dma_ring_destroy(rocker, &rocker->cmd_ring); return err; @@ -659,11 +662,11 @@ err_dma_cmd_ring_bufs_alloc: static void rocker_dma_rings_fini(struct rocker *rocker) { rocker_dma_ring_bufs_free(rocker, &rocker->event_ring, - PCI_DMA_BIDIRECTIONAL); + DMA_BIDIRECTIONAL); rocker_dma_ring_destroy(rocker, &rocker->event_ring); rocker_dma_cmd_ring_waits_free(rocker); rocker_dma_ring_bufs_free(rocker, &rocker->cmd_ring, - PCI_DMA_BIDIRECTIONAL); + DMA_BIDIRECTIONAL); rocker_dma_ring_destroy(rocker, &rocker->cmd_ring); } @@ -675,9 +678,9 @@ static int rocker_dma_rx_ring_skb_map(const struct rocker_port *rocker_port, struct pci_dev *pdev = rocker->pdev; dma_addr_t dma_handle; - dma_handle = pci_map_single(pdev, skb->data, buf_len, - PCI_DMA_FROMDEVICE); - if (pci_dma_mapping_error(pdev, dma_handle)) + dma_handle = dma_map_single(&pdev->dev, skb->data, buf_len, + DMA_FROM_DEVICE); + if (dma_mapping_error(&pdev->dev, dma_handle)) return -EIO; if (rocker_tlv_put_u64(desc_info, ROCKER_TLV_RX_FRAG_ADDR, dma_handle)) goto tlv_put_failure; @@ -686,7 +689,7 @@ static int rocker_dma_rx_ring_skb_map(const struct rocker_port *rocker_port, return 0; tlv_put_failure: - pci_unmap_single(pdev, dma_handle, buf_len, PCI_DMA_FROMDEVICE); + dma_unmap_single(&pdev->dev, dma_handle, buf_len, DMA_FROM_DEVICE); desc_info->tlv_size = 0; return -EMSGSIZE; } @@ -734,7 +737,7 @@ static void rocker_dma_rx_ring_skb_unmap(const struct rocker *rocker, return; dma_handle = rocker_tlv_get_u64(attrs[ROCKER_TLV_RX_FRAG_ADDR]); len = rocker_tlv_get_u16(attrs[ROCKER_TLV_RX_FRAG_MAX_LEN]); - pci_unmap_single(pdev, dma_handle, len, PCI_DMA_FROMDEVICE); + dma_unmap_single(&pdev->dev, dma_handle, len, DMA_FROM_DEVICE); } static void rocker_dma_rx_ring_skb_free(const struct rocker *rocker, @@ -796,7 +799,7 @@ static int rocker_port_dma_rings_init(struct rocker_port *rocker_port) } err = rocker_dma_ring_bufs_alloc(rocker, &rocker_port->tx_ring, - PCI_DMA_TODEVICE, + DMA_TO_DEVICE, ROCKER_DMA_TX_DESC_SIZE); if (err) { netdev_err(rocker_port->dev, "failed to alloc tx dma ring buffers\n"); @@ -813,7 +816,7 @@ static int rocker_port_dma_rings_init(struct rocker_port *rocker_port) } err = rocker_dma_ring_bufs_alloc(rocker, &rocker_port->rx_ring, - PCI_DMA_BIDIRECTIONAL, + DMA_BIDIRECTIONAL, ROCKER_DMA_RX_DESC_SIZE); if (err) { netdev_err(rocker_port->dev, "failed to alloc rx dma ring buffers\n"); @@ -831,12 +834,12 @@ static int rocker_port_dma_rings_init(struct rocker_port *rocker_port) err_dma_rx_ring_skbs_alloc: rocker_dma_ring_bufs_free(rocker, &rocker_port->rx_ring, - PCI_DMA_BIDIRECTIONAL); + DMA_BIDIRECTIONAL); err_dma_rx_ring_bufs_alloc: rocker_dma_ring_destroy(rocker, &rocker_port->rx_ring); err_dma_rx_ring_create: rocker_dma_ring_bufs_free(rocker, &rocker_port->tx_ring, - PCI_DMA_TODEVICE); + DMA_TO_DEVICE); err_dma_tx_ring_bufs_alloc: rocker_dma_ring_destroy(rocker, &rocker_port->tx_ring); return err; @@ -848,10 +851,10 @@ static void rocker_port_dma_rings_fini(struct rocker_port *rocker_port) rocker_dma_rx_ring_skbs_free(rocker_port); rocker_dma_ring_bufs_free(rocker, &rocker_port->rx_ring, - PCI_DMA_BIDIRECTIONAL); + DMA_BIDIRECTIONAL); rocker_dma_ring_destroy(rocker, &rocker_port->rx_ring); rocker_dma_ring_bufs_free(rocker, &rocker_port->tx_ring, - PCI_DMA_TODEVICE); + DMA_TO_DEVICE); rocker_dma_ring_destroy(rocker, &rocker_port->tx_ring); } @@ -1858,7 +1861,7 @@ static void rocker_tx_desc_frags_unmap(const struct rocker_port *rocker_port, continue; dma_handle = rocker_tlv_get_u64(frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_ADDR]); len = rocker_tlv_get_u16(frag_attrs[ROCKER_TLV_TX_FRAG_ATTR_LEN]); - pci_unmap_single(pdev, dma_handle, len, DMA_TO_DEVICE); + dma_unmap_single(&pdev->dev, dma_handle, len, DMA_TO_DEVICE); } } @@ -1871,8 +1874,8 @@ static int rocker_tx_desc_frag_map_put(const struct rocker_port *rocker_port, dma_addr_t dma_handle; struct rocker_tlv *frag; - dma_handle = pci_map_single(pdev, buf, buf_len, DMA_TO_DEVICE); - if (unlikely(pci_dma_mapping_error(pdev, dma_handle))) { + dma_handle = dma_map_single(&pdev->dev, buf, buf_len, DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(&pdev->dev, dma_handle))) { if (net_ratelimit()) netdev_err(rocker_port->dev, "failed to dma map tx frag\n"); return -EIO; @@ -1892,7 +1895,7 @@ static int rocker_tx_desc_frag_map_put(const struct rocker_port *rocker_port, nest_cancel: rocker_tlv_nest_cancel(desc_info, frag); unmap_frag: - pci_unmap_single(pdev, dma_handle, buf_len, DMA_TO_DEVICE); + dma_unmap_single(&pdev->dev, dma_handle, buf_len, DMA_TO_DEVICE); return -EMSGSIZE; } @@ -2905,17 +2908,17 @@ static int rocker_probe(struct pci_dev *pdev, const struct pci_device_id *id) goto err_pci_request_regions; } - err = pci_set_dma_mask(pdev, DMA_BIT_MASK(64)); + err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(64)); if (!err) { - err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64)); + err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(64)); if (err) { - dev_err(&pdev->dev, "pci_set_consistent_dma_mask failed\n"); + dev_err(&pdev->dev, "dma_set_coherent_mask failed\n"); goto err_pci_set_dma_mask; } } else { - err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32)); + err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)); if (err) { - dev_err(&pdev->dev, "pci_set_dma_mask failed\n"); + dev_err(&pdev->dev, "dma_set_mask failed\n"); goto err_pci_set_dma_mask; } } diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c index 2cc8184b7e6b..971f1e54b652 100644 --- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c +++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c @@ -97,7 +97,7 @@ void sxgbe_disable_eee_mode(struct sxgbe_priv_data * const priv) /** * sxgbe_eee_ctrl_timer - * @arg : data hook + * @t: timer list containing a data * Description: * If there is no data transfer and if we are not in LPI state, * then MAC Transmitter can be moved to LPI state. @@ -255,7 +255,7 @@ static void sxgbe_adjust_link(struct net_device *dev) /** * sxgbe_init_phy - PHY initialization - * @dev: net device structure + * @ndev: net device structure * Description: it initializes the driver's PHY state, and attaches the PHY * to the mac driver. * Return value: @@ -364,8 +364,11 @@ static int sxgbe_init_rx_buffers(struct net_device *dev, /** * sxgbe_free_rx_buffers - free what sxgbe_init_rx_buffers() allocated * @dev: net device structure + * @p: dec pointer + * @i: index + * @dma_buf_sz: size * @rx_ring: ring to be freed - * @rx_rsize: ring size + * * Description: this function initializes the DMA RX descriptor */ static void sxgbe_free_rx_buffers(struct net_device *dev, @@ -383,6 +386,7 @@ static void sxgbe_free_rx_buffers(struct net_device *dev, /** * init_tx_ring - init the TX descriptor ring * @dev: net device structure + * @queue_no: queue * @tx_ring: ring to be initialised * @tx_rsize: ring size * Description: this function initializes the DMA TX descriptor @@ -449,6 +453,7 @@ static void free_rx_ring(struct device *dev, struct sxgbe_rx_queue *rx_ring, /** * init_rx_ring - init the RX descriptor ring * @dev: net device structure + * @queue_no: queue * @rx_ring: ring to be initialised * @rx_rsize: ring size * Description: this function initializes the DMA RX descriptor @@ -548,7 +553,7 @@ static void free_tx_ring(struct device *dev, struct sxgbe_tx_queue *tx_ring, /** * init_dma_desc_rings - init the RX/TX descriptor rings - * @dev: net device structure + * @netd: net device structure * Description: this function initializes the DMA RX/TX descriptors * and allocates the socket buffers. It suppors the chained and ring * modes. @@ -724,7 +729,7 @@ static void sxgbe_mtl_operation_mode(struct sxgbe_priv_data *priv) /** * sxgbe_tx_queue_clean: - * @priv: driver private structure + * @tqueue: queue pointer * Description: it reclaims resources after transmission completes. */ static void sxgbe_tx_queue_clean(struct sxgbe_tx_queue *tqueue) @@ -807,6 +812,7 @@ static void sxgbe_tx_all_clean(struct sxgbe_priv_data * const priv) /** * sxgbe_restart_tx_queue: irq tx error mng function * @priv: driver private structure + * @queue_num: queue number * Description: it cleans the descriptors and restarts the transmission * in case of errors. */ @@ -1567,6 +1573,7 @@ static int sxgbe_poll(struct napi_struct *napi, int budget) /** * sxgbe_tx_timeout * @dev : Pointer to net device structure + * @txqueue: index of the hanging queue * Description: this function is called when a packet transmission fails to * complete within a reasonable time. The driver will mark the error in the * netdev structure and arrange for the device to be reset to a sane state diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index 4b0b2cf026a5..da6886dcac37 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -601,10 +601,14 @@ static int efx_ef10_probe(struct efx_nic *efx) efx_ef10_read_licensed_features(efx); /* We can have one VI for each vi_stride-byte region. - * However, until we use TX option descriptors we need two TX queues - * per channel. + * However, until we use TX option descriptors we need up to four + * TX queues per channel for different checksumming combinations. */ - efx->tx_queues_per_channel = 2; + if (nic_data->datapath_caps & + (1 << MC_CMD_GET_CAPABILITIES_OUT_VXLAN_NVGRE_LBN)) + efx->tx_queues_per_channel = 4; + else + efx->tx_queues_per_channel = 2; efx->max_vis = efx_ef10_mem_map_size(efx) / efx->vi_stride; if (!efx->max_vis) { netif_err(efx, drv, efx->net_dev, "error determining max VIs\n"); @@ -1300,6 +1304,7 @@ static void efx_ef10_fini_nic(struct efx_nic *efx) static int efx_ef10_init_nic(struct efx_nic *efx) { struct efx_ef10_nic_data *nic_data = efx->nic_data; + netdev_features_t hw_enc_features = 0; int rc; if (nic_data->must_check_datapath_caps) { @@ -1344,6 +1349,21 @@ static int efx_ef10_init_nic(struct efx_nic *efx) nic_data->must_restore_piobufs = false; } + /* add encapsulated checksum offload features */ + if (efx_has_cap(efx, VXLAN_NVGRE) && !efx_ef10_is_vf(efx)) + hw_enc_features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM; + /* add encapsulated TSO features */ + if (efx_has_cap(efx, TX_TSO_V2_ENCAP)) { + netdev_features_t encap_tso_features; + + encap_tso_features = NETIF_F_GSO_UDP_TUNNEL | NETIF_F_GSO_GRE | + NETIF_F_GSO_UDP_TUNNEL_CSUM | NETIF_F_GSO_GRE_CSUM; + + hw_enc_features |= encap_tso_features | NETIF_F_TSO; + efx->net_dev->features |= encap_tso_features; + } + efx->net_dev->hw_enc_features = hw_enc_features; + /* don't fail init if RSS setup doesn't work */ rc = efx->type->rx_push_rss_config(efx, false, efx->rss_context.rx_indir_table, NULL); @@ -1851,18 +1871,9 @@ static int efx_ef10_try_update_nic_stats_vf(struct efx_nic *efx) spin_unlock_bh(&efx->stats_lock); - if (in_interrupt()) { - /* If in atomic context, cannot update stats. Just update the - * software stats and return so the caller can continue. - */ - spin_lock_bh(&efx->stats_lock); - efx_update_sw_stats(efx, stats); - return 0; - } - efx_ef10_get_stat_mask(efx, mask); - rc = efx_nic_alloc_buffer(efx, &stats_buf, dma_len, GFP_ATOMIC); + rc = efx_nic_alloc_buffer(efx, &stats_buf, dma_len, GFP_KERNEL); if (rc) { spin_lock_bh(&efx->stats_lock); return rc; @@ -1918,6 +1929,18 @@ static size_t efx_ef10_update_stats_vf(struct efx_nic *efx, u64 *full_stats, return efx_ef10_update_stats_common(efx, full_stats, core_stats); } +static size_t efx_ef10_update_stats_atomic_vf(struct efx_nic *efx, u64 *full_stats, + struct rtnl_link_stats64 *core_stats) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + + /* In atomic context, cannot update HW stats. Just update the + * software stats and return so the caller can continue. + */ + efx_update_sw_stats(efx, nic_data->stats); + return efx_ef10_update_stats_common(efx, full_stats, core_stats); +} + static void efx_ef10_push_irq_moderation(struct efx_channel *channel) { struct efx_nic *efx = channel->efx; @@ -2146,6 +2169,9 @@ static int efx_ef10_irq_test_generate(struct efx_nic *efx) static int efx_ef10_tx_probe(struct efx_tx_queue *tx_queue) { + /* low two bits of label are what we want for type */ + BUILD_BUG_ON((EFX_TXQ_TYPE_OUTER_CSUM | EFX_TXQ_TYPE_INNER_CSUM) != 3); + tx_queue->type = tx_queue->label & 3; return efx_nic_alloc_buffer(tx_queue->efx, &tx_queue->txd.buf, (tx_queue->ptr_mask + 1) * sizeof(efx_qword_t), @@ -2168,15 +2194,15 @@ static inline void efx_ef10_push_tx_desc(struct efx_tx_queue *tx_queue, /* Add Firmware-Assisted TSO v2 option descriptors to a queue. */ -static int efx_ef10_tx_tso_desc(struct efx_tx_queue *tx_queue, - struct sk_buff *skb, - bool *data_mapped) +int efx_ef10_tx_tso_desc(struct efx_tx_queue *tx_queue, struct sk_buff *skb, + bool *data_mapped) { struct efx_tx_buffer *buffer; + u16 inner_ipv4_id = 0; + u16 outer_ipv4_id = 0; struct tcphdr *tcp; struct iphdr *ip; - - u16 ipv4_id; + u16 ip_tot_len; u32 seqnum; u32 mss; @@ -2189,21 +2215,43 @@ static int efx_ef10_tx_tso_desc(struct efx_tx_queue *tx_queue, return -EINVAL; } - ip = ip_hdr(skb); + if (skb->encapsulation) { + if (!tx_queue->tso_encap) + return -EINVAL; + ip = ip_hdr(skb); + if (ip->version == 4) + outer_ipv4_id = ntohs(ip->id); + + ip = inner_ip_hdr(skb); + tcp = inner_tcp_hdr(skb); + } else { + ip = ip_hdr(skb); + tcp = tcp_hdr(skb); + } + + /* 8000-series EF10 hardware requires that IP Total Length be + * greater than or equal to the value it will have in each segment + * (which is at most mss + 208 + TCP header length), but also less + * than (0x10000 - inner_network_header). Otherwise the TCP + * checksum calculation will be broken for encapsulated packets. + * We fill in ip->tot_len with 0xff30, which should satisfy the + * first requirement unless the MSS is ridiculously large (which + * should be impossible as the driver max MTU is 9216); it is + * guaranteed to satisfy the second as we only attempt TSO if + * inner_network_header <= 208. + */ + ip_tot_len = -EFX_TSO2_MAX_HDRLEN; + EFX_WARN_ON_ONCE_PARANOID(mss + EFX_TSO2_MAX_HDRLEN + + (tcp->doff << 2u) > ip_tot_len); + if (ip->version == 4) { - /* Modify IPv4 header if needed. */ - ip->tot_len = 0; + ip->tot_len = htons(ip_tot_len); ip->check = 0; - ipv4_id = ntohs(ip->id); + inner_ipv4_id = ntohs(ip->id); } else { - /* Modify IPv6 header if needed. */ - struct ipv6hdr *ipv6 = ipv6_hdr(skb); - - ipv6->payload_len = 0; - ipv4_id = 0; + ((struct ipv6hdr *)ip)->payload_len = htons(ip_tot_len); } - tcp = tcp_hdr(skb); seqnum = ntohl(tcp->seq); buffer = efx_tx_queue_get_insert_buffer(tx_queue); @@ -2216,7 +2264,7 @@ static int efx_ef10_tx_tso_desc(struct efx_tx_queue *tx_queue, ESF_DZ_TX_OPTION_TYPE, ESE_DZ_TX_OPTION_DESC_TSO, ESF_DZ_TX_TSO_OPTION_TYPE, ESE_DZ_TX_TSO_OPTION_DESC_FATSO2A, - ESF_DZ_TX_TSO_IP_ID, ipv4_id, + ESF_DZ_TX_TSO_IP_ID, inner_ipv4_id, ESF_DZ_TX_TSO_TCP_SEQNO, seqnum ); ++tx_queue->insert_count; @@ -2226,11 +2274,12 @@ static int efx_ef10_tx_tso_desc(struct efx_tx_queue *tx_queue, buffer->flags = EFX_TX_BUF_OPTION; buffer->len = 0; buffer->unmap_len = 0; - EFX_POPULATE_QWORD_4(buffer->option, + EFX_POPULATE_QWORD_5(buffer->option, ESF_DZ_TX_DESC_IS_OPT, 1, ESF_DZ_TX_OPTION_TYPE, ESE_DZ_TX_OPTION_DESC_TSO, ESF_DZ_TX_TSO_OPTION_TYPE, ESE_DZ_TX_TSO_OPTION_DESC_FATSO2B, + ESF_DZ_TX_TSO_OUTER_IPID, outer_ipv4_id, ESF_DZ_TX_TSO_TCP_MSS, mss ); ++tx_queue->insert_count; @@ -2254,11 +2303,11 @@ static u32 efx_ef10_tso_versions(struct efx_nic *efx) static void efx_ef10_tx_init(struct efx_tx_queue *tx_queue) { - bool csum_offload = tx_queue->label & EFX_TXQ_TYPE_OFFLOAD; + bool csum_offload = tx_queue->type & EFX_TXQ_TYPE_OUTER_CSUM; + bool inner_csum = tx_queue->type & EFX_TXQ_TYPE_INNER_CSUM; struct efx_channel *channel = tx_queue->channel; struct efx_nic *efx = tx_queue->efx; struct efx_ef10_nic_data *nic_data; - bool tso_v2 = false; efx_qword_t *txd; int rc; @@ -2281,15 +2330,18 @@ static void efx_ef10_tx_init(struct efx_tx_queue *tx_queue) * TSOv2 cannot be used with Hardware timestamping, and is never needed * for XDP tx. */ - if (csum_offload && (nic_data->datapath_caps2 & - (1 << MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_V2_LBN)) && - !tx_queue->timestamping && !tx_queue->xdp_tx) { - tso_v2 = true; - netif_dbg(efx, hw, efx->net_dev, "Using TSOv2 for channel %u\n", - channel->channel); + if (efx_has_cap(efx, TX_TSO_V2)) { + if ((csum_offload || inner_csum) && + !tx_queue->timestamping && !tx_queue->xdp_tx) { + tx_queue->tso_version = 2; + netif_dbg(efx, hw, efx->net_dev, "Using TSOv2 for channel %u\n", + channel->channel); + } + } else if (efx_has_cap(efx, TX_TSO)) { + tx_queue->tso_version = 1; } - rc = efx_mcdi_tx_init(tx_queue, tso_v2); + rc = efx_mcdi_tx_init(tx_queue); if (rc) goto fail; @@ -2302,22 +2354,19 @@ static void efx_ef10_tx_init(struct efx_tx_queue *tx_queue) tx_queue->buffer[0].flags = EFX_TX_BUF_OPTION; tx_queue->insert_count = 1; txd = efx_tx_desc(tx_queue, 0); - EFX_POPULATE_QWORD_5(*txd, + EFX_POPULATE_QWORD_7(*txd, ESF_DZ_TX_DESC_IS_OPT, true, ESF_DZ_TX_OPTION_TYPE, ESE_DZ_TX_OPTION_DESC_CRC_CSUM, ESF_DZ_TX_OPTION_UDP_TCP_CSUM, csum_offload, - ESF_DZ_TX_OPTION_IP_CSUM, csum_offload, + ESF_DZ_TX_OPTION_IP_CSUM, csum_offload && tx_queue->tso_version != 2, + ESF_DZ_TX_OPTION_INNER_UDP_TCP_CSUM, inner_csum, + ESF_DZ_TX_OPTION_INNER_IP_CSUM, inner_csum && tx_queue->tso_version != 2, ESF_DZ_TX_TIMESTAMP, tx_queue->timestamping); tx_queue->write_count = 1; - if (tso_v2) { - tx_queue->handle_tso = efx_ef10_tx_tso_desc; - tx_queue->tso_version = 2; - } else if (nic_data->datapath_caps & - (1 << MC_CMD_GET_CAPABILITIES_OUT_TX_TSO_LBN)) { - tx_queue->tso_version = 1; - } + if (tx_queue->tso_version == 2 && efx_has_cap(efx, TX_TSO_V2_ENCAP)) + tx_queue->tso_encap = true; wmb(); efx_ef10_push_tx_desc(tx_queue, txd); @@ -2367,7 +2416,7 @@ static void efx_ef10_tx_write(struct efx_tx_queue *tx_queue) unsigned int write_ptr; efx_qword_t *txd; - tx_queue->xmit_more_available = false; + tx_queue->xmit_pending = false; if (unlikely(tx_queue->write_count == tx_queue->insert_count)) return; @@ -2880,7 +2929,7 @@ efx_ef10_handle_tx_event(struct efx_channel *channel, efx_qword_t *event) /* Get the transmit queue */ tx_ev_q_label = EFX_QWORD_FIELD(*event, ESF_DZ_TX_QLABEL); tx_queue = efx_channel_get_tx_queue(channel, - tx_ev_q_label % EFX_TXQ_TYPES); + tx_ev_q_label % EFX_MAX_TXQ_PER_CHANNEL); if (!tx_queue->timestamping) { /* Transmit completion */ @@ -3952,10 +4001,10 @@ const struct efx_nic_type efx_hunt_a0_vf_nic_type = { .finish_flr = efx_port_dummy_op_void, .describe_stats = efx_ef10_describe_stats, .update_stats = efx_ef10_update_stats_vf, + .update_stats_atomic = efx_ef10_update_stats_atomic_vf, .start_stats = efx_port_dummy_op_void, .pull_stats = efx_port_dummy_op_void, .stop_stats = efx_port_dummy_op_void, - .set_id_led = efx_mcdi_set_id_led, .push_irq_moderation = efx_ef10_push_irq_moderation, .reconfigure_mac = efx_ef10_mac_reconfigure, .check_mac_fault = efx_mcdi_mac_check_fault, @@ -4066,7 +4115,6 @@ const struct efx_nic_type efx_hunt_a0_nic_type = { .start_stats = efx_mcdi_mac_start_stats, .pull_stats = efx_mcdi_mac_pull_stats, .stop_stats = efx_mcdi_mac_stop_stats, - .set_id_led = efx_mcdi_set_id_led, .push_irq_moderation = efx_ef10_push_irq_moderation, .reconfigure_mac = efx_ef10_mac_reconfigure, .check_mac_fault = efx_mcdi_mac_check_fault, diff --git a/drivers/net/ethernet/sfc/ef100_ethtool.c b/drivers/net/ethernet/sfc/ef100_ethtool.c index 729c425d0f78..835c838b7dfa 100644 --- a/drivers/net/ethernet/sfc/ef100_ethtool.c +++ b/drivers/net/ethernet/sfc/ef100_ethtool.c @@ -17,8 +17,49 @@ #include "ef100_ethtool.h" #include "mcdi_functions.h" +/* This is the maximum number of descriptor rings supported by the QDMA */ +#define EFX_EF100_MAX_DMAQ_SIZE 16384UL + +static void ef100_ethtool_get_ringparam(struct net_device *net_dev, + struct ethtool_ringparam *ring) +{ + struct efx_nic *efx = netdev_priv(net_dev); + + ring->rx_max_pending = EFX_EF100_MAX_DMAQ_SIZE; + ring->tx_max_pending = EFX_EF100_MAX_DMAQ_SIZE; + ring->rx_pending = efx->rxq_entries; + ring->tx_pending = efx->txq_entries; +} + /* Ethtool options available */ const struct ethtool_ops ef100_ethtool_ops = { .get_drvinfo = efx_ethtool_get_drvinfo, + .get_msglevel = efx_ethtool_get_msglevel, + .set_msglevel = efx_ethtool_set_msglevel, + .get_pauseparam = efx_ethtool_get_pauseparam, + .set_pauseparam = efx_ethtool_set_pauseparam, + .get_sset_count = efx_ethtool_get_sset_count, + .self_test = efx_ethtool_self_test, + .get_strings = efx_ethtool_get_strings, + .get_link_ksettings = efx_ethtool_get_link_ksettings, + .set_link_ksettings = efx_ethtool_set_link_ksettings, + .get_link = ethtool_op_get_link, + .get_ringparam = ef100_ethtool_get_ringparam, + .get_fecparam = efx_ethtool_get_fecparam, + .set_fecparam = efx_ethtool_set_fecparam, + .get_ethtool_stats = efx_ethtool_get_stats, + .get_rxnfc = efx_ethtool_get_rxnfc, + .set_rxnfc = efx_ethtool_set_rxnfc, + .reset = efx_ethtool_reset, + + .get_rxfh_indir_size = efx_ethtool_get_rxfh_indir_size, + .get_rxfh_key_size = efx_ethtool_get_rxfh_key_size, + .get_rxfh = efx_ethtool_get_rxfh, + .set_rxfh = efx_ethtool_set_rxfh, + .get_rxfh_context = efx_ethtool_get_rxfh_context, + .set_rxfh_context = efx_ethtool_set_rxfh_context, + + .get_module_info = efx_ethtool_get_module_info, + .get_module_eeprom = efx_ethtool_get_module_eeprom, }; diff --git a/drivers/net/ethernet/sfc/ef100_netdev.c b/drivers/net/ethernet/sfc/ef100_netdev.c index 63c311ba28b9..67fe44db6b61 100644 --- a/drivers/net/ethernet/sfc/ef100_netdev.c +++ b/drivers/net/ethernet/sfc/ef100_netdev.c @@ -217,9 +217,13 @@ static const struct net_device_ops ef100_netdev_ops = { .ndo_open = ef100_net_open, .ndo_stop = ef100_net_stop, .ndo_start_xmit = ef100_hard_start_xmit, + .ndo_tx_timeout = efx_watchdog, .ndo_get_stats64 = efx_net_stats, + .ndo_change_mtu = efx_change_mtu, .ndo_validate_addr = eth_validate_addr, + .ndo_set_mac_address = efx_set_mac_address, .ndo_set_rx_mode = efx_set_rx_mode, /* Lookout */ + .ndo_set_features = efx_set_features, .ndo_get_phys_port_id = efx_get_phys_port_id, .ndo_get_phys_port_name = efx_get_phys_port_name, #ifdef CONFIG_RFS_ACCEL diff --git a/drivers/net/ethernet/sfc/ef100_nic.c b/drivers/net/ethernet/sfc/ef100_nic.c index 19fe86b3b316..3148fe770356 100644 --- a/drivers/net/ethernet/sfc/ef100_nic.c +++ b/drivers/net/ethernet/sfc/ef100_nic.c @@ -428,24 +428,12 @@ static int ef100_reset(struct efx_nic *efx, enum reset_type reset_type) __clear_bit(reset_type, &efx->reset_pending); rc = dev_open(efx->net_dev, NULL); } else if (reset_type == RESET_TYPE_ALL) { - /* A RESET_TYPE_ALL will cause filters to be removed, so we remove filters - * and reprobe after reset to avoid removing filters twice - */ - down_write(&efx->filter_sem); - ef100_filter_table_down(efx); - up_write(&efx->filter_sem); rc = efx_mcdi_reset(efx, reset_type); if (rc) return rc; netif_device_attach(efx->net_dev); - down_write(&efx->filter_sem); - rc = ef100_filter_table_up(efx); - up_write(&efx->filter_sem); - if (rc) - return rc; - rc = dev_open(efx->net_dev, NULL); } else { rc = 1; /* Leave the device closed */ @@ -696,7 +684,7 @@ static unsigned int ef100_check_caps(const struct efx_nic *efx, /* NIC level access functions */ #define EF100_OFFLOAD_FEATURES (NETIF_F_HW_CSUM | NETIF_F_RXCSUM | \ - NETIF_F_HIGHDMA | NETIF_F_SG | NETIF_F_FRAGLIST | \ + NETIF_F_HIGHDMA | NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_NTUPLE | \ NETIF_F_RXHASH | NETIF_F_RXFCS | NETIF_F_TSO_ECN | NETIF_F_RXALL | \ NETIF_F_TSO_MANGLEID | NETIF_F_HW_VLAN_CTAG_TX) @@ -769,6 +757,7 @@ const struct efx_nic_type ef100_pf_nic_type = { .rx_restore_rss_contexts = efx_mcdi_rx_restore_rss_contexts, .reconfigure_mac = ef100_reconfigure_mac, + .reconfigure_port = efx_mcdi_port_reconfigure, .test_nvram = efx_new_mcdi_nvram_test_all, .describe_stats = ef100_describe_stats, .start_stats = efx_mcdi_mac_start_stats, @@ -1172,6 +1161,10 @@ static int ef100_probe_main(struct efx_nic *efx) rc = efx_mcdi_reset(efx, RESET_TYPE_ALL); if (rc) goto fail; + /* Enable event logging */ + rc = efx_mcdi_log_ctrl(efx, true, false, 0); + if (rc) + goto fail; rc = efx_get_pf_index(efx, &nic_data->pf_index); if (rc) @@ -1207,10 +1200,6 @@ static int ef100_probe_main(struct efx_nic *efx) if (rc) goto fail; - rc = efx_init_channels(efx); - if (rc) - goto fail; - down_write(&efx->filter_sem); rc = ef100_filter_table_probe(efx); up_write(&efx->filter_sem); diff --git a/drivers/net/ethernet/sfc/ef100_tx.c b/drivers/net/ethernet/sfc/ef100_tx.c index a09546e43408..a90e5a9d2a37 100644 --- a/drivers/net/ethernet/sfc/ef100_tx.c +++ b/drivers/net/ethernet/sfc/ef100_tx.c @@ -27,7 +27,6 @@ int ef100_tx_probe(struct efx_tx_queue *tx_queue) (tx_queue->ptr_mask + 2) * sizeof(efx_oword_t), GFP_KERNEL); - return 0; } void ef100_tx_init(struct efx_tx_queue *tx_queue) @@ -38,7 +37,14 @@ void ef100_tx_init(struct efx_tx_queue *tx_queue) tx_queue->channel->channel - tx_queue->efx->tx_channel_offset); - if (efx_mcdi_tx_init(tx_queue, false)) + /* This value is purely documentational; as EF100 never passes through + * the switch statement in tx.c:__efx_enqueue_skb(), that switch does + * not handle case 3. EF100's TSOv3 descriptors are generated by + * ef100_make_tso_desc(). + * Meanwhile, all efx_mcdi_tx_init() cares about is that it's not 2. + */ + tx_queue->tso_version = 3; + if (efx_mcdi_tx_init(tx_queue)) netdev_WARN(tx_queue->efx->net_dev, "failed to initialise TXQ %d\n", tx_queue->queue); } @@ -117,11 +123,13 @@ static efx_oword_t *ef100_tx_desc(struct efx_tx_queue *tx_queue, unsigned int in return NULL; } -void ef100_notify_tx_desc(struct efx_tx_queue *tx_queue) +static void ef100_notify_tx_desc(struct efx_tx_queue *tx_queue) { unsigned int write_ptr; efx_dword_t reg; + tx_queue->xmit_pending = false; + if (unlikely(tx_queue->notify_count == tx_queue->write_count)) return; @@ -131,7 +139,6 @@ void ef100_notify_tx_desc(struct efx_tx_queue *tx_queue) efx_writed_page(tx_queue->efx, ®, ER_GZ_TX_RING_DOORBELL, tx_queue->queue); tx_queue->notify_count = tx_queue->write_count; - tx_queue->xmit_more_available = false; } static void ef100_tx_push_buffers(struct efx_tx_queue *tx_queue) @@ -359,28 +366,31 @@ int ef100_enqueue_skb(struct efx_tx_queue *tx_queue, struct sk_buff *skb) goto err; ef100_tx_make_descriptors(tx_queue, skb, segments); - fill_level = efx_channel_tx_fill_level(tx_queue->channel); + fill_level = efx_channel_tx_old_fill_level(tx_queue->channel); if (fill_level > efx->txq_stop_thresh) { + struct efx_tx_queue *txq2; + netif_tx_stop_queue(tx_queue->core_txq); /* Re-read after a memory barrier in case we've raced with * the completion path. Otherwise there's a danger we'll never * restart the queue if all completions have just happened. */ smp_mb(); - fill_level = efx_channel_tx_fill_level(tx_queue->channel); + efx_for_each_channel_tx_queue(txq2, tx_queue->channel) + txq2->old_read_count = READ_ONCE(txq2->read_count); + fill_level = efx_channel_tx_old_fill_level(tx_queue->channel); if (fill_level < efx->txq_stop_thresh) netif_tx_start_queue(tx_queue->core_txq); } - if (__netdev_tx_sent_queue(tx_queue->core_txq, skb->len, xmit_more)) - tx_queue->xmit_more_available = false; /* push doorbell */ - else if (tx_queue->write_count - tx_queue->notify_count > 255) - /* Ensure we never push more than 256 packets at once */ - tx_queue->xmit_more_available = false; /* push */ - else - tx_queue->xmit_more_available = true; /* don't push yet */ + tx_queue->xmit_pending = true; - if (!tx_queue->xmit_more_available) + /* If xmit_more then we don't need to push the doorbell, unless there + * are 256 descriptors already queued in which case we have to push to + * ensure we never push more than 256 at once. + */ + if (__netdev_tx_sent_queue(tx_queue->core_txq, skb->len, xmit_more) || + tx_queue->write_count - tx_queue->notify_count > 255) ef100_tx_push_buffers(tx_queue); if (segments) { @@ -399,10 +409,10 @@ err: /* If we're not expecting another transmit and we had something to push * on this queue then we need to push here to get the previous packets - * out. We only enter this branch from before the 'Update BQL' section - * above, so xmit_more_available still refers to the old state. + * out. We only enter this branch from before the xmit_more handling + * above, so xmit_pending still refers to the old state. */ - if (tx_queue->xmit_more_available && !xmit_more) + if (tx_queue->xmit_pending && !xmit_more) ef100_tx_push_buffers(tx_queue); return rc; } diff --git a/drivers/net/ethernet/sfc/ef100_tx.h b/drivers/net/ethernet/sfc/ef100_tx.h index fa23e430bdd7..ddc4b98fa6db 100644 --- a/drivers/net/ethernet/sfc/ef100_tx.h +++ b/drivers/net/ethernet/sfc/ef100_tx.h @@ -17,7 +17,6 @@ int ef100_tx_probe(struct efx_tx_queue *tx_queue); void ef100_tx_init(struct efx_tx_queue *tx_queue); void ef100_tx_write(struct efx_tx_queue *tx_queue); -void ef100_notify_tx_desc(struct efx_tx_queue *tx_queue); unsigned int ef100_tx_max_skb_descs(struct efx_nic *efx); void ef100_ev_tx(struct efx_channel *channel, const efx_qword_t *p_event); diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c index e06fa89f2d72..718308076341 100644 --- a/drivers/net/ethernet/sfc/efx.c +++ b/drivers/net/ethernet/sfc/efx.c @@ -33,7 +33,7 @@ #include "selftest.h" #include "sriov.h" -#include "mcdi.h" +#include "mcdi_port_common.h" #include "mcdi_pcol.h" #include "workarounds.h" @@ -149,23 +149,17 @@ static int efx_init_port(struct efx_nic *efx) mutex_lock(&efx->mac_lock); - rc = efx->phy_op->init(efx); - if (rc) - goto fail1; - efx->port_initialized = true; /* Ensure the PHY advertises the correct flow control settings */ - rc = efx->phy_op->reconfigure(efx); + rc = efx_mcdi_port_reconfigure(efx); if (rc && rc != -EPERM) - goto fail2; + goto fail; mutex_unlock(&efx->mac_lock); return 0; -fail2: - efx->phy_op->fini(efx); -fail1: +fail: mutex_unlock(&efx->mac_lock); return rc; } @@ -177,7 +171,6 @@ static void efx_fini_port(struct efx_nic *efx) if (!efx->port_initialized) return; - efx->phy_op->fini(efx); efx->port_initialized = false; efx->link_state.up = false; @@ -603,6 +596,7 @@ static const struct net_device_ops efx_netdev_ops = { .ndo_set_mac_address = efx_set_mac_address, .ndo_set_rx_mode = efx_set_rx_mode, .ndo_set_features = efx_set_features, + .ndo_features_check = efx_features_check, .ndo_vlan_rx_add_vid = efx_vlan_rx_add_vid, .ndo_vlan_rx_kill_vid = efx_vlan_rx_kill_vid, #ifdef CONFIG_SFC_SRIOV @@ -1229,7 +1223,7 @@ static int efx_pm_thaw(struct device *dev) goto fail; mutex_lock(&efx->mac_lock); - efx->phy_op->reconfigure(efx); + efx_mcdi_port_reconfigure(efx); mutex_unlock(&efx->mac_lock); efx_start_all(efx); @@ -1336,7 +1330,7 @@ static int __init efx_init_module(void) { int rc; - printk(KERN_INFO "Solarflare NET driver v" EFX_DRIVER_VERSION "\n"); + printk(KERN_INFO "Solarflare NET driver\n"); rc = register_netdevice_notifier(&efx_netdev_notifier); if (rc) @@ -1398,4 +1392,3 @@ MODULE_AUTHOR("Solarflare Communications and " MODULE_DESCRIPTION("Solarflare network driver"); MODULE_LICENSE("GPL"); MODULE_DEVICE_TABLE(pci, efx_pci_table); -MODULE_VERSION(EFX_DRIVER_VERSION); diff --git a/drivers/net/ethernet/sfc/efx_channels.c b/drivers/net/ethernet/sfc/efx_channels.c index dd4f30ea48a8..a4a626e9cd9a 100644 --- a/drivers/net/ethernet/sfc/efx_channels.c +++ b/drivers/net/ethernet/sfc/efx_channels.c @@ -151,7 +151,7 @@ static int efx_allocate_msix_channels(struct efx_nic *efx, */ n_xdp_tx = num_possible_cpus(); - n_xdp_ev = DIV_ROUND_UP(n_xdp_tx, EFX_TXQ_TYPES); + n_xdp_ev = DIV_ROUND_UP(n_xdp_tx, EFX_MAX_TXQ_PER_CHANNEL); vec_count = pci_msix_vec_count(efx->pci_dev); if (vec_count < 0) @@ -179,7 +179,7 @@ static int efx_allocate_msix_channels(struct efx_nic *efx, efx->xdp_tx_queue_count = 0; } else { efx->n_xdp_channels = n_xdp_ev; - efx->xdp_tx_per_channel = EFX_TXQ_TYPES; + efx->xdp_tx_per_channel = EFX_MAX_TXQ_PER_CHANNEL; efx->xdp_tx_queue_count = n_xdp_tx; n_channels += n_xdp_ev; netif_dbg(efx, drv, efx->net_dev, @@ -505,8 +505,7 @@ static void efx_filter_rfs_expire(struct work_struct *data) #endif /* Allocate and initialise a channel structure. */ -struct efx_channel * -efx_alloc_channel(struct efx_nic *efx, int i, struct efx_channel *old_channel) +static struct efx_channel *efx_alloc_channel(struct efx_nic *efx, int i) { struct efx_rx_queue *rx_queue; struct efx_tx_queue *tx_queue; @@ -521,7 +520,7 @@ efx_alloc_channel(struct efx_nic *efx, int i, struct efx_channel *old_channel) channel->channel = i; channel->type = &efx_default_channel_type; - for (j = 0; j < EFX_TXQ_TYPES; j++) { + for (j = 0; j < EFX_MAX_TXQ_PER_CHANNEL; j++) { tx_queue = &channel->tx_queue[j]; tx_queue->efx = efx; tx_queue->queue = -1; @@ -545,7 +544,7 @@ int efx_init_channels(struct efx_nic *efx) unsigned int i; for (i = 0; i < EFX_MAX_CHANNELS; i++) { - efx->channel[i] = efx_alloc_channel(efx, i, NULL); + efx->channel[i] = efx_alloc_channel(efx, i); if (!efx->channel[i]) return -ENOMEM; efx->msi_context[i].efx = efx; @@ -595,7 +594,7 @@ struct efx_channel *efx_copy_channel(const struct efx_channel *old_channel) channel->napi_str.state = 0; memset(&channel->eventq, 0, sizeof(channel->eventq)); - for (j = 0; j < EFX_TXQ_TYPES; j++) { + for (j = 0; j < EFX_MAX_TXQ_PER_CHANNEL; j++) { tx_queue = &channel->tx_queue[j]; if (tx_queue->channel) tx_queue->channel = channel; @@ -895,7 +894,7 @@ int efx_set_channels(struct efx_nic *efx) xdp_queue_number, tx_queue->queue); /* We may have a few left-over XDP TX * queues owing to xdp_tx_queue_count - * not dividing evenly by EFX_TXQ_TYPES. + * not dividing evenly by EFX_MAX_TXQ_PER_CHANNEL. * We still allocate and probe those * TXQs, but never use them. */ diff --git a/drivers/net/ethernet/sfc/efx_channels.h b/drivers/net/ethernet/sfc/efx_channels.h index 2d71dc9a33dd..d77ec1f77fb1 100644 --- a/drivers/net/ethernet/sfc/efx_channels.h +++ b/drivers/net/ethernet/sfc/efx_channels.h @@ -31,8 +31,6 @@ void efx_stop_eventq(struct efx_channel *channel); void efx_fini_eventq(struct efx_channel *channel); void efx_remove_eventq(struct efx_channel *channel); -struct efx_channel * -efx_alloc_channel(struct efx_nic *efx, int i, struct efx_channel *old_channel); int efx_realloc_channels(struct efx_nic *efx, u32 rxq_entries, u32 txq_entries); void efx_get_channel_name(struct efx_channel *channel, char *buf, size_t len); void efx_set_channel_names(struct efx_nic *efx); diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c index dfc6032e75f4..72a3f0e09f52 100644 --- a/drivers/net/ethernet/sfc/efx_common.c +++ b/drivers/net/ethernet/sfc/efx_common.c @@ -11,6 +11,7 @@ #include "net_driver.h" #include <linux/module.h> #include <linux/netdevice.h> +#include <net/gre.h> #include "efx_common.h" #include "efx_channels.h" #include "efx.h" @@ -19,6 +20,7 @@ #include "rx_common.h" #include "tx_common.h" #include "nic.h" +#include "mcdi_port_common.h" #include "io.h" #include "mcdi_pcol.h" @@ -544,7 +546,7 @@ void efx_start_all(struct efx_nic *efx) * to poll now because we could have missed a change */ mutex_lock(&efx->mac_lock); - if (efx->phy_op->poll(efx)) + if (efx_mcdi_phy_poll(efx)) efx_link_status_changed(efx); mutex_unlock(&efx->mac_lock); @@ -600,7 +602,7 @@ void efx_net_stats(struct net_device *net_dev, struct rtnl_link_stats64 *stats) struct efx_nic *efx = netdev_priv(net_dev); spin_lock_bh(&efx->stats_lock); - efx->type->update_stats(efx, NULL, stats); + efx_nic_update_stats_atomic(efx, NULL, stats); spin_unlock_bh(&efx->stats_lock); } @@ -714,9 +716,6 @@ void efx_reset_down(struct efx_nic *efx, enum reset_type method) mutex_lock(&efx->mac_lock); down_write(&efx->filter_sem); mutex_lock(&efx->rss_lock); - if (efx->port_initialized && method != RESET_TYPE_INVISIBLE && - method != RESET_TYPE_DATAPATH) - efx->phy_op->fini(efx); efx->type->fini(efx); } @@ -759,10 +758,7 @@ int efx_reset_up(struct efx_nic *efx, enum reset_type method, bool ok) if (efx->port_initialized && method != RESET_TYPE_INVISIBLE && method != RESET_TYPE_DATAPATH) { - rc = efx->phy_op->init(efx); - if (rc) - goto fail; - rc = efx->phy_op->reconfigure(efx); + rc = efx_mcdi_port_reconfigure(efx); if (rc && rc != -EPERM) netif_err(efx, drv, efx->net_dev, "could not restore PHY settings\n"); @@ -959,7 +955,7 @@ void efx_schedule_reset(struct efx_nic *efx, enum reset_type type) /************************************************************************** * - * Dummy PHY/MAC operations + * Dummy NIC operations * * Can be used for some unimplemented operations * Needed so all function pointers are valid and do not have to be tested @@ -972,18 +968,6 @@ int efx_port_dummy_op_int(struct efx_nic *efx) } void efx_port_dummy_op_void(struct efx_nic *efx) {} -static bool efx_port_dummy_op_poll(struct efx_nic *efx) -{ - return false; -} - -static const struct efx_phy_operations efx_dummy_phy_operations = { - .init = efx_port_dummy_op_int, - .reconfigure = efx_port_dummy_op_int, - .poll = efx_port_dummy_op_poll, - .fini = efx_port_dummy_op_void, -}; - /************************************************************************** * * Data housekeeping @@ -1037,7 +1021,6 @@ int efx_init_struct(struct efx_nic *efx, efx->rps_hash_table = kcalloc(EFX_ARFS_HASH_TABLE_SIZE, sizeof(*efx->rps_hash_table), GFP_KERNEL); #endif - efx->phy_op = &efx_dummy_phy_operations; efx->mdio.dev = net_dev; INIT_WORK(&efx->mac_work, efx_mac_work); init_waitqueue_head(&efx->flush_wq); @@ -1104,17 +1087,7 @@ int efx_init_io(struct efx_nic *efx, int bar, dma_addr_t dma_mask, pci_set_master(pci_dev); - /* Set the PCI DMA mask. Try all possibilities from our - * genuine mask down to 32 bits, because some architectures - * (e.g. x86_64 with iommu_sac_force set) will allow 40 bit - * masks event though they reject 46 bit masks. - */ - while (dma_mask > 0x7fffffffUL) { - rc = dma_set_mask_and_coherent(&pci_dev->dev, dma_mask); - if (rc == 0) - break; - dma_mask >>= 1; - } + rc = dma_set_mask_and_coherent(&pci_dev->dev, dma_mask); if (rc) { netif_err(efx, probe, efx->net_dev, "could not find a suitable DMA mask\n"); @@ -1315,6 +1288,89 @@ const struct pci_error_handlers efx_err_handlers = { .resume = efx_io_resume, }; +/* Determine whether the NIC will be able to handle TX offloads for a given + * encapsulated packet. + */ +static bool efx_can_encap_offloads(struct efx_nic *efx, struct sk_buff *skb) +{ + struct gre_base_hdr *greh; + __be16 dst_port; + u8 ipproto; + + /* Does the NIC support encap offloads? + * If not, we should never get here, because we shouldn't have + * advertised encap offload feature flags in the first place. + */ + if (WARN_ON_ONCE(!efx->type->udp_tnl_has_port)) + return false; + + /* Determine encapsulation protocol in use */ + switch (skb->protocol) { + case htons(ETH_P_IP): + ipproto = ip_hdr(skb)->protocol; + break; + case htons(ETH_P_IPV6): + /* If there are extension headers, this will cause us to + * think we can't offload something that we maybe could have. + */ + ipproto = ipv6_hdr(skb)->nexthdr; + break; + default: + /* Not IP, so can't offload it */ + return false; + } + switch (ipproto) { + case IPPROTO_GRE: + /* We support NVGRE but not IP over GRE or random gretaps. + * Specifically, the NIC will accept GRE as encapsulated if + * the inner protocol is Ethernet, but only handle it + * correctly if the GRE header is 8 bytes long. Moreover, + * it will not update the Checksum or Sequence Number fields + * if they are present. (The Routing Present flag, + * GRE_ROUTING, cannot be set else the header would be more + * than 8 bytes long; so we don't have to worry about it.) + */ + if (skb->inner_protocol_type != ENCAP_TYPE_ETHER) + return false; + if (ntohs(skb->inner_protocol) != ETH_P_TEB) + return false; + if (skb_inner_mac_header(skb) - skb_transport_header(skb) != 8) + return false; + greh = (struct gre_base_hdr *)skb_transport_header(skb); + return !(greh->flags & (GRE_CSUM | GRE_SEQ)); + case IPPROTO_UDP: + /* If the port is registered for a UDP tunnel, we assume the + * packet is for that tunnel, and the NIC will handle it as + * such. If not, the NIC won't know what to do with it. + */ + dst_port = udp_hdr(skb)->dest; + return efx->type->udp_tnl_has_port(efx, dst_port); + default: + return false; + } +} + +netdev_features_t efx_features_check(struct sk_buff *skb, struct net_device *dev, + netdev_features_t features) +{ + struct efx_nic *efx = netdev_priv(dev); + + if (skb->encapsulation) { + if (features & NETIF_F_GSO_MASK) + /* Hardware can only do TSO with at most 208 bytes + * of headers. + */ + if (skb_inner_transport_offset(skb) > + EFX_TSO2_MAX_HDRLEN) + features &= ~(NETIF_F_GSO_MASK); + if (features & (NETIF_F_GSO_MASK | NETIF_F_CSUM_MASK)) + if (!efx_can_encap_offloads(efx, skb)) + features &= ~(NETIF_F_GSO_MASK | + NETIF_F_CSUM_MASK); + } + return features; +} + int efx_get_phys_port_id(struct net_device *net_dev, struct netdev_phys_item_id *ppid) { diff --git a/drivers/net/ethernet/sfc/efx_common.h b/drivers/net/ethernet/sfc/efx_common.h index 4056f68f04e5..65513fd0cf6c 100644 --- a/drivers/net/ethernet/sfc/efx_common.h +++ b/drivers/net/ethernet/sfc/efx_common.h @@ -105,6 +105,9 @@ int efx_change_mtu(struct net_device *net_dev, int new_mtu); extern const struct pci_error_handlers efx_err_handlers; +netdev_features_t efx_features_check(struct sk_buff *skb, struct net_device *dev, + netdev_features_t features); + int efx_get_phys_port_id(struct net_device *net_dev, struct netdev_phys_item_id *ppid); diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c index 4ffda7782f68..12a91c559aa2 100644 --- a/drivers/net/ethernet/sfc/ethtool.c +++ b/drivers/net/ethernet/sfc/ethtool.c @@ -50,8 +50,7 @@ static int efx_ethtool_phys_id(struct net_device *net_dev, return 1; /* cycle on/off once per second */ } - efx->type->set_id_led(efx, mode); - return 0; + return efx_mcdi_set_id_led(efx, mode); } static int efx_ethtool_get_regs_len(struct net_device *net_dev) diff --git a/drivers/net/ethernet/sfc/ethtool_common.c b/drivers/net/ethernet/sfc/ethtool_common.c index 05ac87807929..bf1443539a1a 100644 --- a/drivers/net/ethernet/sfc/ethtool_common.c +++ b/drivers/net/ethernet/sfc/ethtool_common.c @@ -15,6 +15,7 @@ #include "selftest.h" #include "rx_common.h" #include "ethtool_common.h" +#include "mcdi_port_common.h" struct efx_sw_stat_desc { const char *name; @@ -105,7 +106,6 @@ void efx_ethtool_get_drvinfo(struct net_device *net_dev, struct efx_nic *efx = netdev_priv(net_dev); strlcpy(info->driver, KBUILD_MODNAME, sizeof(info->driver)); - strlcpy(info->version, EFX_DRIVER_VERSION, sizeof(info->version)); efx_mcdi_print_fwver(efx, info->fw_version, sizeof(info->fw_version)); strlcpy(info->bus_info, pci_name(efx->pci_dev), sizeof(info->bus_info)); @@ -221,7 +221,7 @@ int efx_ethtool_set_pauseparam(struct net_device *net_dev, efx_link_set_wanted_fc(efx, wanted_fc); if (efx->link_advertising[0] != old_adv || (efx->wanted_fc ^ old_fc) & EFX_FC_AUTO) { - rc = efx->phy_op->reconfigure(efx); + rc = efx_mcdi_port_reconfigure(efx); if (rc) { netif_err(efx, drv, efx->net_dev, "Unable to advertise requested flow " @@ -372,20 +372,15 @@ int efx_ethtool_fill_self_tests(struct efx_nic *efx, efx_fill_test(n++, strings, data, &tests->registers, "core", 0, "registers", NULL); - if (efx->phy_op->run_tests != NULL) { - EFX_WARN_ON_PARANOID(efx->phy_op->test_name == NULL); + for (i = 0; true; ++i) { + const char *name; - for (i = 0; true; ++i) { - const char *name; - - EFX_WARN_ON_PARANOID(i >= EFX_MAX_PHY_TESTS); - name = efx->phy_op->test_name(efx, i); - if (name == NULL) - break; + EFX_WARN_ON_PARANOID(i >= EFX_MAX_PHY_TESTS); + name = efx_mcdi_phy_test_name(efx, i); + if (name == NULL) + break; - efx_fill_test(n++, strings, data, &tests->phy_ext[i], - "phy", 0, name, NULL); - } + efx_fill_test(n++, strings, data, &tests->phy_ext[i], "phy", 0, name, NULL); } /* Loopback tests */ @@ -412,7 +407,7 @@ static size_t efx_describe_per_queue_stats(struct efx_nic *efx, u8 *strings) snprintf(strings, ETH_GSTRING_LEN, "tx-%u.tx_packets", channel->tx_queue[0].queue / - EFX_TXQ_TYPES); + EFX_MAX_TXQ_PER_CHANNEL); strings += ETH_GSTRING_LEN; } @@ -571,7 +566,7 @@ int efx_ethtool_get_link_ksettings(struct net_device *net_dev, u32 supported; mutex_lock(&efx->mac_lock); - efx->phy_op->get_link_ksettings(efx, cmd); + efx_mcdi_phy_get_link_ksettings(efx, cmd); mutex_unlock(&efx->mac_lock); /* Both MACs support pause frames (bidirectional and respond-only) */ @@ -607,7 +602,7 @@ int efx_ethtool_set_link_ksettings(struct net_device *net_dev, } mutex_lock(&efx->mac_lock); - rc = efx->phy_op->set_link_ksettings(efx, cmd); + rc = efx_mcdi_phy_set_link_ksettings(efx, cmd); mutex_unlock(&efx->mac_lock); return rc; } @@ -618,10 +613,8 @@ int efx_ethtool_get_fecparam(struct net_device *net_dev, struct efx_nic *efx = netdev_priv(net_dev); int rc; - if (!efx->phy_op || !efx->phy_op->get_fecparam) - return -EOPNOTSUPP; mutex_lock(&efx->mac_lock); - rc = efx->phy_op->get_fecparam(efx, fecparam); + rc = efx_mcdi_phy_get_fecparam(efx, fecparam); mutex_unlock(&efx->mac_lock); return rc; @@ -633,10 +626,8 @@ int efx_ethtool_set_fecparam(struct net_device *net_dev, struct efx_nic *efx = netdev_priv(net_dev); int rc; - if (!efx->phy_op || !efx->phy_op->get_fecparam) - return -EOPNOTSUPP; mutex_lock(&efx->mac_lock); - rc = efx->phy_op->set_fecparam(efx, fecparam); + rc = efx_mcdi_phy_set_fecparam(efx, fecparam); mutex_unlock(&efx->mac_lock); return rc; @@ -1332,11 +1323,8 @@ int efx_ethtool_get_module_eeprom(struct net_device *net_dev, struct efx_nic *efx = netdev_priv(net_dev); int ret; - if (!efx->phy_op || !efx->phy_op->get_module_eeprom) - return -EOPNOTSUPP; - mutex_lock(&efx->mac_lock); - ret = efx->phy_op->get_module_eeprom(efx, ee, data); + ret = efx_mcdi_phy_get_module_eeprom(efx, ee, data); mutex_unlock(&efx->mac_lock); return ret; @@ -1348,11 +1336,8 @@ int efx_ethtool_get_module_info(struct net_device *net_dev, struct efx_nic *efx = netdev_priv(net_dev); int ret; - if (!efx->phy_op || !efx->phy_op->get_module_info) - return -EOPNOTSUPP; - mutex_lock(&efx->mac_lock); - ret = efx->phy_op->get_module_info(efx, modinfo); + ret = efx_mcdi_phy_get_module_info(efx, modinfo); mutex_unlock(&efx->mac_lock); return ret; diff --git a/drivers/net/ethernet/sfc/falcon/farch.c b/drivers/net/ethernet/sfc/falcon/farch.c index fa1ade856b10..2c91792cec01 100644 --- a/drivers/net/ethernet/sfc/falcon/farch.c +++ b/drivers/net/ethernet/sfc/falcon/farch.c @@ -870,17 +870,12 @@ static u16 ef4_farch_handle_rx_not_ok(struct ef4_rx_queue *rx_queue, { struct ef4_channel *channel = ef4_rx_queue_channel(rx_queue); struct ef4_nic *efx = rx_queue->efx; - bool rx_ev_buf_owner_id_err, rx_ev_ip_hdr_chksum_err; + bool __maybe_unused rx_ev_buf_owner_id_err, rx_ev_ip_hdr_chksum_err; bool rx_ev_tcp_udp_chksum_err, rx_ev_eth_crc_err; bool rx_ev_frm_trunc, rx_ev_drib_nib, rx_ev_tobe_disc; - bool rx_ev_other_err, rx_ev_pause_frm; - bool rx_ev_hdr_type, rx_ev_mcast_pkt; - unsigned rx_ev_pkt_type; + bool rx_ev_pause_frm; - rx_ev_hdr_type = EF4_QWORD_FIELD(*event, FSF_AZ_RX_EV_HDR_TYPE); - rx_ev_mcast_pkt = EF4_QWORD_FIELD(*event, FSF_AZ_RX_EV_MCAST_PKT); rx_ev_tobe_disc = EF4_QWORD_FIELD(*event, FSF_AZ_RX_EV_TOBE_DISC); - rx_ev_pkt_type = EF4_QWORD_FIELD(*event, FSF_AZ_RX_EV_PKT_TYPE); rx_ev_buf_owner_id_err = EF4_QWORD_FIELD(*event, FSF_AZ_RX_EV_BUF_OWNER_ID_ERR); rx_ev_ip_hdr_chksum_err = EF4_QWORD_FIELD(*event, @@ -893,10 +888,6 @@ static u16 ef4_farch_handle_rx_not_ok(struct ef4_rx_queue *rx_queue, 0 : EF4_QWORD_FIELD(*event, FSF_AA_RX_EV_DRIB_NIB)); rx_ev_pause_frm = EF4_QWORD_FIELD(*event, FSF_AZ_RX_EV_PAUSE_FRM_ERR); - /* Every error apart from tobe_disc and pause_frm */ - rx_ev_other_err = (rx_ev_drib_nib | rx_ev_tcp_udp_chksum_err | - rx_ev_buf_owner_id_err | rx_ev_eth_crc_err | - rx_ev_frm_trunc | rx_ev_ip_hdr_chksum_err); /* Count errors that are not in MAC stats. Ignore expected * checksum errors during self-test. */ @@ -916,6 +907,13 @@ static u16 ef4_farch_handle_rx_not_ok(struct ef4_rx_queue *rx_queue, * to a FIFO overflow. */ #ifdef DEBUG + { + /* Every error apart from tobe_disc and pause_frm */ + + bool rx_ev_other_err = (rx_ev_drib_nib | rx_ev_tcp_udp_chksum_err | + rx_ev_buf_owner_id_err | rx_ev_eth_crc_err | + rx_ev_frm_trunc | rx_ev_ip_hdr_chksum_err); + if (rx_ev_other_err && net_ratelimit()) { netif_dbg(efx, rx_err, efx->net_dev, " RX queue %d unexpected RX event " @@ -932,6 +930,7 @@ static u16 ef4_farch_handle_rx_not_ok(struct ef4_rx_queue *rx_queue, rx_ev_tobe_disc ? " [TOBE_DISC]" : "", rx_ev_pause_frm ? " [PAUSE]" : ""); } + } #endif /* The frame must be discarded if any of these are true. */ @@ -1643,15 +1642,11 @@ void ef4_farch_rx_push_indir_table(struct ef4_nic *efx) */ void ef4_farch_dimension_resources(struct ef4_nic *efx, unsigned sram_lim_qw) { - unsigned vi_count, buftbl_min; + unsigned vi_count; /* Account for the buffer table entries backing the datapath channels * and the descriptor caches for those channels. */ - buftbl_min = ((efx->n_rx_channels * EF4_MAX_DMAQ_SIZE + - efx->n_tx_channels * EF4_TXQ_TYPES * EF4_MAX_DMAQ_SIZE + - efx->n_channels * EF4_MAX_EVQ_SIZE) - * sizeof(ef4_qword_t) / EF4_BUF_SIZE); vi_count = max(efx->n_channels, efx->n_tx_channels * EF4_TXQ_TYPES); efx->tx_dc_base = sram_lim_qw - vi_count * TX_DC_ENTRIES; @@ -2532,7 +2527,6 @@ int ef4_farch_filter_remove_safe(struct ef4_nic *efx, enum ef4_farch_filter_table_id table_id; struct ef4_farch_filter_table *table; unsigned int filter_idx; - struct ef4_farch_filter_spec *spec; int rc; table_id = ef4_farch_filter_id_table_id(filter_id); @@ -2543,7 +2537,6 @@ int ef4_farch_filter_remove_safe(struct ef4_nic *efx, filter_idx = ef4_farch_filter_id_index(filter_id); if (filter_idx >= table->size) return -ENOENT; - spec = &table->spec[filter_idx]; spin_lock_bh(&efx->filter_lock); rc = ef4_farch_filter_remove(efx, table, filter_idx, priority); diff --git a/drivers/net/ethernet/sfc/falcon/rx.c b/drivers/net/ethernet/sfc/falcon/rx.c index 05ea3523890a..966f13e7475d 100644 --- a/drivers/net/ethernet/sfc/falcon/rx.c +++ b/drivers/net/ethernet/sfc/falcon/rx.c @@ -140,6 +140,7 @@ static struct page *ef4_reuse_page(struct ef4_rx_queue *rx_queue) * ef4_init_rx_buffers - create EF4_RX_BATCH page-based RX buffers * * @rx_queue: Efx RX queue + * @atomic: control memory allocation flags * * This allocates a batch of pages, maps them for DMA, and populates * struct ef4_rx_buffers for each one. Return a negative error code or @@ -316,6 +317,7 @@ static void ef4_discard_rx_packet(struct ef4_channel *channel, * This will aim to fill the RX descriptor queue up to * @rx_queue->@max_fill. If there is insufficient atomic * memory to do so, a slow fill will be scheduled. + * @atomic: control memory allocation flags * * The caller must provide serialisation (none is used here). In practise, * this means this function must run from the NAPI handler, or be called diff --git a/drivers/net/ethernet/sfc/falcon/selftest.c b/drivers/net/ethernet/sfc/falcon/selftest.c index 147677c7c72f..6a454ac6f876 100644 --- a/drivers/net/ethernet/sfc/falcon/selftest.c +++ b/drivers/net/ethernet/sfc/falcon/selftest.c @@ -65,7 +65,7 @@ static const char *const ef4_interrupt_mode_names[] = { STRING_TABLE_LOOKUP(efx->interrupt_mode, ef4_interrupt_mode) /** - * ef4_loopback_state - persistent state during a loopback selftest + * struct ef4_loopback_state - persistent state during a loopback selftest * @flush: Drop all packets in ef4_loopback_rx_packet * @packet_count: Number of packets being used in this test * @skbs: An array of skbs transmitted diff --git a/drivers/net/ethernet/sfc/farch.c b/drivers/net/ethernet/sfc/farch.c index 4002f9a3ae90..d75cf5ff5686 100644 --- a/drivers/net/ethernet/sfc/farch.c +++ b/drivers/net/ethernet/sfc/farch.c @@ -320,7 +320,7 @@ void efx_farch_tx_write(struct efx_tx_queue *tx_queue) unsigned write_ptr; unsigned old_write_count = tx_queue->write_count; - tx_queue->xmit_more_available = false; + tx_queue->xmit_pending = false; if (unlikely(tx_queue->write_count == tx_queue->insert_count)) return; @@ -372,6 +372,8 @@ int efx_farch_tx_probe(struct efx_tx_queue *tx_queue) struct efx_nic *efx = tx_queue->efx; unsigned entries; + tx_queue->type = ((tx_queue->label & 1) ? EFX_TXQ_TYPE_OUTER_CSUM : 0) | + ((tx_queue->label & 2) ? EFX_TXQ_TYPE_HIGHPRI : 0); entries = tx_queue->ptr_mask + 1; return efx_alloc_special_buffer(efx, &tx_queue->txd, entries * sizeof(efx_qword_t)); @@ -379,7 +381,7 @@ int efx_farch_tx_probe(struct efx_tx_queue *tx_queue) void efx_farch_tx_init(struct efx_tx_queue *tx_queue) { - int csum = tx_queue->label & EFX_TXQ_TYPE_OFFLOAD; + int csum = tx_queue->type & EFX_TXQ_TYPE_OUTER_CSUM; struct efx_nic *efx = tx_queue->efx; efx_oword_t reg; @@ -409,10 +411,12 @@ void efx_farch_tx_init(struct efx_tx_queue *tx_queue) EFX_POPULATE_OWORD_1(reg, FRF_BZ_TX_PACE, - (tx_queue->label & EFX_TXQ_TYPE_HIGHPRI) ? + (tx_queue->type & EFX_TXQ_TYPE_HIGHPRI) ? FFE_BZ_TX_PACE_OFF : FFE_BZ_TX_PACE_RESERVED); efx_writeo_table(efx, ®, FR_BZ_TX_PACE_TBL, tx_queue->queue); + + tx_queue->tso_version = 1; } static void efx_farch_flush_tx_queue(struct efx_tx_queue *tx_queue) @@ -832,13 +836,13 @@ efx_farch_handle_tx_event(struct efx_channel *channel, efx_qword_t *event) tx_ev_desc_ptr = EFX_QWORD_FIELD(*event, FSF_AZ_TX_EV_DESC_PTR); tx_ev_q_label = EFX_QWORD_FIELD(*event, FSF_AZ_TX_EV_Q_LABEL); tx_queue = efx_channel_get_tx_queue( - channel, tx_ev_q_label % EFX_TXQ_TYPES); + channel, tx_ev_q_label % EFX_MAX_TXQ_PER_CHANNEL); efx_xmit_done(tx_queue, tx_ev_desc_ptr); } else if (EFX_QWORD_FIELD(*event, FSF_AZ_TX_EV_WQ_FF_FULL)) { /* Rewrite the FIFO write pointer */ tx_ev_q_label = EFX_QWORD_FIELD(*event, FSF_AZ_TX_EV_Q_LABEL); tx_queue = efx_channel_get_tx_queue( - channel, tx_ev_q_label % EFX_TXQ_TYPES); + channel, tx_ev_q_label % EFX_MAX_TXQ_PER_CHANNEL); netif_tx_lock(efx->net_dev); efx_farch_notify_tx_desc(tx_queue); @@ -863,13 +867,8 @@ static u16 efx_farch_handle_rx_not_ok(struct efx_rx_queue *rx_queue, bool rx_ev_tcp_udp_chksum_err, rx_ev_eth_crc_err; bool rx_ev_frm_trunc, rx_ev_tobe_disc; bool rx_ev_other_err, rx_ev_pause_frm; - bool rx_ev_hdr_type, rx_ev_mcast_pkt; - unsigned rx_ev_pkt_type; - rx_ev_hdr_type = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_HDR_TYPE); - rx_ev_mcast_pkt = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_MCAST_PKT); rx_ev_tobe_disc = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_TOBE_DISC); - rx_ev_pkt_type = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_PKT_TYPE); rx_ev_buf_owner_id_err = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_BUF_OWNER_ID_ERR); rx_ev_ip_hdr_chksum_err = EFX_QWORD_FIELD(*event, @@ -918,6 +917,8 @@ static u16 efx_farch_handle_rx_not_ok(struct efx_rx_queue *rx_queue, rx_ev_tobe_disc ? " [TOBE_DISC]" : "", rx_ev_pause_frm ? " [PAUSE]" : ""); } +#else + (void) rx_ev_other_err; #endif if (efx->net_dev->features & NETIF_F_RXALL) @@ -1083,9 +1084,9 @@ efx_farch_handle_tx_flush_done(struct efx_nic *efx, efx_qword_t *event) int qid; qid = EFX_QWORD_FIELD(*event, FSF_AZ_DRIVER_EV_SUBDATA); - if (qid < EFX_TXQ_TYPES * (efx->n_tx_channels + efx->n_extra_tx_channels)) { - tx_queue = efx_get_tx_queue(efx, qid / EFX_TXQ_TYPES, - qid % EFX_TXQ_TYPES); + if (qid < EFX_MAX_TXQ_PER_CHANNEL * (efx->n_tx_channels + efx->n_extra_tx_channels)) { + tx_queue = efx_get_tx_queue(efx, qid / EFX_MAX_TXQ_PER_CHANNEL, + qid % EFX_MAX_TXQ_PER_CHANNEL); if (atomic_cmpxchg(&tx_queue->flush_outstanding, 1, 0)) { efx_farch_magic_event(tx_queue->channel, EFX_CHANNEL_MAGIC_TX_DRAIN(tx_queue)); @@ -1678,10 +1679,10 @@ void efx_farch_dimension_resources(struct efx_nic *efx, unsigned sram_lim_qw) * and the descriptor caches for those channels. */ buftbl_min = ((efx->n_rx_channels * EFX_MAX_DMAQ_SIZE + - total_tx_channels * EFX_TXQ_TYPES * EFX_MAX_DMAQ_SIZE + + total_tx_channels * EFX_MAX_TXQ_PER_CHANNEL * EFX_MAX_DMAQ_SIZE + efx->n_channels * EFX_MAX_EVQ_SIZE) * sizeof(efx_qword_t) / EFX_BUF_SIZE); - vi_count = max(efx->n_channels, total_tx_channels * EFX_TXQ_TYPES); + vi_count = max(efx->n_channels, total_tx_channels * EFX_MAX_TXQ_PER_CHANNEL); #ifdef CONFIG_SFC_SRIOV if (efx->type->sriov_wanted) { @@ -2592,7 +2593,6 @@ int efx_farch_filter_remove_safe(struct efx_nic *efx, enum efx_farch_filter_table_id table_id; struct efx_farch_filter_table *table; unsigned int filter_idx; - struct efx_farch_filter_spec *spec; int rc; table_id = efx_farch_filter_id_table_id(filter_id); @@ -2604,7 +2604,6 @@ int efx_farch_filter_remove_safe(struct efx_nic *efx, if (filter_idx >= table->size) return -ENOENT; down_write(&state->lock); - spec = &table->spec[filter_idx]; rc = efx_farch_filter_remove(efx, table, filter_idx, priority); up_write(&state->lock); diff --git a/drivers/net/ethernet/sfc/mcdi.c b/drivers/net/ethernet/sfc/mcdi.c index 5467819aef6e..be6bfd6b7ec7 100644 --- a/drivers/net/ethernet/sfc/mcdi.c +++ b/drivers/net/ethernet/sfc/mcdi.c @@ -1868,10 +1868,9 @@ int efx_mcdi_handle_assertion(struct efx_nic *efx) return efx_mcdi_exit_assertion(efx); } -void efx_mcdi_set_id_led(struct efx_nic *efx, enum efx_led_mode mode) +int efx_mcdi_set_id_led(struct efx_nic *efx, enum efx_led_mode mode) { MCDI_DECLARE_BUF(inbuf, MC_CMD_SET_ID_LED_IN_LEN); - int rc; BUILD_BUG_ON(EFX_LED_OFF != MC_CMD_LED_OFF); BUILD_BUG_ON(EFX_LED_ON != MC_CMD_LED_ON); @@ -1881,8 +1880,7 @@ void efx_mcdi_set_id_led(struct efx_nic *efx, enum efx_led_mode mode) MCDI_SET_DWORD(inbuf, SET_ID_LED_IN_STATE, mode); - rc = efx_mcdi_rpc(efx, MC_CMD_SET_ID_LED, inbuf, sizeof(inbuf), - NULL, 0, NULL); + return efx_mcdi_rpc(efx, MC_CMD_SET_ID_LED, inbuf, sizeof(inbuf), NULL, 0, NULL); } static int efx_mcdi_reset_func(struct efx_nic *efx) diff --git a/drivers/net/ethernet/sfc/mcdi.h b/drivers/net/ethernet/sfc/mcdi.h index 658cf345420d..69c2924a147c 100644 --- a/drivers/net/ethernet/sfc/mcdi.h +++ b/drivers/net/ethernet/sfc/mcdi.h @@ -190,6 +190,7 @@ void efx_mcdi_sensor_event(struct efx_nic *efx, efx_qword_t *ev); * 32-bit-aligned. Also, on Siena we must copy to the MC shared * memory strictly 32 bits at a time, so add any necessary padding. */ +#define MCDI_TX_BUF_LEN(_len) DIV_ROUND_UP((_len), 4) #define _MCDI_DECLARE_BUF(_name, _len) \ efx_dword_t _name[DIV_ROUND_UP(_len, 4)] #define MCDI_DECLARE_BUF(_name, _len) \ @@ -348,14 +349,13 @@ int efx_mcdi_nvram_info(struct efx_nic *efx, unsigned int type, int efx_new_mcdi_nvram_test_all(struct efx_nic *efx); int efx_mcdi_nvram_test_all(struct efx_nic *efx); int efx_mcdi_handle_assertion(struct efx_nic *efx); -void efx_mcdi_set_id_led(struct efx_nic *efx, enum efx_led_mode mode); +int efx_mcdi_set_id_led(struct efx_nic *efx, enum efx_led_mode mode); int efx_mcdi_wol_filter_set_magic(struct efx_nic *efx, const u8 *mac, int *id_out); int efx_mcdi_wol_filter_get_magic(struct efx_nic *efx, int *id_out); int efx_mcdi_wol_filter_remove(struct efx_nic *efx, int id); int efx_mcdi_wol_filter_reset(struct efx_nic *efx); int efx_mcdi_flush_rxqs(struct efx_nic *efx); -int efx_mcdi_port_reconfigure(struct efx_nic *efx); void efx_mcdi_process_link_change(struct efx_nic *efx, efx_qword_t *ev); void efx_mcdi_mac_start_stats(struct efx_nic *efx); void efx_mcdi_mac_stop_stats(struct efx_nic *efx); diff --git a/drivers/net/ethernet/sfc/mcdi_functions.c b/drivers/net/ethernet/sfc/mcdi_functions.c index d8a3af86ef78..d3e6d8239f5c 100644 --- a/drivers/net/ethernet/sfc/mcdi_functions.c +++ b/drivers/net/ethernet/sfc/mcdi_functions.c @@ -160,11 +160,12 @@ fail: outbuf, outlen, rc); } -int efx_mcdi_tx_init(struct efx_tx_queue *tx_queue, bool tso_v2) +int efx_mcdi_tx_init(struct efx_tx_queue *tx_queue) { MCDI_DECLARE_BUF(inbuf, MC_CMD_INIT_TXQ_IN_LEN(EFX_MAX_DMAQ_SIZE * 8 / EFX_BUF_SIZE)); - bool csum_offload = tx_queue->label & EFX_TXQ_TYPE_OFFLOAD; + bool csum_offload = tx_queue->type & EFX_TXQ_TYPE_OUTER_CSUM; + bool inner_csum = tx_queue->type & EFX_TXQ_TYPE_INNER_CSUM; size_t entries = tx_queue->txd.buf.len / EFX_BUF_SIZE; struct efx_channel *channel = tx_queue->channel; struct efx_nic *efx = tx_queue->efx; @@ -194,22 +195,31 @@ int efx_mcdi_tx_init(struct efx_tx_queue *tx_queue, bool tso_v2) inlen = MC_CMD_INIT_TXQ_IN_LEN(entries); do { - MCDI_POPULATE_DWORD_4(inbuf, INIT_TXQ_IN_FLAGS, + bool tso_v2 = tx_queue->tso_version == 2; + + /* TSOv2 implies IP header checksum offload for TSO frames, + * so we can safely disable IP header checksum offload for + * everything else. If we don't have TSOv2, then we have to + * enable IP header checksum offload, which is strictly + * incorrect but better than breaking TSO. + */ + MCDI_POPULATE_DWORD_6(inbuf, INIT_TXQ_IN_FLAGS, /* This flag was removed from mcdi_pcol.h for * the non-_EXT version of INIT_TXQ. However, * firmware still honours it. */ INIT_TXQ_EXT_IN_FLAG_TSOV2_EN, tso_v2, - INIT_TXQ_IN_FLAG_IP_CSUM_DIS, !csum_offload, + INIT_TXQ_IN_FLAG_IP_CSUM_DIS, !(csum_offload && tso_v2), INIT_TXQ_IN_FLAG_TCP_CSUM_DIS, !csum_offload, - INIT_TXQ_EXT_IN_FLAG_TIMESTAMP, - tx_queue->timestamping); + INIT_TXQ_EXT_IN_FLAG_TIMESTAMP, tx_queue->timestamping, + INIT_TXQ_IN_FLAG_INNER_IP_CSUM_EN, inner_csum && !tso_v2, + INIT_TXQ_IN_FLAG_INNER_TCP_CSUM_EN, inner_csum); rc = efx_mcdi_rpc_quiet(efx, MC_CMD_INIT_TXQ, inbuf, inlen, NULL, 0, NULL); if (rc == -ENOSPC && tso_v2) { /* Retry without TSOv2 if we're short on contexts. */ - tso_v2 = false; + tx_queue->tso_version = 0; netif_warn(efx, probe, efx->net_dev, "TSOv2 context not available to segment in " "hardware. TCP performance may be reduced.\n" diff --git a/drivers/net/ethernet/sfc/mcdi_functions.h b/drivers/net/ethernet/sfc/mcdi_functions.h index 687be8b00cd8..b0e2f53a0d9b 100644 --- a/drivers/net/ethernet/sfc/mcdi_functions.h +++ b/drivers/net/ethernet/sfc/mcdi_functions.h @@ -19,7 +19,7 @@ int efx_mcdi_ev_probe(struct efx_channel *channel); int efx_mcdi_ev_init(struct efx_channel *channel, bool v1_cut_thru, bool v2); void efx_mcdi_ev_remove(struct efx_channel *channel); void efx_mcdi_ev_fini(struct efx_channel *channel); -int efx_mcdi_tx_init(struct efx_tx_queue *tx_queue, bool tso_v2); +int efx_mcdi_tx_init(struct efx_tx_queue *tx_queue); void efx_mcdi_tx_remove(struct efx_tx_queue *tx_queue); void efx_mcdi_tx_fini(struct efx_tx_queue *tx_queue); int efx_mcdi_rx_probe(struct efx_rx_queue *rx_queue); diff --git a/drivers/net/ethernet/sfc/mcdi_port.c b/drivers/net/ethernet/sfc/mcdi_port.c index 98eeb404f68d..94c6a345c0b1 100644 --- a/drivers/net/ethernet/sfc/mcdi_port.c +++ b/drivers/net/ethernet/sfc/mcdi_port.c @@ -70,592 +70,6 @@ static int efx_mcdi_mdio_write(struct net_device *net_dev, return 0; } -static int efx_mcdi_phy_probe(struct efx_nic *efx) -{ - struct efx_mcdi_phy_data *phy_data; - MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LINK_OUT_LEN); - u32 caps; - int rc; - - /* Initialise and populate phy_data */ - phy_data = kzalloc(sizeof(*phy_data), GFP_KERNEL); - if (phy_data == NULL) - return -ENOMEM; - - rc = efx_mcdi_get_phy_cfg(efx, phy_data); - if (rc != 0) - goto fail; - - /* Read initial link advertisement */ - BUILD_BUG_ON(MC_CMD_GET_LINK_IN_LEN != 0); - rc = efx_mcdi_rpc(efx, MC_CMD_GET_LINK, NULL, 0, - outbuf, sizeof(outbuf), NULL); - if (rc) - goto fail; - - /* Fill out nic state */ - efx->phy_data = phy_data; - efx->phy_type = phy_data->type; - - efx->mdio_bus = phy_data->channel; - efx->mdio.prtad = phy_data->port; - efx->mdio.mmds = phy_data->mmd_mask & ~(1 << MC_CMD_MMD_CLAUSE22); - efx->mdio.mode_support = 0; - if (phy_data->mmd_mask & (1 << MC_CMD_MMD_CLAUSE22)) - efx->mdio.mode_support |= MDIO_SUPPORTS_C22; - if (phy_data->mmd_mask & ~(1 << MC_CMD_MMD_CLAUSE22)) - efx->mdio.mode_support |= MDIO_SUPPORTS_C45 | MDIO_EMULATE_C22; - - caps = MCDI_DWORD(outbuf, GET_LINK_OUT_CAP); - if (caps & (1 << MC_CMD_PHY_CAP_AN_LBN)) - mcdi_to_ethtool_linkset(phy_data->media, caps, - efx->link_advertising); - else - phy_data->forced_cap = caps; - - /* Assert that we can map efx -> mcdi loopback modes */ - BUILD_BUG_ON(LOOPBACK_NONE != MC_CMD_LOOPBACK_NONE); - BUILD_BUG_ON(LOOPBACK_DATA != MC_CMD_LOOPBACK_DATA); - BUILD_BUG_ON(LOOPBACK_GMAC != MC_CMD_LOOPBACK_GMAC); - BUILD_BUG_ON(LOOPBACK_XGMII != MC_CMD_LOOPBACK_XGMII); - BUILD_BUG_ON(LOOPBACK_XGXS != MC_CMD_LOOPBACK_XGXS); - BUILD_BUG_ON(LOOPBACK_XAUI != MC_CMD_LOOPBACK_XAUI); - BUILD_BUG_ON(LOOPBACK_GMII != MC_CMD_LOOPBACK_GMII); - BUILD_BUG_ON(LOOPBACK_SGMII != MC_CMD_LOOPBACK_SGMII); - BUILD_BUG_ON(LOOPBACK_XGBR != MC_CMD_LOOPBACK_XGBR); - BUILD_BUG_ON(LOOPBACK_XFI != MC_CMD_LOOPBACK_XFI); - BUILD_BUG_ON(LOOPBACK_XAUI_FAR != MC_CMD_LOOPBACK_XAUI_FAR); - BUILD_BUG_ON(LOOPBACK_GMII_FAR != MC_CMD_LOOPBACK_GMII_FAR); - BUILD_BUG_ON(LOOPBACK_SGMII_FAR != MC_CMD_LOOPBACK_SGMII_FAR); - BUILD_BUG_ON(LOOPBACK_XFI_FAR != MC_CMD_LOOPBACK_XFI_FAR); - BUILD_BUG_ON(LOOPBACK_GPHY != MC_CMD_LOOPBACK_GPHY); - BUILD_BUG_ON(LOOPBACK_PHYXS != MC_CMD_LOOPBACK_PHYXS); - BUILD_BUG_ON(LOOPBACK_PCS != MC_CMD_LOOPBACK_PCS); - BUILD_BUG_ON(LOOPBACK_PMAPMD != MC_CMD_LOOPBACK_PMAPMD); - BUILD_BUG_ON(LOOPBACK_XPORT != MC_CMD_LOOPBACK_XPORT); - BUILD_BUG_ON(LOOPBACK_XGMII_WS != MC_CMD_LOOPBACK_XGMII_WS); - BUILD_BUG_ON(LOOPBACK_XAUI_WS != MC_CMD_LOOPBACK_XAUI_WS); - BUILD_BUG_ON(LOOPBACK_XAUI_WS_FAR != MC_CMD_LOOPBACK_XAUI_WS_FAR); - BUILD_BUG_ON(LOOPBACK_XAUI_WS_NEAR != MC_CMD_LOOPBACK_XAUI_WS_NEAR); - BUILD_BUG_ON(LOOPBACK_GMII_WS != MC_CMD_LOOPBACK_GMII_WS); - BUILD_BUG_ON(LOOPBACK_XFI_WS != MC_CMD_LOOPBACK_XFI_WS); - BUILD_BUG_ON(LOOPBACK_XFI_WS_FAR != MC_CMD_LOOPBACK_XFI_WS_FAR); - BUILD_BUG_ON(LOOPBACK_PHYXS_WS != MC_CMD_LOOPBACK_PHYXS_WS); - - rc = efx_mcdi_loopback_modes(efx, &efx->loopback_modes); - if (rc != 0) - goto fail; - /* The MC indicates that LOOPBACK_NONE is a valid loopback mode, - * but by convention we don't */ - efx->loopback_modes &= ~(1 << LOOPBACK_NONE); - - /* Set the initial link mode */ - efx_mcdi_phy_decode_link( - efx, &efx->link_state, - MCDI_DWORD(outbuf, GET_LINK_OUT_LINK_SPEED), - MCDI_DWORD(outbuf, GET_LINK_OUT_FLAGS), - MCDI_DWORD(outbuf, GET_LINK_OUT_FCNTL)); - - /* Record the initial FEC configuration (or nearest approximation - * representable in the ethtool configuration space) - */ - efx->fec_config = mcdi_fec_caps_to_ethtool(caps, - efx->link_state.speed == 25000 || - efx->link_state.speed == 50000); - - /* Default to Autonegotiated flow control if the PHY supports it */ - efx->wanted_fc = EFX_FC_RX | EFX_FC_TX; - if (phy_data->supported_cap & (1 << MC_CMD_PHY_CAP_AN_LBN)) - efx->wanted_fc |= EFX_FC_AUTO; - efx_link_set_wanted_fc(efx, efx->wanted_fc); - - return 0; - -fail: - kfree(phy_data); - return rc; -} - -static void efx_mcdi_phy_remove(struct efx_nic *efx) -{ - struct efx_mcdi_phy_data *phy_data = efx->phy_data; - - efx->phy_data = NULL; - kfree(phy_data); -} - -static void efx_mcdi_phy_get_link_ksettings(struct efx_nic *efx, - struct ethtool_link_ksettings *cmd) -{ - struct efx_mcdi_phy_data *phy_cfg = efx->phy_data; - MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LINK_OUT_LEN); - int rc; - - cmd->base.speed = efx->link_state.speed; - cmd->base.duplex = efx->link_state.fd; - cmd->base.port = mcdi_to_ethtool_media(phy_cfg->media); - cmd->base.phy_address = phy_cfg->port; - cmd->base.autoneg = !!(efx->link_advertising[0] & ADVERTISED_Autoneg); - cmd->base.mdio_support = (efx->mdio.mode_support & - (MDIO_SUPPORTS_C45 | MDIO_SUPPORTS_C22)); - - mcdi_to_ethtool_linkset(phy_cfg->media, phy_cfg->supported_cap, - cmd->link_modes.supported); - memcpy(cmd->link_modes.advertising, efx->link_advertising, - sizeof(__ETHTOOL_DECLARE_LINK_MODE_MASK())); - - BUILD_BUG_ON(MC_CMD_GET_LINK_IN_LEN != 0); - rc = efx_mcdi_rpc(efx, MC_CMD_GET_LINK, NULL, 0, - outbuf, sizeof(outbuf), NULL); - if (rc) - return; - mcdi_to_ethtool_linkset(phy_cfg->media, - MCDI_DWORD(outbuf, GET_LINK_OUT_LP_CAP), - cmd->link_modes.lp_advertising); -} - -static int -efx_mcdi_phy_set_link_ksettings(struct efx_nic *efx, - const struct ethtool_link_ksettings *cmd) -{ - struct efx_mcdi_phy_data *phy_cfg = efx->phy_data; - u32 caps; - int rc; - - if (cmd->base.autoneg) { - caps = (ethtool_linkset_to_mcdi_cap(cmd->link_modes.advertising) | - 1 << MC_CMD_PHY_CAP_AN_LBN); - } else if (cmd->base.duplex) { - switch (cmd->base.speed) { - case 10: caps = 1 << MC_CMD_PHY_CAP_10FDX_LBN; break; - case 100: caps = 1 << MC_CMD_PHY_CAP_100FDX_LBN; break; - case 1000: caps = 1 << MC_CMD_PHY_CAP_1000FDX_LBN; break; - case 10000: caps = 1 << MC_CMD_PHY_CAP_10000FDX_LBN; break; - case 40000: caps = 1 << MC_CMD_PHY_CAP_40000FDX_LBN; break; - case 100000: caps = 1 << MC_CMD_PHY_CAP_100000FDX_LBN; break; - case 25000: caps = 1 << MC_CMD_PHY_CAP_25000FDX_LBN; break; - case 50000: caps = 1 << MC_CMD_PHY_CAP_50000FDX_LBN; break; - default: return -EINVAL; - } - } else { - switch (cmd->base.speed) { - case 10: caps = 1 << MC_CMD_PHY_CAP_10HDX_LBN; break; - case 100: caps = 1 << MC_CMD_PHY_CAP_100HDX_LBN; break; - case 1000: caps = 1 << MC_CMD_PHY_CAP_1000HDX_LBN; break; - default: return -EINVAL; - } - } - - caps |= ethtool_fec_caps_to_mcdi(efx->fec_config); - - rc = efx_mcdi_set_link(efx, caps, efx_get_mcdi_phy_flags(efx), - efx->loopback_mode, 0); - if (rc) - return rc; - - if (cmd->base.autoneg) { - efx_link_set_advertising(efx, cmd->link_modes.advertising); - phy_cfg->forced_cap = 0; - } else { - efx_link_clear_advertising(efx); - phy_cfg->forced_cap = caps; - } - return 0; -} - -static int efx_mcdi_phy_set_fecparam(struct efx_nic *efx, - const struct ethtool_fecparam *fec) -{ - struct efx_mcdi_phy_data *phy_cfg = efx->phy_data; - u32 caps; - int rc; - - /* Work out what efx_mcdi_phy_set_link_ksettings() would produce from - * saved advertising bits - */ - if (test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT, efx->link_advertising)) - caps = (ethtool_linkset_to_mcdi_cap(efx->link_advertising) | - 1 << MC_CMD_PHY_CAP_AN_LBN); - else - caps = phy_cfg->forced_cap; - - caps |= ethtool_fec_caps_to_mcdi(fec->fec); - rc = efx_mcdi_set_link(efx, caps, efx_get_mcdi_phy_flags(efx), - efx->loopback_mode, 0); - if (rc) - return rc; - - /* Record the new FEC setting for subsequent set_link calls */ - efx->fec_config = fec->fec; - return 0; -} - -static const char *const mcdi_sft9001_cable_diag_names[] = { - "cable.pairA.length", - "cable.pairB.length", - "cable.pairC.length", - "cable.pairD.length", - "cable.pairA.status", - "cable.pairB.status", - "cable.pairC.status", - "cable.pairD.status", -}; - -static int efx_mcdi_bist(struct efx_nic *efx, unsigned int bist_mode, - int *results) -{ - unsigned int retry, i, count = 0; - size_t outlen; - u32 status; - MCDI_DECLARE_BUF(inbuf, MC_CMD_START_BIST_IN_LEN); - MCDI_DECLARE_BUF(outbuf, MC_CMD_POLL_BIST_OUT_SFT9001_LEN); - u8 *ptr; - int rc; - - BUILD_BUG_ON(MC_CMD_START_BIST_OUT_LEN != 0); - MCDI_SET_DWORD(inbuf, START_BIST_IN_TYPE, bist_mode); - rc = efx_mcdi_rpc(efx, MC_CMD_START_BIST, - inbuf, MC_CMD_START_BIST_IN_LEN, NULL, 0, NULL); - if (rc) - goto out; - - /* Wait up to 10s for BIST to finish */ - for (retry = 0; retry < 100; ++retry) { - BUILD_BUG_ON(MC_CMD_POLL_BIST_IN_LEN != 0); - rc = efx_mcdi_rpc(efx, MC_CMD_POLL_BIST, NULL, 0, - outbuf, sizeof(outbuf), &outlen); - if (rc) - goto out; - - status = MCDI_DWORD(outbuf, POLL_BIST_OUT_RESULT); - if (status != MC_CMD_POLL_BIST_RUNNING) - goto finished; - - msleep(100); - } - - rc = -ETIMEDOUT; - goto out; - -finished: - results[count++] = (status == MC_CMD_POLL_BIST_PASSED) ? 1 : -1; - - /* SFT9001 specific cable diagnostics output */ - if (efx->phy_type == PHY_TYPE_SFT9001B && - (bist_mode == MC_CMD_PHY_BIST_CABLE_SHORT || - bist_mode == MC_CMD_PHY_BIST_CABLE_LONG)) { - ptr = MCDI_PTR(outbuf, POLL_BIST_OUT_SFT9001_CABLE_LENGTH_A); - if (status == MC_CMD_POLL_BIST_PASSED && - outlen >= MC_CMD_POLL_BIST_OUT_SFT9001_LEN) { - for (i = 0; i < 8; i++) { - results[count + i] = - EFX_DWORD_FIELD(((efx_dword_t *)ptr)[i], - EFX_DWORD_0); - } - } - count += 8; - } - rc = count; - -out: - return rc; -} - -static int efx_mcdi_phy_run_tests(struct efx_nic *efx, int *results, - unsigned flags) -{ - struct efx_mcdi_phy_data *phy_cfg = efx->phy_data; - u32 mode; - int rc; - - if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_OUT_BIST_LBN)) { - rc = efx_mcdi_bist(efx, MC_CMD_PHY_BIST, results); - if (rc < 0) - return rc; - - results += rc; - } - - /* If we support both LONG and SHORT, then run each in response to - * break or not. Otherwise, run the one we support */ - mode = 0; - if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_OUT_BIST_CABLE_SHORT_LBN)) { - if ((flags & ETH_TEST_FL_OFFLINE) && - (phy_cfg->flags & - (1 << MC_CMD_GET_PHY_CFG_OUT_BIST_CABLE_LONG_LBN))) - mode = MC_CMD_PHY_BIST_CABLE_LONG; - else - mode = MC_CMD_PHY_BIST_CABLE_SHORT; - } else if (phy_cfg->flags & - (1 << MC_CMD_GET_PHY_CFG_OUT_BIST_CABLE_LONG_LBN)) - mode = MC_CMD_PHY_BIST_CABLE_LONG; - - if (mode != 0) { - rc = efx_mcdi_bist(efx, mode, results); - if (rc < 0) - return rc; - results += rc; - } - - return 0; -} - -static const char *efx_mcdi_phy_test_name(struct efx_nic *efx, - unsigned int index) -{ - struct efx_mcdi_phy_data *phy_cfg = efx->phy_data; - - if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_OUT_BIST_LBN)) { - if (index == 0) - return "bist"; - --index; - } - - if (phy_cfg->flags & ((1 << MC_CMD_GET_PHY_CFG_OUT_BIST_CABLE_SHORT_LBN) | - (1 << MC_CMD_GET_PHY_CFG_OUT_BIST_CABLE_LONG_LBN))) { - if (index == 0) - return "cable"; - --index; - - if (efx->phy_type == PHY_TYPE_SFT9001B) { - if (index < ARRAY_SIZE(mcdi_sft9001_cable_diag_names)) - return mcdi_sft9001_cable_diag_names[index]; - index -= ARRAY_SIZE(mcdi_sft9001_cable_diag_names); - } - } - - return NULL; -} - -#define SFP_PAGE_SIZE 128 -#define SFF_DIAG_TYPE_OFFSET 92 -#define SFF_DIAG_ADDR_CHANGE BIT(2) -#define SFF_8079_NUM_PAGES 2 -#define SFF_8472_NUM_PAGES 4 -#define SFF_8436_NUM_PAGES 5 -#define SFF_DMT_LEVEL_OFFSET 94 - -/** efx_mcdi_phy_get_module_eeprom_page() - Get a single page of module eeprom - * @efx: NIC context - * @page: EEPROM page number - * @data: Destination data pointer - * @offset: Offset in page to copy from in to data - * @space: Space available in data - * - * Return: - * >=0 - amount of data copied - * <0 - error - */ -static int efx_mcdi_phy_get_module_eeprom_page(struct efx_nic *efx, - unsigned int page, - u8 *data, ssize_t offset, - ssize_t space) -{ - MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PHY_MEDIA_INFO_OUT_LENMAX); - MCDI_DECLARE_BUF(inbuf, MC_CMD_GET_PHY_MEDIA_INFO_IN_LEN); - size_t outlen; - unsigned int payload_len; - unsigned int to_copy; - int rc; - - if (offset > SFP_PAGE_SIZE) - return -EINVAL; - - to_copy = min(space, SFP_PAGE_SIZE - offset); - - MCDI_SET_DWORD(inbuf, GET_PHY_MEDIA_INFO_IN_PAGE, page); - rc = efx_mcdi_rpc_quiet(efx, MC_CMD_GET_PHY_MEDIA_INFO, - inbuf, sizeof(inbuf), - outbuf, sizeof(outbuf), - &outlen); - - if (rc) - return rc; - - if (outlen < (MC_CMD_GET_PHY_MEDIA_INFO_OUT_DATA_OFST + - SFP_PAGE_SIZE)) - return -EIO; - - payload_len = MCDI_DWORD(outbuf, GET_PHY_MEDIA_INFO_OUT_DATALEN); - if (payload_len != SFP_PAGE_SIZE) - return -EIO; - - memcpy(data, MCDI_PTR(outbuf, GET_PHY_MEDIA_INFO_OUT_DATA) + offset, - to_copy); - - return to_copy; -} - -static int efx_mcdi_phy_get_module_eeprom_byte(struct efx_nic *efx, - unsigned int page, - u8 byte) -{ - int rc; - u8 data; - - rc = efx_mcdi_phy_get_module_eeprom_page(efx, page, &data, byte, 1); - if (rc == 1) - return data; - - return rc; -} - -static int efx_mcdi_phy_diag_type(struct efx_nic *efx) -{ - /* Page zero of the EEPROM includes the diagnostic type at byte 92. */ - return efx_mcdi_phy_get_module_eeprom_byte(efx, 0, - SFF_DIAG_TYPE_OFFSET); -} - -static int efx_mcdi_phy_sff_8472_level(struct efx_nic *efx) -{ - /* Page zero of the EEPROM includes the DMT level at byte 94. */ - return efx_mcdi_phy_get_module_eeprom_byte(efx, 0, - SFF_DMT_LEVEL_OFFSET); -} - -static u32 efx_mcdi_phy_module_type(struct efx_nic *efx) -{ - struct efx_mcdi_phy_data *phy_data = efx->phy_data; - - if (phy_data->media != MC_CMD_MEDIA_QSFP_PLUS) - return phy_data->media; - - /* A QSFP+ NIC may actually have an SFP+ module attached. - * The ID is page 0, byte 0. - */ - switch (efx_mcdi_phy_get_module_eeprom_byte(efx, 0, 0)) { - case 0x3: - return MC_CMD_MEDIA_SFP_PLUS; - case 0xc: - case 0xd: - return MC_CMD_MEDIA_QSFP_PLUS; - default: - return 0; - } -} - -static int efx_mcdi_phy_get_module_eeprom(struct efx_nic *efx, - struct ethtool_eeprom *ee, u8 *data) -{ - int rc; - ssize_t space_remaining = ee->len; - unsigned int page_off; - bool ignore_missing; - int num_pages; - int page; - - switch (efx_mcdi_phy_module_type(efx)) { - case MC_CMD_MEDIA_SFP_PLUS: - num_pages = efx_mcdi_phy_sff_8472_level(efx) > 0 ? - SFF_8472_NUM_PAGES : SFF_8079_NUM_PAGES; - page = 0; - ignore_missing = false; - break; - case MC_CMD_MEDIA_QSFP_PLUS: - num_pages = SFF_8436_NUM_PAGES; - page = -1; /* We obtain the lower page by asking for -1. */ - ignore_missing = true; /* Ignore missing pages after page 0. */ - break; - default: - return -EOPNOTSUPP; - } - - page_off = ee->offset % SFP_PAGE_SIZE; - page += ee->offset / SFP_PAGE_SIZE; - - while (space_remaining && (page < num_pages)) { - rc = efx_mcdi_phy_get_module_eeprom_page(efx, page, - data, page_off, - space_remaining); - - if (rc > 0) { - space_remaining -= rc; - data += rc; - page_off = 0; - page++; - } else if (rc == 0) { - space_remaining = 0; - } else if (ignore_missing && (page > 0)) { - int intended_size = SFP_PAGE_SIZE - page_off; - - space_remaining -= intended_size; - if (space_remaining < 0) { - space_remaining = 0; - } else { - memset(data, 0, intended_size); - data += intended_size; - page_off = 0; - page++; - rc = 0; - } - } else { - return rc; - } - } - - return 0; -} - -static int efx_mcdi_phy_get_module_info(struct efx_nic *efx, - struct ethtool_modinfo *modinfo) -{ - int sff_8472_level; - int diag_type; - - switch (efx_mcdi_phy_module_type(efx)) { - case MC_CMD_MEDIA_SFP_PLUS: - sff_8472_level = efx_mcdi_phy_sff_8472_level(efx); - - /* If we can't read the diagnostics level we have none. */ - if (sff_8472_level < 0) - return -EOPNOTSUPP; - - /* Check if this module requires the (unsupported) address - * change operation. - */ - diag_type = efx_mcdi_phy_diag_type(efx); - - if ((sff_8472_level == 0) || - (diag_type & SFF_DIAG_ADDR_CHANGE)) { - modinfo->type = ETH_MODULE_SFF_8079; - modinfo->eeprom_len = ETH_MODULE_SFF_8079_LEN; - } else { - modinfo->type = ETH_MODULE_SFF_8472; - modinfo->eeprom_len = ETH_MODULE_SFF_8472_LEN; - } - break; - - case MC_CMD_MEDIA_QSFP_PLUS: - modinfo->type = ETH_MODULE_SFF_8436; - modinfo->eeprom_len = ETH_MODULE_SFF_8436_LEN; - break; - - default: - return -EOPNOTSUPP; - } - - return 0; -} - -static const struct efx_phy_operations efx_mcdi_phy_ops = { - .probe = efx_mcdi_phy_probe, - .init = efx_port_dummy_op_int, - .reconfigure = efx_mcdi_port_reconfigure, - .poll = efx_mcdi_phy_poll, - .fini = efx_port_dummy_op_void, - .remove = efx_mcdi_phy_remove, - .get_link_ksettings = efx_mcdi_phy_get_link_ksettings, - .set_link_ksettings = efx_mcdi_phy_set_link_ksettings, - .get_fecparam = efx_mcdi_phy_get_fecparam, - .set_fecparam = efx_mcdi_phy_set_fecparam, - .test_alive = efx_mcdi_phy_test_alive, - .run_tests = efx_mcdi_phy_run_tests, - .test_name = efx_mcdi_phy_test_name, - .get_module_eeprom = efx_mcdi_phy_get_module_eeprom, - .get_module_info = efx_mcdi_phy_get_module_info, -}; - u32 efx_mcdi_phy_get_caps(struct efx_nic *efx) { struct efx_mcdi_phy_data *phy_data = efx->phy_data; @@ -683,16 +97,13 @@ int efx_mcdi_port_probe(struct efx_nic *efx) { int rc; - /* Hook in PHY operations table */ - efx->phy_op = &efx_mcdi_phy_ops; - /* Set up MDIO structure for PHY */ efx->mdio.mode_support = MDIO_SUPPORTS_C45 | MDIO_EMULATE_C22; efx->mdio.mdio_read = efx_mcdi_mdio_read; efx->mdio.mdio_write = efx_mcdi_mdio_write; /* Fill out MDIO structure, loopback modes, and initial link state */ - rc = efx->phy_op->probe(efx); + rc = efx_mcdi_phy_probe(efx); if (rc != 0) return rc; @@ -701,6 +112,6 @@ int efx_mcdi_port_probe(struct efx_nic *efx) void efx_mcdi_port_remove(struct efx_nic *efx) { - efx->phy_op->remove(efx); + efx_mcdi_phy_remove(efx); efx_mcdi_mac_fini_stats(efx); } diff --git a/drivers/net/ethernet/sfc/mcdi_port_common.c b/drivers/net/ethernet/sfc/mcdi_port_common.c index 714d7f937212..4bd3ef8f3384 100644 --- a/drivers/net/ethernet/sfc/mcdi_port_common.c +++ b/drivers/net/ethernet/sfc/mcdi_port_common.c @@ -308,7 +308,7 @@ void efx_mcdi_phy_decode_link(struct efx_nic *efx, * Both RS and BASER (whether AUTO or not) means use FEC if cable and link * partner support it, preferring RS to BASER. */ -u32 ethtool_fec_caps_to_mcdi(u32 ethtool_cap) +u32 ethtool_fec_caps_to_mcdi(u32 supported_cap, u32 ethtool_cap) { u32 ret = 0; @@ -316,17 +316,21 @@ u32 ethtool_fec_caps_to_mcdi(u32 ethtool_cap) return 0; if (ethtool_cap & ETHTOOL_FEC_AUTO) - ret |= (1 << MC_CMD_PHY_CAP_BASER_FEC_LBN) | - (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_LBN) | - (1 << MC_CMD_PHY_CAP_RS_FEC_LBN); - if (ethtool_cap & ETHTOOL_FEC_RS) + ret |= ((1 << MC_CMD_PHY_CAP_BASER_FEC_LBN) | + (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_LBN) | + (1 << MC_CMD_PHY_CAP_RS_FEC_LBN)) & supported_cap; + if (ethtool_cap & ETHTOOL_FEC_RS && + supported_cap & (1 << MC_CMD_PHY_CAP_RS_FEC_LBN)) ret |= (1 << MC_CMD_PHY_CAP_RS_FEC_LBN) | (1 << MC_CMD_PHY_CAP_RS_FEC_REQUESTED_LBN); - if (ethtool_cap & ETHTOOL_FEC_BASER) - ret |= (1 << MC_CMD_PHY_CAP_BASER_FEC_LBN) | - (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_LBN) | - (1 << MC_CMD_PHY_CAP_BASER_FEC_REQUESTED_LBN) | - (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_REQUESTED_LBN); + if (ethtool_cap & ETHTOOL_FEC_BASER) { + if (supported_cap & (1 << MC_CMD_PHY_CAP_BASER_FEC_LBN)) + ret |= (1 << MC_CMD_PHY_CAP_BASER_FEC_LBN) | + (1 << MC_CMD_PHY_CAP_BASER_FEC_REQUESTED_LBN); + if (supported_cap & (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_LBN)) + ret |= (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_LBN) | + (1 << MC_CMD_PHY_CAP_25G_BASER_FEC_REQUESTED_LBN); + } return ret; } @@ -404,6 +408,196 @@ bool efx_mcdi_phy_poll(struct efx_nic *efx) return !efx_link_state_equal(&efx->link_state, &old_state); } +int efx_mcdi_phy_probe(struct efx_nic *efx) +{ + struct efx_mcdi_phy_data *phy_data; + MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LINK_OUT_LEN); + u32 caps; + int rc; + + /* Initialise and populate phy_data */ + phy_data = kzalloc(sizeof(*phy_data), GFP_KERNEL); + if (phy_data == NULL) + return -ENOMEM; + + rc = efx_mcdi_get_phy_cfg(efx, phy_data); + if (rc != 0) + goto fail; + + /* Read initial link advertisement */ + BUILD_BUG_ON(MC_CMD_GET_LINK_IN_LEN != 0); + rc = efx_mcdi_rpc(efx, MC_CMD_GET_LINK, NULL, 0, + outbuf, sizeof(outbuf), NULL); + if (rc) + goto fail; + + /* Fill out nic state */ + efx->phy_data = phy_data; + efx->phy_type = phy_data->type; + + efx->mdio_bus = phy_data->channel; + efx->mdio.prtad = phy_data->port; + efx->mdio.mmds = phy_data->mmd_mask & ~(1 << MC_CMD_MMD_CLAUSE22); + efx->mdio.mode_support = 0; + if (phy_data->mmd_mask & (1 << MC_CMD_MMD_CLAUSE22)) + efx->mdio.mode_support |= MDIO_SUPPORTS_C22; + if (phy_data->mmd_mask & ~(1 << MC_CMD_MMD_CLAUSE22)) + efx->mdio.mode_support |= MDIO_SUPPORTS_C45 | MDIO_EMULATE_C22; + + caps = MCDI_DWORD(outbuf, GET_LINK_OUT_CAP); + if (caps & (1 << MC_CMD_PHY_CAP_AN_LBN)) + mcdi_to_ethtool_linkset(phy_data->media, caps, + efx->link_advertising); + else + phy_data->forced_cap = caps; + + /* Assert that we can map efx -> mcdi loopback modes */ + BUILD_BUG_ON(LOOPBACK_NONE != MC_CMD_LOOPBACK_NONE); + BUILD_BUG_ON(LOOPBACK_DATA != MC_CMD_LOOPBACK_DATA); + BUILD_BUG_ON(LOOPBACK_GMAC != MC_CMD_LOOPBACK_GMAC); + BUILD_BUG_ON(LOOPBACK_XGMII != MC_CMD_LOOPBACK_XGMII); + BUILD_BUG_ON(LOOPBACK_XGXS != MC_CMD_LOOPBACK_XGXS); + BUILD_BUG_ON(LOOPBACK_XAUI != MC_CMD_LOOPBACK_XAUI); + BUILD_BUG_ON(LOOPBACK_GMII != MC_CMD_LOOPBACK_GMII); + BUILD_BUG_ON(LOOPBACK_SGMII != MC_CMD_LOOPBACK_SGMII); + BUILD_BUG_ON(LOOPBACK_XGBR != MC_CMD_LOOPBACK_XGBR); + BUILD_BUG_ON(LOOPBACK_XFI != MC_CMD_LOOPBACK_XFI); + BUILD_BUG_ON(LOOPBACK_XAUI_FAR != MC_CMD_LOOPBACK_XAUI_FAR); + BUILD_BUG_ON(LOOPBACK_GMII_FAR != MC_CMD_LOOPBACK_GMII_FAR); + BUILD_BUG_ON(LOOPBACK_SGMII_FAR != MC_CMD_LOOPBACK_SGMII_FAR); + BUILD_BUG_ON(LOOPBACK_XFI_FAR != MC_CMD_LOOPBACK_XFI_FAR); + BUILD_BUG_ON(LOOPBACK_GPHY != MC_CMD_LOOPBACK_GPHY); + BUILD_BUG_ON(LOOPBACK_PHYXS != MC_CMD_LOOPBACK_PHYXS); + BUILD_BUG_ON(LOOPBACK_PCS != MC_CMD_LOOPBACK_PCS); + BUILD_BUG_ON(LOOPBACK_PMAPMD != MC_CMD_LOOPBACK_PMAPMD); + BUILD_BUG_ON(LOOPBACK_XPORT != MC_CMD_LOOPBACK_XPORT); + BUILD_BUG_ON(LOOPBACK_XGMII_WS != MC_CMD_LOOPBACK_XGMII_WS); + BUILD_BUG_ON(LOOPBACK_XAUI_WS != MC_CMD_LOOPBACK_XAUI_WS); + BUILD_BUG_ON(LOOPBACK_XAUI_WS_FAR != MC_CMD_LOOPBACK_XAUI_WS_FAR); + BUILD_BUG_ON(LOOPBACK_XAUI_WS_NEAR != MC_CMD_LOOPBACK_XAUI_WS_NEAR); + BUILD_BUG_ON(LOOPBACK_GMII_WS != MC_CMD_LOOPBACK_GMII_WS); + BUILD_BUG_ON(LOOPBACK_XFI_WS != MC_CMD_LOOPBACK_XFI_WS); + BUILD_BUG_ON(LOOPBACK_XFI_WS_FAR != MC_CMD_LOOPBACK_XFI_WS_FAR); + BUILD_BUG_ON(LOOPBACK_PHYXS_WS != MC_CMD_LOOPBACK_PHYXS_WS); + + rc = efx_mcdi_loopback_modes(efx, &efx->loopback_modes); + if (rc != 0) + goto fail; + /* The MC indicates that LOOPBACK_NONE is a valid loopback mode, + * but by convention we don't + */ + efx->loopback_modes &= ~(1 << LOOPBACK_NONE); + + /* Set the initial link mode */ + efx_mcdi_phy_decode_link(efx, &efx->link_state, + MCDI_DWORD(outbuf, GET_LINK_OUT_LINK_SPEED), + MCDI_DWORD(outbuf, GET_LINK_OUT_FLAGS), + MCDI_DWORD(outbuf, GET_LINK_OUT_FCNTL)); + + /* Record the initial FEC configuration (or nearest approximation + * representable in the ethtool configuration space) + */ + efx->fec_config = mcdi_fec_caps_to_ethtool(caps, + efx->link_state.speed == 25000 || + efx->link_state.speed == 50000); + + /* Default to Autonegotiated flow control if the PHY supports it */ + efx->wanted_fc = EFX_FC_RX | EFX_FC_TX; + if (phy_data->supported_cap & (1 << MC_CMD_PHY_CAP_AN_LBN)) + efx->wanted_fc |= EFX_FC_AUTO; + efx_link_set_wanted_fc(efx, efx->wanted_fc); + + return 0; + +fail: + kfree(phy_data); + return rc; +} + +void efx_mcdi_phy_remove(struct efx_nic *efx) +{ + struct efx_mcdi_phy_data *phy_data = efx->phy_data; + + efx->phy_data = NULL; + kfree(phy_data); +} + +void efx_mcdi_phy_get_link_ksettings(struct efx_nic *efx, struct ethtool_link_ksettings *cmd) +{ + struct efx_mcdi_phy_data *phy_cfg = efx->phy_data; + MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LINK_OUT_LEN); + int rc; + + cmd->base.speed = efx->link_state.speed; + cmd->base.duplex = efx->link_state.fd; + cmd->base.port = mcdi_to_ethtool_media(phy_cfg->media); + cmd->base.phy_address = phy_cfg->port; + cmd->base.autoneg = !!(efx->link_advertising[0] & ADVERTISED_Autoneg); + cmd->base.mdio_support = (efx->mdio.mode_support & + (MDIO_SUPPORTS_C45 | MDIO_SUPPORTS_C22)); + + mcdi_to_ethtool_linkset(phy_cfg->media, phy_cfg->supported_cap, + cmd->link_modes.supported); + memcpy(cmd->link_modes.advertising, efx->link_advertising, + sizeof(__ETHTOOL_DECLARE_LINK_MODE_MASK())); + + BUILD_BUG_ON(MC_CMD_GET_LINK_IN_LEN != 0); + rc = efx_mcdi_rpc(efx, MC_CMD_GET_LINK, NULL, 0, + outbuf, sizeof(outbuf), NULL); + if (rc) + return; + mcdi_to_ethtool_linkset(phy_cfg->media, + MCDI_DWORD(outbuf, GET_LINK_OUT_LP_CAP), + cmd->link_modes.lp_advertising); +} + +int efx_mcdi_phy_set_link_ksettings(struct efx_nic *efx, const struct ethtool_link_ksettings *cmd) +{ + struct efx_mcdi_phy_data *phy_cfg = efx->phy_data; + u32 caps; + int rc; + + if (cmd->base.autoneg) { + caps = (ethtool_linkset_to_mcdi_cap(cmd->link_modes.advertising) | + 1 << MC_CMD_PHY_CAP_AN_LBN); + } else if (cmd->base.duplex) { + switch (cmd->base.speed) { + case 10: caps = 1 << MC_CMD_PHY_CAP_10FDX_LBN; break; + case 100: caps = 1 << MC_CMD_PHY_CAP_100FDX_LBN; break; + case 1000: caps = 1 << MC_CMD_PHY_CAP_1000FDX_LBN; break; + case 10000: caps = 1 << MC_CMD_PHY_CAP_10000FDX_LBN; break; + case 40000: caps = 1 << MC_CMD_PHY_CAP_40000FDX_LBN; break; + case 100000: caps = 1 << MC_CMD_PHY_CAP_100000FDX_LBN; break; + case 25000: caps = 1 << MC_CMD_PHY_CAP_25000FDX_LBN; break; + case 50000: caps = 1 << MC_CMD_PHY_CAP_50000FDX_LBN; break; + default: return -EINVAL; + } + } else { + switch (cmd->base.speed) { + case 10: caps = 1 << MC_CMD_PHY_CAP_10HDX_LBN; break; + case 100: caps = 1 << MC_CMD_PHY_CAP_100HDX_LBN; break; + case 1000: caps = 1 << MC_CMD_PHY_CAP_1000HDX_LBN; break; + default: return -EINVAL; + } + } + + caps |= ethtool_fec_caps_to_mcdi(phy_cfg->supported_cap, efx->fec_config); + + rc = efx_mcdi_set_link(efx, caps, efx_get_mcdi_phy_flags(efx), + efx->loopback_mode, 0); + if (rc) + return rc; + + if (cmd->base.autoneg) { + efx_link_set_advertising(efx, cmd->link_modes.advertising); + phy_cfg->forced_cap = 0; + } else { + efx_link_clear_advertising(efx); + phy_cfg->forced_cap = caps; + } + return 0; +} + int efx_mcdi_phy_get_fecparam(struct efx_nic *efx, struct ethtool_fecparam *fec) { MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_LINK_OUT_V2_LEN); @@ -455,6 +649,50 @@ int efx_mcdi_phy_get_fecparam(struct efx_nic *efx, struct ethtool_fecparam *fec) return 0; } +/* Basic validation to ensure that the caps we are going to attempt to set are + * in fact supported by the adapter. Note that 'no FEC' is always supported. + */ +static int ethtool_fec_supported(u32 supported_cap, u32 ethtool_cap) +{ + if (ethtool_cap & ETHTOOL_FEC_OFF) + return 0; + + if (ethtool_cap && + !ethtool_fec_caps_to_mcdi(supported_cap, ethtool_cap)) + return -EINVAL; + return 0; +} + +int efx_mcdi_phy_set_fecparam(struct efx_nic *efx, const struct ethtool_fecparam *fec) +{ + struct efx_mcdi_phy_data *phy_cfg = efx->phy_data; + u32 caps; + int rc; + + rc = ethtool_fec_supported(phy_cfg->supported_cap, fec->fec); + if (rc) + return rc; + + /* Work out what efx_mcdi_phy_set_link_ksettings() would produce from + * saved advertising bits + */ + if (test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT, efx->link_advertising)) + caps = (ethtool_linkset_to_mcdi_cap(efx->link_advertising) | + 1 << MC_CMD_PHY_CAP_AN_LBN); + else + caps = phy_cfg->forced_cap; + + caps |= ethtool_fec_caps_to_mcdi(phy_cfg->supported_cap, fec->fec); + rc = efx_mcdi_set_link(efx, caps, efx_get_mcdi_phy_flags(efx), + efx->loopback_mode, 0); + if (rc) + return rc; + + /* Record the new FEC setting for subsequent set_link calls */ + efx->fec_config = fec->fec; + return 0; +} + int efx_mcdi_phy_test_alive(struct efx_nic *efx) { MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PHY_STATE_OUT_LEN); @@ -483,12 +721,357 @@ int efx_mcdi_port_reconfigure(struct efx_nic *efx) ethtool_linkset_to_mcdi_cap(efx->link_advertising) : phy_cfg->forced_cap); - caps |= ethtool_fec_caps_to_mcdi(efx->fec_config); + caps |= ethtool_fec_caps_to_mcdi(phy_cfg->supported_cap, efx->fec_config); return efx_mcdi_set_link(efx, caps, efx_get_mcdi_phy_flags(efx), efx->loopback_mode, 0); } +static const char *const mcdi_sft9001_cable_diag_names[] = { + "cable.pairA.length", + "cable.pairB.length", + "cable.pairC.length", + "cable.pairD.length", + "cable.pairA.status", + "cable.pairB.status", + "cable.pairC.status", + "cable.pairD.status", +}; + +static int efx_mcdi_bist(struct efx_nic *efx, unsigned int bist_mode, + int *results) +{ + unsigned int retry, i, count = 0; + size_t outlen; + u32 status; + MCDI_DECLARE_BUF(inbuf, MC_CMD_START_BIST_IN_LEN); + MCDI_DECLARE_BUF(outbuf, MC_CMD_POLL_BIST_OUT_SFT9001_LEN); + u8 *ptr; + int rc; + + BUILD_BUG_ON(MC_CMD_START_BIST_OUT_LEN != 0); + MCDI_SET_DWORD(inbuf, START_BIST_IN_TYPE, bist_mode); + rc = efx_mcdi_rpc(efx, MC_CMD_START_BIST, + inbuf, MC_CMD_START_BIST_IN_LEN, NULL, 0, NULL); + if (rc) + goto out; + + /* Wait up to 10s for BIST to finish */ + for (retry = 0; retry < 100; ++retry) { + BUILD_BUG_ON(MC_CMD_POLL_BIST_IN_LEN != 0); + rc = efx_mcdi_rpc(efx, MC_CMD_POLL_BIST, NULL, 0, + outbuf, sizeof(outbuf), &outlen); + if (rc) + goto out; + + status = MCDI_DWORD(outbuf, POLL_BIST_OUT_RESULT); + if (status != MC_CMD_POLL_BIST_RUNNING) + goto finished; + + msleep(100); + } + + rc = -ETIMEDOUT; + goto out; + +finished: + results[count++] = (status == MC_CMD_POLL_BIST_PASSED) ? 1 : -1; + + /* SFT9001 specific cable diagnostics output */ + if (efx->phy_type == PHY_TYPE_SFT9001B && + (bist_mode == MC_CMD_PHY_BIST_CABLE_SHORT || + bist_mode == MC_CMD_PHY_BIST_CABLE_LONG)) { + ptr = MCDI_PTR(outbuf, POLL_BIST_OUT_SFT9001_CABLE_LENGTH_A); + if (status == MC_CMD_POLL_BIST_PASSED && + outlen >= MC_CMD_POLL_BIST_OUT_SFT9001_LEN) { + for (i = 0; i < 8; i++) { + results[count + i] = + EFX_DWORD_FIELD(((efx_dword_t *)ptr)[i], + EFX_DWORD_0); + } + } + count += 8; + } + rc = count; + +out: + return rc; +} + +int efx_mcdi_phy_run_tests(struct efx_nic *efx, int *results, unsigned int flags) +{ + struct efx_mcdi_phy_data *phy_cfg = efx->phy_data; + u32 mode; + int rc; + + if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_OUT_BIST_LBN)) { + rc = efx_mcdi_bist(efx, MC_CMD_PHY_BIST, results); + if (rc < 0) + return rc; + + results += rc; + } + + /* If we support both LONG and SHORT, then run each in response to + * break or not. Otherwise, run the one we support + */ + mode = 0; + if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_OUT_BIST_CABLE_SHORT_LBN)) { + if ((flags & ETH_TEST_FL_OFFLINE) && + (phy_cfg->flags & + (1 << MC_CMD_GET_PHY_CFG_OUT_BIST_CABLE_LONG_LBN))) + mode = MC_CMD_PHY_BIST_CABLE_LONG; + else + mode = MC_CMD_PHY_BIST_CABLE_SHORT; + } else if (phy_cfg->flags & + (1 << MC_CMD_GET_PHY_CFG_OUT_BIST_CABLE_LONG_LBN)) + mode = MC_CMD_PHY_BIST_CABLE_LONG; + + if (mode != 0) { + rc = efx_mcdi_bist(efx, mode, results); + if (rc < 0) + return rc; + results += rc; + } + + return 0; +} + +const char *efx_mcdi_phy_test_name(struct efx_nic *efx, unsigned int index) +{ + struct efx_mcdi_phy_data *phy_cfg = efx->phy_data; + + if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_OUT_BIST_LBN)) { + if (index == 0) + return "bist"; + --index; + } + + if (phy_cfg->flags & ((1 << MC_CMD_GET_PHY_CFG_OUT_BIST_CABLE_SHORT_LBN) | + (1 << MC_CMD_GET_PHY_CFG_OUT_BIST_CABLE_LONG_LBN))) { + if (index == 0) + return "cable"; + --index; + + if (efx->phy_type == PHY_TYPE_SFT9001B) { + if (index < ARRAY_SIZE(mcdi_sft9001_cable_diag_names)) + return mcdi_sft9001_cable_diag_names[index]; + index -= ARRAY_SIZE(mcdi_sft9001_cable_diag_names); + } + } + + return NULL; +} + +#define SFP_PAGE_SIZE 128 +#define SFF_DIAG_TYPE_OFFSET 92 +#define SFF_DIAG_ADDR_CHANGE BIT(2) +#define SFF_8079_NUM_PAGES 2 +#define SFF_8472_NUM_PAGES 4 +#define SFF_8436_NUM_PAGES 5 +#define SFF_DMT_LEVEL_OFFSET 94 + +/** efx_mcdi_phy_get_module_eeprom_page() - Get a single page of module eeprom + * @efx: NIC context + * @page: EEPROM page number + * @data: Destination data pointer + * @offset: Offset in page to copy from in to data + * @space: Space available in data + * + * Return: + * >=0 - amount of data copied + * <0 - error + */ +static int efx_mcdi_phy_get_module_eeprom_page(struct efx_nic *efx, + unsigned int page, + u8 *data, ssize_t offset, + ssize_t space) +{ + MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PHY_MEDIA_INFO_OUT_LENMAX); + MCDI_DECLARE_BUF(inbuf, MC_CMD_GET_PHY_MEDIA_INFO_IN_LEN); + unsigned int payload_len; + unsigned int to_copy; + size_t outlen; + int rc; + + if (offset > SFP_PAGE_SIZE) + return -EINVAL; + + to_copy = min(space, SFP_PAGE_SIZE - offset); + + MCDI_SET_DWORD(inbuf, GET_PHY_MEDIA_INFO_IN_PAGE, page); + rc = efx_mcdi_rpc_quiet(efx, MC_CMD_GET_PHY_MEDIA_INFO, + inbuf, sizeof(inbuf), + outbuf, sizeof(outbuf), + &outlen); + + if (rc) + return rc; + + if (outlen < (MC_CMD_GET_PHY_MEDIA_INFO_OUT_DATA_OFST + + SFP_PAGE_SIZE)) + return -EIO; + + payload_len = MCDI_DWORD(outbuf, GET_PHY_MEDIA_INFO_OUT_DATALEN); + if (payload_len != SFP_PAGE_SIZE) + return -EIO; + + memcpy(data, MCDI_PTR(outbuf, GET_PHY_MEDIA_INFO_OUT_DATA) + offset, + to_copy); + + return to_copy; +} + +static int efx_mcdi_phy_get_module_eeprom_byte(struct efx_nic *efx, + unsigned int page, + u8 byte) +{ + u8 data; + int rc; + + rc = efx_mcdi_phy_get_module_eeprom_page(efx, page, &data, byte, 1); + if (rc == 1) + return data; + + return rc; +} + +static int efx_mcdi_phy_diag_type(struct efx_nic *efx) +{ + /* Page zero of the EEPROM includes the diagnostic type at byte 92. */ + return efx_mcdi_phy_get_module_eeprom_byte(efx, 0, + SFF_DIAG_TYPE_OFFSET); +} + +static int efx_mcdi_phy_sff_8472_level(struct efx_nic *efx) +{ + /* Page zero of the EEPROM includes the DMT level at byte 94. */ + return efx_mcdi_phy_get_module_eeprom_byte(efx, 0, + SFF_DMT_LEVEL_OFFSET); +} + +static u32 efx_mcdi_phy_module_type(struct efx_nic *efx) +{ + struct efx_mcdi_phy_data *phy_data = efx->phy_data; + + if (phy_data->media != MC_CMD_MEDIA_QSFP_PLUS) + return phy_data->media; + + /* A QSFP+ NIC may actually have an SFP+ module attached. + * The ID is page 0, byte 0. + */ + switch (efx_mcdi_phy_get_module_eeprom_byte(efx, 0, 0)) { + case 0x3: + return MC_CMD_MEDIA_SFP_PLUS; + case 0xc: + case 0xd: + return MC_CMD_MEDIA_QSFP_PLUS; + default: + return 0; + } +} + +int efx_mcdi_phy_get_module_eeprom(struct efx_nic *efx, struct ethtool_eeprom *ee, u8 *data) +{ + int rc; + ssize_t space_remaining = ee->len; + unsigned int page_off; + bool ignore_missing; + int num_pages; + int page; + + switch (efx_mcdi_phy_module_type(efx)) { + case MC_CMD_MEDIA_SFP_PLUS: + num_pages = efx_mcdi_phy_sff_8472_level(efx) > 0 ? + SFF_8472_NUM_PAGES : SFF_8079_NUM_PAGES; + page = 0; + ignore_missing = false; + break; + case MC_CMD_MEDIA_QSFP_PLUS: + num_pages = SFF_8436_NUM_PAGES; + page = -1; /* We obtain the lower page by asking for -1. */ + ignore_missing = true; /* Ignore missing pages after page 0. */ + break; + default: + return -EOPNOTSUPP; + } + + page_off = ee->offset % SFP_PAGE_SIZE; + page += ee->offset / SFP_PAGE_SIZE; + + while (space_remaining && (page < num_pages)) { + rc = efx_mcdi_phy_get_module_eeprom_page(efx, page, + data, page_off, + space_remaining); + + if (rc > 0) { + space_remaining -= rc; + data += rc; + page_off = 0; + page++; + } else if (rc == 0) { + space_remaining = 0; + } else if (ignore_missing && (page > 0)) { + int intended_size = SFP_PAGE_SIZE - page_off; + + space_remaining -= intended_size; + if (space_remaining < 0) { + space_remaining = 0; + } else { + memset(data, 0, intended_size); + data += intended_size; + page_off = 0; + page++; + rc = 0; + } + } else { + return rc; + } + } + + return 0; +} + +int efx_mcdi_phy_get_module_info(struct efx_nic *efx, struct ethtool_modinfo *modinfo) +{ + int sff_8472_level; + int diag_type; + + switch (efx_mcdi_phy_module_type(efx)) { + case MC_CMD_MEDIA_SFP_PLUS: + sff_8472_level = efx_mcdi_phy_sff_8472_level(efx); + + /* If we can't read the diagnostics level we have none. */ + if (sff_8472_level < 0) + return -EOPNOTSUPP; + + /* Check if this module requires the (unsupported) address + * change operation. + */ + diag_type = efx_mcdi_phy_diag_type(efx); + + if (sff_8472_level == 0 || + (diag_type & SFF_DIAG_ADDR_CHANGE)) { + modinfo->type = ETH_MODULE_SFF_8079; + modinfo->eeprom_len = ETH_MODULE_SFF_8079_LEN; + } else { + modinfo->type = ETH_MODULE_SFF_8472; + modinfo->eeprom_len = ETH_MODULE_SFF_8472_LEN; + } + break; + + case MC_CMD_MEDIA_QSFP_PLUS: + modinfo->type = ETH_MODULE_SFF_8436; + modinfo->eeprom_len = ETH_MODULE_SFF_8436_LEN; + break; + + default: + return -EOPNOTSUPP; + } + + return 0; +} + static unsigned int efx_calc_mac_mtu(struct efx_nic *efx) { return EFX_MAX_FRAME_LEN(efx->net_dev->mtu); diff --git a/drivers/net/ethernet/sfc/mcdi_port_common.h b/drivers/net/ethernet/sfc/mcdi_port_common.h index 9dbeee83266f..ed31690e591c 100644 --- a/drivers/net/ethernet/sfc/mcdi_port_common.h +++ b/drivers/net/ethernet/sfc/mcdi_port_common.h @@ -41,13 +41,22 @@ u8 mcdi_to_ethtool_media(u32 media); void efx_mcdi_phy_decode_link(struct efx_nic *efx, struct efx_link_state *link_state, u32 speed, u32 flags, u32 fcntl); -u32 ethtool_fec_caps_to_mcdi(u32 ethtool_cap); +u32 ethtool_fec_caps_to_mcdi(u32 supported_cap, u32 ethtool_cap); u32 mcdi_fec_caps_to_ethtool(u32 caps, bool is_25g); void efx_mcdi_phy_check_fcntl(struct efx_nic *efx, u32 lpa); bool efx_mcdi_phy_poll(struct efx_nic *efx); -int efx_mcdi_phy_get_fecparam(struct efx_nic *efx, - struct ethtool_fecparam *fec); +int efx_mcdi_phy_probe(struct efx_nic *efx); +void efx_mcdi_phy_remove(struct efx_nic *efx); +void efx_mcdi_phy_get_link_ksettings(struct efx_nic *efx, struct ethtool_link_ksettings *cmd); +int efx_mcdi_phy_set_link_ksettings(struct efx_nic *efx, const struct ethtool_link_ksettings *cmd); +int efx_mcdi_phy_get_fecparam(struct efx_nic *efx, struct ethtool_fecparam *fec); +int efx_mcdi_phy_set_fecparam(struct efx_nic *efx, const struct ethtool_fecparam *fec); int efx_mcdi_phy_test_alive(struct efx_nic *efx); +int efx_mcdi_port_reconfigure(struct efx_nic *efx); +int efx_mcdi_phy_run_tests(struct efx_nic *efx, int *results, unsigned int flags); +const char *efx_mcdi_phy_test_name(struct efx_nic *efx, unsigned int index); +int efx_mcdi_phy_get_module_eeprom(struct efx_nic *efx, struct ethtool_eeprom *ee, u8 *data); +int efx_mcdi_phy_get_module_info(struct efx_nic *efx, struct ethtool_modinfo *modinfo); int efx_mcdi_set_mac(struct efx_nic *efx); int efx_mcdi_set_mtu(struct efx_nic *efx); int efx_mcdi_mac_init_stats(struct efx_nic *efx); diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h index 062462a13847..9f7dfdf708cf 100644 --- a/drivers/net/ethernet/sfc/net_driver.h +++ b/drivers/net/ethernet/sfc/net_driver.h @@ -38,8 +38,6 @@ * **************************************************************************/ -#define EFX_DRIVER_VERSION "4.1" - #ifdef DEBUG #define EFX_WARN_ON_ONCE_PARANOID(x) WARN_ON_ONCE(x) #define EFX_WARN_ON_PARANOID(x) WARN_ON(x) @@ -65,10 +63,13 @@ * queues. */ #define EFX_MAX_TX_TC 2 #define EFX_MAX_CORE_TX_QUEUES (EFX_MAX_TX_TC * EFX_MAX_CHANNELS) -#define EFX_TXQ_TYPE_OFFLOAD 1 /* flag */ -#define EFX_TXQ_TYPE_HIGHPRI 2 /* flag */ -#define EFX_TXQ_TYPES 4 -#define EFX_MAX_TX_QUEUES (EFX_TXQ_TYPES * EFX_MAX_CHANNELS) +#define EFX_TXQ_TYPE_OUTER_CSUM 1 /* Outer checksum offload */ +#define EFX_TXQ_TYPE_INNER_CSUM 2 /* Inner checksum offload */ +#define EFX_TXQ_TYPE_HIGHPRI 4 /* High-priority (for TC) */ +#define EFX_TXQ_TYPES 8 +/* HIGHPRI is Siena-only, and INNER_CSUM is EF10, so no need for both */ +#define EFX_MAX_TXQ_PER_CHANNEL 4 +#define EFX_MAX_TX_QUEUES (EFX_MAX_TXQ_PER_CHANNEL * EFX_MAX_CHANNELS) /* Maximum possible MTU the driver supports */ #define EFX_MAX_MTU (9 * 1024) @@ -76,6 +77,9 @@ /* Minimum MTU, from RFC791 (IP) */ #define EFX_MIN_MTU 68 +/* Maximum total header length for TSOv2 */ +#define EFX_TSO2_MAX_HDRLEN 208 + /* Size of an RX scatter buffer. Small enough to pack 2 into a 4K page, * and should be a multiple of the cache line size. */ @@ -192,7 +196,9 @@ struct efx_tx_buffer { * @queue: DMA queue number * @label: Label for TX completion events. * Is our index within @channel->tx_queue array. + * @type: configuration type of this TX queue. A bitmask of %EFX_TXQ_TYPE_* flags. * @tso_version: Version of TSO in use for this queue. + * @tso_encap: Is encapsulated TSO supported? Supported in TSOv2 on 8000 series. * @channel: The associated channel * @core_txq: The networking core TX queue structure * @buffer: The software buffer ring @@ -206,8 +212,6 @@ struct efx_tx_buffer { * @initialised: Has hardware queue been initialised? * @timestamping: Is timestamping enabled for this channel? * @xdp_tx: Is this an XDP tx queue? - * @handle_tso: TSO xmit preparation handler. Sets up the TSO metadata and - * may also map tx data, depending on the nature of the TSO implementation. * @read_count: Current read pointer. * This is the number of buffers that have been removed from both rings. * @old_write_count: The value of @write_count when last checked. @@ -244,7 +248,7 @@ struct efx_tx_buffer { * @tso_fallbacks: Number of times TSO fallback used * @pushes: Number of times the TX push feature has been used * @pio_packets: Number of times the TX PIO feature has been used - * @xmit_more_available: Are any packets waiting to be pushed to the NIC + * @xmit_pending: Are any packets waiting to be pushed to the NIC * @cb_packets: Number of times the TX copybreak feature has been used * @notify_count: Count of notified descriptors to the NIC * @empty_read_count: If the completion path has seen the queue as empty @@ -256,7 +260,9 @@ struct efx_tx_queue { struct efx_nic *efx ____cacheline_aligned_in_smp; unsigned int queue; unsigned int label; + unsigned int type; unsigned int tso_version; + bool tso_encap; struct efx_channel *channel; struct netdev_queue *core_txq; struct efx_tx_buffer *buffer; @@ -269,9 +275,6 @@ struct efx_tx_queue { bool timestamping; bool xdp_tx; - /* Function pointers used in the fast path. */ - int (*handle_tso)(struct efx_tx_queue*, struct sk_buff*, bool *); - /* Members used mainly on the completion path */ unsigned int read_count ____cacheline_aligned_in_smp; unsigned int old_write_count; @@ -292,7 +295,7 @@ struct efx_tx_queue { unsigned int tso_fallbacks; unsigned int pushes; unsigned int pio_packets; - bool xmit_more_available; + bool xmit_pending; unsigned int cb_packets; unsigned int notify_count; /* Statistics to supplement MAC stats */ @@ -455,7 +458,7 @@ enum efx_sync_events_state { * were checked for expiry * @rfs_expire_index: next accelerated RFS filter ID to check for expiry * @n_rfs_succeeded: number of successful accelerated RFS filter insertions - * @n_rfs_failed; number of failed accelerated RFS filter insertions + * @n_rfs_failed: number of failed accelerated RFS filter insertions * @filter_work: Work item for efx_filter_rfs_expire() * @rps_flow_id: Flow IDs of filters allocated for accelerated RFS, * indexed by filter ID @@ -481,6 +484,7 @@ enum efx_sync_events_state { * @rx_list: list of SKBs from current RX, awaiting processing * @rx_queue: RX queue for this channel * @tx_queue: TX queues for this channel + * @tx_queue_by_type: pointers into @tx_queue, or %NULL, indexed by txq type * @sync_events_state: Current state of sync events on this channel * @sync_timestamp_major: Major part of the last ptp sync event * @sync_timestamp_minor: Minor part of the last ptp sync event @@ -542,7 +546,8 @@ struct efx_channel { struct list_head *rx_list; struct efx_rx_queue rx_queue; - struct efx_tx_queue tx_queue[EFX_TXQ_TYPES]; + struct efx_tx_queue tx_queue[EFX_MAX_TXQ_PER_CHANNEL]; + struct efx_tx_queue *tx_queue_by_type[EFX_TXQ_TYPES]; enum efx_sync_events_state sync_events_state; u32 sync_timestamp_major; @@ -658,51 +663,6 @@ static inline bool efx_link_state_equal(const struct efx_link_state *left, } /** - * struct efx_phy_operations - Efx PHY operations table - * @probe: Probe PHY and initialise efx->mdio.mode_support, efx->mdio.mmds, - * efx->loopback_modes. - * @init: Initialise PHY - * @fini: Shut down PHY - * @reconfigure: Reconfigure PHY (e.g. for new link parameters) - * @poll: Update @link_state and report whether it changed. - * Serialised by the mac_lock. - * @get_link_ksettings: Get ethtool settings. Serialised by the mac_lock. - * @set_link_ksettings: Set ethtool settings. Serialised by the mac_lock. - * @get_fecparam: Get Forward Error Correction settings. Serialised by mac_lock. - * @set_fecparam: Set Forward Error Correction settings. Serialised by mac_lock. - * @set_npage_adv: Set abilities advertised in (Extended) Next Page - * (only needed where AN bit is set in mmds) - * @test_alive: Test that PHY is 'alive' (online) - * @test_name: Get the name of a PHY-specific test/result - * @run_tests: Run tests and record results as appropriate (offline). - * Flags are the ethtool tests flags. - */ -struct efx_phy_operations { - int (*probe) (struct efx_nic *efx); - int (*init) (struct efx_nic *efx); - void (*fini) (struct efx_nic *efx); - void (*remove) (struct efx_nic *efx); - int (*reconfigure) (struct efx_nic *efx); - bool (*poll) (struct efx_nic *efx); - void (*get_link_ksettings)(struct efx_nic *efx, - struct ethtool_link_ksettings *cmd); - int (*set_link_ksettings)(struct efx_nic *efx, - const struct ethtool_link_ksettings *cmd); - int (*get_fecparam)(struct efx_nic *efx, struct ethtool_fecparam *fec); - int (*set_fecparam)(struct efx_nic *efx, - const struct ethtool_fecparam *fec); - void (*set_npage_adv) (struct efx_nic *efx, u32); - int (*test_alive) (struct efx_nic *efx); - const char *(*test_name) (struct efx_nic *efx, unsigned int index); - int (*run_tests) (struct efx_nic *efx, int *results, unsigned flags); - int (*get_module_eeprom) (struct efx_nic *efx, - struct ethtool_eeprom *ee, - u8 *data); - int (*get_module_info) (struct efx_nic *efx, - struct ethtool_modinfo *modinfo); -}; - -/** * enum efx_phy_mode - PHY operating mode flags * @PHY_MODE_NORMAL: on and should pass traffic * @PHY_MODE_TX_DISABLED: on with TX disabled @@ -920,7 +880,6 @@ struct efx_async_filter_insertion { * field of %MC_CMD_GET_CAPABILITIES_V4 response, or %MC_CMD_MAC_NSTATS) * @stats_buffer: DMA buffer for statistics * @phy_type: PHY type - * @phy_op: PHY interface * @phy_data: PHY private data (including PHY-specific stats) * @mdio: PHY MDIO interface * @mdio_bus: PHY MDIO bus ID (only used by Siena) @@ -1094,7 +1053,6 @@ struct efx_nic { bool rx_nodesc_drops_prev_state; unsigned int phy_type; - const struct efx_phy_operations *phy_op; void *phy_data; struct mdio_if_info mdio; unsigned int mdio_bus; @@ -1214,10 +1172,12 @@ struct efx_udp_tunnel { * @describe_stats: Describe statistics for ethtool * @update_stats: Update statistics not provided by event handling. * Either argument may be %NULL. + * @update_stats_atomic: Update statistics while in atomic context, if that + * is more limiting than @update_stats. Otherwise, leave %NULL and + * driver core will call @update_stats. * @start_stats: Start the regular fetching of statistics * @pull_stats: Pull stats from the NIC and wait until they arrive. * @stop_stats: Stop the regular fetching of statistics - * @set_id_led: Set state of identifying LED or revert to automatic function * @push_irq_moderation: Apply interrupt moderation value * @reconfigure_port: Push loopback/power/txdis changes to the MAC and PHY * @prepare_enable_fc_tx: Prepare MAC to enable pause frame TX (may be %NULL) @@ -1250,7 +1210,7 @@ struct efx_udp_tunnel { * a pointer to the &struct efx_msi_context for the channel. * @irq_handle_legacy: Handle legacy interrupt. The @dev_id argument * is a pointer to the &struct efx_nic. - * @tx_probe: Allocate resources for TX queue + * @tx_probe: Allocate resources for TX queue (and select TXQ type) * @tx_init: Initialise TX queue on the NIC * @tx_remove: Free resources for TX queue * @tx_write: Write TX descriptors and doorbell @@ -1359,10 +1319,11 @@ struct efx_nic_type { size_t (*describe_stats)(struct efx_nic *efx, u8 *names); size_t (*update_stats)(struct efx_nic *efx, u64 *full_stats, struct rtnl_link_stats64 *core_stats); + size_t (*update_stats_atomic)(struct efx_nic *efx, u64 *full_stats, + struct rtnl_link_stats64 *core_stats); void (*start_stats)(struct efx_nic *efx); void (*pull_stats)(struct efx_nic *efx); void (*stop_stats)(struct efx_nic *efx); - void (*set_id_led)(struct efx_nic *efx, enum efx_led_mode mode); void (*push_irq_moderation)(struct efx_channel *channel); int (*reconfigure_port)(struct efx_nic *efx); void (*prepare_enable_fc_tx)(struct efx_nic *efx); @@ -1546,14 +1507,6 @@ efx_get_tx_channel(struct efx_nic *efx, unsigned int index) return efx->channel[efx->tx_channel_offset + index]; } -static inline struct efx_tx_queue * -efx_get_tx_queue(struct efx_nic *efx, unsigned index, unsigned type) -{ - EFX_WARN_ON_ONCE_PARANOID(index >= efx->n_tx_channels || - type >= efx->tx_queues_per_channel); - return &efx->channel[efx->tx_channel_offset + index]->tx_queue[type]; -} - static inline struct efx_channel * efx_get_xdp_channel(struct efx_nic *efx, unsigned int index) { @@ -1580,10 +1533,18 @@ static inline unsigned int efx_channel_num_tx_queues(struct efx_channel *channel } static inline struct efx_tx_queue * -efx_channel_get_tx_queue(struct efx_channel *channel, unsigned type) +efx_channel_get_tx_queue(struct efx_channel *channel, unsigned int type) +{ + EFX_WARN_ON_ONCE_PARANOID(type >= EFX_TXQ_TYPES); + return channel->tx_queue_by_type[type]; +} + +static inline struct efx_tx_queue * +efx_get_tx_queue(struct efx_nic *efx, unsigned int index, unsigned int type) { - EFX_WARN_ON_ONCE_PARANOID(type >= efx_channel_num_tx_queues(channel)); - return &channel->tx_queue[type]; + struct efx_channel *channel = efx_get_tx_channel(efx, index); + + return efx_channel_get_tx_queue(channel, type); } /* Iterate over all TX queues belonging to a channel */ @@ -1683,10 +1644,6 @@ efx_channel_tx_fill_level(struct efx_channel *channel) struct efx_tx_queue *tx_queue; unsigned int fill_level = 0; - /* This function is currently only used by EF100, which maybe - * could do something simpler and just compute the fill level - * of the single TXQ that's really in use. - */ efx_for_each_channel_tx_queue(tx_queue, channel) fill_level = max(fill_level, tx_queue->insert_count - tx_queue->read_count); @@ -1694,6 +1651,20 @@ efx_channel_tx_fill_level(struct efx_channel *channel) return fill_level; } +/* Conservative approximation of efx_channel_tx_fill_level using cached value */ +static inline unsigned int +efx_channel_tx_old_fill_level(struct efx_channel *channel) +{ + struct efx_tx_queue *tx_queue; + unsigned int fill_level = 0; + + efx_for_each_channel_tx_queue(tx_queue, channel) + fill_level = max(fill_level, + tx_queue->insert_count - tx_queue->old_read_count); + + return fill_level; +} + /* Get all supported features. * If a feature is not fixed, it is present in hw_features. * If a feature is fixed, it does not present in hw_features, but diff --git a/drivers/net/ethernet/sfc/nic.h b/drivers/net/ethernet/sfc/nic.h index 724e2776b585..5c2fe3ce3f4d 100644 --- a/drivers/net/ethernet/sfc/nic.h +++ b/drivers/net/ethernet/sfc/nic.h @@ -297,6 +297,10 @@ struct efx_ef10_nic_data { u64 licensed_features; }; +/* TSOv2 */ +int efx_ef10_tx_tso_desc(struct efx_tx_queue *tx_queue, struct sk_buff *skb, + bool *data_mapped); + int efx_init_sriov(void); void efx_fini_sriov(void); diff --git a/drivers/net/ethernet/sfc/nic_common.h b/drivers/net/ethernet/sfc/nic_common.h index 974107354087..b9cafe9cd568 100644 --- a/drivers/net/ethernet/sfc/nic_common.h +++ b/drivers/net/ethernet/sfc/nic_common.h @@ -65,8 +65,7 @@ efx_tx_desc(struct efx_tx_queue *tx_queue, unsigned int index) /* Report whether this TX queue would be empty for the given write_count. * May return false negative. */ -static inline bool __efx_nic_tx_is_empty(struct efx_tx_queue *tx_queue, - unsigned int write_count) +static inline bool efx_nic_tx_is_empty(struct efx_tx_queue *tx_queue, unsigned int write_count) { unsigned int empty_read_count = READ_ONCE(tx_queue->empty_read_count); @@ -76,41 +75,6 @@ static inline bool __efx_nic_tx_is_empty(struct efx_tx_queue *tx_queue, return ((empty_read_count ^ write_count) & ~EFX_EMPTY_COUNT_VALID) == 0; } -/* Report whether the NIC considers this TX queue empty, using - * packet_write_count (the write count recorded for the last completable - * doorbell push). May return false negative. EF10 only, which is OK - * because only EF10 supports PIO. - */ -static inline bool efx_nic_tx_is_empty(struct efx_tx_queue *tx_queue) -{ - EFX_WARN_ON_ONCE_PARANOID(!tx_queue->efx->type->option_descriptors); - return __efx_nic_tx_is_empty(tx_queue, tx_queue->packet_write_count); -} - -/* Get partner of a TX queue, seen as part of the same net core queue */ -/* XXX is this a thing on EF100? */ -static inline struct efx_tx_queue *efx_tx_queue_partner(struct efx_tx_queue *tx_queue) -{ - if (tx_queue->label & EFX_TXQ_TYPE_OFFLOAD) - return tx_queue - EFX_TXQ_TYPE_OFFLOAD; - else - return tx_queue + EFX_TXQ_TYPE_OFFLOAD; -} - -/* Decide whether we can use TX PIO, ie. write packet data directly into - * a buffer on the device. This can reduce latency at the expense of - * throughput, so we only do this if both hardware and software TX rings - * are empty. This also ensures that only one packet at a time can be - * using the PIO buffer. - */ -static inline bool efx_nic_may_tx_pio(struct efx_tx_queue *tx_queue) -{ - struct efx_tx_queue *partner = efx_tx_queue_partner(tx_queue); - - return tx_queue->piobuf && efx_nic_tx_is_empty(tx_queue) && - efx_nic_tx_is_empty(partner); -} - int efx_enqueue_skb_tso(struct efx_tx_queue *tx_queue, struct sk_buff *skb, bool *data_mapped); @@ -125,7 +89,7 @@ int efx_enqueue_skb_tso(struct efx_tx_queue *tx_queue, struct sk_buff *skb, static inline bool efx_nic_may_push_tx_desc(struct efx_tx_queue *tx_queue, unsigned int write_count) { - bool was_empty = __efx_nic_tx_is_empty(tx_queue, write_count); + bool was_empty = efx_nic_tx_is_empty(tx_queue, write_count); tx_queue->empty_read_count = 0; return was_empty && tx_queue->write_count - write_count == 1; @@ -280,6 +244,13 @@ void efx_nic_update_stats(const struct efx_hw_stat_desc *desc, size_t count, const unsigned long *mask, u64 *stats, const void *dma_buf, bool accumulate); void efx_nic_fix_nodesc_drop_stat(struct efx_nic *efx, u64 *stat); +static inline size_t efx_nic_update_stats_atomic(struct efx_nic *efx, u64 *full_stats, + struct rtnl_link_stats64 *core_stats) +{ + if (efx->type->update_stats_atomic) + return efx->type->update_stats_atomic(efx, full_stats, core_stats); + return efx->type->update_stats(efx, full_stats, core_stats); +} #define EFX_MAX_FLUSH_TIME 5000 diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c index bea4725a4499..a39c5143b386 100644 --- a/drivers/net/ethernet/sfc/ptp.c +++ b/drivers/net/ethernet/sfc/ptp.c @@ -43,6 +43,7 @@ #include "mcdi_pcol.h" #include "io.h" #include "farch_regs.h" +#include "tx.h" #include "nic.h" /* indirectly includes ptp.h */ /* Maximum number of events expected to make up a PTP event */ @@ -172,9 +173,11 @@ struct efx_ptp_match { /** * struct efx_ptp_event_rx - A PTP receive event (from MC) + * @link: list of events * @seq0: First part of (PTP) UUID * @seq1: Second part of (PTP) UUID and sequence number * @hwtimestamp: Event timestamp + * @expiry: Time which the packet arrived */ struct efx_ptp_event_rx { struct list_head link; @@ -222,11 +225,13 @@ struct efx_ptp_timeset { * reset (disable, enable). * @rxfilter_event: Receive filter when operating * @rxfilter_general: Receive filter when operating + * @rxfilter_installed: Receive filter installed * @config: Current timestamp configuration * @enabled: PTP operation enabled * @mode: Mode in which PTP operating (PTP version) * @ns_to_nic_time: Function to convert from scalar nanoseconds to NIC time * @nic_to_kernel_time: Function to convert from NIC to kernel time + * @nic_time: contains time details * @nic_time.minor_max: Wrap point for NIC minor times * @nic_time.sync_event_diff_min: Minimum acceptable difference between time * in packet prefix and last MCDI time sync event i.e. how much earlier than @@ -238,6 +243,7 @@ struct efx_ptp_timeset { * field in MCDI time sync event. * @min_synchronisation_ns: Minimum acceptable corrected sync window * @capabilities: Capabilities flags from the NIC + * @ts_corrections: contains corrections details * @ts_corrections.ptp_tx: Required driver correction of PTP packet transmit * timestamps * @ts_corrections.ptp_rx: Required driver correction of PTP packet receive @@ -325,7 +331,7 @@ struct efx_ptp_data { struct work_struct pps_work; struct workqueue_struct *pps_workwq; bool nic_ts_enabled; - _MCDI_DECLARE_BUF(txbuf, MC_CMD_PTP_IN_TRANSMIT_LENMAX); + efx_dword_t txbuf[MCDI_TX_BUF_LEN(MC_CMD_PTP_IN_TRANSMIT_LENMAX)]; unsigned int good_syncs; unsigned int fast_syncs; @@ -1082,10 +1088,10 @@ static int efx_ptp_synchronize(struct efx_nic *efx, unsigned int num_readings) static void efx_ptp_xmit_skb_queue(struct efx_nic *efx, struct sk_buff *skb) { struct efx_ptp_data *ptp_data = efx->ptp_data; + u8 type = efx_tx_csum_type_skb(skb); struct efx_tx_queue *tx_queue; - u8 type = skb->ip_summed == CHECKSUM_PARTIAL ? EFX_TXQ_TYPE_OFFLOAD : 0; - tx_queue = &ptp_data->channel->tx_queue[type]; + tx_queue = efx_channel_get_tx_queue(ptp_data->channel, type); if (tx_queue && tx_queue->timestamping) { efx_enqueue_skb(tx_queue, skb); } else { diff --git a/drivers/net/ethernet/sfc/selftest.c b/drivers/net/ethernet/sfc/selftest.c index e71d6d37a317..3c5227afd497 100644 --- a/drivers/net/ethernet/sfc/selftest.c +++ b/drivers/net/ethernet/sfc/selftest.c @@ -21,6 +21,7 @@ #include "efx_common.h" #include "efx_channels.h" #include "nic.h" +#include "mcdi_port_common.h" #include "selftest.h" #include "workarounds.h" @@ -67,7 +68,7 @@ static const char *const efx_interrupt_mode_names[] = { STRING_TABLE_LOOKUP(efx->interrupt_mode, efx_interrupt_mode) /** - * efx_loopback_state - persistent state during a loopback selftest + * struct efx_loopback_state - persistent state during a loopback selftest * @flush: Drop all packets in efx_loopback_rx_packet * @packet_count: Number of packets being used in this test * @skbs: An array of skbs transmitted @@ -99,10 +100,8 @@ static int efx_test_phy_alive(struct efx_nic *efx, struct efx_self_tests *tests) { int rc = 0; - if (efx->phy_op->test_alive) { - rc = efx->phy_op->test_alive(efx); - tests->phy_alive = rc ? -1 : 1; - } + rc = efx_mcdi_phy_test_alive(efx); + tests->phy_alive = rc ? -1 : 1; return rc; } @@ -257,11 +256,8 @@ static int efx_test_phy(struct efx_nic *efx, struct efx_self_tests *tests, { int rc; - if (!efx->phy_op->run_tests) - return 0; - mutex_lock(&efx->mac_lock); - rc = efx->phy_op->run_tests(efx, tests->phy_ext, flags); + rc = efx_mcdi_phy_run_tests(efx, tests->phy_ext, flags); mutex_unlock(&efx->mac_lock); if (rc == -EPERM) rc = 0; @@ -660,8 +656,8 @@ static int efx_test_loopbacks(struct efx_nic *efx, struct efx_self_tests *tests, /* Test all enabled types of TX queue */ efx_for_each_channel_tx_queue(tx_queue, channel) { - state->offload_csum = (tx_queue->label & - EFX_TXQ_TYPE_OFFLOAD); + state->offload_csum = (tx_queue->type & + EFX_TXQ_TYPE_OUTER_CSUM); rc = efx_test_loopback(tx_queue, &tests->loopback[mode]); if (rc) diff --git a/drivers/net/ethernet/sfc/selftest.h b/drivers/net/ethernet/sfc/selftest.h index ca88ebb4f6b1..a23f085bf298 100644 --- a/drivers/net/ethernet/sfc/selftest.h +++ b/drivers/net/ethernet/sfc/selftest.h @@ -15,8 +15,8 @@ */ struct efx_loopback_self_tests { - int tx_sent[EFX_TXQ_TYPES]; - int tx_done[EFX_TXQ_TYPES]; + int tx_sent[EFX_MAX_TXQ_PER_CHANNEL]; + int tx_done[EFX_MAX_TXQ_PER_CHANNEL]; int rx_good; int rx_bad; }; diff --git a/drivers/net/ethernet/sfc/siena.c b/drivers/net/ethernet/sfc/siena.c index a7ea630bb5e6..16347a6d0c47 100644 --- a/drivers/net/ethernet/sfc/siena.c +++ b/drivers/net/ethernet/sfc/siena.c @@ -994,7 +994,6 @@ const struct efx_nic_type siena_a0_nic_type = { .start_stats = efx_mcdi_mac_start_stats, .pull_stats = efx_mcdi_mac_pull_stats, .stop_stats = efx_mcdi_mac_stop_stats, - .set_id_led = efx_mcdi_set_id_led, .push_irq_moderation = siena_push_irq_moderation, .reconfigure_mac = siena_mac_reconfigure, .check_mac_fault = efx_mcdi_mac_check_fault, diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c index 727201d5eb24..1665529a7271 100644 --- a/drivers/net/ethernet/sfc/tx.c +++ b/drivers/net/ethernet/sfc/tx.c @@ -59,13 +59,12 @@ u8 *efx_tx_get_copy_buffer_limited(struct efx_tx_queue *tx_queue, static void efx_tx_maybe_stop_queue(struct efx_tx_queue *txq1) { - /* We need to consider both queues that the net core sees as one */ - struct efx_tx_queue *txq2 = efx_tx_queue_partner(txq1); + /* We need to consider all queues that the net core sees as one */ struct efx_nic *efx = txq1->efx; + struct efx_tx_queue *txq2; unsigned int fill_level; - fill_level = max(txq1->insert_count - txq1->old_read_count, - txq2->insert_count - txq2->old_read_count); + fill_level = efx_channel_tx_old_fill_level(txq1->channel); if (likely(fill_level < efx->txq_stop_thresh)) return; @@ -85,11 +84,10 @@ static void efx_tx_maybe_stop_queue(struct efx_tx_queue *txq1) */ netif_tx_stop_queue(txq1->core_txq); smp_mb(); - txq1->old_read_count = READ_ONCE(txq1->read_count); - txq2->old_read_count = READ_ONCE(txq2->read_count); + efx_for_each_channel_tx_queue(txq2, txq1->channel) + txq2->old_read_count = READ_ONCE(txq2->read_count); - fill_level = max(txq1->insert_count - txq1->old_read_count, - txq2->insert_count - txq2->old_read_count); + fill_level = efx_channel_tx_old_fill_level(txq1->channel); EFX_WARN_ON_ONCE_PARANOID(fill_level >= efx->txq_entries); if (likely(fill_level < efx->txq_stop_thresh)) { smp_mb(); @@ -266,8 +264,45 @@ static int efx_enqueue_skb_pio(struct efx_tx_queue *tx_queue, ++tx_queue->insert_count; return 0; } + +/* Decide whether we can use TX PIO, ie. write packet data directly into + * a buffer on the device. This can reduce latency at the expense of + * throughput, so we only do this if both hardware and software TX rings + * are empty, including all queues for the channel. This also ensures that + * only one packet at a time can be using the PIO buffer. If the xmit_more + * flag is set then we don't use this - there'll be another packet along + * shortly and we want to hold off the doorbell. + */ +static bool efx_tx_may_pio(struct efx_tx_queue *tx_queue) +{ + struct efx_channel *channel = tx_queue->channel; + + if (!tx_queue->piobuf) + return false; + + EFX_WARN_ON_ONCE_PARANOID(!channel->efx->type->option_descriptors); + + efx_for_each_channel_tx_queue(tx_queue, channel) + if (!efx_nic_tx_is_empty(tx_queue, tx_queue->packet_write_count)) + return false; + + return true; +} #endif /* EFX_USE_PIO */ +/* Send any pending traffic for a channel. xmit_more is shared across all + * queues for a channel, so we must check all of them. + */ +static void efx_tx_send_pending(struct efx_channel *channel) +{ + struct efx_tx_queue *q; + + efx_for_each_channel_tx_queue(q, channel) { + if (q->xmit_pending) + efx_nic_push_buffers(q); + } +} + /* * Add a socket buffer to a TX queue * @@ -303,8 +338,18 @@ netdev_tx_t __efx_enqueue_skb(struct efx_tx_queue *tx_queue, struct sk_buff *skb * size limit. */ if (segments) { - EFX_WARN_ON_ONCE_PARANOID(!tx_queue->handle_tso); - rc = tx_queue->handle_tso(tx_queue, skb, &data_mapped); + switch (tx_queue->tso_version) { + case 1: + rc = efx_enqueue_skb_tso(tx_queue, skb, &data_mapped); + break; + case 2: + rc = efx_ef10_tx_tso_desc(tx_queue, skb, &data_mapped); + break; + case 0: /* No TSO on this queue, SW fallback needed */ + default: + rc = -EINVAL; + break; + } if (rc == -EINVAL) { rc = efx_tx_tso_fallback(tx_queue, skb); tx_queue->tso_fallbacks++; @@ -315,7 +360,7 @@ netdev_tx_t __efx_enqueue_skb(struct efx_tx_queue *tx_queue, struct sk_buff *skb goto err; #ifdef EFX_USE_PIO } else if (skb_len <= efx_piobuf_size && !xmit_more && - efx_nic_may_tx_pio(tx_queue)) { + efx_tx_may_pio(tx_queue)) { /* Use PIO for short packets with an empty queue. */ if (efx_enqueue_skb_pio(tx_queue, skb)) goto err; @@ -336,21 +381,11 @@ netdev_tx_t __efx_enqueue_skb(struct efx_tx_queue *tx_queue, struct sk_buff *skb efx_tx_maybe_stop_queue(tx_queue); - /* Pass off to hardware */ - if (__netdev_tx_sent_queue(tx_queue->core_txq, skb_len, xmit_more)) { - struct efx_tx_queue *txq2 = efx_tx_queue_partner(tx_queue); + tx_queue->xmit_pending = true; - /* There could be packets left on the partner queue if - * xmit_more was set. If we do not push those they - * could be left for a long time and cause a netdev watchdog. - */ - if (txq2->xmit_more_available) - efx_nic_push_buffers(txq2); - - efx_nic_push_buffers(tx_queue); - } else { - tx_queue->xmit_more_available = xmit_more; - } + /* Pass off to hardware */ + if (__netdev_tx_sent_queue(tx_queue->core_txq, skb_len, xmit_more)) + efx_tx_send_pending(tx_queue->channel); if (segments) { tx_queue->tso_bursts++; @@ -371,14 +406,8 @@ err: * on this queue or a partner queue then we need to push here to get the * previous packets out. */ - if (!xmit_more) { - struct efx_tx_queue *txq2 = efx_tx_queue_partner(tx_queue); - - if (txq2->xmit_more_available) - efx_nic_push_buffers(txq2); - - efx_nic_push_buffers(tx_queue); - } + if (!xmit_more) + efx_tx_send_pending(tx_queue->channel); return NETDEV_TX_OK; } @@ -472,13 +501,10 @@ int efx_xdp_tx_buffers(struct efx_nic *efx, int n, struct xdp_frame **xdpfs, } /* Initiate a packet transmission. We use one channel per CPU - * (sharing when we have more CPUs than channels). On Falcon, the TX - * completion events will be directed back to the CPU that transmitted - * the packet, which should be cache-efficient. + * (sharing when we have more CPUs than channels). * * Context: non-blocking. - * Note that returning anything other than NETDEV_TX_OK will cause the - * OS to free the skb. + * Should always return NETDEV_TX_OK and consume the skb. */ netdev_tx_t efx_hard_start_xmit(struct sk_buff *skb, struct net_device *net_dev) @@ -489,19 +515,39 @@ netdev_tx_t efx_hard_start_xmit(struct sk_buff *skb, EFX_WARN_ON_PARANOID(!netif_device_present(net_dev)); + index = skb_get_queue_mapping(skb); + type = efx_tx_csum_type_skb(skb); + if (index >= efx->n_tx_channels) { + index -= efx->n_tx_channels; + type |= EFX_TXQ_TYPE_HIGHPRI; + } + /* PTP "event" packet */ if (unlikely(efx_xmit_with_hwtstamp(skb)) && unlikely(efx_ptp_is_ptp_tx(efx, skb))) { + /* There may be existing transmits on the channel that are + * waiting for this packet to trigger the doorbell write. + * We need to send the packets at this point. + */ + efx_tx_send_pending(efx_get_tx_channel(efx, index)); return efx_ptp_tx(efx, skb); } - index = skb_get_queue_mapping(skb); - type = skb->ip_summed == CHECKSUM_PARTIAL ? EFX_TXQ_TYPE_OFFLOAD : 0; - if (index >= efx->n_tx_channels) { - index -= efx->n_tx_channels; - type |= EFX_TXQ_TYPE_HIGHPRI; - } tx_queue = efx_get_tx_queue(efx, index, type); + if (WARN_ON_ONCE(!tx_queue)) { + /* We don't have a TXQ of the right type. + * This should never happen, as we don't advertise offload + * features unless we can support them. + */ + dev_kfree_skb_any(skb); + /* If we're not expecting another transmit and we had something to push + * on this queue or a partner queue then we need to push here to get the + * previous packets out. + */ + if (!netdev_xmit_more()) + efx_tx_send_pending(tx_queue->channel); + return NETDEV_TX_OK; + } return __efx_enqueue_skb(tx_queue, skb); } @@ -552,7 +598,7 @@ void efx_init_tx_queue_core_txq(struct efx_tx_queue *tx_queue) tx_queue->core_txq = netdev_get_tx_queue(efx->net_dev, tx_queue->channel->channel + - ((tx_queue->label & EFX_TXQ_TYPE_HIGHPRI) ? + ((tx_queue->type & EFX_TXQ_TYPE_HIGHPRI) ? efx->n_tx_channels : 0)); } diff --git a/drivers/net/ethernet/sfc/tx.h b/drivers/net/ethernet/sfc/tx.h index a3cf06c5570d..f2c4d2f89919 100644 --- a/drivers/net/ethernet/sfc/tx.h +++ b/drivers/net/ethernet/sfc/tx.h @@ -18,4 +18,30 @@ unsigned int efx_tx_limit_len(struct efx_tx_queue *tx_queue, u8 *efx_tx_get_copy_buffer_limited(struct efx_tx_queue *tx_queue, struct efx_tx_buffer *buffer, size_t len); +/* What TXQ type will satisfy the checksum offloads required for this skb? */ +static inline unsigned int efx_tx_csum_type_skb(struct sk_buff *skb) +{ + if (skb->ip_summed != CHECKSUM_PARTIAL) + return 0; /* no checksum offload */ + + if (skb->encapsulation && + skb_checksum_start_offset(skb) == skb_inner_transport_offset(skb)) { + /* we only advertise features for IPv4 and IPv6 checksums on + * encapsulated packets, so if the checksum is for the inner + * packet, it must be one of them; no further checking required. + */ + + /* Do we also need to offload the outer header checksum? */ + if (skb_shinfo(skb)->gso_segs > 1 && + !(skb_shinfo(skb)->gso_type & SKB_GSO_PARTIAL) && + (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL_CSUM)) + return EFX_TXQ_TYPE_OUTER_CSUM | EFX_TXQ_TYPE_INNER_CSUM; + return EFX_TXQ_TYPE_INNER_CSUM; + } + + /* similarly, we only advertise features for IPv4 and IPv6 checksums, + * so it must be one of them. No need for further checks. + */ + return EFX_TXQ_TYPE_OUTER_CSUM; +} #endif /* EFX_TX_H */ diff --git a/drivers/net/ethernet/sfc/tx_common.c b/drivers/net/ethernet/sfc/tx_common.c index 793e234819a8..d530cde2b864 100644 --- a/drivers/net/ethernet/sfc/tx_common.c +++ b/drivers/net/ethernet/sfc/tx_common.c @@ -47,11 +47,12 @@ int efx_probe_tx_queue(struct efx_tx_queue *tx_queue) goto fail1; } - /* Allocate hardware ring */ + /* Allocate hardware ring, determine TXQ type */ rc = efx_nic_probe_tx(tx_queue); if (rc) goto fail2; + tx_queue->channel->tx_queue_by_type[tx_queue->type] = tx_queue; return 0; fail2: @@ -78,18 +79,14 @@ void efx_init_tx_queue(struct efx_tx_queue *tx_queue) tx_queue->read_count = 0; tx_queue->old_read_count = 0; tx_queue->empty_read_count = 0 | EFX_EMPTY_COUNT_VALID; - tx_queue->xmit_more_available = false; + tx_queue->xmit_pending = false; tx_queue->timestamping = (efx_ptp_use_mac_tx_timestamps(efx) && tx_queue->channel == efx_ptp_channel(efx)); tx_queue->completed_timestamp_major = 0; tx_queue->completed_timestamp_minor = 0; tx_queue->xdp_tx = efx_channel_is_xdp_tx(tx_queue->channel); - - /* Set up default function pointers. These may get replaced by - * efx_nic_init_tx() based off NIC/queue capabilities. - */ - tx_queue->handle_tso = efx_enqueue_skb_tso; + tx_queue->tso_version = 0; /* Set up TX descriptor ring */ efx_nic_init_tx(tx_queue); @@ -116,7 +113,7 @@ void efx_fini_tx_queue(struct efx_tx_queue *tx_queue) ++tx_queue->read_count; } - tx_queue->xmit_more_available = false; + tx_queue->xmit_pending = false; netdev_tx_reset_queue(tx_queue->core_txq); } @@ -141,6 +138,7 @@ void efx_remove_tx_queue(struct efx_tx_queue *tx_queue) kfree(tx_queue->buffer); tx_queue->buffer = NULL; + tx_queue->channel->tx_queue_by_type[tx_queue->type] = NULL; } void efx_dequeue_buffer(struct efx_tx_queue *tx_queue, @@ -242,7 +240,6 @@ void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index) { unsigned int fill_level, pkts_compl = 0, bytes_compl = 0; struct efx_nic *efx = tx_queue->efx; - struct efx_tx_queue *txq2; EFX_WARN_ON_ONCE_PARANOID(index > tx_queue->ptr_mask); @@ -261,9 +258,7 @@ void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index) if (unlikely(netif_tx_queue_stopped(tx_queue->core_txq)) && likely(efx->port_enabled) && likely(netif_device_present(efx->net_dev))) { - txq2 = efx_tx_queue_partner(tx_queue); - fill_level = max(tx_queue->insert_count - tx_queue->read_count, - txq2->insert_count - txq2->read_count); + fill_level = efx_channel_tx_fill_level(tx_queue->channel); if (fill_level <= efx->txq_wake_thresh) netif_tx_wake_queue(tx_queue->core_txq); } diff --git a/drivers/net/ethernet/silan/sc92031.c b/drivers/net/ethernet/silan/sc92031.c index f94078f8ebe5..1fd08a04bd4e 100644 --- a/drivers/net/ethernet/silan/sc92031.c +++ b/drivers/net/ethernet/silan/sc92031.c @@ -301,6 +301,7 @@ struct sc92031_priv { /* for dev->get_stats */ long rx_value; + struct net_device *ndev; }; /* I don't know which registers can be safely read; however, I can guess @@ -829,10 +830,10 @@ static void _sc92031_link_tasklet(struct net_device *dev) } } -static void sc92031_tasklet(unsigned long data) +static void sc92031_tasklet(struct tasklet_struct *t) { - struct net_device *dev = (struct net_device *)data; - struct sc92031_priv *priv = netdev_priv(dev); + struct sc92031_priv *priv = from_tasklet(priv, t, tasklet); + struct net_device *dev = priv->ndev; void __iomem *port_base = priv->port_base; u32 intr_status, intr_mask; @@ -993,15 +994,15 @@ static int sc92031_open(struct net_device *dev) struct sc92031_priv *priv = netdev_priv(dev); struct pci_dev *pdev = priv->pdev; - priv->rx_ring = pci_alloc_consistent(pdev, RX_BUF_LEN, - &priv->rx_ring_dma_addr); + priv->rx_ring = dma_alloc_coherent(&pdev->dev, RX_BUF_LEN, + &priv->rx_ring_dma_addr, GFP_KERNEL); if (unlikely(!priv->rx_ring)) { err = -ENOMEM; goto out_alloc_rx_ring; } - priv->tx_bufs = pci_alloc_consistent(pdev, TX_BUF_TOT_LEN, - &priv->tx_bufs_dma_addr); + priv->tx_bufs = dma_alloc_coherent(&pdev->dev, TX_BUF_TOT_LEN, + &priv->tx_bufs_dma_addr, GFP_KERNEL); if (unlikely(!priv->tx_bufs)) { err = -ENOMEM; goto out_alloc_tx_bufs; @@ -1031,11 +1032,11 @@ static int sc92031_open(struct net_device *dev) return 0; out_request_irq: - pci_free_consistent(pdev, TX_BUF_TOT_LEN, priv->tx_bufs, - priv->tx_bufs_dma_addr); + dma_free_coherent(&pdev->dev, TX_BUF_TOT_LEN, priv->tx_bufs, + priv->tx_bufs_dma_addr); out_alloc_tx_bufs: - pci_free_consistent(pdev, RX_BUF_LEN, priv->rx_ring, - priv->rx_ring_dma_addr); + dma_free_coherent(&pdev->dev, RX_BUF_LEN, priv->rx_ring, + priv->rx_ring_dma_addr); out_alloc_rx_ring: return err; } @@ -1058,10 +1059,10 @@ static int sc92031_stop(struct net_device *dev) spin_unlock_bh(&priv->lock); free_irq(pdev->irq, dev); - pci_free_consistent(pdev, TX_BUF_TOT_LEN, priv->tx_bufs, - priv->tx_bufs_dma_addr); - pci_free_consistent(pdev, RX_BUF_LEN, priv->rx_ring, - priv->rx_ring_dma_addr); + dma_free_coherent(&pdev->dev, TX_BUF_TOT_LEN, priv->tx_bufs, + priv->tx_bufs_dma_addr); + dma_free_coherent(&pdev->dev, RX_BUF_LEN, priv->rx_ring, + priv->rx_ring_dma_addr); return 0; } @@ -1108,7 +1109,7 @@ static void sc92031_poll_controller(struct net_device *dev) disable_irq(irq); if (sc92031_interrupt(irq, dev) != IRQ_NONE) - sc92031_tasklet((unsigned long)dev); + sc92031_tasklet(&priv->tasklet); enable_irq(irq); } #endif @@ -1407,11 +1408,11 @@ static int sc92031_probe(struct pci_dev *pdev, const struct pci_device_id *id) pci_set_master(pdev); - err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32)); + err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)); if (unlikely(err < 0)) goto out_set_dma_mask; - err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32)); + err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32)); if (unlikely(err < 0)) goto out_set_dma_mask; @@ -1443,10 +1444,11 @@ static int sc92031_probe(struct pci_dev *pdev, const struct pci_device_id *id) dev->ethtool_ops = &sc92031_ethtool_ops; priv = netdev_priv(dev); + priv->ndev = dev; spin_lock_init(&priv->lock); priv->port_base = port_base; priv->pdev = pdev; - tasklet_init(&priv->tasklet, sc92031_tasklet, (unsigned long)dev); + tasklet_setup(&priv->tasklet, sc92031_tasklet); /* Fudge tasklet count so the call to sc92031_enable_interrupts at * sc92031_open will work correctly */ tasklet_disable_nosync(&priv->tasklet); diff --git a/drivers/net/ethernet/sis/sis900.c b/drivers/net/ethernet/sis/sis900.c index cfa460c7db23..620c26f71be8 100644 --- a/drivers/net/ethernet/sis/sis900.c +++ b/drivers/net/ethernet/sis/sis900.c @@ -789,10 +789,9 @@ static u16 sis900_default_phy(struct net_device * net_dev) static void sis900_set_capability(struct net_device *net_dev, struct mii_phy *phy) { u16 cap; - u16 status; - status = mdio_read(net_dev, phy->phy_addr, MII_STATUS); - status = mdio_read(net_dev, phy->phy_addr, MII_STATUS); + mdio_read(net_dev, phy->phy_addr, MII_STATUS); + mdio_read(net_dev, phy->phy_addr, MII_STATUS); cap = MII_NWAY_CSMA_CD | ((phy->status & MII_STAT_CAN_TX_FDX)? MII_NWAY_TX_FDX:0) | @@ -1302,7 +1301,7 @@ static void sis630_set_eq(struct net_device *net_dev, u8 revision) /** * sis900_timer - sis900 timer routine - * @data: pointer to sis900 net device + * @t: timer list containing a pointer to sis900 net device * * On each timer ticks we check two things, * link status (ON/OFF) and link mode (10/100/Full/Half) @@ -1536,6 +1535,7 @@ static void sis900_read_mode(struct net_device *net_dev, int *speed, int *duplex /** * sis900_tx_timeout - sis900 transmit timeout routine * @net_dev: the net device to transmit + * @txqueue: index of hanging queue * * print transmit timeout status * disable interrupts and do some tasks diff --git a/drivers/net/ethernet/smsc/epic100.c b/drivers/net/ethernet/smsc/epic100.c index d950b312c418..51cd7dca91cd 100644 --- a/drivers/net/ethernet/smsc/epic100.c +++ b/drivers/net/ethernet/smsc/epic100.c @@ -374,13 +374,15 @@ static int epic_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) ep->mii.phy_id_mask = 0x1f; ep->mii.reg_num_mask = 0x1f; - ring_space = pci_alloc_consistent(pdev, TX_TOTAL_SIZE, &ring_dma); + ring_space = dma_alloc_coherent(&pdev->dev, TX_TOTAL_SIZE, &ring_dma, + GFP_KERNEL); if (!ring_space) goto err_out_iounmap; ep->tx_ring = ring_space; ep->tx_ring_dma = ring_dma; - ring_space = pci_alloc_consistent(pdev, RX_TOTAL_SIZE, &ring_dma); + ring_space = dma_alloc_coherent(&pdev->dev, RX_TOTAL_SIZE, &ring_dma, + GFP_KERNEL); if (!ring_space) goto err_out_unmap_tx; ep->rx_ring = ring_space; @@ -493,9 +495,11 @@ out: return ret; err_out_unmap_rx: - pci_free_consistent(pdev, RX_TOTAL_SIZE, ep->rx_ring, ep->rx_ring_dma); + dma_free_coherent(&pdev->dev, RX_TOTAL_SIZE, ep->rx_ring, + ep->rx_ring_dma); err_out_unmap_tx: - pci_free_consistent(pdev, TX_TOTAL_SIZE, ep->tx_ring, ep->tx_ring_dma); + dma_free_coherent(&pdev->dev, TX_TOTAL_SIZE, ep->tx_ring, + ep->tx_ring_dma); err_out_iounmap: pci_iounmap(pdev, ioaddr); err_out_free_netdev: @@ -918,8 +922,10 @@ static void epic_init_ring(struct net_device *dev) if (skb == NULL) break; skb_reserve(skb, 2); /* 16 byte align the IP header. */ - ep->rx_ring[i].bufaddr = pci_map_single(ep->pci_dev, - skb->data, ep->rx_buf_sz, PCI_DMA_FROMDEVICE); + ep->rx_ring[i].bufaddr = dma_map_single(&ep->pci_dev->dev, + skb->data, + ep->rx_buf_sz, + DMA_FROM_DEVICE); ep->rx_ring[i].rxstatus = DescOwn; } ep->dirty_rx = (unsigned int)(i - RX_RING_SIZE); @@ -955,8 +961,9 @@ static netdev_tx_t epic_start_xmit(struct sk_buff *skb, struct net_device *dev) entry = ep->cur_tx % TX_RING_SIZE; ep->tx_skbuff[entry] = skb; - ep->tx_ring[entry].bufaddr = pci_map_single(ep->pci_dev, skb->data, - skb->len, PCI_DMA_TODEVICE); + ep->tx_ring[entry].bufaddr = dma_map_single(&ep->pci_dev->dev, + skb->data, skb->len, + DMA_TO_DEVICE); if (free_count < TX_QUEUE_LEN/2) {/* Typical path */ ctrl_word = 0x100000; /* No interrupt */ } else if (free_count == TX_QUEUE_LEN/2) { @@ -1036,8 +1043,9 @@ static void epic_tx(struct net_device *dev, struct epic_private *ep) /* Free the original skb. */ skb = ep->tx_skbuff[entry]; - pci_unmap_single(ep->pci_dev, ep->tx_ring[entry].bufaddr, - skb->len, PCI_DMA_TODEVICE); + dma_unmap_single(&ep->pci_dev->dev, + ep->tx_ring[entry].bufaddr, skb->len, + DMA_TO_DEVICE); dev_consume_skb_irq(skb); ep->tx_skbuff[entry] = NULL; } @@ -1178,20 +1186,21 @@ static int epic_rx(struct net_device *dev, int budget) if (pkt_len < rx_copybreak && (skb = netdev_alloc_skb(dev, pkt_len + 2)) != NULL) { skb_reserve(skb, 2); /* 16 byte align the IP header */ - pci_dma_sync_single_for_cpu(ep->pci_dev, - ep->rx_ring[entry].bufaddr, - ep->rx_buf_sz, - PCI_DMA_FROMDEVICE); + dma_sync_single_for_cpu(&ep->pci_dev->dev, + ep->rx_ring[entry].bufaddr, + ep->rx_buf_sz, + DMA_FROM_DEVICE); skb_copy_to_linear_data(skb, ep->rx_skbuff[entry]->data, pkt_len); skb_put(skb, pkt_len); - pci_dma_sync_single_for_device(ep->pci_dev, - ep->rx_ring[entry].bufaddr, - ep->rx_buf_sz, - PCI_DMA_FROMDEVICE); + dma_sync_single_for_device(&ep->pci_dev->dev, + ep->rx_ring[entry].bufaddr, + ep->rx_buf_sz, + DMA_FROM_DEVICE); } else { - pci_unmap_single(ep->pci_dev, - ep->rx_ring[entry].bufaddr, - ep->rx_buf_sz, PCI_DMA_FROMDEVICE); + dma_unmap_single(&ep->pci_dev->dev, + ep->rx_ring[entry].bufaddr, + ep->rx_buf_sz, + DMA_FROM_DEVICE); skb_put(skb = ep->rx_skbuff[entry], pkt_len); ep->rx_skbuff[entry] = NULL; } @@ -1213,8 +1222,10 @@ static int epic_rx(struct net_device *dev, int budget) if (skb == NULL) break; skb_reserve(skb, 2); /* Align IP on 16 byte boundaries */ - ep->rx_ring[entry].bufaddr = pci_map_single(ep->pci_dev, - skb->data, ep->rx_buf_sz, PCI_DMA_FROMDEVICE); + ep->rx_ring[entry].bufaddr = dma_map_single(&ep->pci_dev->dev, + skb->data, + ep->rx_buf_sz, + DMA_FROM_DEVICE); work_done++; } /* AV: shouldn't we add a barrier here? */ @@ -1294,8 +1305,8 @@ static int epic_close(struct net_device *dev) ep->rx_ring[i].rxstatus = 0; /* Not owned by Epic chip. */ ep->rx_ring[i].buflength = 0; if (skb) { - pci_unmap_single(pdev, ep->rx_ring[i].bufaddr, - ep->rx_buf_sz, PCI_DMA_FROMDEVICE); + dma_unmap_single(&pdev->dev, ep->rx_ring[i].bufaddr, + ep->rx_buf_sz, DMA_FROM_DEVICE); dev_kfree_skb(skb); } ep->rx_ring[i].bufaddr = 0xBADF00D0; /* An invalid address. */ @@ -1305,8 +1316,8 @@ static int epic_close(struct net_device *dev) ep->tx_skbuff[i] = NULL; if (!skb) continue; - pci_unmap_single(pdev, ep->tx_ring[i].bufaddr, skb->len, - PCI_DMA_TODEVICE); + dma_unmap_single(&pdev->dev, ep->tx_ring[i].bufaddr, skb->len, + DMA_TO_DEVICE); dev_kfree_skb(skb); } @@ -1502,8 +1513,10 @@ static void epic_remove_one(struct pci_dev *pdev) struct net_device *dev = pci_get_drvdata(pdev); struct epic_private *ep = netdev_priv(dev); - pci_free_consistent(pdev, TX_TOTAL_SIZE, ep->tx_ring, ep->tx_ring_dma); - pci_free_consistent(pdev, RX_TOTAL_SIZE, ep->rx_ring, ep->rx_ring_dma); + dma_free_coherent(&pdev->dev, TX_TOTAL_SIZE, ep->tx_ring, + ep->tx_ring_dma); + dma_free_coherent(&pdev->dev, RX_TOTAL_SIZE, ep->rx_ring, + ep->rx_ring_dma); unregister_netdev(dev); pci_iounmap(pdev, ep->ioaddr); pci_release_regions(pdev); diff --git a/drivers/net/ethernet/smsc/smc91x.c b/drivers/net/ethernet/smsc/smc91x.c index 1c4fea9c3ec4..f6b73afd1879 100644 --- a/drivers/net/ethernet/smsc/smc91x.c +++ b/drivers/net/ethernet/smsc/smc91x.c @@ -535,10 +535,10 @@ static inline void smc_rcv(struct net_device *dev) /* * This is called to actually send a packet to the chip. */ -static void smc_hardware_send_pkt(unsigned long data) +static void smc_hardware_send_pkt(struct tasklet_struct *t) { - struct net_device *dev = (struct net_device *)data; - struct smc_local *lp = netdev_priv(dev); + struct smc_local *lp = from_tasklet(lp, t, tx_task); + struct net_device *dev = lp->dev; void __iomem *ioaddr = lp->base; struct sk_buff *skb; unsigned int packet_no, len; @@ -688,7 +688,7 @@ smc_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) * Allocation succeeded: push packet to the chip's own memory * immediately. */ - smc_hardware_send_pkt((unsigned long)dev); + smc_hardware_send_pkt(&lp->tx_task); } return NETDEV_TX_OK; @@ -1036,7 +1036,6 @@ static void smc_phy_configure(struct work_struct *work) int phyaddr = lp->mii.phy_id; int my_phy_caps; /* My PHY capabilities */ int my_ad_caps; /* My Advertised capabilities */ - int status; DBG(3, dev, "smc_program_phy()\n"); @@ -1110,7 +1109,7 @@ static void smc_phy_configure(struct work_struct *work) * auto-negotiation is restarted, sometimes it isn't ready and * the link does not come up. */ - status = smc_phy_read(dev, phyaddr, MII_ADVERTISE); + smc_phy_read(dev, phyaddr, MII_ADVERTISE); DBG(2, dev, "phy caps=%x\n", my_phy_caps); DBG(2, dev, "phy advertised caps=%x\n", my_ad_caps); @@ -1965,7 +1964,7 @@ static int smc_probe(struct net_device *dev, void __iomem *ioaddr, dev->netdev_ops = &smc_netdev_ops; dev->ethtool_ops = &smc_ethtool_ops; - tasklet_init(&lp->tx_task, smc_hardware_send_pkt, (unsigned long)dev); + tasklet_setup(&lp->tx_task, smc_hardware_send_pkt); INIT_WORK(&lp->phy_configure, smc_phy_configure); lp->dev = dev; lp->mii.phy_id_mask = 0x1f; diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index fc168f85e7af..823d9a7184fe 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -1196,9 +1196,8 @@ smsc911x_rx_fastforward(struct smsc911x_data *pdata, unsigned int pktwords) SMSC_WARN(pdata, hw, "Timed out waiting for " "RX FFWD to finish, RX_DP_CTRL: 0x%08X", val); } else { - unsigned int temp; while (pktwords--) - temp = smsc911x_reg_read(pdata, RX_DATA_FIFO); + smsc911x_reg_read(pdata, RX_DATA_FIFO); } } @@ -2055,7 +2054,6 @@ static int smsc911x_eeprom_write_location(struct smsc911x_data *pdata, u8 address, u8 data) { u32 op = E2P_CMD_EPC_CMD_ERASE_ | address; - u32 temp; int ret; SMSC_TRACE(pdata, drv, "address 0x%x, data 0x%x", address, data); @@ -2066,7 +2064,7 @@ static int smsc911x_eeprom_write_location(struct smsc911x_data *pdata, smsc911x_reg_write(pdata, E2P_DATA, (u32)data); /* Workaround for hardware read-after-write restriction */ - temp = smsc911x_reg_read(pdata, BYTE_TEST); + smsc911x_reg_read(pdata, BYTE_TEST); ret = smsc911x_eeprom_send_cmd(pdata, op); } diff --git a/drivers/net/ethernet/smsc/smsc9420.c b/drivers/net/ethernet/smsc/smsc9420.c index 42bef04d65ba..c1dab009415d 100644 --- a/drivers/net/ethernet/smsc/smsc9420.c +++ b/drivers/net/ethernet/smsc/smsc9420.c @@ -497,8 +497,9 @@ static void smsc9420_free_tx_ring(struct smsc9420_pdata *pd) if (skb) { BUG_ON(!pd->tx_buffers[i].mapping); - pci_unmap_single(pd->pdev, pd->tx_buffers[i].mapping, - skb->len, PCI_DMA_TODEVICE); + dma_unmap_single(&pd->pdev->dev, + pd->tx_buffers[i].mapping, skb->len, + DMA_TO_DEVICE); dev_kfree_skb_any(skb); } @@ -530,8 +531,9 @@ static void smsc9420_free_rx_ring(struct smsc9420_pdata *pd) dev_kfree_skb_any(pd->rx_buffers[i].skb); if (pd->rx_buffers[i].mapping) - pci_unmap_single(pd->pdev, pd->rx_buffers[i].mapping, - PKT_BUF_SZ, PCI_DMA_FROMDEVICE); + dma_unmap_single(&pd->pdev->dev, + pd->rx_buffers[i].mapping, + PKT_BUF_SZ, DMA_FROM_DEVICE); pd->rx_ring[i].status = 0; pd->rx_ring[i].length = 0; @@ -749,8 +751,8 @@ static void smsc9420_rx_handoff(struct smsc9420_pdata *pd, const int index, dev->stats.rx_packets++; dev->stats.rx_bytes += packet_length; - pci_unmap_single(pd->pdev, pd->rx_buffers[index].mapping, - PKT_BUF_SZ, PCI_DMA_FROMDEVICE); + dma_unmap_single(&pd->pdev->dev, pd->rx_buffers[index].mapping, + PKT_BUF_SZ, DMA_FROM_DEVICE); pd->rx_buffers[index].mapping = 0; skb = pd->rx_buffers[index].skb; @@ -782,9 +784,9 @@ static int smsc9420_alloc_rx_buffer(struct smsc9420_pdata *pd, int index) if (unlikely(!skb)) return -ENOMEM; - mapping = pci_map_single(pd->pdev, skb_tail_pointer(skb), - PKT_BUF_SZ, PCI_DMA_FROMDEVICE); - if (pci_dma_mapping_error(pd->pdev, mapping)) { + mapping = dma_map_single(&pd->pdev->dev, skb_tail_pointer(skb), + PKT_BUF_SZ, DMA_FROM_DEVICE); + if (dma_mapping_error(&pd->pdev->dev, mapping)) { dev_kfree_skb_any(skb); netif_warn(pd, rx_err, pd->dev, "pci_map_single failed!\n"); return -ENOMEM; @@ -901,8 +903,10 @@ static void smsc9420_complete_tx(struct net_device *dev) BUG_ON(!pd->tx_buffers[index].skb); BUG_ON(!pd->tx_buffers[index].mapping); - pci_unmap_single(pd->pdev, pd->tx_buffers[index].mapping, - pd->tx_buffers[index].skb->len, PCI_DMA_TODEVICE); + dma_unmap_single(&pd->pdev->dev, + pd->tx_buffers[index].mapping, + pd->tx_buffers[index].skb->len, + DMA_TO_DEVICE); pd->tx_buffers[index].mapping = 0; dev_kfree_skb_any(pd->tx_buffers[index].skb); @@ -932,9 +936,9 @@ static netdev_tx_t smsc9420_hard_start_xmit(struct sk_buff *skb, BUG_ON(pd->tx_buffers[index].skb); BUG_ON(pd->tx_buffers[index].mapping); - mapping = pci_map_single(pd->pdev, skb->data, - skb->len, PCI_DMA_TODEVICE); - if (pci_dma_mapping_error(pd->pdev, mapping)) { + mapping = dma_map_single(&pd->pdev->dev, skb->data, skb->len, + DMA_TO_DEVICE); + if (dma_mapping_error(&pd->pdev->dev, mapping)) { netif_warn(pd, tx_err, pd->dev, "pci_map_single failed, dropping packet\n"); return NETDEV_TX_BUSY; @@ -1522,7 +1526,7 @@ smsc9420_probe(struct pci_dev *pdev, const struct pci_device_id *id) goto out_free_netdev_2; } - if (pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) { + if (dma_set_mask(&pdev->dev, DMA_BIT_MASK(32))) { netdev_err(dev, "No usable DMA configuration, aborting\n"); goto out_free_regions_3; } @@ -1540,10 +1544,9 @@ smsc9420_probe(struct pci_dev *pdev, const struct pci_device_id *id) pd = netdev_priv(dev); /* pci descriptors are created in the PCI consistent area */ - pd->rx_ring = pci_alloc_consistent(pdev, - sizeof(struct smsc9420_dma_desc) * RX_RING_SIZE + - sizeof(struct smsc9420_dma_desc) * TX_RING_SIZE, - &pd->rx_dma_addr); + pd->rx_ring = dma_alloc_coherent(&pdev->dev, + sizeof(struct smsc9420_dma_desc) * (RX_RING_SIZE + TX_RING_SIZE), + &pd->rx_dma_addr, GFP_KERNEL); if (!pd->rx_ring) goto out_free_io_4; @@ -1599,8 +1602,9 @@ smsc9420_probe(struct pci_dev *pdev, const struct pci_device_id *id) return 0; out_free_dmadesc_5: - pci_free_consistent(pdev, sizeof(struct smsc9420_dma_desc) * - (RX_RING_SIZE + TX_RING_SIZE), pd->rx_ring, pd->rx_dma_addr); + dma_free_coherent(&pdev->dev, + sizeof(struct smsc9420_dma_desc) * (RX_RING_SIZE + TX_RING_SIZE), + pd->rx_ring, pd->rx_dma_addr); out_free_io_4: iounmap(virt_addr - LAN9420_CPSR_ENDIAN_OFFSET); out_free_regions_3: @@ -1632,8 +1636,9 @@ static void smsc9420_remove(struct pci_dev *pdev) BUG_ON(!pd->tx_ring); BUG_ON(!pd->rx_ring); - pci_free_consistent(pdev, sizeof(struct smsc9420_dma_desc) * - (RX_RING_SIZE + TX_RING_SIZE), pd->rx_ring, pd->rx_dma_addr); + dma_free_coherent(&pdev->dev, + sizeof(struct smsc9420_dma_desc) * (RX_RING_SIZE + TX_RING_SIZE), + pd->rx_ring, pd->rx_dma_addr); iounmap(pd->ioaddr - LAN9420_CPSR_ENDIAN_OFFSET); pci_release_regions(pdev); diff --git a/drivers/net/ethernet/socionext/sni_ave.c b/drivers/net/ethernet/socionext/sni_ave.c index 81b554dd7221..501b9c7aba56 100644 --- a/drivers/net/ethernet/socionext/sni_ave.c +++ b/drivers/net/ethernet/socionext/sni_ave.c @@ -1585,7 +1585,7 @@ static int ave_probe(struct platform_device *pdev) if (IS_ERR(base)) return PTR_ERR(base); - ndev = alloc_etherdev(sizeof(struct ave_private)); + ndev = devm_alloc_etherdev(dev, sizeof(struct ave_private)); if (!ndev) { dev_err(dev, "can't allocate ethernet device\n"); return -ENOMEM; @@ -1632,7 +1632,7 @@ static int ave_probe(struct platform_device *pdev) } ret = dma_set_mask(dev, dma_mask); if (ret) - goto out_free_netdev; + return ret; priv->tx.ndesc = AVE_NR_TXDESC; priv->rx.ndesc = AVE_NR_RXDESC; @@ -1645,10 +1645,8 @@ static int ave_probe(struct platform_device *pdev) if (!name) break; priv->clk[i] = devm_clk_get(dev, name); - if (IS_ERR(priv->clk[i])) { - ret = PTR_ERR(priv->clk[i]); - goto out_free_netdev; - } + if (IS_ERR(priv->clk[i])) + return PTR_ERR(priv->clk[i]); priv->nclks++; } @@ -1657,10 +1655,8 @@ static int ave_probe(struct platform_device *pdev) if (!name) break; priv->rst[i] = devm_reset_control_get_shared(dev, name); - if (IS_ERR(priv->rst[i])) { - ret = PTR_ERR(priv->rst[i]); - goto out_free_netdev; - } + if (IS_ERR(priv->rst[i])) + return PTR_ERR(priv->rst[i]); priv->nrsts++; } @@ -1669,26 +1665,23 @@ static int ave_probe(struct platform_device *pdev) 1, 0, &args); if (ret) { dev_err(dev, "can't get syscon-phy-mode property\n"); - goto out_free_netdev; + return ret; } priv->regmap = syscon_node_to_regmap(args.np); of_node_put(args.np); if (IS_ERR(priv->regmap)) { dev_err(dev, "can't map syscon-phy-mode\n"); - ret = PTR_ERR(priv->regmap); - goto out_free_netdev; + return PTR_ERR(priv->regmap); } ret = priv->data->get_pinmode(priv, phy_mode, args.args[0]); if (ret) { dev_err(dev, "invalid phy-mode setting\n"); - goto out_free_netdev; + return ret; } priv->mdio = devm_mdiobus_alloc(dev); - if (!priv->mdio) { - ret = -ENOMEM; - goto out_free_netdev; - } + if (!priv->mdio) + return -ENOMEM; priv->mdio->priv = ndev; priv->mdio->parent = dev; priv->mdio->read = ave_mdiobus_read; @@ -1725,8 +1718,6 @@ static int ave_probe(struct platform_device *pdev) out_del_napi: netif_napi_del(&priv->napi_rx); netif_napi_del(&priv->napi_tx); -out_free_netdev: - free_netdev(ndev); return ret; } @@ -1739,7 +1730,6 @@ static int ave_remove(struct platform_device *pdev) unregister_netdev(ndev); netif_napi_del(&priv->napi_rx); netif_napi_del(&priv->napi_tx); - free_netdev(ndev); return 0; } diff --git a/drivers/net/ethernet/stmicro/stmmac/Kconfig b/drivers/net/ethernet/stmicro/stmmac/Kconfig index 9a47c5aec91a..53f14c5a9e02 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Kconfig +++ b/drivers/net/ethernet/stmicro/stmmac/Kconfig @@ -3,7 +3,7 @@ config STMMAC_ETH tristate "STMicroelectronics Multi-Gigabit Ethernet driver" depends on HAS_IOMEM && HAS_DMA select MII - select MDIO_XPCS + select PCS_XPCS select PAGE_POOL select PHYLINK select CRC32 @@ -209,6 +209,16 @@ config DWMAC_IMX8 device driver. This driver is used for i.MX8 series like iMX8MP/iMX8DXL GMAC ethernet controller. +config DWMAC_INTEL_PLAT + tristate "Intel dwmac support" + depends on OF && COMMON_CLK + depends on STMMAC_ETH + help + Support for ethernet controllers on Intel SoCs + + This selects the Intel platform specific glue layer support for + the stmmac device driver. This driver is used for the Intel Keem Bay + SoC. endif config DWMAC_INTEL diff --git a/drivers/net/ethernet/stmicro/stmmac/Makefile b/drivers/net/ethernet/stmicro/stmmac/Makefile index 295615ab36a7..24e6145d4eae 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Makefile +++ b/drivers/net/ethernet/stmicro/stmmac/Makefile @@ -26,6 +26,7 @@ obj-$(CONFIG_DWMAC_STM32) += dwmac-stm32.o obj-$(CONFIG_DWMAC_SUNXI) += dwmac-sunxi.o obj-$(CONFIG_DWMAC_SUN8I) += dwmac-sun8i.o obj-$(CONFIG_DWMAC_DWC_QOS_ETH) += dwmac-dwc-qos-eth.o +obj-$(CONFIG_DWMAC_INTEL_PLAT) += dwmac-intel-plat.o obj-$(CONFIG_DWMAC_GENERIC) += dwmac-generic.o obj-$(CONFIG_DWMAC_IMX8) += dwmac-imx.o stmmac-platform-objs:= stmmac_platform.o diff --git a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c index 52971f5293aa..d2cdc02d9f94 100644 --- a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c +++ b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c @@ -46,7 +46,7 @@ static int jumbo_frm(void *p, struct sk_buff *skb, int csum) while (len != 0) { tx_q->tx_skbuff[entry] = NULL; - entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE); + entry = STMMAC_GET_ENTRY(entry, priv->dma_tx_size); desc = tx_q->dma_tx + entry; if (len > bmax) { @@ -137,7 +137,7 @@ static void refill_desc3(void *priv_ptr, struct dma_desc *p) */ p->des3 = cpu_to_le32((unsigned int)(rx_q->dma_rx_phy + (((rx_q->dirty_rx) + 1) % - DMA_RX_SIZE) * + priv->dma_rx_size) * sizeof(struct dma_desc))); } @@ -154,7 +154,8 @@ static void clean_desc3(void *priv_ptr, struct dma_desc *p) * to keep explicit chaining in the descriptor. */ p->des3 = cpu_to_le32((unsigned int)((tx_q->dma_tx_phy + - ((tx_q->dirty_tx + 1) % DMA_TX_SIZE)) + ((tx_q->dirty_tx + 1) % + priv->dma_tx_size)) * sizeof(struct dma_desc))); } diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 127f75862962..df7de50497a0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -15,7 +15,7 @@ #include <linux/netdevice.h> #include <linux/stmmac.h> #include <linux/phy.h> -#include <linux/mdio-xpcs.h> +#include <linux/pcs/pcs-xpcs.h> #include <linux/module.h> #if IS_ENABLED(CONFIG_VLAN_8021Q) #define STMMAC_VLAN_TAG_USED @@ -42,9 +42,16 @@ #define STMMAC_CHAN0 0 /* Always supported and default for all chips */ -/* These need to be power of two, and >= 4 */ -#define DMA_TX_SIZE 512 -#define DMA_RX_SIZE 512 +/* TX and RX Descriptor Length, these need to be power of two. + * TX descriptor length less than 64 may cause transmit queue timed out error. + * RX descriptor length less than 64 may cause inconsistent Rx chain error. + */ +#define DMA_MIN_TX_SIZE 64 +#define DMA_MAX_TX_SIZE 1024 +#define DMA_DEFAULT_TX_SIZE 512 +#define DMA_MIN_RX_SIZE 64 +#define DMA_MAX_RX_SIZE 1024 +#define DMA_DEFAULT_RX_SIZE 512 #define STMMAC_GET_ENTRY(x, size) ((x + 1) & (size - 1)) #undef FRAME_FILTER_DEBUG @@ -474,6 +481,8 @@ struct mac_device_info { unsigned int num_vlan; u32 vlan_filter[32]; unsigned int promisc; + bool vlan_fail_q_en; + u8 vlan_fail_q; }; struct stmmac_rx_routing { diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c index 3c5df5eeed6c..efef5476a577 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c @@ -129,8 +129,7 @@ static void imx_dwmac_exit(struct platform_device *pdev, void *priv) { struct imx_priv_data *dwmac = priv; - if (dwmac->clk_tx) - clk_disable_unprepare(dwmac->clk_tx); + clk_disable_unprepare(dwmac->clk_tx); clk_disable_unprepare(dwmac->clk_mem); } diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel-plat.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel-plat.c new file mode 100644 index 000000000000..f61cb997a8f6 --- /dev/null +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel-plat.c @@ -0,0 +1,196 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Intel DWMAC platform driver + * + * Copyright(C) 2020 Intel Corporation + */ + +#include <linux/ethtool.h> +#include <linux/module.h> +#include <linux/of.h> +#include <linux/of_device.h> +#include <linux/platform_device.h> +#include <linux/stmmac.h> + +#include "dwmac4.h" +#include "stmmac.h" +#include "stmmac_platform.h" + +struct intel_dwmac { + struct device *dev; + struct clk *tx_clk; + const struct intel_dwmac_data *data; +}; + +struct intel_dwmac_data { + void (*fix_mac_speed)(void *priv, unsigned int speed); + unsigned long ptp_ref_clk_rate; + unsigned long tx_clk_rate; + bool tx_clk_en; +}; + +static void kmb_eth_fix_mac_speed(void *priv, unsigned int speed) +{ + struct intel_dwmac *dwmac = priv; + unsigned long rate; + int ret; + + rate = clk_get_rate(dwmac->tx_clk); + + switch (speed) { + case SPEED_1000: + rate = 125000000; + break; + + case SPEED_100: + rate = 25000000; + break; + + case SPEED_10: + rate = 2500000; + break; + + default: + dev_err(dwmac->dev, "Invalid speed\n"); + break; + } + + ret = clk_set_rate(dwmac->tx_clk, rate); + if (ret) + dev_err(dwmac->dev, "Failed to configure tx clock rate\n"); +} + +static const struct intel_dwmac_data kmb_data = { + .fix_mac_speed = kmb_eth_fix_mac_speed, + .ptp_ref_clk_rate = 200000000, + .tx_clk_rate = 125000000, + .tx_clk_en = true, +}; + +static const struct of_device_id intel_eth_plat_match[] = { + { .compatible = "intel,keembay-dwmac", .data = &kmb_data }, + { } +}; +MODULE_DEVICE_TABLE(of, intel_eth_plat_match); + +static int intel_eth_plat_probe(struct platform_device *pdev) +{ + struct net_device *ndev = platform_get_drvdata(pdev); + struct stmmac_priv *priv = netdev_priv(ndev); + struct plat_stmmacenet_data *plat_dat; + struct stmmac_resources stmmac_res; + const struct of_device_id *match; + struct intel_dwmac *dwmac; + unsigned long rate; + int ret; + + plat_dat = priv->plat; + ret = stmmac_get_platform_resources(pdev, &stmmac_res); + if (ret) + return ret; + + plat_dat = stmmac_probe_config_dt(pdev, &stmmac_res.mac); + if (IS_ERR(plat_dat)) { + dev_err(&pdev->dev, "dt configuration failed\n"); + return PTR_ERR(plat_dat); + } + + dwmac = devm_kzalloc(&pdev->dev, sizeof(*dwmac), GFP_KERNEL); + if (!dwmac) { + ret = -ENOMEM; + goto err_remove_config_dt; + } + + dwmac->dev = &pdev->dev; + dwmac->tx_clk = NULL; + + match = of_match_device(intel_eth_plat_match, &pdev->dev); + if (match && match->data) { + dwmac->data = (const struct intel_dwmac_data *)match->data; + + if (dwmac->data->fix_mac_speed) + plat_dat->fix_mac_speed = dwmac->data->fix_mac_speed; + + /* Enable TX clock */ + if (dwmac->data->tx_clk_en) { + dwmac->tx_clk = devm_clk_get(&pdev->dev, "tx_clk"); + if (IS_ERR(dwmac->tx_clk)) + goto err_remove_config_dt; + + clk_prepare_enable(dwmac->tx_clk); + + /* Check and configure TX clock rate */ + rate = clk_get_rate(dwmac->tx_clk); + if (dwmac->data->tx_clk_rate && + rate != dwmac->data->tx_clk_rate) { + rate = dwmac->data->tx_clk_rate; + ret = clk_set_rate(dwmac->tx_clk, rate); + if (ret) { + dev_err(&pdev->dev, + "Failed to set tx_clk\n"); + return ret; + } + } + } + + /* Check and configure PTP ref clock rate */ + rate = clk_get_rate(plat_dat->clk_ptp_ref); + if (dwmac->data->ptp_ref_clk_rate && + rate != dwmac->data->ptp_ref_clk_rate) { + rate = dwmac->data->ptp_ref_clk_rate; + ret = clk_set_rate(plat_dat->clk_ptp_ref, rate); + if (ret) { + dev_err(&pdev->dev, + "Failed to set clk_ptp_ref\n"); + return ret; + } + } + } + + plat_dat->bsp_priv = dwmac; + plat_dat->eee_usecs_rate = plat_dat->clk_ptp_rate; + + if (plat_dat->eee_usecs_rate > 0) { + u32 tx_lpi_usec; + + tx_lpi_usec = (plat_dat->eee_usecs_rate / 1000000) - 1; + writel(tx_lpi_usec, stmmac_res.addr + GMAC_1US_TIC_COUNTER); + } + + ret = stmmac_dvr_probe(&pdev->dev, plat_dat, &stmmac_res); + if (ret) { + clk_disable_unprepare(dwmac->tx_clk); + goto err_remove_config_dt; + } + + return 0; + +err_remove_config_dt: + stmmac_remove_config_dt(pdev, plat_dat); + + return ret; +} + +static int intel_eth_plat_remove(struct platform_device *pdev) +{ + struct intel_dwmac *dwmac = get_stmmac_bsp_priv(&pdev->dev); + int ret; + + ret = stmmac_pltfr_remove(pdev); + clk_disable_unprepare(dwmac->tx_clk); + + return ret; +} + +static struct platform_driver intel_eth_plat_driver = { + .probe = intel_eth_plat_probe, + .remove = intel_eth_plat_remove, + .driver = { + .name = "intel-eth-plat", + .pm = &stmmac_pltfr_pm_ops, + .of_match_table = intel_eth_plat_match, + }, +}; +module_platform_driver(intel_eth_plat_driver); + +MODULE_LICENSE("GPL v2"); +MODULE_DESCRIPTION("Intel DWMAC platform driver"); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c index 9e6d60e75f85..b6e5e3e36b63 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c @@ -6,6 +6,7 @@ #include <linux/pci.h> #include <linux/dmi.h> #include "dwmac-intel.h" +#include "dwmac4.h" #include "stmmac.h" struct intel_priv_data { @@ -295,6 +296,7 @@ static int intel_mgbe_common_data(struct pci_dev *pdev, plat->axi->axi_blen[2] = 16; plat->ptp_max_adj = plat->clk_ptp_rate; + plat->eee_usecs_rate = plat->clk_ptp_rate; /* Set system clock */ plat->stmmac_clk = clk_register_fixed_rate(&pdev->dev, @@ -321,6 +323,11 @@ static int intel_mgbe_common_data(struct pci_dev *pdev, /* Set the maxmtu to a default of JUMBO_LEN */ plat->maxmtu = JUMBO_LEN; + plat->vlan_fail_q_en = true; + + /* Use the last Rx queue */ + plat->vlan_fail_q = plat->rx_queues_to_use - 1; + return 0; } @@ -618,6 +625,13 @@ static int intel_eth_pci_probe(struct pci_dev *pdev, if (ret) return ret; + if (plat->eee_usecs_rate > 0) { + u32 tx_lpi_usec; + + tx_lpi_usec = (plat->eee_usecs_rate / 1000000) - 1; + writel(tx_lpi_usec, res.addr + GMAC_1US_TIC_COUNTER); + } + ret = pci_alloc_irq_vectors(pdev, 1, 1, PCI_IRQ_ALL_TYPES); if (ret < 0) return ret; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c index 2d5573b3dee1..6ef30252bfe0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0-or-later /** - * dwmac-rk.c - Rockchip RK3288 DWMAC specific glue layer + * DOC: dwmac-rk.c - Rockchip RK3288 DWMAC specific glue layer * * Copyright (C) 2014 Chen-Zhi (Roger Chen) * diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h index 61f3249bd724..592b043f9676 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h @@ -76,6 +76,7 @@ #define GMAC_PACKET_FILTER_HPF BIT(10) #define GMAC_PACKET_FILTER_VTFE BIT(16) #define GMAC_PACKET_FILTER_IPFE BIT(20) +#define GMAC_PACKET_FILTER_RA BIT(31) #define GMAC_MAX_PERFECT_ADDRESSES 128 diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index ecd834e0e121..002791b77356 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -618,7 +618,18 @@ static void dwmac4_set_filter(struct mac_device_info *hw, value &= ~GMAC_PACKET_FILTER_PM; value &= ~GMAC_PACKET_FILTER_PR; if (dev->flags & IFF_PROMISC) { - value = GMAC_PACKET_FILTER_PR | GMAC_PACKET_FILTER_PCF; + /* VLAN Tag Filter Fail Packets Queuing */ + if (hw->vlan_fail_q_en) { + value = readl(ioaddr + GMAC_RXQ_CTRL4); + value &= ~GMAC_RXQCTRL_VFFQ_MASK; + value |= GMAC_RXQCTRL_VFFQE | + (hw->vlan_fail_q << GMAC_RXQCTRL_VFFQ_SHIFT); + writel(value, ioaddr + GMAC_RXQ_CTRL4); + value = GMAC_PACKET_FILTER_PR | GMAC_PACKET_FILTER_RA; + } else { + value = GMAC_PACKET_FILTER_PR | GMAC_PACKET_FILTER_PCF; + } + } else if ((dev->flags & IFF_ALLMULTI) || (netdev_mc_count(dev) > hw->multicast_filter_bins)) { /* Pass all multi */ @@ -680,7 +691,7 @@ static void dwmac4_set_filter(struct mac_device_info *hw, writel(value, ioaddr + GMAC_PACKET_FILTER); - if (dev->flags & IFF_PROMISC) { + if (dev->flags & IFF_PROMISC && !hw->vlan_fail_q_en) { if (!hw->promisc) { hw->promisc = 1; dwmac4_vlan_promisc_enable(dev, hw); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c index eff82065a501..c6540b003b43 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c @@ -494,10 +494,9 @@ static void dwmac4_set_vlan(struct dma_desc *p, u32 type) p->des2 |= cpu_to_le32(type & TDES2_VLAN_TAG_MASK); } -static int dwmac4_get_rx_header_len(struct dma_desc *p, unsigned int *len) +static void dwmac4_get_rx_header_len(struct dma_desc *p, unsigned int *len) { *len = le32_to_cpu(p->des2) & RDES2_HL; - return 0; } static void dwmac4_set_sec_addr(struct dma_desc *p, dma_addr_t addr) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac5.h b/drivers/net/ethernet/stmicro/stmmac/dwmac5.h index 3e8faa96b4d4..56b0762c1276 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac5.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac5.h @@ -92,6 +92,12 @@ #define TCEIE BIT(0) #define DMA_ECC_INT_STATUS 0x00001088 +/* EQoS version 5.xx VLAN Tag Filter Fail Packets Queuing */ +#define GMAC_RXQ_CTRL4 0x00000094 +#define GMAC_RXQCTRL_VFFQ_MASK GENMASK(19, 17) +#define GMAC_RXQCTRL_VFFQ_SHIFT 17 +#define GMAC_RXQCTRL_VFFQE BIT(16) + int dwmac5_safety_feat_config(void __iomem *ioaddr, unsigned int asp); int dwmac5_safety_feat_irq_status(struct net_device *ndev, void __iomem *ioaddr, unsigned int asp, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_descs.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_descs.c index c3d654cfa9ef..0aaf19ab5672 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_descs.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_descs.c @@ -286,11 +286,10 @@ static int dwxgmac2_get_rx_hash(struct dma_desc *p, u32 *hash, return -EINVAL; } -static int dwxgmac2_get_rx_header_len(struct dma_desc *p, unsigned int *len) +static void dwxgmac2_get_rx_header_len(struct dma_desc *p, unsigned int *len) { if (le32_to_cpu(p->des3) & XGMAC_RDES3_L34T) *len = le32_to_cpu(p->des2) & XGMAC_RDES2_HL; - return 0; } static void dwxgmac2_set_sec_addr(struct dma_desc *p, dma_addr_t addr) diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h index ffe2d63389b8..e2dca9b6e992 100644 --- a/drivers/net/ethernet/stmicro/stmmac/hwif.h +++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h @@ -90,7 +90,7 @@ struct stmmac_desc_ops { /* RSS */ int (*get_rx_hash)(struct dma_desc *p, u32 *hash, enum pkt_hash_types *type); - int (*get_rx_header_len)(struct dma_desc *p, unsigned int *len); + void (*get_rx_header_len)(struct dma_desc *p, unsigned int *len); void (*set_sec_addr)(struct dma_desc *p, dma_addr_t addr); void (*set_sarc)(struct dma_desc *p, u32 sarc_type); void (*set_vlan_tag)(struct dma_desc *p, u16 tag, u16 inner_tag, @@ -150,7 +150,7 @@ struct stmmac_desc_ops { #define stmmac_get_rx_hash(__priv, __args...) \ stmmac_do_callback(__priv, desc, get_rx_hash, __args) #define stmmac_get_rx_header_len(__priv, __args...) \ - stmmac_do_callback(__priv, desc, get_rx_header_len, __args) + stmmac_do_void_callback(__priv, desc, get_rx_header_len, __args) #define stmmac_set_desc_sec_addr(__priv, __args...) \ stmmac_do_void_callback(__priv, desc, set_sec_addr, __args) #define stmmac_set_desc_sarc(__priv, __args...) \ diff --git a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c index 14bd5e7b9875..8ad900949dc8 100644 --- a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c +++ b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c @@ -51,7 +51,7 @@ static int jumbo_frm(void *p, struct sk_buff *skb, int csum) stmmac_prepare_tx_desc(priv, desc, 1, bmax, csum, STMMAC_RING_MODE, 0, false, skb->len); tx_q->tx_skbuff[entry] = NULL; - entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE); + entry = STMMAC_GET_ENTRY(entry, priv->dma_tx_size); if (priv->extend_desc) desc = (struct dma_desc *)(tx_q->dma_etx + entry); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index 545696971f65..727e68dfaf1c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -171,9 +171,11 @@ struct stmmac_priv { /* RX Queue */ struct stmmac_rx_queue rx_queue[MTL_MAX_RX_QUEUES]; + unsigned int dma_rx_size; /* TX Queue */ struct stmmac_tx_queue tx_queue[MTL_MAX_TX_QUEUES]; + unsigned int dma_tx_size; /* Generic channel for NAPI */ struct stmmac_channel channel[STMMAC_CH_MAX]; @@ -266,6 +268,8 @@ int stmmac_dvr_probe(struct device *device, struct stmmac_resources *res); void stmmac_disable_eee_mode(struct stmmac_priv *priv); bool stmmac_eee_init(struct stmmac_priv *priv); +int stmmac_reinit_queues(struct net_device *dev, u32 rx_cnt, u32 tx_cnt); +int stmmac_reinit_ringparam(struct net_device *dev, u32 rx_size, u32 tx_size); #if IS_ENABLED(CONFIG_STMMAC_SELFTESTS) void stmmac_selftest_run(struct net_device *dev, diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c index 814879f91f76..9e54f953634b 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c @@ -440,6 +440,33 @@ static int stmmac_nway_reset(struct net_device *dev) return phylink_ethtool_nway_reset(priv->phylink); } +static void stmmac_get_ringparam(struct net_device *netdev, + struct ethtool_ringparam *ring) +{ + struct stmmac_priv *priv = netdev_priv(netdev); + + ring->rx_max_pending = DMA_MAX_RX_SIZE; + ring->tx_max_pending = DMA_MAX_TX_SIZE; + ring->rx_pending = priv->dma_rx_size; + ring->tx_pending = priv->dma_tx_size; +} + +static int stmmac_set_ringparam(struct net_device *netdev, + struct ethtool_ringparam *ring) +{ + if (ring->rx_mini_pending || ring->rx_jumbo_pending || + ring->rx_pending < DMA_MIN_RX_SIZE || + ring->rx_pending > DMA_MAX_RX_SIZE || + !is_power_of_2(ring->rx_pending) || + ring->tx_pending < DMA_MIN_TX_SIZE || + ring->tx_pending > DMA_MAX_TX_SIZE || + !is_power_of_2(ring->tx_pending)) + return -EINVAL; + + return stmmac_reinit_ringparam(netdev, ring->rx_pending, + ring->tx_pending); +} + static void stmmac_get_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) @@ -843,6 +870,30 @@ static int stmmac_set_rxfh(struct net_device *dev, const u32 *indir, priv->plat->rx_queues_to_use); } +static void stmmac_get_channels(struct net_device *dev, + struct ethtool_channels *chan) +{ + struct stmmac_priv *priv = netdev_priv(dev); + + chan->rx_count = priv->plat->rx_queues_to_use; + chan->tx_count = priv->plat->tx_queues_to_use; + chan->max_rx = priv->dma_cap.number_rx_queues; + chan->max_tx = priv->dma_cap.number_tx_queues; +} + +static int stmmac_set_channels(struct net_device *dev, + struct ethtool_channels *chan) +{ + struct stmmac_priv *priv = netdev_priv(dev); + + if (chan->rx_count > priv->dma_cap.number_rx_queues || + chan->tx_count > priv->dma_cap.number_tx_queues || + !chan->rx_count || !chan->tx_count) + return -EINVAL; + + return stmmac_reinit_queues(dev, chan->rx_count, chan->tx_count); +} + static int stmmac_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info) { @@ -926,6 +977,8 @@ static const struct ethtool_ops stmmac_ethtool_ops = { .get_regs_len = stmmac_ethtool_get_regs_len, .get_link = ethtool_op_get_link, .nway_reset = stmmac_nway_reset, + .get_ringparam = stmmac_get_ringparam, + .set_ringparam = stmmac_set_ringparam, .get_pauseparam = stmmac_get_pauseparam, .set_pauseparam = stmmac_set_pauseparam, .self_test = stmmac_selftest_run, @@ -944,6 +997,8 @@ static const struct ethtool_ops stmmac_ethtool_ops = { .get_ts_info = stmmac_get_ts_info, .get_coalesce = stmmac_get_coalesce, .set_coalesce = stmmac_set_coalesce, + .get_channels = stmmac_get_channels, + .set_channels = stmmac_set_channels, .get_tunable = stmmac_get_tunable, .set_tunable = stmmac_set_tunable, .get_link_ksettings = stmmac_ethtool_get_link_ksettings, diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index b56b13d64ab4..220626a8d499 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -63,8 +63,8 @@ static int phyaddr = -1; module_param(phyaddr, int, 0444); MODULE_PARM_DESC(phyaddr, "Physical device address"); -#define STMMAC_TX_THRESH (DMA_TX_SIZE / 4) -#define STMMAC_RX_THRESH (DMA_RX_SIZE / 4) +#define STMMAC_TX_THRESH(x) ((x)->dma_tx_size / 4) +#define STMMAC_RX_THRESH(x) ((x)->dma_rx_size / 4) static int flow_ctrl = FLOW_AUTO; module_param(flow_ctrl, int, 0644); @@ -176,32 +176,6 @@ static void stmmac_enable_all_queues(struct stmmac_priv *priv) } } -/** - * stmmac_stop_all_queues - Stop all queues - * @priv: driver private structure - */ -static void stmmac_stop_all_queues(struct stmmac_priv *priv) -{ - u32 tx_queues_cnt = priv->plat->tx_queues_to_use; - u32 queue; - - for (queue = 0; queue < tx_queues_cnt; queue++) - netif_tx_stop_queue(netdev_get_tx_queue(priv->dev, queue)); -} - -/** - * stmmac_start_all_queues - Start all queues - * @priv: driver private structure - */ -static void stmmac_start_all_queues(struct stmmac_priv *priv) -{ - u32 tx_queues_cnt = priv->plat->tx_queues_to_use; - u32 queue; - - for (queue = 0; queue < tx_queues_cnt; queue++) - netif_tx_start_queue(netdev_get_tx_queue(priv->dev, queue)); -} - static void stmmac_service_event_schedule(struct stmmac_priv *priv) { if (!test_bit(STMMAC_DOWN, &priv->state) && @@ -297,7 +271,7 @@ static inline u32 stmmac_tx_avail(struct stmmac_priv *priv, u32 queue) if (tx_q->dirty_tx > tx_q->cur_tx) avail = tx_q->dirty_tx - tx_q->cur_tx - 1; else - avail = DMA_TX_SIZE - tx_q->cur_tx + tx_q->dirty_tx - 1; + avail = priv->dma_tx_size - tx_q->cur_tx + tx_q->dirty_tx - 1; return avail; } @@ -315,7 +289,7 @@ static inline u32 stmmac_rx_dirty(struct stmmac_priv *priv, u32 queue) if (rx_q->dirty_rx <= rx_q->cur_rx) dirty = rx_q->cur_rx - rx_q->dirty_rx; else - dirty = DMA_RX_SIZE - rx_q->dirty_rx + rx_q->cur_rx; + dirty = priv->dma_rx_size - rx_q->dirty_rx + rx_q->cur_rx; return dirty; } @@ -360,7 +334,7 @@ void stmmac_disable_eee_mode(struct stmmac_priv *priv) /** * stmmac_eee_ctrl_timer - EEE TX SW timer. - * @arg : data hook + * @t: timer_list struct containing private info * Description: * if there is no data transfer and if we are not in LPI state, * then MAC Transmitter can be moved to LPI state. @@ -736,7 +710,7 @@ static int stmmac_hwtstamp_set(struct net_device *dev, struct ifreq *ifr) * a proprietary structure used to pass information to the driver. * Description: * This function obtain the current hardware timestamping settings - as requested. + * as requested. */ static int stmmac_hwtstamp_get(struct net_device *dev, struct ifreq *ifr) { @@ -789,14 +763,14 @@ static int stmmac_init_ptp(struct stmmac_priv *priv) static void stmmac_release_ptp(struct stmmac_priv *priv) { - if (priv->plat->clk_ptp_ref) - clk_disable_unprepare(priv->plat->clk_ptp_ref); + clk_disable_unprepare(priv->plat->clk_ptp_ref); stmmac_ptp_unregister(priv); } /** * stmmac_mac_flow_ctrl - Configure flow control in all queues * @priv: driver private structure + * @duplex: duplex passed to the next function * Description: It is used for configuring the flow control in all queues */ static void stmmac_mac_flow_ctrl(struct stmmac_priv *priv, u32 duplex) @@ -1150,7 +1124,7 @@ static void stmmac_display_rx_rings(struct stmmac_priv *priv) head_rx = (void *)rx_q->dma_rx; /* Display RX ring */ - stmmac_display_ring(priv, head_rx, DMA_RX_SIZE, true); + stmmac_display_ring(priv, head_rx, priv->dma_rx_size, true); } } @@ -1173,7 +1147,7 @@ static void stmmac_display_tx_rings(struct stmmac_priv *priv) else head_tx = (void *)tx_q->dma_tx; - stmmac_display_ring(priv, head_tx, DMA_TX_SIZE, false); + stmmac_display_ring(priv, head_tx, priv->dma_tx_size, false); } } @@ -1217,16 +1191,16 @@ static void stmmac_clear_rx_descriptors(struct stmmac_priv *priv, u32 queue) int i; /* Clear the RX descriptors */ - for (i = 0; i < DMA_RX_SIZE; i++) + for (i = 0; i < priv->dma_rx_size; i++) if (priv->extend_desc) stmmac_init_rx_desc(priv, &rx_q->dma_erx[i].basic, priv->use_riwt, priv->mode, - (i == DMA_RX_SIZE - 1), + (i == priv->dma_rx_size - 1), priv->dma_buf_sz); else stmmac_init_rx_desc(priv, &rx_q->dma_rx[i], priv->use_riwt, priv->mode, - (i == DMA_RX_SIZE - 1), + (i == priv->dma_rx_size - 1), priv->dma_buf_sz); } @@ -1243,8 +1217,8 @@ static void stmmac_clear_tx_descriptors(struct stmmac_priv *priv, u32 queue) int i; /* Clear the TX descriptors */ - for (i = 0; i < DMA_TX_SIZE; i++) { - int last = (i == (DMA_TX_SIZE - 1)); + for (i = 0; i < priv->dma_tx_size; i++) { + int last = (i == (priv->dma_tx_size - 1)); struct dma_desc *p; if (priv->extend_desc) @@ -1398,7 +1372,7 @@ static int init_dma_rx_desc_rings(struct net_device *dev, gfp_t flags) stmmac_clear_rx_descriptors(priv, queue); - for (i = 0; i < DMA_RX_SIZE; i++) { + for (i = 0; i < priv->dma_rx_size; i++) { struct dma_desc *p; if (priv->extend_desc) @@ -1413,16 +1387,18 @@ static int init_dma_rx_desc_rings(struct net_device *dev, gfp_t flags) } rx_q->cur_rx = 0; - rx_q->dirty_rx = (unsigned int)(i - DMA_RX_SIZE); + rx_q->dirty_rx = (unsigned int)(i - priv->dma_rx_size); /* Setup the chained descriptor addresses */ if (priv->mode == STMMAC_CHAIN_MODE) { if (priv->extend_desc) stmmac_mode_init(priv, rx_q->dma_erx, - rx_q->dma_rx_phy, DMA_RX_SIZE, 1); + rx_q->dma_rx_phy, + priv->dma_rx_size, 1); else stmmac_mode_init(priv, rx_q->dma_rx, - rx_q->dma_rx_phy, DMA_RX_SIZE, 0); + rx_q->dma_rx_phy, + priv->dma_rx_size, 0); } } @@ -1436,7 +1412,7 @@ err_init_rx_buffers: if (queue == 0) break; - i = DMA_RX_SIZE; + i = priv->dma_rx_size; queue--; } @@ -1468,13 +1444,15 @@ static int init_dma_tx_desc_rings(struct net_device *dev) if (priv->mode == STMMAC_CHAIN_MODE) { if (priv->extend_desc) stmmac_mode_init(priv, tx_q->dma_etx, - tx_q->dma_tx_phy, DMA_TX_SIZE, 1); + tx_q->dma_tx_phy, + priv->dma_tx_size, 1); else if (!(tx_q->tbs & STMMAC_TBS_AVAIL)) stmmac_mode_init(priv, tx_q->dma_tx, - tx_q->dma_tx_phy, DMA_TX_SIZE, 0); + tx_q->dma_tx_phy, + priv->dma_tx_size, 0); } - for (i = 0; i < DMA_TX_SIZE; i++) { + for (i = 0; i < priv->dma_tx_size; i++) { struct dma_desc *p; if (priv->extend_desc) p = &((tx_q->dma_etx + i)->basic); @@ -1538,7 +1516,7 @@ static void dma_free_rx_skbufs(struct stmmac_priv *priv, u32 queue) { int i; - for (i = 0; i < DMA_RX_SIZE; i++) + for (i = 0; i < priv->dma_rx_size; i++) stmmac_free_rx_buffer(priv, queue, i); } @@ -1551,7 +1529,7 @@ static void dma_free_tx_skbufs(struct stmmac_priv *priv, u32 queue) { int i; - for (i = 0; i < DMA_TX_SIZE; i++) + for (i = 0; i < priv->dma_tx_size; i++) stmmac_free_tx_buffer(priv, queue, i); } @@ -1573,11 +1551,11 @@ static void free_dma_rx_desc_resources(struct stmmac_priv *priv) /* Free DMA regions of consistent memory previously allocated */ if (!priv->extend_desc) - dma_free_coherent(priv->device, - DMA_RX_SIZE * sizeof(struct dma_desc), + dma_free_coherent(priv->device, priv->dma_rx_size * + sizeof(struct dma_desc), rx_q->dma_rx, rx_q->dma_rx_phy); else - dma_free_coherent(priv->device, DMA_RX_SIZE * + dma_free_coherent(priv->device, priv->dma_rx_size * sizeof(struct dma_extended_desc), rx_q->dma_erx, rx_q->dma_rx_phy); @@ -1616,7 +1594,7 @@ static void free_dma_tx_desc_resources(struct stmmac_priv *priv) addr = tx_q->dma_tx; } - size *= DMA_TX_SIZE; + size *= priv->dma_tx_size; dma_free_coherent(priv->device, size, addr, tx_q->dma_tx_phy); @@ -1649,7 +1627,7 @@ static int alloc_dma_rx_desc_resources(struct stmmac_priv *priv) rx_q->priv_data = priv; pp_params.flags = PP_FLAG_DMA_MAP; - pp_params.pool_size = DMA_RX_SIZE; + pp_params.pool_size = priv->dma_rx_size; num_pages = DIV_ROUND_UP(priv->dma_buf_sz, PAGE_SIZE); pp_params.order = ilog2(num_pages); pp_params.nid = dev_to_node(priv->device); @@ -1663,14 +1641,16 @@ static int alloc_dma_rx_desc_resources(struct stmmac_priv *priv) goto err_dma; } - rx_q->buf_pool = kcalloc(DMA_RX_SIZE, sizeof(*rx_q->buf_pool), + rx_q->buf_pool = kcalloc(priv->dma_rx_size, + sizeof(*rx_q->buf_pool), GFP_KERNEL); if (!rx_q->buf_pool) goto err_dma; if (priv->extend_desc) { rx_q->dma_erx = dma_alloc_coherent(priv->device, - DMA_RX_SIZE * sizeof(struct dma_extended_desc), + priv->dma_rx_size * + sizeof(struct dma_extended_desc), &rx_q->dma_rx_phy, GFP_KERNEL); if (!rx_q->dma_erx) @@ -1678,7 +1658,8 @@ static int alloc_dma_rx_desc_resources(struct stmmac_priv *priv) } else { rx_q->dma_rx = dma_alloc_coherent(priv->device, - DMA_RX_SIZE * sizeof(struct dma_desc), + priv->dma_rx_size * + sizeof(struct dma_desc), &rx_q->dma_rx_phy, GFP_KERNEL); if (!rx_q->dma_rx) @@ -1717,13 +1698,13 @@ static int alloc_dma_tx_desc_resources(struct stmmac_priv *priv) tx_q->queue_index = queue; tx_q->priv_data = priv; - tx_q->tx_skbuff_dma = kcalloc(DMA_TX_SIZE, + tx_q->tx_skbuff_dma = kcalloc(priv->dma_tx_size, sizeof(*tx_q->tx_skbuff_dma), GFP_KERNEL); if (!tx_q->tx_skbuff_dma) goto err_dma; - tx_q->tx_skbuff = kcalloc(DMA_TX_SIZE, + tx_q->tx_skbuff = kcalloc(priv->dma_tx_size, sizeof(struct sk_buff *), GFP_KERNEL); if (!tx_q->tx_skbuff) @@ -1736,7 +1717,7 @@ static int alloc_dma_tx_desc_resources(struct stmmac_priv *priv) else size = sizeof(struct dma_desc); - size *= DMA_TX_SIZE; + size *= priv->dma_tx_size; addr = dma_alloc_coherent(priv->device, size, &tx_q->dma_tx_phy, GFP_KERNEL); @@ -1965,6 +1946,7 @@ static void stmmac_dma_operation_mode(struct stmmac_priv *priv) /** * stmmac_tx_clean - to manage the transmission completion * @priv: driver private structure + * @budget: napi budget limiting this functions packet handling * @queue: TX queue index * Description: it reclaims the transmit resources after transmission completes. */ @@ -2046,7 +2028,7 @@ static int stmmac_tx_clean(struct stmmac_priv *priv, int budget, u32 queue) stmmac_release_tx_desc(priv, p, priv->mode); - entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE); + entry = STMMAC_GET_ENTRY(entry, priv->dma_tx_size); } tx_q->dirty_tx = entry; @@ -2055,7 +2037,7 @@ static int stmmac_tx_clean(struct stmmac_priv *priv, int budget, u32 queue) if (unlikely(netif_tx_queue_stopped(netdev_get_tx_queue(priv->dev, queue))) && - stmmac_tx_avail(priv, queue) > STMMAC_TX_THRESH) { + stmmac_tx_avail(priv, queue) > STMMAC_TX_THRESH(priv)) { netif_dbg(priv, tx_done, priv->dev, "%s: restart transmit\n", __func__); @@ -2328,7 +2310,8 @@ static int stmmac_init_dma_engine(struct stmmac_priv *priv) rx_q->dma_rx_phy, chan); rx_q->rx_tail_addr = rx_q->dma_rx_phy + - (DMA_RX_SIZE * sizeof(struct dma_desc)); + (priv->dma_rx_size * + sizeof(struct dma_desc)); stmmac_set_rx_tail_ptr(priv, priv->ioaddr, rx_q->rx_tail_addr, chan); } @@ -2357,7 +2340,7 @@ static void stmmac_tx_timer_arm(struct stmmac_priv *priv, u32 queue) /** * stmmac_tx_timer - mitigation sw timer for tx. - * @data: data pointer + * @t: data pointer * Description: * This is the timer handler to directly invoke the stmmac_tx_clean. */ @@ -2412,12 +2395,12 @@ static void stmmac_set_rings_length(struct stmmac_priv *priv) /* set TX ring length */ for (chan = 0; chan < tx_channels_count; chan++) stmmac_set_tx_ring_len(priv, priv->ioaddr, - (DMA_TX_SIZE - 1), chan); + (priv->dma_tx_size - 1), chan); /* set RX ring length */ for (chan = 0; chan < rx_channels_count; chan++) stmmac_set_rx_ring_len(priv, priv->ioaddr, - (DMA_RX_SIZE - 1), chan); + (priv->dma_rx_size - 1), chan); } /** @@ -2620,6 +2603,7 @@ static void stmmac_safety_feat_configuration(struct stmmac_priv *priv) /** * stmmac_hw_setup - setup mac in a usable state. * @dev : pointer to the device structure. + * @init_ptp: initialize PTP if set * Description: * this is the main function to setup the HW in a usable state because the * dma engine is reset, the core registers are configured (e.g. AXI, @@ -2740,6 +2724,10 @@ static int stmmac_hw_setup(struct net_device *dev, bool init_ptp) stmmac_enable_tbs(priv, priv->ioaddr, enable, chan); } + /* Configure real RX and TX queues */ + netif_set_real_num_rx_queues(dev, priv->plat->rx_queues_to_use); + netif_set_real_num_tx_queues(dev, priv->plat->tx_queues_to_use); + /* Start the ball rolling... */ stmmac_start_all_dma(priv); @@ -2797,6 +2785,11 @@ static int stmmac_open(struct net_device *dev) priv->rx_copybreak = STMMAC_RX_COPYBREAK; + if (!priv->dma_tx_size) + priv->dma_tx_size = DMA_DEFAULT_TX_SIZE; + if (!priv->dma_rx_size) + priv->dma_rx_size = DMA_DEFAULT_RX_SIZE; + /* Earlier check for TBS */ for (chan = 0; chan < priv->plat->tx_queues_to_use; chan++) { struct stmmac_tx_queue *tx_q = &priv->tx_queue[chan]; @@ -2868,7 +2861,7 @@ static int stmmac_open(struct net_device *dev) } stmmac_enable_all_queues(priv); - stmmac_start_all_queues(priv); + netif_tx_start_all_queues(priv->dev); return 0; @@ -2911,8 +2904,6 @@ static int stmmac_release(struct net_device *dev) phylink_stop(priv->phylink); phylink_disconnect_phy(priv->phylink); - stmmac_stop_all_queues(priv); - stmmac_disable_all_queues(priv); for (chan = 0; chan < priv->plat->tx_queues_to_use; chan++) @@ -2968,7 +2959,7 @@ static bool stmmac_vlan_insert(struct stmmac_priv *priv, struct sk_buff *skb, return false; stmmac_set_tx_owner(priv, p); - tx_q->cur_tx = STMMAC_GET_ENTRY(tx_q->cur_tx, DMA_TX_SIZE); + tx_q->cur_tx = STMMAC_GET_ENTRY(tx_q->cur_tx, priv->dma_tx_size); return true; } @@ -2977,7 +2968,7 @@ static bool stmmac_vlan_insert(struct stmmac_priv *priv, struct sk_buff *skb, * @priv: driver private structure * @des: buffer start address * @total_len: total length to fill in descriptors - * @last_segmant: condition for the last descriptor + * @last_segment: condition for the last descriptor * @queue: TX queue index * Description: * This function fills descriptor and request new descriptors according to @@ -2996,7 +2987,8 @@ static void stmmac_tso_allocator(struct stmmac_priv *priv, dma_addr_t des, while (tmp_len > 0) { dma_addr_t curr_addr; - tx_q->cur_tx = STMMAC_GET_ENTRY(tx_q->cur_tx, DMA_TX_SIZE); + tx_q->cur_tx = STMMAC_GET_ENTRY(tx_q->cur_tx, + priv->dma_tx_size); WARN_ON(tx_q->tx_skbuff[tx_q->cur_tx]); if (tx_q->tbs & STMMAC_TBS_AVAIL) @@ -3103,7 +3095,8 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev) stmmac_set_mss(priv, mss_desc, mss); tx_q->mss = mss; - tx_q->cur_tx = STMMAC_GET_ENTRY(tx_q->cur_tx, DMA_TX_SIZE); + tx_q->cur_tx = STMMAC_GET_ENTRY(tx_q->cur_tx, + priv->dma_tx_size); WARN_ON(tx_q->tx_skbuff[tx_q->cur_tx]); } @@ -3210,7 +3203,7 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev) * ndo_start_xmit will fill this descriptor the next time it's * called and stmmac_tx_clean may clean up to this descriptor. */ - tx_q->cur_tx = STMMAC_GET_ENTRY(tx_q->cur_tx, DMA_TX_SIZE); + tx_q->cur_tx = STMMAC_GET_ENTRY(tx_q->cur_tx, priv->dma_tx_size); if (unlikely(stmmac_tx_avail(priv, queue) <= (MAX_SKB_FRAGS + 1))) { netif_dbg(priv, hw, priv->dev, "%s: stop transmitted packets\n", @@ -3373,7 +3366,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev) int len = skb_frag_size(frag); bool last_segment = (i == (nfrags - 1)); - entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE); + entry = STMMAC_GET_ENTRY(entry, priv->dma_tx_size); WARN_ON(tx_q->tx_skbuff[entry]); if (likely(priv->extend_desc)) @@ -3441,7 +3434,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev) * ndo_start_xmit will fill this descriptor the next time it's * called and stmmac_tx_clean may clean up to this descriptor. */ - entry = STMMAC_GET_ENTRY(entry, DMA_TX_SIZE); + entry = STMMAC_GET_ENTRY(entry, priv->dma_tx_size); tx_q->cur_tx = entry; if (netif_msg_pktdata(priv)) { @@ -3626,7 +3619,7 @@ static inline void stmmac_rx_refill(struct stmmac_priv *priv, u32 queue) dma_wmb(); stmmac_set_rx_owner(priv, p, use_rx_wd); - entry = STMMAC_GET_ENTRY(entry, DMA_RX_SIZE); + entry = STMMAC_GET_ENTRY(entry, priv->dma_rx_size); } rx_q->dirty_rx = entry; rx_q->rx_tail_addr = rx_q->dma_rx_phy + @@ -3638,15 +3631,15 @@ static unsigned int stmmac_rx_buf1_len(struct stmmac_priv *priv, struct dma_desc *p, int status, unsigned int len) { - int ret, coe = priv->hw->rx_csum; unsigned int plen = 0, hlen = 0; + int coe = priv->hw->rx_csum; /* Not first descriptor, buffer is always zero */ if (priv->sph && len) return 0; /* First descriptor, get split header length */ - ret = stmmac_get_rx_header_len(priv, p, &hlen); + stmmac_get_rx_header_len(priv, p, &hlen); if (priv->sph && hlen) { priv->xstats.rx_split_hdr_pkt_n++; return hlen; @@ -3709,7 +3702,7 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit, u32 queue) else rx_head = (void *)rx_q->dma_rx; - stmmac_display_ring(priv, rx_head, DMA_RX_SIZE, true); + stmmac_display_ring(priv, rx_head, priv->dma_rx_size, true); } while (count < limit) { unsigned int buf1_len = 0, buf2_len = 0; @@ -3751,7 +3744,8 @@ read_again: if (unlikely(status & dma_own)) break; - rx_q->cur_rx = STMMAC_GET_ENTRY(rx_q->cur_rx, DMA_RX_SIZE); + rx_q->cur_rx = STMMAC_GET_ENTRY(rx_q->cur_rx, + priv->dma_rx_size); next_entry = rx_q->cur_rx; if (priv->extend_desc) @@ -3926,7 +3920,7 @@ static int stmmac_napi_poll_tx(struct napi_struct *napi, int budget) priv->xstats.napi_poll++; - work_done = stmmac_tx_clean(priv, DMA_TX_SIZE, chan); + work_done = stmmac_tx_clean(priv, priv->dma_tx_size, chan); work_done = min(work_done, budget); if (work_done < budget && napi_complete_done(napi, work_done)) { @@ -3943,6 +3937,7 @@ static int stmmac_napi_poll_tx(struct napi_struct *napi, int budget) /** * stmmac_tx_timeout * @dev : Pointer to net device structure + * @txqueue: the index of the hanging transmit queue * Description: this function is called when a packet transmission fails to * complete within a reasonable time. The driver will mark the error in the * netdev structure and arrange for the device to be reset to a sane state @@ -4319,11 +4314,11 @@ static int stmmac_rings_status_show(struct seq_file *seq, void *v) if (priv->extend_desc) { seq_printf(seq, "Extended descriptor ring:\n"); sysfs_display_ring((void *)rx_q->dma_erx, - DMA_RX_SIZE, 1, seq); + priv->dma_rx_size, 1, seq); } else { seq_printf(seq, "Descriptor ring:\n"); sysfs_display_ring((void *)rx_q->dma_rx, - DMA_RX_SIZE, 0, seq); + priv->dma_rx_size, 0, seq); } } @@ -4335,11 +4330,11 @@ static int stmmac_rings_status_show(struct seq_file *seq, void *v) if (priv->extend_desc) { seq_printf(seq, "Extended descriptor ring:\n"); sysfs_display_ring((void *)tx_q->dma_etx, - DMA_TX_SIZE, 1, seq); + priv->dma_tx_size, 1, seq); } else if (!(tx_q->tbs & STMMAC_TBS_AVAIL)) { seq_printf(seq, "Descriptor ring:\n"); sysfs_display_ring((void *)tx_q->dma_tx, - DMA_TX_SIZE, 0, seq); + priv->dma_tx_size, 0, seq); } } @@ -4725,6 +4720,9 @@ static int stmmac_hw_init(struct stmmac_priv *priv) if (priv->dma_cap.tsoen) dev_info(priv->device, "TSO supported\n"); + priv->hw->vlan_fail_q_en = priv->plat->vlan_fail_q_en; + priv->hw->vlan_fail_q = priv->plat->vlan_fail_q; + /* Run HW quirks, if any */ if (priv->hwif_quirks) { ret = priv->hwif_quirks(priv); @@ -4747,6 +4745,86 @@ static int stmmac_hw_init(struct stmmac_priv *priv) return 0; } +static void stmmac_napi_add(struct net_device *dev) +{ + struct stmmac_priv *priv = netdev_priv(dev); + u32 queue, maxq; + + maxq = max(priv->plat->rx_queues_to_use, priv->plat->tx_queues_to_use); + + for (queue = 0; queue < maxq; queue++) { + struct stmmac_channel *ch = &priv->channel[queue]; + + ch->priv_data = priv; + ch->index = queue; + + if (queue < priv->plat->rx_queues_to_use) { + netif_napi_add(dev, &ch->rx_napi, stmmac_napi_poll_rx, + NAPI_POLL_WEIGHT); + } + if (queue < priv->plat->tx_queues_to_use) { + netif_tx_napi_add(dev, &ch->tx_napi, + stmmac_napi_poll_tx, + NAPI_POLL_WEIGHT); + } + } +} + +static void stmmac_napi_del(struct net_device *dev) +{ + struct stmmac_priv *priv = netdev_priv(dev); + u32 queue, maxq; + + maxq = max(priv->plat->rx_queues_to_use, priv->plat->tx_queues_to_use); + + for (queue = 0; queue < maxq; queue++) { + struct stmmac_channel *ch = &priv->channel[queue]; + + if (queue < priv->plat->rx_queues_to_use) + netif_napi_del(&ch->rx_napi); + if (queue < priv->plat->tx_queues_to_use) + netif_napi_del(&ch->tx_napi); + } +} + +int stmmac_reinit_queues(struct net_device *dev, u32 rx_cnt, u32 tx_cnt) +{ + struct stmmac_priv *priv = netdev_priv(dev); + int ret = 0; + + if (netif_running(dev)) + stmmac_release(dev); + + stmmac_napi_del(dev); + + priv->plat->rx_queues_to_use = rx_cnt; + priv->plat->tx_queues_to_use = tx_cnt; + + stmmac_napi_add(dev); + + if (netif_running(dev)) + ret = stmmac_open(dev); + + return ret; +} + +int stmmac_reinit_ringparam(struct net_device *dev, u32 rx_size, u32 tx_size) +{ + struct stmmac_priv *priv = netdev_priv(dev); + int ret = 0; + + if (netif_running(dev)) + stmmac_release(dev); + + priv->dma_rx_size = rx_size; + priv->dma_tx_size = tx_size; + + if (netif_running(dev)) + ret = stmmac_open(dev); + + return ret; +} + /** * stmmac_dvr_probe * @device: device pointer @@ -4763,7 +4841,7 @@ int stmmac_dvr_probe(struct device *device, { struct net_device *ndev = NULL; struct stmmac_priv *priv; - u32 queue, rxq, maxq; + u32 rxq; int i, ret = 0; ndev = devm_alloc_etherdev_mqs(device, sizeof(struct stmmac_priv), @@ -4827,10 +4905,6 @@ int stmmac_dvr_probe(struct device *device, stmmac_check_ether_addr(priv); - /* Configure real RX and TX queues */ - netif_set_real_num_rx_queues(ndev, priv->plat->rx_queues_to_use); - netif_set_real_num_tx_queues(ndev, priv->plat->tx_queues_to_use); - ndev->netdev_ops = &stmmac_netdev_ops; ndev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | @@ -4928,25 +5002,7 @@ int stmmac_dvr_probe(struct device *device, priv->flow_ctrl = FLOW_AUTO; /* RX/TX pause on */ /* Setup channels NAPI */ - maxq = max(priv->plat->rx_queues_to_use, priv->plat->tx_queues_to_use); - - for (queue = 0; queue < maxq; queue++) { - struct stmmac_channel *ch = &priv->channel[queue]; - - spin_lock_init(&ch->lock); - ch->priv_data = priv; - ch->index = queue; - - if (queue < priv->plat->rx_queues_to_use) { - netif_napi_add(ndev, &ch->rx_napi, stmmac_napi_poll_rx, - NAPI_POLL_WEIGHT); - } - if (queue < priv->plat->tx_queues_to_use) { - netif_tx_napi_add(ndev, &ch->tx_napi, - stmmac_napi_poll_tx, - NAPI_POLL_WEIGHT); - } - } + stmmac_napi_add(ndev); mutex_init(&priv->lock); @@ -5011,14 +5067,7 @@ error_phy_setup: priv->hw->pcs != STMMAC_PCS_RTBI) stmmac_mdio_unregister(ndev); error_mdio_register: - for (queue = 0; queue < maxq; queue++) { - struct stmmac_channel *ch = &priv->channel[queue]; - - if (queue < priv->plat->rx_queues_to_use) - netif_napi_del(&ch->rx_napi); - if (queue < priv->plat->tx_queues_to_use) - netif_napi_del(&ch->tx_napi); - } + stmmac_napi_del(ndev); error_hw_init: destroy_workqueue(priv->wq); @@ -5086,7 +5135,6 @@ int stmmac_suspend(struct device *dev) mutex_lock(&priv->lock); netif_device_detach(ndev); - stmmac_stop_all_queues(priv); stmmac_disable_all_queues(priv); @@ -5115,8 +5163,7 @@ int stmmac_suspend(struct device *dev) stmmac_mac_set(priv, priv->ioaddr, false); pinctrl_pm_select_sleep_state(priv->device); /* Disable clock in case of PWM is off */ - if (priv->plat->clk_ptp_ref) - clk_disable_unprepare(priv->plat->clk_ptp_ref); + clk_disable_unprepare(priv->plat->clk_ptp_ref); clk_disable_unprepare(priv->plat->pclk); clk_disable_unprepare(priv->plat->stmmac_clk); } @@ -5129,7 +5176,7 @@ EXPORT_SYMBOL_GPL(stmmac_suspend); /** * stmmac_reset_queues_param - reset queue parameters - * @dev: device pointer + * @priv: device pointer */ static void stmmac_reset_queues_param(struct stmmac_priv *priv) { @@ -5213,8 +5260,6 @@ int stmmac_resume(struct device *dev) stmmac_enable_all_queues(priv); - stmmac_start_all_queues(priv); - mutex_unlock(&priv->lock); if (!device_may_wakeup(priv->device) || !priv->plat->pmt) { diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index f32317fa75c8..af34a4cadbb0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -125,6 +125,7 @@ static struct stmmac_axi *stmmac_axi_setup(struct platform_device *pdev) /** * stmmac_mtl_setup - parse DT parameters for multiple queues configuration * @pdev: platform device + * @plat: enet data */ static int stmmac_mtl_setup(struct platform_device *pdev, struct plat_stmmacenet_data *plat) @@ -360,7 +361,7 @@ static int stmmac_dt_phy(struct plat_stmmacenet_data *plat, /** * stmmac_of_get_mac_mode - retrieves the interface of the MAC - * @np - device-tree node + * @np: - device-tree node * Description: * Similar to `of_get_phy_mode()`, this function will retrieve (from * the device-tree) the interface mode on the MAC side. This assumes diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c index bf195adee393..0462dcc93e53 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c @@ -796,7 +796,7 @@ static int stmmac_test_flowctrl(struct stmmac_priv *priv) u32 tail; tail = priv->rx_queue[i].dma_rx_phy + - (DMA_RX_SIZE * sizeof(struct dma_desc)); + (priv->dma_rx_size * sizeof(struct dma_desc)); stmmac_set_rx_tail_ptr(priv, priv->ioaddr, tail, i); stmmac_start_rx(priv, priv->ioaddr, i); diff --git a/drivers/net/ethernet/sun/cassini.c b/drivers/net/ethernet/sun/cassini.c index b624e177ec71..9ff894ba8d3e 100644 --- a/drivers/net/ethernet/sun/cassini.c +++ b/drivers/net/ethernet/sun/cassini.c @@ -454,8 +454,8 @@ static int cas_page_free(struct cas *cp, cas_page_t *page) #define RX_USED_ADD(x, y) ((x)->used += (y)) #define RX_USED_SET(x, y) ((x)->used = (y)) #else -#define RX_USED_ADD(x, y) -#define RX_USED_SET(x, y) +#define RX_USED_ADD(x, y) do { } while(0) +#define RX_USED_SET(x, y) do { } while(0) #endif /* local page allocation routines for the receive buffers. jumbo pages diff --git a/drivers/net/ethernet/sun/sunbmac.c b/drivers/net/ethernet/sun/sunbmac.c index 34fdbc6d6031..c646575e79d5 100644 --- a/drivers/net/ethernet/sun/sunbmac.c +++ b/drivers/net/ethernet/sun/sunbmac.c @@ -209,13 +209,13 @@ static void bigmac_clean_rings(struct bigmac *bp) } } -static void bigmac_init_rings(struct bigmac *bp, int from_irq) +static void bigmac_init_rings(struct bigmac *bp, bool non_blocking) { struct bmac_init_block *bb = bp->bmac_block; int i; gfp_t gfp_flags = GFP_KERNEL; - if (from_irq || in_interrupt()) + if (non_blocking) gfp_flags = GFP_ATOMIC; bp->rx_new = bp->rx_old = bp->tx_new = bp->tx_old = 0; @@ -489,7 +489,7 @@ static void bigmac_tcvr_init(struct bigmac *bp) } } -static int bigmac_init_hw(struct bigmac *, int); +static int bigmac_init_hw(struct bigmac *, bool); static int try_next_permutation(struct bigmac *bp, void __iomem *tregs) { @@ -549,7 +549,7 @@ static void bigmac_timer(struct timer_list *t) if (ret == -1) { printk(KERN_ERR "%s: Link down, cable problem?\n", bp->dev->name); - ret = bigmac_init_hw(bp, 0); + ret = bigmac_init_hw(bp, true); if (ret) { printk(KERN_ERR "%s: Error, cannot re-init the " "BigMAC.\n", bp->dev->name); @@ -617,7 +617,7 @@ static void bigmac_begin_auto_negotiation(struct bigmac *bp) add_timer(&bp->bigmac_timer); } -static int bigmac_init_hw(struct bigmac *bp, int from_irq) +static int bigmac_init_hw(struct bigmac *bp, bool non_blocking) { void __iomem *gregs = bp->gregs; void __iomem *cregs = bp->creg; @@ -635,7 +635,7 @@ static int bigmac_init_hw(struct bigmac *bp, int from_irq) qec_init(bp); /* Alloc and reset the tx/rx descriptor chains. */ - bigmac_init_rings(bp, from_irq); + bigmac_init_rings(bp, non_blocking); /* Initialize the PHY. */ bigmac_tcvr_init(bp); @@ -749,7 +749,7 @@ static void bigmac_is_medium_rare(struct bigmac *bp, u32 qec_status, u32 bmac_st } printk(" RESET\n"); - bigmac_init_hw(bp, 1); + bigmac_init_hw(bp, true); } /* BigMAC transmit complete service routines. */ @@ -921,7 +921,7 @@ static int bigmac_open(struct net_device *dev) return ret; } timer_setup(&bp->bigmac_timer, bigmac_timer, 0); - ret = bigmac_init_hw(bp, 0); + ret = bigmac_init_hw(bp, false); if (ret) free_irq(dev->irq, bp); return ret; @@ -945,7 +945,7 @@ static void bigmac_tx_timeout(struct net_device *dev, unsigned int txqueue) { struct bigmac *bp = netdev_priv(dev); - bigmac_init_hw(bp, 0); + bigmac_init_hw(bp, true); netif_wake_queue(dev); } diff --git a/drivers/net/ethernet/sun/sungem.c b/drivers/net/ethernet/sun/sungem.c index 8deb943ca5de..58f142ee78a3 100644 --- a/drivers/net/ethernet/sun/sungem.c +++ b/drivers/net/ethernet/sun/sungem.c @@ -2965,9 +2965,8 @@ static int gem_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) /* It is guaranteed that the returned buffer will be at least * PAGE_SIZE aligned. */ - gp->init_block = (struct gem_init_block *) - dma_alloc_coherent(&pdev->dev, sizeof(struct gem_init_block), - &gp->gblock_dvma, GFP_KERNEL); + gp->init_block = dma_alloc_coherent(&pdev->dev, sizeof(struct gem_init_block), + &gp->gblock_dvma, GFP_KERNEL); if (!gp->init_block) { pr_err("Cannot allocate init block, aborting\n"); err = -ENOMEM; diff --git a/drivers/net/ethernet/synopsys/dwc-xlgmac-common.c b/drivers/net/ethernet/synopsys/dwc-xlgmac-common.c index eb1c6b03c329..df26cea45904 100644 --- a/drivers/net/ethernet/synopsys/dwc-xlgmac-common.c +++ b/drivers/net/ethernet/synopsys/dwc-xlgmac-common.c @@ -513,7 +513,7 @@ void xlgmac_get_all_hw_features(struct xlgmac_pdata *pdata) void xlgmac_print_all_hw_features(struct xlgmac_pdata *pdata) { - char *str = NULL; + char __maybe_unused *str = NULL; XLGMAC_PR("\n"); XLGMAC_PR("=====================================================\n"); diff --git a/drivers/net/ethernet/tehuti/tehuti.c b/drivers/net/ethernet/tehuti/tehuti.c index e28727297563..b8f4f419173f 100644 --- a/drivers/net/ethernet/tehuti/tehuti.c +++ b/drivers/net/ethernet/tehuti/tehuti.c @@ -138,7 +138,10 @@ static void print_eth_id(struct net_device *ndev) * @priv: NIC private structure * @f: fifo to initialize * @fsz_type: fifo size type: 0-4KB, 1-8KB, 2-16KB, 3-32KB - * @reg_XXX: offsets of registers relative to base address + * @reg_CFG0: offsets of registers relative to base address + * @reg_CFG1: offsets of registers relative to base address + * @reg_RPTR: offsets of registers relative to base address + * @reg_WPTR: offsets of registers relative to base address * * 1K extra space is allocated at the end of the fifo to simplify * processing of descriptors that wraps around fifo's end @@ -153,11 +156,11 @@ bdx_fifo_init(struct bdx_priv *priv, struct fifo *f, int fsz_type, u16 memsz = FIFO_SIZE * (1 << fsz_type); memset(f, 0, sizeof(struct fifo)); - /* pci_alloc_consistent gives us 4k-aligned memory */ - f->va = pci_alloc_consistent(priv->pdev, - memsz + FIFO_EXTRA_SPACE, &f->da); + /* dma_alloc_coherent gives us 4k-aligned memory */ + f->va = dma_alloc_coherent(&priv->pdev->dev, memsz + FIFO_EXTRA_SPACE, + &f->da, GFP_ATOMIC); if (!f->va) { - pr_err("pci_alloc_consistent failed\n"); + pr_err("dma_alloc_coherent failed\n"); RET(-ENOMEM); } f->reg_CFG0 = reg_CFG0; @@ -183,8 +186,8 @@ static void bdx_fifo_free(struct bdx_priv *priv, struct fifo *f) { ENTER; if (f->va) { - pci_free_consistent(priv->pdev, - f->memsz + FIFO_EXTRA_SPACE, f->va, f->da); + dma_free_coherent(&priv->pdev->dev, + f->memsz + FIFO_EXTRA_SPACE, f->va, f->da); f->va = NULL; } RET(); @@ -558,7 +561,7 @@ static int bdx_reset(struct bdx_priv *priv) /** * bdx_close - Disables a network interface - * @netdev: network interface device structure + * @ndev: network interface device structure * * Returns 0, this is not allowed to fail * @@ -585,7 +588,7 @@ static int bdx_close(struct net_device *ndev) /** * bdx_open - Called when a network interface is made active - * @netdev: network interface device structure + * @ndev: network interface device structure * * Returns 0 on success, negative value on failure * @@ -698,7 +701,7 @@ static int bdx_ioctl(struct net_device *ndev, struct ifreq *ifr, int cmd) * __bdx_vlan_rx_vid - private helper for adding/killing VLAN vid * @ndev: network device * @vid: VLAN vid - * @op: add or kill operation + * @enable: enable or disable vlan * * Passes VLAN filter table to hardware */ @@ -729,6 +732,7 @@ static void __bdx_vlan_rx_vid(struct net_device *ndev, uint16_t vid, int enable) /** * bdx_vlan_rx_add_vid - kernel hook for adding VLAN vid to hw filtering table * @ndev: network device + * @proto: unused * @vid: VLAN vid to add */ static int bdx_vlan_rx_add_vid(struct net_device *ndev, __be16 proto, u16 vid) @@ -740,6 +744,7 @@ static int bdx_vlan_rx_add_vid(struct net_device *ndev, __be16 proto, u16 vid) /** * bdx_vlan_rx_kill_vid - kernel hook for killing VLAN vid in hw filtering table * @ndev: network device + * @proto: unused * @vid: VLAN vid to kill */ static int bdx_vlan_rx_kill_vid(struct net_device *ndev, __be16 proto, u16 vid) @@ -750,7 +755,7 @@ static int bdx_vlan_rx_kill_vid(struct net_device *ndev, __be16 proto, u16 vid) /** * bdx_change_mtu - Change the Maximum Transfer Unit - * @netdev: network interface device structure + * @ndev: network interface device structure * @new_mtu: new value for maximum frame size * * Returns 0 on success, negative on failure @@ -1033,9 +1038,8 @@ static void bdx_rx_free_skbs(struct bdx_priv *priv, struct rxf_fifo *f) for (i = 0; i < db->nelem; i++) { dm = bdx_rxdb_addr_elem(db, i); if (dm->dma) { - pci_unmap_single(priv->pdev, - dm->dma, f->m.pktsz, - PCI_DMA_FROMDEVICE); + dma_unmap_single(&priv->pdev->dev, dm->dma, + f->m.pktsz, DMA_FROM_DEVICE); dev_kfree_skb(dm->skb); } } @@ -1097,9 +1101,8 @@ static void bdx_rx_alloc_skbs(struct bdx_priv *priv, struct rxf_fifo *f) idx = bdx_rxdb_alloc_elem(db); dm = bdx_rxdb_addr_elem(db, idx); - dm->dma = pci_map_single(priv->pdev, - skb->data, f->m.pktsz, - PCI_DMA_FROMDEVICE); + dm->dma = dma_map_single(&priv->pdev->dev, skb->data, + f->m.pktsz, DMA_FROM_DEVICE); dm->skb = skb; rxfd = (struct rxf_desc *)(f->m.va + f->m.wptr); rxfd->info = CPU_CHIP_SWAP32(0x10003); /* INFO=1 BC=3 */ @@ -1259,16 +1262,15 @@ static int bdx_rx_receive(struct bdx_priv *priv, struct rxd_fifo *f, int budget) (skb2 = netdev_alloc_skb(priv->ndev, len + NET_IP_ALIGN))) { skb_reserve(skb2, NET_IP_ALIGN); /*skb_put(skb2, len); */ - pci_dma_sync_single_for_cpu(priv->pdev, - dm->dma, rxf_fifo->m.pktsz, - PCI_DMA_FROMDEVICE); + dma_sync_single_for_cpu(&priv->pdev->dev, dm->dma, + rxf_fifo->m.pktsz, + DMA_FROM_DEVICE); memcpy(skb2->data, skb->data, len); bdx_recycle_skb(priv, rxdd); skb = skb2; } else { - pci_unmap_single(priv->pdev, - dm->dma, rxf_fifo->m.pktsz, - PCI_DMA_FROMDEVICE); + dma_unmap_single(&priv->pdev->dev, dm->dma, + rxf_fifo->m.pktsz, DMA_FROM_DEVICE); bdx_rxdb_free_elem(db, rxdd->va_lo); } @@ -1478,8 +1480,8 @@ bdx_tx_map_skb(struct bdx_priv *priv, struct sk_buff *skb, int i; db->wptr->len = skb_headlen(skb); - db->wptr->addr.dma = pci_map_single(priv->pdev, skb->data, - db->wptr->len, PCI_DMA_TODEVICE); + db->wptr->addr.dma = dma_map_single(&priv->pdev->dev, skb->data, + db->wptr->len, DMA_TO_DEVICE); pbl->len = CPU_CHIP_SWAP32(db->wptr->len); pbl->pa_lo = CPU_CHIP_SWAP32(L32_64(db->wptr->addr.dma)); pbl->pa_hi = CPU_CHIP_SWAP32(H32_64(db->wptr->addr.dma)); @@ -1716,8 +1718,8 @@ static void bdx_tx_cleanup(struct bdx_priv *priv) BDX_ASSERT(db->rptr->len == 0); do { BDX_ASSERT(db->rptr->addr.dma == 0); - pci_unmap_page(priv->pdev, db->rptr->addr.dma, - db->rptr->len, PCI_DMA_TODEVICE); + dma_unmap_page(&priv->pdev->dev, db->rptr->addr.dma, + db->rptr->len, DMA_TO_DEVICE); bdx_tx_db_inc_rptr(db); } while (db->rptr->len > 0); tx_level -= db->rptr->len; /* '-' koz len is negative */ @@ -1756,6 +1758,8 @@ static void bdx_tx_cleanup(struct bdx_priv *priv) /** * bdx_tx_free_skbs - frees all skbs from TXD fifo. + * @priv: NIC private structure + * * It gets called when OS stops this dev, eg upon "ifconfig down" or rmmod */ static void bdx_tx_free_skbs(struct bdx_priv *priv) @@ -1765,8 +1769,8 @@ static void bdx_tx_free_skbs(struct bdx_priv *priv) ENTER; while (db->rptr != db->wptr) { if (likely(db->rptr->len)) - pci_unmap_page(priv->pdev, db->rptr->addr.dma, - db->rptr->len, PCI_DMA_TODEVICE); + dma_unmap_page(&priv->pdev->dev, db->rptr->addr.dma, + db->rptr->len, DMA_TO_DEVICE); else dev_kfree_skb(db->rptr->addr.skb); bdx_tx_db_inc_rptr(db); @@ -1902,12 +1906,12 @@ bdx_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (err) /* it triggers interrupt, dunno why. */ goto err_pci; /* it's not a problem though */ - if (!(err = pci_set_dma_mask(pdev, DMA_BIT_MASK(64))) && - !(err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64)))) { + if (!(err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(64))) && + !(err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(64)))) { pci_using_dac = 1; } else { - if ((err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32))) || - (err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32)))) { + if ((err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32))) || + (err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32)))) { pr_err("No usable DMA configuration, aborting\n"); goto err_dma; } diff --git a/drivers/net/ethernet/ti/am65-cpsw-ethtool.c b/drivers/net/ethernet/ti/am65-cpsw-ethtool.c index 496dafb25128..6e4d4f9e32e0 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-ethtool.c +++ b/drivers/net/ethernet/ti/am65-cpsw-ethtool.c @@ -572,13 +572,14 @@ static int am65_cpsw_nway_reset(struct net_device *ndev) static int am65_cpsw_get_regs_len(struct net_device *ndev) { struct am65_cpsw_common *common = am65_ndev_to_common(ndev); - u32 i, regdump_len = 0; + u32 ale_entries, i, regdump_len = 0; + ale_entries = cpsw_ale_get_num_entries(common->ale); for (i = 0; i < ARRAY_SIZE(am65_cpsw_regdump); i++) { if (am65_cpsw_regdump[i].hdr.module_id == AM65_CPSW_REGDUMP_MOD_CPSW_ALE_TBL) { regdump_len += sizeof(struct am65_cpsw_regdump_hdr); - regdump_len += common->ale->params.ale_entries * + regdump_len += ale_entries * ALE_ENTRY_WORDS * sizeof(u32); continue; } @@ -592,10 +593,11 @@ static void am65_cpsw_get_regs(struct net_device *ndev, struct ethtool_regs *regs, void *p) { struct am65_cpsw_common *common = am65_ndev_to_common(ndev); - u32 i, j, pos, *reg = p; + u32 ale_entries, i, j, pos, *reg = p; /* update CPSW IP version */ regs->version = AM65_CPSW_REGDUMP_VER; + ale_entries = cpsw_ale_get_num_entries(common->ale); pos = 0; for (i = 0; i < ARRAY_SIZE(am65_cpsw_regdump); i++) { @@ -603,7 +605,7 @@ static void am65_cpsw_get_regs(struct net_device *ndev, if (am65_cpsw_regdump[i].hdr.module_id == AM65_CPSW_REGDUMP_MOD_CPSW_ALE_TBL) { - u32 ale_tbl_len = common->ale->params.ale_entries * + u32 ale_tbl_len = ale_entries * ALE_ENTRY_WORDS * sizeof(u32) + sizeof(struct am65_cpsw_regdump_hdr); reg[pos++] = ale_tbl_len; diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c index 9baf3f3da91e..501d676fd88b 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c +++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c @@ -5,6 +5,7 @@ * */ +#include <linux/clk.h> #include <linux/etherdevice.h> #include <linux/if_vlan.h> #include <linux/interrupt.h> @@ -2038,6 +2039,7 @@ static int am65_cpsw_nuss_probe(struct platform_device *pdev) struct am65_cpsw_common *common; struct device_node *node; struct resource *res; + struct clk *clk; int ret, i; common = devm_kzalloc(dev, sizeof(struct am65_cpsw_common), GFP_KERNEL); @@ -2086,6 +2088,16 @@ static int am65_cpsw_nuss_probe(struct platform_device *pdev) if (!common->ports) return -ENOMEM; + clk = devm_clk_get(dev, "fck"); + if (IS_ERR(clk)) { + ret = PTR_ERR(clk); + + if (ret != -EPROBE_DEFER) + dev_err(dev, "error getting fck clock %d\n", ret); + return ret; + } + common->bus_freq = clk_get_rate(clk); + pm_runtime_enable(dev); ret = pm_runtime_get_sync(dev); if (ret < 0) { @@ -2131,10 +2143,10 @@ static int am65_cpsw_nuss_probe(struct platform_device *pdev) /* init common data */ ale_params.dev = dev; ale_params.ale_ageout = AM65_CPSW_ALE_AGEOUT_DEFAULT; - ale_params.ale_entries = 0; ale_params.ale_ports = common->port_num + 1; ale_params.ale_regs = common->cpsw_base + AM65_CPSW_NU_ALE_BASE; - ale_params.nu_switch_ale = true; + ale_params.dev_id = "am65x-cpsw2g"; + ale_params.bus_freq = common->bus_freq; common->ale = cpsw_ale_create(&ale_params); if (IS_ERR(common->ale)) { diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.h b/drivers/net/ethernet/ti/am65-cpsw-nuss.h index 94f666ea0e53..993e1d4d3222 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-nuss.h +++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.h @@ -106,6 +106,7 @@ struct am65_cpsw_common { u32 nuss_ver; u32 cpsw_ver; + unsigned long bus_freq; bool pf_p0_rx_ptype_rrobin; struct am65_cpts *cpts; int est_enabled; diff --git a/drivers/net/ethernet/ti/am65-cpts.c b/drivers/net/ethernet/ti/am65-cpts.c index c59a289e428c..75056c14b161 100644 --- a/drivers/net/ethernet/ti/am65-cpts.c +++ b/drivers/net/ethernet/ti/am65-cpts.c @@ -83,6 +83,8 @@ struct am65_cpts_regs { #define AM65_CPTS_CONTROL_HW8_TS_PUSH_EN BIT(15) #define AM65_CPTS_CONTROL_HW1_TS_PUSH_OFFSET (8) +#define AM65_CPTS_CONTROL_TX_GENF_CLR_EN BIT(17) + #define AM65_CPTS_CONTROL_TS_SYNC_SEL_MASK (0xF) #define AM65_CPTS_CONTROL_TS_SYNC_SEL_SHIFT (28) @@ -748,42 +750,23 @@ EXPORT_SYMBOL_GPL(am65_cpts_rx_enable); static int am65_skb_get_mtype_seqid(struct sk_buff *skb, u32 *mtype_seqid) { unsigned int ptp_class = ptp_classify_raw(skb); - u8 *msgtype, *data = skb->data; - unsigned int offset = 0; - __be16 *seqid; + struct ptp_header *hdr; + u8 msgtype; + u16 seqid; if (ptp_class == PTP_CLASS_NONE) return 0; - if (ptp_class & PTP_CLASS_VLAN) - offset += VLAN_HLEN; - - switch (ptp_class & PTP_CLASS_PMASK) { - case PTP_CLASS_IPV4: - offset += ETH_HLEN + IPV4_HLEN(data + offset) + UDP_HLEN; - break; - case PTP_CLASS_IPV6: - offset += ETH_HLEN + IP6_HLEN + UDP_HLEN; - break; - case PTP_CLASS_L2: - offset += ETH_HLEN; - break; - default: - return 0; - } - - if (skb->len + ETH_HLEN < offset + OFF_PTP_SEQUENCE_ID + sizeof(*seqid)) + hdr = ptp_parse_header(skb, ptp_class); + if (!hdr) return 0; - if (unlikely(ptp_class & PTP_CLASS_V1)) - msgtype = data + offset + OFF_PTP_CONTROL; - else - msgtype = data + offset; + msgtype = ptp_get_msgtype(hdr, ptp_class); + seqid = ntohs(hdr->sequence_id); - seqid = (__be16 *)(data + offset + OFF_PTP_SEQUENCE_ID); - *mtype_seqid = (*msgtype << AM65_CPTS_EVENT_1_MESSAGE_TYPE_SHIFT) & + *mtype_seqid = (msgtype << AM65_CPTS_EVENT_1_MESSAGE_TYPE_SHIFT) & AM65_CPTS_EVENT_1_MESSAGE_TYPE_MASK; - *mtype_seqid |= (ntohs(*seqid) & AM65_CPTS_EVENT_1_SEQUENCE_ID_MASK); + *mtype_seqid |= (seqid & AM65_CPTS_EVENT_1_SEQUENCE_ID_MASK); return 1; } @@ -1005,7 +988,9 @@ struct am65_cpts *am65_cpts_create(struct device *dev, void __iomem *regs, am65_cpts_set_add_val(cpts); - am65_cpts_write32(cpts, AM65_CPTS_CONTROL_EN | AM65_CPTS_CONTROL_64MODE, + am65_cpts_write32(cpts, AM65_CPTS_CONTROL_EN | + AM65_CPTS_CONTROL_64MODE | + AM65_CPTS_CONTROL_TX_GENF_CLR_EN, control); am65_cpts_write32(cpts, AM65_CPTS_INT_ENABLE_TS_PEND_EN, int_enable); diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index 4a65edc5a375..9fd1f77190ad 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -1278,12 +1278,6 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data, } data->channels = prop; - if (of_property_read_u32(node, "ale_entries", &prop)) { - dev_err(&pdev->dev, "Missing ale_entries property in the DT.\n"); - return -EINVAL; - } - data->ale_entries = prop; - if (of_property_read_u32(node, "bd_ram_size", &prop)) { dev_err(&pdev->dev, "Missing bd_ram_size property in the DT.\n"); return -EINVAL; @@ -1297,7 +1291,7 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data, data->mac_control = prop; if (of_property_read_bool(node, "dual_emac")) - data->dual_emac = 1; + data->dual_emac = true; /* * Populate all the child nodes here... @@ -1596,7 +1590,7 @@ static int cpsw_probe(struct platform_device *pdev) soc = soc_device_match(cpsw_soc_devices); if (soc) - cpsw->quirk_irq = 1; + cpsw->quirk_irq = true; data = &cpsw->data; cpsw->slaves = devm_kcalloc(dev, diff --git a/drivers/net/ethernet/ti/cpsw_ale.c b/drivers/net/ethernet/ti/cpsw_ale.c index 9ad872bfae3a..a6a455c32628 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.c +++ b/drivers/net/ethernet/ti/cpsw_ale.c @@ -32,6 +32,7 @@ #define ALE_STATUS 0x04 #define ALE_CONTROL 0x08 #define ALE_PRESCALE 0x10 +#define ALE_AGING_TIMER 0x14 #define ALE_UNKNOWNVLAN 0x18 #define ALE_TABLE_CONTROL 0x20 #define ALE_TABLE 0x34 @@ -46,6 +47,46 @@ #define AM65_CPSW_ALE_THREAD_DEF_REG 0x134 +/* ALE_AGING_TIMER */ +#define ALE_AGING_TIMER_MASK GENMASK(23, 0) + +/** + * struct ale_entry_fld - The ALE tbl entry field description + * @start_bit: field start bit + * @num_bits: field bit length + * @flags: field flags + */ +struct ale_entry_fld { + u8 start_bit; + u8 num_bits; + u8 flags; +}; + +enum { + CPSW_ALE_F_STATUS_REG = BIT(0), /* Status register present */ + CPSW_ALE_F_HW_AUTOAGING = BIT(1), /* HW auto aging */ + + CPSW_ALE_F_COUNT +}; + +/** + * struct ale_dev_id - The ALE version/SoC specific configuration + * @dev_id: ALE version/SoC id + * @features: features supported by ALE + * @tbl_entries: number of ALE entries + * @major_ver_mask: mask of ALE Major Version Value in ALE_IDVER reg. + * @nu_switch_ale: NU Switch ALE + * @vlan_entry_tbl: ALE vlan entry fields description tbl + */ +struct cpsw_ale_dev_id { + const char *dev_id; + u32 features; + u32 tbl_entries; + u32 major_ver_mask; + bool nu_switch_ale; + const struct ale_entry_fld *vlan_entry_tbl; +}; + #define ALE_TABLE_WRITE BIT(31) #define ALE_TYPE_FREE 0 @@ -60,7 +101,6 @@ #define ALE_TABLE_SIZE_MULTIPLIER 1024 #define ALE_STATUS_SIZE_MASK 0x1f -#define ALE_TABLE_SIZE_DEFAULT 64 static inline int cpsw_ale_get_field(u32 *ale_entry, u32 start, u32 bits) { @@ -106,6 +146,59 @@ static inline void cpsw_ale_set_##name(u32 *ale_entry, u32 value, \ cpsw_ale_set_field(ale_entry, start, bits, value); \ } +enum { + ALE_ENT_VID_MEMBER_LIST = 0, + ALE_ENT_VID_UNREG_MCAST_MSK, + ALE_ENT_VID_REG_MCAST_MSK, + ALE_ENT_VID_FORCE_UNTAGGED_MSK, + ALE_ENT_VID_UNREG_MCAST_IDX, + ALE_ENT_VID_REG_MCAST_IDX, + ALE_ENT_VID_LAST, +}; + +#define ALE_FLD_ALLOWED BIT(0) +#define ALE_FLD_SIZE_PORT_MASK_BITS BIT(1) +#define ALE_FLD_SIZE_PORT_NUM_BITS BIT(2) + +#define ALE_ENTRY_FLD(id, start, bits) \ +[id] = { \ + .start_bit = start, \ + .num_bits = bits, \ + .flags = ALE_FLD_ALLOWED, \ +} + +#define ALE_ENTRY_FLD_DYN_MSK_SIZE(id, start) \ +[id] = { \ + .start_bit = start, \ + .num_bits = 0, \ + .flags = ALE_FLD_ALLOWED | \ + ALE_FLD_SIZE_PORT_MASK_BITS, \ +} + +/* dm814x, am3/am4/am5, k2hk */ +static const struct ale_entry_fld vlan_entry_cpsw[ALE_ENT_VID_LAST] = { + ALE_ENTRY_FLD(ALE_ENT_VID_MEMBER_LIST, 0, 3), + ALE_ENTRY_FLD(ALE_ENT_VID_UNREG_MCAST_MSK, 8, 3), + ALE_ENTRY_FLD(ALE_ENT_VID_REG_MCAST_MSK, 16, 3), + ALE_ENTRY_FLD(ALE_ENT_VID_FORCE_UNTAGGED_MSK, 24, 3), +}; + +/* k2e/k2l, k3 am65/j721e cpsw2g */ +static const struct ale_entry_fld vlan_entry_nu[ALE_ENT_VID_LAST] = { + ALE_ENTRY_FLD_DYN_MSK_SIZE(ALE_ENT_VID_MEMBER_LIST, 0), + ALE_ENTRY_FLD(ALE_ENT_VID_UNREG_MCAST_IDX, 20, 3), + ALE_ENTRY_FLD_DYN_MSK_SIZE(ALE_ENT_VID_FORCE_UNTAGGED_MSK, 24), + ALE_ENTRY_FLD(ALE_ENT_VID_REG_MCAST_IDX, 44, 3), +}; + +/* K3 j721e/j7200 cpsw9g/5g, am64x cpsw3g */ +static const struct ale_entry_fld vlan_entry_k3_cpswxg[] = { + ALE_ENTRY_FLD_DYN_MSK_SIZE(ALE_ENT_VID_MEMBER_LIST, 0), + ALE_ENTRY_FLD_DYN_MSK_SIZE(ALE_ENT_VID_UNREG_MCAST_MSK, 12), + ALE_ENTRY_FLD_DYN_MSK_SIZE(ALE_ENT_VID_FORCE_UNTAGGED_MSK, 24), + ALE_ENTRY_FLD_DYN_MSK_SIZE(ALE_ENT_VID_REG_MCAST_MSK, 36), +}; + DEFINE_ALE_FIELD(entry_type, 60, 2) DEFINE_ALE_FIELD(vlan_id, 48, 12) DEFINE_ALE_FIELD(mcast_state, 62, 2) @@ -115,17 +208,76 @@ DEFINE_ALE_FIELD(ucast_type, 62, 2) DEFINE_ALE_FIELD1(port_num, 66) DEFINE_ALE_FIELD(blocked, 65, 1) DEFINE_ALE_FIELD(secure, 64, 1) -DEFINE_ALE_FIELD1(vlan_untag_force, 24) -DEFINE_ALE_FIELD1(vlan_reg_mcast, 16) -DEFINE_ALE_FIELD1(vlan_unreg_mcast, 8) -DEFINE_ALE_FIELD1(vlan_member_list, 0) DEFINE_ALE_FIELD(mcast, 40, 1) -/* ALE NetCP nu switch specific */ -DEFINE_ALE_FIELD(vlan_unreg_mcast_idx, 20, 3) -DEFINE_ALE_FIELD(vlan_reg_mcast_idx, 44, 3) #define NU_VLAN_UNREG_MCAST_IDX 1 +static int cpsw_ale_entry_get_fld(struct cpsw_ale *ale, + u32 *ale_entry, + const struct ale_entry_fld *entry_tbl, + int fld_id) +{ + const struct ale_entry_fld *entry_fld; + u32 bits; + + if (!ale || !ale_entry) + return -EINVAL; + + entry_fld = &entry_tbl[fld_id]; + if (!(entry_fld->flags & ALE_FLD_ALLOWED)) { + dev_err(ale->params.dev, "get: wrong ale fld id %d\n", fld_id); + return -ENOENT; + } + + bits = entry_fld->num_bits; + if (entry_fld->flags & ALE_FLD_SIZE_PORT_MASK_BITS) + bits = ale->port_mask_bits; + + return cpsw_ale_get_field(ale_entry, entry_fld->start_bit, bits); +} + +static void cpsw_ale_entry_set_fld(struct cpsw_ale *ale, + u32 *ale_entry, + const struct ale_entry_fld *entry_tbl, + int fld_id, + u32 value) +{ + const struct ale_entry_fld *entry_fld; + u32 bits; + + if (!ale || !ale_entry) + return; + + entry_fld = &entry_tbl[fld_id]; + if (!(entry_fld->flags & ALE_FLD_ALLOWED)) { + dev_err(ale->params.dev, "set: wrong ale fld id %d\n", fld_id); + return; + } + + bits = entry_fld->num_bits; + if (entry_fld->flags & ALE_FLD_SIZE_PORT_MASK_BITS) + bits = ale->port_mask_bits; + + cpsw_ale_set_field(ale_entry, entry_fld->start_bit, bits, value); +} + +static int cpsw_ale_vlan_get_fld(struct cpsw_ale *ale, + u32 *ale_entry, + int fld_id) +{ + return cpsw_ale_entry_get_fld(ale, ale_entry, + ale->vlan_entry_tbl, fld_id); +} + +static void cpsw_ale_vlan_set_fld(struct cpsw_ale *ale, + u32 *ale_entry, + int fld_id, + u32 value) +{ + cpsw_ale_entry_set_fld(ale, ale_entry, + ale->vlan_entry_tbl, fld_id, value); +} + /* The MAC address field in the ALE entry cannot be macroized as above */ static inline void cpsw_ale_get_addr(u32 *ale_entry, u8 *addr) { @@ -420,19 +572,22 @@ static void cpsw_ale_set_vlan_mcast(struct cpsw_ale *ale, u32 *ale_entry, int idx; /* Set VLAN registered multicast flood mask */ - idx = cpsw_ale_get_vlan_reg_mcast_idx(ale_entry); + idx = cpsw_ale_vlan_get_fld(ale, ale_entry, + ALE_ENT_VID_REG_MCAST_IDX); writel(reg_mcast, ale->params.ale_regs + ALE_VLAN_MASK_MUX(idx)); /* Set VLAN unregistered multicast flood mask */ - idx = cpsw_ale_get_vlan_unreg_mcast_idx(ale_entry); + idx = cpsw_ale_vlan_get_fld(ale, ale_entry, + ALE_ENT_VID_UNREG_MCAST_IDX); writel(unreg_mcast, ale->params.ale_regs + ALE_VLAN_MASK_MUX(idx)); } static void cpsw_ale_set_vlan_untag(struct cpsw_ale *ale, u32 *ale_entry, u16 vid, int untag_mask) { - cpsw_ale_set_vlan_untag_force(ale_entry, - untag_mask, ale->vlan_field_bits); + cpsw_ale_vlan_set_fld(ale, ale_entry, + ALE_ENT_VID_FORCE_UNTAGGED_MSK, + untag_mask); if (untag_mask & ALE_PORT_HOST) bitmap_set(ale->p0_untag_vid_mask, vid, 1); else @@ -454,17 +609,19 @@ int cpsw_ale_add_vlan(struct cpsw_ale *ale, u16 vid, int port_mask, int untag, cpsw_ale_set_vlan_untag(ale, ale_entry, vid, untag); if (!ale->params.nu_switch_ale) { - cpsw_ale_set_vlan_reg_mcast(ale_entry, reg_mcast, - ale->vlan_field_bits); - cpsw_ale_set_vlan_unreg_mcast(ale_entry, unreg_mcast, - ale->vlan_field_bits); + cpsw_ale_vlan_set_fld(ale, ale_entry, + ALE_ENT_VID_REG_MCAST_MSK, reg_mcast); + cpsw_ale_vlan_set_fld(ale, ale_entry, + ALE_ENT_VID_UNREG_MCAST_MSK, unreg_mcast); } else { - cpsw_ale_set_vlan_unreg_mcast_idx(ale_entry, - NU_VLAN_UNREG_MCAST_IDX); + cpsw_ale_vlan_set_fld(ale, ale_entry, + ALE_ENT_VID_UNREG_MCAST_IDX, + NU_VLAN_UNREG_MCAST_IDX); cpsw_ale_set_vlan_mcast(ale, ale_entry, reg_mcast, unreg_mcast); } - cpsw_ale_set_vlan_member_list(ale_entry, port_mask, - ale->vlan_field_bits); + + cpsw_ale_vlan_set_fld(ale, ale_entry, + ALE_ENT_VID_MEMBER_LIST, port_mask); if (idx < 0) idx = cpsw_ale_match_free(ale); @@ -483,20 +640,20 @@ static void cpsw_ale_del_vlan_modify(struct cpsw_ale *ale, u32 *ale_entry, int reg_mcast, unreg_mcast; int members, untag; - members = cpsw_ale_get_vlan_member_list(ale_entry, - ale->vlan_field_bits); + members = cpsw_ale_vlan_get_fld(ale, ale_entry, + ALE_ENT_VID_MEMBER_LIST); members &= ~port_mask; if (!members) { cpsw_ale_set_entry_type(ale_entry, ALE_TYPE_FREE); return; } - untag = cpsw_ale_get_vlan_untag_force(ale_entry, - ale->vlan_field_bits); - reg_mcast = cpsw_ale_get_vlan_reg_mcast(ale_entry, - ale->vlan_field_bits); - unreg_mcast = cpsw_ale_get_vlan_unreg_mcast(ale_entry, - ale->vlan_field_bits); + untag = cpsw_ale_vlan_get_fld(ale, ale_entry, + ALE_ENT_VID_FORCE_UNTAGGED_MSK); + reg_mcast = cpsw_ale_vlan_get_fld(ale, ale_entry, + ALE_ENT_VID_REG_MCAST_MSK); + unreg_mcast = cpsw_ale_vlan_get_fld(ale, ale_entry, + ALE_ENT_VID_UNREG_MCAST_MSK); untag &= members; reg_mcast &= members; unreg_mcast &= members; @@ -504,16 +661,16 @@ static void cpsw_ale_del_vlan_modify(struct cpsw_ale *ale, u32 *ale_entry, cpsw_ale_set_vlan_untag(ale, ale_entry, vid, untag); if (!ale->params.nu_switch_ale) { - cpsw_ale_set_vlan_reg_mcast(ale_entry, reg_mcast, - ale->vlan_field_bits); - cpsw_ale_set_vlan_unreg_mcast(ale_entry, unreg_mcast, - ale->vlan_field_bits); + cpsw_ale_vlan_set_fld(ale, ale_entry, + ALE_ENT_VID_REG_MCAST_MSK, reg_mcast); + cpsw_ale_vlan_set_fld(ale, ale_entry, + ALE_ENT_VID_UNREG_MCAST_MSK, unreg_mcast); } else { cpsw_ale_set_vlan_mcast(ale, ale_entry, reg_mcast, unreg_mcast); } - cpsw_ale_set_vlan_member_list(ale_entry, members, - ale->vlan_field_bits); + cpsw_ale_vlan_set_fld(ale, ale_entry, + ALE_ENT_VID_MEMBER_LIST, members); } int cpsw_ale_del_vlan(struct cpsw_ale *ale, u16 vid, int port_mask) @@ -551,15 +708,15 @@ int cpsw_ale_vlan_add_modify(struct cpsw_ale *ale, u16 vid, int port_mask, if (idx >= 0) cpsw_ale_read(ale, idx, ale_entry); - vlan_members = cpsw_ale_get_vlan_member_list(ale_entry, - ale->vlan_field_bits); - reg_mcast_members = cpsw_ale_get_vlan_reg_mcast(ale_entry, - ale->vlan_field_bits); + vlan_members = cpsw_ale_vlan_get_fld(ale, ale_entry, + ALE_ENT_VID_MEMBER_LIST); + reg_mcast_members = cpsw_ale_vlan_get_fld(ale, ale_entry, + ALE_ENT_VID_REG_MCAST_MSK); unreg_mcast_members = - cpsw_ale_get_vlan_unreg_mcast(ale_entry, - ale->vlan_field_bits); - untag_members = cpsw_ale_get_vlan_untag_force(ale_entry, - ale->vlan_field_bits); + cpsw_ale_vlan_get_fld(ale, ale_entry, + ALE_ENT_VID_UNREG_MCAST_MSK); + untag_members = cpsw_ale_vlan_get_fld(ale, ale_entry, + ALE_ENT_VID_FORCE_UNTAGGED_MSK); vlan_members |= port_mask; untag_members = (untag_members & ~port_mask) | untag_mask; @@ -592,14 +749,15 @@ void cpsw_ale_set_unreg_mcast(struct cpsw_ale *ale, int unreg_mcast_mask, continue; unreg_members = - cpsw_ale_get_vlan_unreg_mcast(ale_entry, - ale->vlan_field_bits); + cpsw_ale_vlan_get_fld(ale, ale_entry, + ALE_ENT_VID_UNREG_MCAST_MSK); if (add) unreg_members |= unreg_mcast_mask; else unreg_members &= ~unreg_mcast_mask; - cpsw_ale_set_vlan_unreg_mcast(ale_entry, unreg_members, - ale->vlan_field_bits); + cpsw_ale_vlan_set_fld(ale, ale_entry, + ALE_ENT_VID_UNREG_MCAST_MSK, + unreg_members); cpsw_ale_write(ale, idx, ale_entry); } } @@ -609,15 +767,15 @@ static void cpsw_ale_vlan_set_unreg_mcast(struct cpsw_ale *ale, u32 *ale_entry, { int unreg_mcast; - unreg_mcast = - cpsw_ale_get_vlan_unreg_mcast(ale_entry, - ale->vlan_field_bits); + unreg_mcast = cpsw_ale_vlan_get_fld(ale, ale_entry, + ALE_ENT_VID_UNREG_MCAST_MSK); if (allmulti) unreg_mcast |= ALE_PORT_HOST; else unreg_mcast &= ~ALE_PORT_HOST; - cpsw_ale_set_vlan_unreg_mcast(ale_entry, unreg_mcast, - ale->vlan_field_bits); + + cpsw_ale_vlan_set_fld(ale, ale_entry, + ALE_ENT_VID_UNREG_MCAST_MSK, unreg_mcast); } static void @@ -627,7 +785,8 @@ cpsw_ale_vlan_set_unreg_mcast_idx(struct cpsw_ale *ale, u32 *ale_entry, int unreg_mcast; int idx; - idx = cpsw_ale_get_vlan_unreg_mcast_idx(ale_entry); + idx = cpsw_ale_vlan_get_fld(ale, ale_entry, + ALE_ENT_VID_UNREG_MCAST_IDX); unreg_mcast = readl(ale->params.ale_regs + ALE_VLAN_MASK_MUX(idx)); @@ -651,9 +810,9 @@ void cpsw_ale_set_allmulti(struct cpsw_ale *ale, int allmulti, int port) type = cpsw_ale_get_entry_type(ale_entry); if (type != ALE_TYPE_VLAN) continue; - vlan_members = - cpsw_ale_get_vlan_member_list(ale_entry, - ale->vlan_field_bits); + + vlan_members = cpsw_ale_vlan_get_fld(ale, ale_entry, + ALE_ENT_VID_MEMBER_LIST); if (port != -1 && !(vlan_members & BIT(port))) continue; @@ -960,30 +1119,146 @@ static void cpsw_ale_timer(struct timer_list *t) } } +static void cpsw_ale_hw_aging_timer_start(struct cpsw_ale *ale) +{ + u32 aging_timer; + + aging_timer = ale->params.bus_freq / 1000000; + aging_timer *= ale->params.ale_ageout; + + if (aging_timer & ~ALE_AGING_TIMER_MASK) { + aging_timer = ALE_AGING_TIMER_MASK; + dev_warn(ale->params.dev, + "ALE aging timer overflow, set to max\n"); + } + + writel(aging_timer, ale->params.ale_regs + ALE_AGING_TIMER); +} + +static void cpsw_ale_hw_aging_timer_stop(struct cpsw_ale *ale) +{ + writel(0, ale->params.ale_regs + ALE_AGING_TIMER); +} + +static void cpsw_ale_aging_start(struct cpsw_ale *ale) +{ + if (!ale->params.ale_ageout) + return; + + if (ale->features & CPSW_ALE_F_HW_AUTOAGING) { + cpsw_ale_hw_aging_timer_start(ale); + return; + } + + timer_setup(&ale->timer, cpsw_ale_timer, 0); + ale->timer.expires = jiffies + ale->ageout; + add_timer(&ale->timer); +} + +static void cpsw_ale_aging_stop(struct cpsw_ale *ale) +{ + if (!ale->params.ale_ageout) + return; + + if (ale->features & CPSW_ALE_F_HW_AUTOAGING) { + cpsw_ale_hw_aging_timer_stop(ale); + return; + } + + del_timer_sync(&ale->timer); +} + void cpsw_ale_start(struct cpsw_ale *ale) { cpsw_ale_control_set(ale, 0, ALE_ENABLE, 1); cpsw_ale_control_set(ale, 0, ALE_CLEAR, 1); - timer_setup(&ale->timer, cpsw_ale_timer, 0); - if (ale->ageout) { - ale->timer.expires = jiffies + ale->ageout; - add_timer(&ale->timer); - } + cpsw_ale_aging_start(ale); } void cpsw_ale_stop(struct cpsw_ale *ale) { - del_timer_sync(&ale->timer); + cpsw_ale_aging_stop(ale); cpsw_ale_control_set(ale, 0, ALE_CLEAR, 1); cpsw_ale_control_set(ale, 0, ALE_ENABLE, 0); } +static const struct cpsw_ale_dev_id cpsw_ale_id_match[] = { + { + /* am3/4/5, dra7. dm814x, 66ak2hk-gbe */ + .dev_id = "cpsw", + .tbl_entries = 1024, + .major_ver_mask = 0xff, + .vlan_entry_tbl = vlan_entry_cpsw, + }, + { + /* 66ak2h_xgbe */ + .dev_id = "66ak2h-xgbe", + .tbl_entries = 2048, + .major_ver_mask = 0xff, + .vlan_entry_tbl = vlan_entry_cpsw, + }, + { + .dev_id = "66ak2el", + .features = CPSW_ALE_F_STATUS_REG, + .major_ver_mask = 0x7, + .nu_switch_ale = true, + .vlan_entry_tbl = vlan_entry_nu, + }, + { + .dev_id = "66ak2g", + .features = CPSW_ALE_F_STATUS_REG, + .tbl_entries = 64, + .major_ver_mask = 0x7, + .nu_switch_ale = true, + .vlan_entry_tbl = vlan_entry_nu, + }, + { + .dev_id = "am65x-cpsw2g", + .features = CPSW_ALE_F_STATUS_REG | CPSW_ALE_F_HW_AUTOAGING, + .tbl_entries = 64, + .major_ver_mask = 0x7, + .nu_switch_ale = true, + .vlan_entry_tbl = vlan_entry_nu, + }, + { + .dev_id = "j721e-cpswxg", + .features = CPSW_ALE_F_STATUS_REG | CPSW_ALE_F_HW_AUTOAGING, + .major_ver_mask = 0x7, + .vlan_entry_tbl = vlan_entry_k3_cpswxg, + }, + { }, +}; + +static const struct +cpsw_ale_dev_id *cpsw_ale_match_id(const struct cpsw_ale_dev_id *id, + const char *dev_id) +{ + if (!dev_id) + return NULL; + + while (id->dev_id) { + if (strcmp(dev_id, id->dev_id) == 0) + return id; + id++; + } + return NULL; +} + struct cpsw_ale *cpsw_ale_create(struct cpsw_ale_params *params) { + const struct cpsw_ale_dev_id *ale_dev_id; struct cpsw_ale *ale; u32 rev, ale_entries; + ale_dev_id = cpsw_ale_match_id(cpsw_ale_id_match, params->dev_id); + if (!ale_dev_id) + return ERR_PTR(-EINVAL); + + params->ale_entries = ale_dev_id->tbl_entries; + params->major_ver_mask = ale_dev_id->major_ver_mask; + params->nu_switch_ale = ale_dev_id->nu_switch_ale; + ale = devm_kzalloc(params->dev, sizeof(*ale), GFP_KERNEL); if (!ale) return ERR_PTR(-ENOMEM); @@ -997,10 +1272,10 @@ struct cpsw_ale *cpsw_ale_create(struct cpsw_ale_params *params) ale->params = *params; ale->ageout = ale->params.ale_ageout * HZ; + ale->features = ale_dev_id->features; + ale->vlan_entry_tbl = ale_dev_id->vlan_entry_tbl; rev = readl_relaxed(ale->params.ale_regs + ALE_IDVER); - if (!ale->params.major_ver_mask) - ale->params.major_ver_mask = 0xff; ale->version = (ALE_VERSION_MAJOR(rev, ale->params.major_ver_mask) << 8) | ALE_VERSION_MINOR(rev); @@ -1008,7 +1283,8 @@ struct cpsw_ale *cpsw_ale_create(struct cpsw_ale_params *params) ALE_VERSION_MAJOR(rev, ale->params.major_ver_mask), ALE_VERSION_MINOR(rev)); - if (!ale->params.ale_entries) { + if (ale->features & CPSW_ALE_F_STATUS_REG && + !ale->params.ale_entries) { ale_entries = readl_relaxed(ale->params.ale_regs + ALE_STATUS) & ALE_STATUS_SIZE_MASK; @@ -1017,16 +1293,12 @@ struct cpsw_ale *cpsw_ale_create(struct cpsw_ale_params *params) * table which shows the size as a multiple of 1024 entries. * For these, params.ale_entries will be set to zero. So * read the register and update the value of ale_entries. - * ALE table on NetCP lite, is much smaller and is indicated - * by a value of zero in ALE_STATUS. So use a default value - * of ALE_TABLE_SIZE_DEFAULT for this. Caller is expected - * to set the value of ale_entries for all other versions - * of ALE. + * return error if ale_entries is zero in ALE_STATUS. */ if (!ale_entries) - ale_entries = ALE_TABLE_SIZE_DEFAULT; - else - ale_entries *= ALE_TABLE_SIZE_MULTIPLIER; + return ERR_PTR(-EINVAL); + + ale_entries *= ALE_TABLE_SIZE_MULTIPLIER; ale->params.ale_entries = ale_entries; } dev_info(ale->params.dev, @@ -1079,3 +1351,8 @@ void cpsw_ale_dump(struct cpsw_ale *ale, u32 *data) data += ALE_ENTRY_WORDS; } } + +u32 cpsw_ale_get_num_entries(struct cpsw_ale *ale) +{ + return ale ? ale->params.ale_entries : 0; +} diff --git a/drivers/net/ethernet/ti/cpsw_ale.h b/drivers/net/ethernet/ti/cpsw_ale.h index 6a3cb6898728..5e4a69662c5f 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.h +++ b/drivers/net/ethernet/ti/cpsw_ale.h @@ -24,18 +24,24 @@ struct cpsw_ale_params { * pass it from caller. */ u32 major_ver_mask; + const char *dev_id; + unsigned long bus_freq; }; +struct ale_entry_fld; + struct cpsw_ale { struct cpsw_ale_params params; struct timer_list timer; unsigned long ageout; u32 version; + u32 features; /* These bits are different on NetCP NU Switch ALE */ u32 port_mask_bits; u32 port_num_bits; u32 vlan_field_bits; unsigned long *p0_untag_vid_mask; + const struct ale_entry_fld *vlan_entry_tbl; }; enum cpsw_ale_control { @@ -119,6 +125,7 @@ int cpsw_ale_control_get(struct cpsw_ale *ale, int port, int control); int cpsw_ale_control_set(struct cpsw_ale *ale, int port, int control, int value); void cpsw_ale_dump(struct cpsw_ale *ale, u32 *data); +u32 cpsw_ale_get_num_entries(struct cpsw_ale *ale); static inline int cpsw_ale_get_vlan_p0_untag(struct cpsw_ale *ale, u16 vid) { diff --git a/drivers/net/ethernet/ti/cpsw_ethtool.c b/drivers/net/ethernet/ti/cpsw_ethtool.c index fa54efe3be63..4d02c5135611 100644 --- a/drivers/net/ethernet/ti/cpsw_ethtool.c +++ b/drivers/net/ethernet/ti/cpsw_ethtool.c @@ -339,7 +339,8 @@ int cpsw_get_regs_len(struct net_device *ndev) { struct cpsw_common *cpsw = ndev_to_cpsw(ndev); - return cpsw->data.ale_entries * ALE_ENTRY_WORDS * sizeof(u32); + return cpsw_ale_get_num_entries(cpsw->ale) * + ALE_ENTRY_WORDS * sizeof(u32); } void cpsw_get_regs(struct net_device *ndev, struct ethtool_regs *regs, void *p) diff --git a/drivers/net/ethernet/ti/cpsw_new.c b/drivers/net/ethernet/ti/cpsw_new.c index 15672d0a4de6..f779d2e1b5c5 100644 --- a/drivers/net/ethernet/ti/cpsw_new.c +++ b/drivers/net/ethernet/ti/cpsw_new.c @@ -1244,7 +1244,6 @@ static int cpsw_probe_dt(struct cpsw_common *cpsw) data->active_slave = 0; data->channels = CPSW_MAX_QUEUES; - data->ale_entries = CPSW_ALE_NUM_ENTRIES; data->dual_emac = true; data->bd_ram_size = CPSW_BD_RAM_SIZE; data->mac_control = 0; @@ -1661,12 +1660,10 @@ static int cpsw_dl_switch_mode_set(struct devlink *dl, u32 id, for (i = 0; i < cpsw->data.slaves; i++) { struct cpsw_slave *slave = &cpsw->slaves[i]; struct net_device *sl_ndev = slave->ndev; - struct cpsw_priv *priv; if (!sl_ndev) continue; - priv = netdev_priv(sl_ndev); if (switch_en) vlan = cpsw->data.default_vlan; else diff --git a/drivers/net/ethernet/ti/cpsw_priv.c b/drivers/net/ethernet/ti/cpsw_priv.c index 482a1a451e43..51cc29f39038 100644 --- a/drivers/net/ethernet/ti/cpsw_priv.c +++ b/drivers/net/ethernet/ti/cpsw_priv.c @@ -500,8 +500,8 @@ int cpsw_init_common(struct cpsw_common *cpsw, void __iomem *ss_regs, ale_params.dev = dev; ale_params.ale_ageout = ale_ageout; - ale_params.ale_entries = data->ale_entries; ale_params.ale_ports = CPSW_ALE_PORTS_NUM; + ale_params.dev_id = "cpsw"; cpsw->ale = cpsw_ale_create(&ale_params); if (IS_ERR(cpsw->ale)) { diff --git a/drivers/net/ethernet/ti/cpsw_priv.h b/drivers/net/ethernet/ti/cpsw_priv.h index bf4e179b4ca4..7b7f3596b20d 100644 --- a/drivers/net/ethernet/ti/cpsw_priv.h +++ b/drivers/net/ethernet/ti/cpsw_priv.h @@ -117,7 +117,6 @@ do { \ #define CPSW_MAX_QUEUES 8 #define CPSW_CPDMA_DESCS_POOL_SIZE_DEFAULT 256 #define CPSW_ALE_AGEOUT_DEFAULT 10 /* sec */ -#define CPSW_ALE_NUM_ENTRIES 1024 #define CPSW_FIFO_QUEUE_TYPE_SHIFT 16 #define CPSW_FIFO_SHAPE_EN_SHIFT 16 #define CPSW_FIFO_RATE_EN_SHIFT 20 @@ -294,7 +293,6 @@ struct cpsw_platform_data { u32 channels; /* number of cpdma channels (symmetric) */ u32 slaves; /* number of slave cpgmac ports */ u32 active_slave;/* time stamping, ethtool and SIOCGMIIPHY slave */ - u32 ale_entries; /* ale table size */ u32 bd_ram_size; /*buffer descriptor ram size */ u32 mac_control; /* Mac control register */ u16 default_vlan; /* Def VLAN for ALE lookup in VLAN aware mode*/ diff --git a/drivers/net/ethernet/ti/cpts.c b/drivers/net/ethernet/ti/cpts.c index 7c55d395de2c..d1fc7955d422 100644 --- a/drivers/net/ethernet/ti/cpts.c +++ b/drivers/net/ethernet/ti/cpts.c @@ -446,41 +446,22 @@ static const struct ptp_clock_info cpts_info = { static int cpts_skb_get_mtype_seqid(struct sk_buff *skb, u32 *mtype_seqid) { unsigned int ptp_class = ptp_classify_raw(skb); - u8 *msgtype, *data = skb->data; - unsigned int offset = 0; - u16 *seqid; + struct ptp_header *hdr; + u8 msgtype; + u16 seqid; if (ptp_class == PTP_CLASS_NONE) return 0; - if (ptp_class & PTP_CLASS_VLAN) - offset += VLAN_HLEN; - - switch (ptp_class & PTP_CLASS_PMASK) { - case PTP_CLASS_IPV4: - offset += ETH_HLEN + IPV4_HLEN(data + offset) + UDP_HLEN; - break; - case PTP_CLASS_IPV6: - offset += ETH_HLEN + IP6_HLEN + UDP_HLEN; - break; - case PTP_CLASS_L2: - offset += ETH_HLEN; - break; - default: - return 0; - } - - if (skb->len + ETH_HLEN < offset + OFF_PTP_SEQUENCE_ID + sizeof(*seqid)) + hdr = ptp_parse_header(skb, ptp_class); + if (!hdr) return 0; - if (unlikely(ptp_class & PTP_CLASS_V1)) - msgtype = data + offset + OFF_PTP_CONTROL; - else - msgtype = data + offset; + msgtype = ptp_get_msgtype(hdr, ptp_class); + seqid = ntohs(hdr->sequence_id); - seqid = (u16 *)(data + offset + OFF_PTP_SEQUENCE_ID); - *mtype_seqid = (*msgtype & MESSAGE_TYPE_MASK) << MESSAGE_TYPE_SHIFT; - *mtype_seqid |= (ntohs(*seqid) & SEQUENCE_ID_MASK) << SEQUENCE_ID_SHIFT; + *mtype_seqid = (msgtype & MESSAGE_TYPE_MASK) << MESSAGE_TYPE_SHIFT; + *mtype_seqid |= (seqid & SEQUENCE_ID_MASK) << SEQUENCE_ID_SHIFT; return 1; } @@ -528,6 +509,11 @@ void cpts_rx_timestamp(struct cpts *cpts, struct sk_buff *skb) int ret; u64 ns; + /* cpts_rx_timestamp() is called before eth_type_trans(), so + * skb MAC Hdr properties are not configured yet. Hence need to + * reset skb MAC header here + */ + skb_reset_mac_header(skb); ret = cpts_skb_get_mtype_seqid(skb, &skb_cb->skb_mtype_seqid); if (!ret) return; diff --git a/drivers/net/ethernet/ti/davinci_cpdma.c b/drivers/net/ethernet/ti/davinci_cpdma.c index 6614fa3089b2..d2eab5cd1e0c 100644 --- a/drivers/net/ethernet/ti/davinci_cpdma.c +++ b/drivers/net/ethernet/ti/davinci_cpdma.c @@ -718,7 +718,7 @@ static void cpdma_chan_set_descs(struct cpdma_ctlr *ctlr, most_chan->desc_num += desc_cnt; } -/** +/* * cpdma_chan_split_pool - Splits ctrl pool between all channels. * Has to be called under ctlr lock */ diff --git a/drivers/net/ethernet/ti/davinci_emac.c b/drivers/net/ethernet/ti/davinci_emac.c index de282531f68b..c7031e1960d4 100644 --- a/drivers/net/ethernet/ti/davinci_emac.c +++ b/drivers/net/ethernet/ti/davinci_emac.c @@ -671,7 +671,7 @@ static int emac_hash_del(struct emac_priv *priv, u8 *mac_addr) * emac_add_mcast - Set multicast address in the EMAC adapter (Internal) * @priv: The DaVinci EMAC private adapter structure * @action: multicast operation to perform - * mac_addr: mac address to set + * @mac_addr: mac address to set * * Set multicast addresses in EMAC adapter - internal function * @@ -977,6 +977,7 @@ fail_tx: /** * emac_dev_tx_timeout - EMAC Transmit timeout function * @ndev: The DaVinci EMAC network adapter + * @txqueue: the index of the hung transmit queue * * Called when system detects that a skb timeout period has expired * potentially due to a fault in the adapter in not being able to send @@ -1209,7 +1210,7 @@ static int emac_hw_enable(struct emac_priv *priv) /** * emac_poll - EMAC NAPI Poll function - * @ndev: The DaVinci EMAC network adapter + * @napi: pointer to the napi_struct containing The DaVinci EMAC network adapter * @budget: Number of receive packets to process (as told by NAPI layer) * * NAPI Poll function implemented to process packets as per budget. We check @@ -1227,7 +1228,7 @@ static int emac_poll(struct napi_struct *napi, int budget) struct net_device *ndev = priv->ndev; struct device *emac_dev = &ndev->dev; u32 status = 0; - u32 num_tx_pkts = 0, num_rx_pkts = 0; + u32 num_rx_pkts = 0; /* Check interrupt vectors and call packet processing */ status = emac_read(EMAC_MACINVECTOR); @@ -1238,8 +1239,7 @@ static int emac_poll(struct napi_struct *napi, int budget) mask = EMAC_DM646X_MAC_IN_VECTOR_TX_INT_VEC; if (status & mask) { - num_tx_pkts = cpdma_chan_process(priv->txchan, - EMAC_DEF_TX_MAX_SERVICE); + cpdma_chan_process(priv->txchan, EMAC_DEF_TX_MAX_SERVICE); } /* TX processing */ mask = EMAC_DM644X_MAC_IN_VECTOR_RX_INT_VEC; diff --git a/drivers/net/ethernet/ti/netcp_ethss.c b/drivers/net/ethernet/ti/netcp_ethss.c index 28093923a7fb..33c1592d5381 100644 --- a/drivers/net/ethernet/ti/netcp_ethss.c +++ b/drivers/net/ethernet/ti/netcp_ethss.c @@ -51,7 +51,6 @@ #define GBE13_CPTS_OFFSET 0x500 #define GBE13_ALE_OFFSET 0x600 #define GBE13_HOST_PORT_NUM 0 -#define GBE13_NUM_ALE_ENTRIES 1024 /* 1G Ethernet NU SS defines */ #define GBENU_MODULE_NAME "netcp-gbenu" @@ -101,7 +100,6 @@ #define XGBE10_ALE_OFFSET 0x700 #define XGBE10_HW_STATS_OFFSET 0x800 #define XGBE10_HOST_PORT_NUM 0 -#define XGBE10_NUM_ALE_ENTRIES 2048 #define GBE_TIMER_INTERVAL (HZ / 2) @@ -711,7 +709,6 @@ struct gbe_priv { struct netcp_device *netcp_device; struct timer_list timer; u32 num_slaves; - u32 ale_entries; u32 ale_ports; bool enable_ale; u8 max_num_slaves; @@ -3309,7 +3306,6 @@ static int set_xgbe_ethss10_priv(struct gbe_priv *gbe_dev, gbe_dev->cpts_reg = gbe_dev->switch_regs + XGBE10_CPTS_OFFSET; gbe_dev->ale_ports = gbe_dev->max_num_ports; gbe_dev->host_port = XGBE10_HOST_PORT_NUM; - gbe_dev->ale_entries = XGBE10_NUM_ALE_ENTRIES; gbe_dev->stats_en_mask = (1 << (gbe_dev->max_num_ports)) - 1; /* Subsystem registers */ @@ -3433,7 +3429,6 @@ static int set_gbe_ethss14_priv(struct gbe_priv *gbe_dev, gbe_dev->ale_reg = gbe_dev->switch_regs + GBE13_ALE_OFFSET; gbe_dev->ale_ports = gbe_dev->max_num_ports; gbe_dev->host_port = GBE13_HOST_PORT_NUM; - gbe_dev->ale_entries = GBE13_NUM_ALE_ENTRIES; gbe_dev->stats_en_mask = GBE13_REG_VAL_STAT_ENABLE_ALL; /* Subsystem registers */ @@ -3697,12 +3692,15 @@ static int gbe_probe(struct netcp_device *netcp_device, struct device *dev, ale_params.dev = gbe_dev->dev; ale_params.ale_regs = gbe_dev->ale_reg; ale_params.ale_ageout = GBE_DEFAULT_ALE_AGEOUT; - ale_params.ale_entries = gbe_dev->ale_entries; ale_params.ale_ports = gbe_dev->ale_ports; - if (IS_SS_ID_MU(gbe_dev)) { - ale_params.major_ver_mask = 0x7; - ale_params.nu_switch_ale = true; - } + ale_params.dev_id = "cpsw"; + if (IS_SS_ID_NU(gbe_dev)) + ale_params.dev_id = "66ak2el"; + else if (IS_SS_ID_2U(gbe_dev)) + ale_params.dev_id = "66ak2g"; + else if (IS_SS_ID_XGBE(gbe_dev)) + ale_params.dev_id = "66ak2h-xgbe"; + gbe_dev->ale = cpsw_ale_create(&ale_params); if (IS_ERR(gbe_dev->ale)) { dev_err(gbe_dev->dev, "error initializing ale engine\n"); diff --git a/drivers/net/ethernet/ti/tlan.c b/drivers/net/ethernet/ti/tlan.c index 76a342ea3797..267c080ee084 100644 --- a/drivers/net/ethernet/ti/tlan.c +++ b/drivers/net/ethernet/ti/tlan.c @@ -305,9 +305,8 @@ static void tlan_remove_one(struct pci_dev *pdev) unregister_netdev(dev); if (priv->dma_storage) { - pci_free_consistent(priv->pci_dev, - priv->dma_size, priv->dma_storage, - priv->dma_storage_dma); + dma_free_coherent(&priv->pci_dev->dev, priv->dma_size, + priv->dma_storage, priv->dma_storage_dma); } #ifdef CONFIG_PCI @@ -482,7 +481,7 @@ static int tlan_probe1(struct pci_dev *pdev, long ioaddr, int irq, int rev, priv->adapter = &board_info[ent->driver_data]; - rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(32)); + rc = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)); if (rc) { pr_err("No suitable PCI mapping available\n"); goto err_out_free_dev; @@ -584,8 +583,8 @@ static int tlan_probe1(struct pci_dev *pdev, long ioaddr, int irq, int rev, return 0; err_out_uninit: - pci_free_consistent(priv->pci_dev, priv->dma_size, priv->dma_storage, - priv->dma_storage_dma); + dma_free_coherent(&priv->pci_dev->dev, priv->dma_size, + priv->dma_storage, priv->dma_storage_dma); err_out_free_dev: free_netdev(dev); err_out_regions: @@ -609,9 +608,9 @@ static void tlan_eisa_cleanup(void) dev = tlan_eisa_devices; priv = netdev_priv(dev); if (priv->dma_storage) { - pci_free_consistent(priv->pci_dev, priv->dma_size, - priv->dma_storage, - priv->dma_storage_dma); + dma_free_coherent(&priv->pci_dev->dev, priv->dma_size, + priv->dma_storage, + priv->dma_storage_dma); } release_region(dev->base_addr, 0x10); unregister_netdev(dev); @@ -654,7 +653,6 @@ module_exit(tlan_exit); static void __init tlan_eisa_probe(void) { long ioaddr; - int rc = -ENODEV; int irq; u16 device_id; @@ -719,8 +717,7 @@ static void __init tlan_eisa_probe(void) /* Setup the newly found eisa adapter */ - rc = tlan_probe1(NULL, ioaddr, irq, - 12, NULL); + tlan_probe1(NULL, ioaddr, irq, 12, NULL); continue; out: @@ -826,9 +823,8 @@ static int tlan_init(struct net_device *dev) dma_size = (TLAN_NUM_RX_LISTS + TLAN_NUM_TX_LISTS) * (sizeof(struct tlan_list)); - priv->dma_storage = pci_alloc_consistent(priv->pci_dev, - dma_size, - &priv->dma_storage_dma); + priv->dma_storage = dma_alloc_coherent(&priv->pci_dev->dev, dma_size, + &priv->dma_storage_dma, GFP_KERNEL); priv->dma_size = dma_size; if (priv->dma_storage == NULL) { @@ -1069,9 +1065,9 @@ static netdev_tx_t tlan_start_tx(struct sk_buff *skb, struct net_device *dev) tail_list->forward = 0; - tail_list->buffer[0].address = pci_map_single(priv->pci_dev, + tail_list->buffer[0].address = dma_map_single(&priv->pci_dev->dev, skb->data, txlen, - PCI_DMA_TODEVICE); + DMA_TO_DEVICE); tlan_store_skb(tail_list, skb); tail_list->frame_size = (u16) txlen; @@ -1365,10 +1361,10 @@ static u32 tlan_handle_tx_eof(struct net_device *dev, u16 host_int) struct sk_buff *skb = tlan_get_skb(head_list); ack++; - pci_unmap_single(priv->pci_dev, head_list->buffer[0].address, - max(skb->len, - (unsigned int)TLAN_MIN_FRAME_SIZE), - PCI_DMA_TODEVICE); + dma_unmap_single(&priv->pci_dev->dev, + head_list->buffer[0].address, + max(skb->len, (unsigned int)TLAN_MIN_FRAME_SIZE), + DMA_TO_DEVICE); dev_kfree_skb_any(skb); head_list->buffer[8].address = 0; head_list->buffer[9].address = 0; @@ -1511,8 +1507,8 @@ static u32 tlan_handle_rx_eof(struct net_device *dev, u16 host_int) goto drop_and_reuse; skb = tlan_get_skb(head_list); - pci_unmap_single(priv->pci_dev, frame_dma, - TLAN_MAX_FRAME_SIZE, PCI_DMA_FROMDEVICE); + dma_unmap_single(&priv->pci_dev->dev, frame_dma, + TLAN_MAX_FRAME_SIZE, DMA_FROM_DEVICE); skb_put(skb, frame_size); dev->stats.rx_bytes += frame_size; @@ -1521,8 +1517,8 @@ static u32 tlan_handle_rx_eof(struct net_device *dev, u16 host_int) netif_rx(skb); head_list->buffer[0].address = - pci_map_single(priv->pci_dev, new_skb->data, - TLAN_MAX_FRAME_SIZE, PCI_DMA_FROMDEVICE); + dma_map_single(&priv->pci_dev->dev, new_skb->data, + TLAN_MAX_FRAME_SIZE, DMA_FROM_DEVICE); tlan_store_skb(head_list, new_skb); drop_and_reuse: @@ -1923,10 +1919,10 @@ static void tlan_reset_lists(struct net_device *dev) if (!skb) break; - list->buffer[0].address = pci_map_single(priv->pci_dev, + list->buffer[0].address = dma_map_single(&priv->pci_dev->dev, skb->data, TLAN_MAX_FRAME_SIZE, - PCI_DMA_FROMDEVICE); + DMA_FROM_DEVICE); tlan_store_skb(list, skb); list->buffer[1].count = 0; list->buffer[1].address = 0; @@ -1954,12 +1950,10 @@ static void tlan_free_lists(struct net_device *dev) list = priv->tx_list + i; skb = tlan_get_skb(list); if (skb) { - pci_unmap_single( - priv->pci_dev, - list->buffer[0].address, - max(skb->len, - (unsigned int)TLAN_MIN_FRAME_SIZE), - PCI_DMA_TODEVICE); + dma_unmap_single(&priv->pci_dev->dev, + list->buffer[0].address, + max(skb->len, (unsigned int)TLAN_MIN_FRAME_SIZE), + DMA_TO_DEVICE); dev_kfree_skb_any(skb); list->buffer[8].address = 0; list->buffer[9].address = 0; @@ -1970,10 +1964,9 @@ static void tlan_free_lists(struct net_device *dev) list = priv->rx_list + i; skb = tlan_get_skb(list); if (skb) { - pci_unmap_single(priv->pci_dev, + dma_unmap_single(&priv->pci_dev->dev, list->buffer[0].address, - TLAN_MAX_FRAME_SIZE, - PCI_DMA_FROMDEVICE); + TLAN_MAX_FRAME_SIZE, DMA_FROM_DEVICE); dev_kfree_skb_any(skb); list->buffer[8].address = 0; list->buffer[9].address = 0; @@ -2511,7 +2504,7 @@ static void tlan_phy_power_down(struct net_device *dev) } /* Wait for 50 ms and powerup - * This is abitrary. It is intended to make sure the + * This is arbitrary. It is intended to make sure the * transceiver settles. */ tlan_set_timer(dev, msecs_to_jiffies(50), TLAN_TIMER_PHY_PUP); diff --git a/drivers/net/ethernet/toshiba/tc35815.c b/drivers/net/ethernet/toshiba/tc35815.c index 6bcda20ed7e7..7a6e5ff8e5d4 100644 --- a/drivers/net/ethernet/toshiba/tc35815.c +++ b/drivers/net/ethernet/toshiba/tc35815.c @@ -454,9 +454,9 @@ static struct sk_buff *alloc_rxbuf_skb(struct net_device *dev, skb = netdev_alloc_skb(dev, RX_BUF_SIZE); if (!skb) return NULL; - *dma_handle = pci_map_single(hwdev, skb->data, RX_BUF_SIZE, - PCI_DMA_FROMDEVICE); - if (pci_dma_mapping_error(hwdev, *dma_handle)) { + *dma_handle = dma_map_single(&hwdev->dev, skb->data, RX_BUF_SIZE, + DMA_FROM_DEVICE); + if (dma_mapping_error(&hwdev->dev, *dma_handle)) { dev_kfree_skb_any(skb); return NULL; } @@ -466,8 +466,8 @@ static struct sk_buff *alloc_rxbuf_skb(struct net_device *dev, static void free_rxbuf_skb(struct pci_dev *hwdev, struct sk_buff *skb, dma_addr_t dma_handle) { - pci_unmap_single(hwdev, dma_handle, RX_BUF_SIZE, - PCI_DMA_FROMDEVICE); + dma_unmap_single(&hwdev->dev, dma_handle, RX_BUF_SIZE, + DMA_FROM_DEVICE); dev_kfree_skb_any(skb); } @@ -876,9 +876,9 @@ tc35815_init_queues(struct net_device *dev) sizeof(struct TxFD) * TX_FD_NUM > PAGE_SIZE * FD_PAGE_NUM); - lp->fd_buf = pci_alloc_consistent(lp->pci_dev, - PAGE_SIZE * FD_PAGE_NUM, - &lp->fd_buf_dma); + lp->fd_buf = dma_alloc_coherent(&lp->pci_dev->dev, + PAGE_SIZE * FD_PAGE_NUM, + &lp->fd_buf_dma, GFP_ATOMIC); if (!lp->fd_buf) return -ENOMEM; for (i = 0; i < RX_BUF_NUM; i++) { @@ -892,10 +892,9 @@ tc35815_init_queues(struct net_device *dev) lp->rx_skbs[i].skb_dma); lp->rx_skbs[i].skb = NULL; } - pci_free_consistent(lp->pci_dev, - PAGE_SIZE * FD_PAGE_NUM, - lp->fd_buf, - lp->fd_buf_dma); + dma_free_coherent(&lp->pci_dev->dev, + PAGE_SIZE * FD_PAGE_NUM, + lp->fd_buf, lp->fd_buf_dma); lp->fd_buf = NULL; return -ENOMEM; } @@ -990,7 +989,9 @@ tc35815_clear_queues(struct net_device *dev) BUG_ON(lp->tx_skbs[i].skb != skb); #endif if (skb) { - pci_unmap_single(lp->pci_dev, lp->tx_skbs[i].skb_dma, skb->len, PCI_DMA_TODEVICE); + dma_unmap_single(&lp->pci_dev->dev, + lp->tx_skbs[i].skb_dma, skb->len, + DMA_TO_DEVICE); lp->tx_skbs[i].skb = NULL; lp->tx_skbs[i].skb_dma = 0; dev_kfree_skb_any(skb); @@ -1022,7 +1023,9 @@ tc35815_free_queues(struct net_device *dev) BUG_ON(lp->tx_skbs[i].skb != skb); #endif if (skb) { - pci_unmap_single(lp->pci_dev, lp->tx_skbs[i].skb_dma, skb->len, PCI_DMA_TODEVICE); + dma_unmap_single(&lp->pci_dev->dev, + lp->tx_skbs[i].skb_dma, + skb->len, DMA_TO_DEVICE); dev_kfree_skb(skb); lp->tx_skbs[i].skb = NULL; lp->tx_skbs[i].skb_dma = 0; @@ -1044,8 +1047,8 @@ tc35815_free_queues(struct net_device *dev) } } if (lp->fd_buf) { - pci_free_consistent(lp->pci_dev, PAGE_SIZE * FD_PAGE_NUM, - lp->fd_buf, lp->fd_buf_dma); + dma_free_coherent(&lp->pci_dev->dev, PAGE_SIZE * FD_PAGE_NUM, + lp->fd_buf, lp->fd_buf_dma); lp->fd_buf = NULL; } } @@ -1292,7 +1295,10 @@ tc35815_send_packet(struct sk_buff *skb, struct net_device *dev) BUG_ON(lp->tx_skbs[lp->tfd_start].skb); #endif lp->tx_skbs[lp->tfd_start].skb = skb; - lp->tx_skbs[lp->tfd_start].skb_dma = pci_map_single(lp->pci_dev, skb->data, skb->len, PCI_DMA_TODEVICE); + lp->tx_skbs[lp->tfd_start].skb_dma = dma_map_single(&lp->pci_dev->dev, + skb->data, + skb->len, + DMA_TO_DEVICE); /*add to ring */ txfd = &lp->tfd_base[lp->tfd_start]; @@ -1500,9 +1506,9 @@ tc35815_rx(struct net_device *dev, int limit) skb = lp->rx_skbs[cur_bd].skb; prefetch(skb->data); lp->rx_skbs[cur_bd].skb = NULL; - pci_unmap_single(lp->pci_dev, + dma_unmap_single(&lp->pci_dev->dev, lp->rx_skbs[cur_bd].skb_dma, - RX_BUF_SIZE, PCI_DMA_FROMDEVICE); + RX_BUF_SIZE, DMA_FROM_DEVICE); if (!HAVE_DMA_RXALIGN(lp) && NET_IP_ALIGN != 0) memmove(skb->data, skb->data - NET_IP_ALIGN, pkt_len); @@ -1756,7 +1762,9 @@ tc35815_txdone(struct net_device *dev) #endif if (skb) { dev->stats.tx_bytes += skb->len; - pci_unmap_single(lp->pci_dev, lp->tx_skbs[lp->tfd_end].skb_dma, skb->len, PCI_DMA_TODEVICE); + dma_unmap_single(&lp->pci_dev->dev, + lp->tx_skbs[lp->tfd_end].skb_dma, + skb->len, DMA_TO_DEVICE); lp->tx_skbs[lp->tfd_end].skb = NULL; lp->tx_skbs[lp->tfd_end].skb_dma = 0; dev_kfree_skb_any(skb); diff --git a/drivers/net/ethernet/via/via-rhine.c b/drivers/net/ethernet/via/via-rhine.c index 55b0ddab1776..73ca597ebd1b 100644 --- a/drivers/net/ethernet/via/via-rhine.c +++ b/drivers/net/ethernet/via/via-rhine.c @@ -1504,7 +1504,7 @@ static void rhine_init_cam_filter(struct net_device *dev) /** * rhine_update_vcam - update VLAN CAM filters - * @rp: rhine_private data of this Rhine + * @dev: rhine_private data of this Rhine * * Update VLAN CAM filters to match configuration change. */ diff --git a/drivers/net/ethernet/via/via-velocity.c b/drivers/net/ethernet/via/via-velocity.c index 6d2a31488a74..b65767f9e499 100644 --- a/drivers/net/ethernet/via/via-velocity.c +++ b/drivers/net/ethernet/via/via-velocity.c @@ -372,7 +372,7 @@ static const struct pci_device_id velocity_pci_id_table[] = { MODULE_DEVICE_TABLE(pci, velocity_pci_id_table); -/** +/* * Describe the OF device identifiers that we support in this * device driver. Used for devicetree nodes. */ @@ -384,7 +384,7 @@ MODULE_DEVICE_TABLE(of, velocity_of_ids); /** * get_chip_name - identifier to name - * @id: chip identifier + * @chip_id: chip identifier * * Given a chip identifier return a suitable description. Returns * a pointer a static string valid while the driver is loaded. @@ -748,7 +748,7 @@ static u32 mii_check_media_mode(struct mac_regs __iomem *regs) /** * velocity_mii_write - write MII data * @regs: velocity registers - * @index: MII register index + * @mii_addr: MII register index * @data: 16bit data for the MII register * * Perform a single write to an MII 16bit register. Returns zero @@ -869,6 +869,7 @@ static u32 check_connection_type(struct mac_regs __iomem *regs) /** * velocity_set_media_mode - set media mode + * @vptr: velocity adapter * @mii_status: old MII link state * * Check the media link state and configure the flow control @@ -877,26 +878,13 @@ static u32 check_connection_type(struct mac_regs __iomem *regs) */ static int velocity_set_media_mode(struct velocity_info *vptr, u32 mii_status) { - u32 curr_status; struct mac_regs __iomem *regs = vptr->mac_regs; vptr->mii_status = mii_check_media_mode(vptr->mac_regs); - curr_status = vptr->mii_status & (~VELOCITY_LINK_FAIL); /* Set mii link status */ set_mii_flow_control(vptr); - /* - Check if new status is consistent with current status - if (((mii_status & curr_status) & VELOCITY_AUTONEG_ENABLE) || - (mii_status==curr_status)) { - vptr->mii_status=mii_check_media_mode(vptr->mac_regs); - vptr->mii_status=check_connection_type(vptr->mac_regs); - netdev_info(vptr->netdev, "Velocity link no change\n"); - return 0; - } - */ - if (PHYID_GET_PHY_ID(vptr->phy_id) == PHYID_CICADA_CS8201) MII_REG_BITS_ON(AUXCR_MDPPS, MII_NCONFIG, vptr->mac_regs); @@ -1269,6 +1257,7 @@ static void mii_init(struct velocity_info *vptr, u32 mii_status) /** * setup_queue_timers - Setup interrupt timers + * @vptr: velocity adapter * * Setup interrupt frequency during suppression (timeout if the frame * count isn't filled). @@ -1293,8 +1282,7 @@ static void setup_queue_timers(struct velocity_info *vptr) /** * setup_adaptive_interrupts - Setup interrupt suppression - * - * @vptr velocity adapter + * @vptr: velocity adapter * * The velocity is able to suppress interrupt during high interrupt load. * This function turns on that feature. @@ -1735,6 +1723,7 @@ err_free_dma_rings_0: * velocity_free_tx_buf - free transmit buffer * @vptr: velocity * @tdinfo: buffer + * @td: transmit descriptor to free * * Release an transmit buffer. If the buffer was preallocated then * recycle it, if not then unmap the buffer. @@ -1909,7 +1898,7 @@ static void velocity_error(struct velocity_info *vptr, int status) /** * tx_srv - transmit interrupt service - * @vptr; Velocity + * @vptr: Velocity * * Scan the queues looking for transmitted packets that * we can complete and clean up. Update any statistics as @@ -2003,8 +1992,7 @@ static inline void velocity_rx_csum(struct rx_desc *rd, struct sk_buff *skb) * velocity_rx_copy - in place Rx copy for small packets * @rx_skb: network layer packet buffer candidate * @pkt_size: received data size - * @rd: receive packet descriptor - * @dev: network device + * @vptr: velocity adapter * * Replace the current skb that is scheduled for Rx processing by a * shorter, immediately allocated skb, if the received packet is small @@ -2110,6 +2098,7 @@ static int velocity_receive_frame(struct velocity_info *vptr, int idx) /** * velocity_rx_srv - service RX interrupt * @vptr: velocity + * @budget_left: remaining budget * * Walk the receive ring of the velocity adapter and remove * any received packets from the receive queue. Hand the ring @@ -2658,7 +2647,6 @@ static const struct net_device_ops velocity_netdev_ops = { /** * velocity_init_info - init private data - * @pdev: PCI device * @vptr: Velocity info * @info: Board type * @@ -2677,7 +2665,6 @@ static void velocity_init_info(struct velocity_info *vptr, /** * velocity_get_pci_info - retrieve PCI info for device * @vptr: velocity device - * @pdev: PCI device it matches * * Retrieve the PCI configuration space data that interests us from * the kernel PCI layer @@ -2714,7 +2701,6 @@ static int velocity_get_pci_info(struct velocity_info *vptr) /** * velocity_get_platform_info - retrieve platform info for device * @vptr: velocity device - * @pdev: platform device it matches * * Retrieve the Platform configuration data that interests us */ @@ -2764,8 +2750,9 @@ static u32 velocity_get_link(struct net_device *dev) /** * velocity_probe - set up discovered velocity device - * @pdev: PCI device - * @ent: PCI device table entry that matched + * @dev: PCI device + * @info: table of match + * @irq: interrupt info * @bustype: bus that device is connected to * * Configure a discovered adapter from scratch. Return a negative @@ -2982,6 +2969,7 @@ static int velocity_platform_remove(struct platform_device *pdev) #ifdef CONFIG_PM_SLEEP /** * wol_calc_crc - WOL CRC + * @size: size of the wake mask * @pattern: data pattern * @mask_pattern: mask * diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c index 9a15f14daa47..60c199fcb91e 100644 --- a/drivers/net/ethernet/xilinx/ll_temac_main.c +++ b/drivers/net/ethernet/xilinx/ll_temac_main.c @@ -106,7 +106,7 @@ static bool hard_acs_rdy_or_timeout(struct temac_local *lp, ktime_t timeout) */ #define HARD_ACS_RDY_POLL_NS (20 * NSEC_PER_MSEC) -/** +/* * temac_indirect_busywait - Wait for current indirect register access * to complete. */ @@ -121,7 +121,7 @@ int temac_indirect_busywait(struct temac_local *lp) return 0; } -/** +/* * temac_indirect_in32 - Indirect register read access. This function * must be called without lp->indirect_lock being held. */ @@ -136,7 +136,7 @@ u32 temac_indirect_in32(struct temac_local *lp, int reg) return val; } -/** +/* * temac_indirect_in32_locked - Indirect register read access. This * function must be called with lp->indirect_lock being held. Use * this together with spin_lock_irqsave/spin_lock_irqrestore to avoid @@ -164,7 +164,7 @@ u32 temac_indirect_in32_locked(struct temac_local *lp, int reg) return temac_ior(lp, XTE_LSW0_OFFSET); } -/** +/* * temac_indirect_out32 - Indirect register write access. This function * must be called without lp->indirect_lock being held. */ @@ -177,7 +177,7 @@ void temac_indirect_out32(struct temac_local *lp, int reg, u32 value) spin_unlock_irqrestore(lp->indirect_lock, flags); } -/** +/* * temac_indirect_out32_locked - Indirect register write access. This * function must be called with lp->indirect_lock being held. Use * this together with spin_lock_irqsave/spin_lock_irqrestore to avoid @@ -202,7 +202,7 @@ void temac_indirect_out32_locked(struct temac_local *lp, int reg, u32 value) WARN_ON(temac_indirect_busywait(lp)); } -/** +/* * temac_dma_in32_* - Memory mapped DMA read, these function expects a * register input that is based on DCR word addresses which are then * converted to memory mapped byte addresses. To be assigned to @@ -218,7 +218,7 @@ static u32 temac_dma_in32_le(struct temac_local *lp, int reg) return ioread32(lp->sdma_regs + (reg << 2)); } -/** +/* * temac_dma_out32_* - Memory mapped DMA read, these function expects * a register input that is based on DCR word addresses which are then * converted to memory mapped byte addresses. To be assigned to @@ -240,7 +240,7 @@ static void temac_dma_out32_le(struct temac_local *lp, int reg, u32 value) */ #ifdef CONFIG_PPC_DCR -/** +/* * temac_dma_dcr_in32 - DCR based DMA read */ static u32 temac_dma_dcr_in(struct temac_local *lp, int reg) @@ -248,7 +248,7 @@ static u32 temac_dma_dcr_in(struct temac_local *lp, int reg) return dcr_read(lp->sdma_dcrs, reg); } -/** +/* * temac_dma_dcr_out32 - DCR based DMA write */ static void temac_dma_dcr_out(struct temac_local *lp, int reg, u32 value) @@ -256,7 +256,7 @@ static void temac_dma_dcr_out(struct temac_local *lp, int reg, u32 value) dcr_write(lp->sdma_dcrs, reg, value); } -/** +/* * temac_dcr_setup - If the DMA is DCR based, then setup the address and * I/O functions */ @@ -293,7 +293,7 @@ static int temac_dcr_setup(struct temac_local *lp, struct platform_device *op, #endif -/** +/* * temac_dma_bd_release - Release buffer descriptor rings */ static void temac_dma_bd_release(struct net_device *ndev) @@ -323,7 +323,7 @@ static void temac_dma_bd_release(struct net_device *ndev) lp->tx_bd_v, lp->tx_bd_p); } -/** +/* * temac_dma_bd_init - Setup buffer descriptor rings */ static int temac_dma_bd_init(struct net_device *ndev) @@ -593,7 +593,7 @@ static struct temac_option { {} }; -/** +/* * temac_setoptions */ static u32 temac_setoptions(struct net_device *ndev, u32 options) diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c index fa5dc2993520..9aafd3ecdaa4 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c @@ -2038,8 +2038,7 @@ static int axienet_remove(struct platform_device *pdev) axienet_mdio_teardown(lp); - if (lp->clk) - clk_disable_unprepare(lp->clk); + clk_disable_unprepare(lp->clk); of_node_put(lp->phy_node); lp->phy_node = NULL; |