/************************************************************************** * * Copyright 2000-2006 Alacritech, Inc. All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. 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. * * Alternatively, this software may be distributed under the terms of the * GNU General Public License ("GPL") version 2 as published by the Free * Software Foundation. * * THIS SOFTWARE IS PROVIDED BY ALACRITECH, INC. ``AS IS'' AND ANY * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL ALACRITECH, INC. OR * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. * * The views and conclusions contained in the software and documentation * are those of the authors and should not be interpreted as representing * official policies, either expressed or implied, of Alacritech, Inc. * **************************************************************************/ /* * FILENAME: slicoss.c * * The SLICOSS driver for Alacritech's IS-NIC products. * * This driver is supposed to support: * * Mojave cards (single port PCI Gigabit) both copper and fiber * Oasis cards (single and dual port PCI-x Gigabit) copper and fiber * Kalahari cards (dual and quad port PCI-e Gigabit) copper and fiber * * The driver was actually tested on Oasis and Kalahari cards. * * * NOTE: This is the standard, non-accelerated version of Alacritech's * IS-NIC driver. */ #define KLUDGE_FOR_4GB_BOUNDARY 1 #define DEBUG_MICROCODE 1 #define DBG 1 #define SLIC_INTERRUPT_PROCESS_LIMIT 1 #define SLIC_OFFLOAD_IP_CHECKSUM 1 #define STATS_TIMER_INTERVAL 2 #define PING_TIMER_INTERVAL 1 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include "slichw.h" #include "slic.h" static uint slic_first_init = 1; static char *slic_banner = "Alacritech SLIC Technology(tm) Server " "and Storage Accelerator (Non-Accelerated)"; static char *slic_proc_version = "2.0.351 2006/07/14 12:26:00"; static struct net_device *head_netdevice; static struct base_driver slic_global = { {}, 0, 0, 0, 1, NULL, NULL }; static int intagg_delay = 100; static u32 dynamic_intagg; static unsigned int rcv_count; #define DRV_NAME "slicoss" #define DRV_VERSION "2.0.1" #define DRV_AUTHOR "Alacritech, Inc. Engineering" #define DRV_DESCRIPTION "Alacritech SLIC Techonology(tm) "\ "Non-Accelerated Driver" #define DRV_COPYRIGHT "Copyright 2000-2006 Alacritech, Inc. "\ "All rights reserved." #define PFX DRV_NAME " " MODULE_AUTHOR(DRV_AUTHOR); MODULE_DESCRIPTION(DRV_DESCRIPTION); MODULE_LICENSE("Dual BSD/GPL"); module_param(dynamic_intagg, int, 0); MODULE_PARM_DESC(dynamic_intagg, "Dynamic Interrupt Aggregation Setting"); module_param(intagg_delay, int, 0); MODULE_PARM_DESC(intagg_delay, "uSec Interrupt Aggregation Delay"); static const struct pci_device_id slic_pci_tbl[] = { { PCI_DEVICE(PCI_VENDOR_ID_ALACRITECH, SLIC_1GB_DEVICE_ID) }, { PCI_DEVICE(PCI_VENDOR_ID_ALACRITECH, SLIC_2GB_DEVICE_ID) }, { 0 } }; MODULE_DEVICE_TABLE(pci, slic_pci_tbl); static inline void slic_reg32_write(void __iomem *reg, u32 value, bool flush) { writel(value, reg); if (flush) mb(); } static inline void slic_reg64_write(struct adapter *adapter, void __iomem *reg, u32 value, void __iomem *regh, u32 paddrh, bool flush) { spin_lock_irqsave(&adapter->bit64reglock.lock, adapter->bit64reglock.flags); if (paddrh != adapter->curaddrupper) { adapter->curaddrupper = paddrh; writel(paddrh, regh); } writel(value, reg); if (flush) mb(); spin_unlock_irqrestore(&adapter->bit64reglock.lock, adapter->bit64reglock.flags); } static void slic_mcast_set_bit(struct adapter *adapter, char *address) { unsigned char crcpoly; /* Get the CRC polynomial for the mac address */ /* we use bits 1-8 (lsb), bitwise reversed, * msb (= lsb bit 0 before bitrev) is automatically discarded */ crcpoly = (ether_crc(ETH_ALEN, address)>>23); /* We only have space on the SLIC for 64 entries. Lop * off the top two bits. (2^6 = 64) */ crcpoly &= 0x3F; /* OR in the new bit into our 64 bit mask. */ adapter->mcastmask |= (u64) 1 << crcpoly; } static void slic_mcast_set_mask(struct adapter *adapter) { __iomem struct slic_regs *slic_regs = adapter->slic_regs; if (adapter->macopts & (MAC_ALLMCAST | MAC_PROMISC)) { /* Turn on all multicast addresses. We have to do this for * promiscuous mode as well as ALLMCAST mode. It saves the * Microcode from having to keep state about the MAC * configuration. */ slic_reg32_write(&slic_regs->slic_mcastlow, 0xFFFFFFFF, FLUSH); slic_reg32_write(&slic_regs->slic_mcasthigh, 0xFFFFFFFF, FLUSH); } else { /* Commit our multicast mast to the SLIC by writing to the * multicast address mask registers */ slic_reg32_write(&slic_regs->slic_mcastlow, (u32)(adapter->mcastmask & 0xFFFFFFFF), FLUSH); slic_reg32_write(&slic_regs->slic_mcasthigh, (u32)((adapter->mcastmask >> 32) & 0xFFFFFFFF), FLUSH); } } static void slic_timer_ping(ulong dev) { struct adapter *adapter; struct sliccard *card; adapter = netdev_priv((struct net_device *)dev); card = adapter->card; adapter->pingtimer.expires = jiffies + (PING_TIMER_INTERVAL * HZ); add_timer(&adapter->pingtimer); } static void slic_unmap_mmio_space(struct adapter *adapter) { if (adapter->slic_regs) iounmap(adapter->slic_regs); adapter->slic_regs = NULL; } /* * slic_link_config * * Write phy control to configure link duplex/speed * */ static void slic_link_config(struct adapter *adapter, u32 linkspeed, u32 linkduplex) { u32 __iomem *wphy; u32 speed; u32 duplex; u32 phy_config; u32 phy_advreg; u32 phy_gctlreg; if (adapter->state != ADAPT_UP) return; if (linkspeed > LINK_1000MB) linkspeed = LINK_AUTOSPEED; if (linkduplex > LINK_AUTOD) linkduplex = LINK_AUTOD; wphy = &adapter->slic_regs->slic_wphy; if ((linkspeed == LINK_AUTOSPEED) || (linkspeed == LINK_1000MB)) { if (adapter->flags & ADAPT_FLAGS_FIBERMEDIA) { /* We've got a fiber gigabit interface, and register * 4 is different in fiber mode than in copper mode */ /* advertise FD only @1000 Mb */ phy_advreg = (MIICR_REG_4 | (PAR_ADV1000XFD)); /* enable PAUSE frames */ phy_advreg |= PAR_ASYMPAUSE_FIBER; slic_reg32_write(wphy, phy_advreg, FLUSH); if (linkspeed == LINK_AUTOSPEED) { /* reset phy, enable auto-neg */ phy_config = (MIICR_REG_PCR | (PCR_RESET | PCR_AUTONEG | PCR_AUTONEG_RST)); slic_reg32_write(wphy, phy_config, FLUSH); } else { /* forced 1000 Mb FD*/ /* power down phy to break link this may not work) */ phy_config = (MIICR_REG_PCR | PCR_POWERDOWN); slic_reg32_write(wphy, phy_config, FLUSH); /* wait, Marvell says 1 sec, try to get away with 10 ms */ mdelay(10); /* disable auto-neg, set speed/duplex, soft reset phy, powerup */ phy_config = (MIICR_REG_PCR | (PCR_RESET | PCR_SPEED_1000 | PCR_DUPLEX_FULL)); slic_reg32_write(wphy, phy_config, FLUSH); } } else { /* copper gigabit */ /* Auto-Negotiate or 1000 Mb must be auto negotiated * We've got a copper gigabit interface, and * register 4 is different in copper mode than * in fiber mode */ if (linkspeed == LINK_AUTOSPEED) { /* advertise 10/100 Mb modes */ phy_advreg = (MIICR_REG_4 | (PAR_ADV100FD | PAR_ADV100HD | PAR_ADV10FD | PAR_ADV10HD)); } else { /* linkspeed == LINK_1000MB - don't advertise 10/100 Mb modes */ phy_advreg = MIICR_REG_4; } /* enable PAUSE frames */ phy_advreg |= PAR_ASYMPAUSE; /* required by the Cicada PHY */ phy_advreg |= PAR_802_3; slic_reg32_write(wphy, phy_advreg, FLUSH); /* advertise FD only @1000 Mb */ phy_gctlreg = (MIICR_REG_9 | (PGC_ADV1000FD)); slic_reg32_write(wphy, phy_gctlreg, FLUSH); if (adapter->subsysid != SLIC_1GB_CICADA_SUBSYS_ID) { /* if a Marvell PHY enable auto crossover */ phy_config = (MIICR_REG_16 | (MRV_REG16_XOVERON)); slic_reg32_write(wphy, phy_config, FLUSH); /* reset phy, enable auto-neg */ phy_config = (MIICR_REG_PCR | (PCR_RESET | PCR_AUTONEG | PCR_AUTONEG_RST)); slic_reg32_write(wphy, phy_config, FLUSH); } else { /* it's a Cicada PHY */ /* enable and restart auto-neg (don't reset) */ phy_config = (MIICR_REG_PCR | (PCR_AUTONEG | PCR_AUTONEG_RST)); slic_reg32_write(wphy, phy_config, FLUSH); } } } else { /* Forced 10/100 */ if (linkspeed == LINK_10MB) speed = 0; else speed = PCR_SPEED_100; if (linkduplex == LINK_HALFD) duplex = 0; else duplex = PCR_DUPLEX_FULL; if (adapter->subsysid != SLIC_1GB_CICADA_SUBSYS_ID) { /* if a Marvell PHY disable auto crossover */ phy_config = (MIICR_REG_16 | (MRV_REG16_XOVEROFF)); slic_reg32_write(wphy, phy_config, FLUSH); } /* power down phy to break link (this may not work) */ phy_config = (MIICR_REG_PCR | (PCR_POWERDOWN | speed | duplex)); slic_reg32_write(wphy, phy_config, FLUSH); /* wait, Marvell says 1 sec, try to get away with 10 ms */ mdelay(10); if (adapter->subsysid != SLIC_1GB_CICADA_SUBSYS_ID) { /* if a Marvell PHY disable auto-neg, set speed, soft reset phy, powerup */ phy_config = (MIICR_REG_PCR | (PCR_RESET | speed | duplex)); slic_reg32_write(wphy, phy_config, FLUSH); } else { /* it's a Cicada PHY */ /* disable auto-neg, set speed, powerup */ phy_config = (MIICR_REG_PCR | (speed | duplex)); slic_reg32_write(wphy, phy_config, FLUSH); } } } static int slic_card_download_gbrcv(struct adapter *adapter) { const struct firmware *fw; const char *file = ""; int ret; __iomem struct slic_regs *slic_regs = adapter->slic_regs; u32 codeaddr; u32 instruction; int index = 0; u32 rcvucodelen = 0; switch (adapter->devid) { case SLIC_2GB_DEVICE_ID: file = "slicoss/oasisrcvucode.sys"; break; case SLIC_1GB_DEVICE_ID: file = "slicoss/gbrcvucode.sys"; break; default: return -ENOENT; } ret = request_firmware(&fw, file, &adapter->pcidev->dev); if (ret) { dev_err(&adapter->pcidev->dev, "SLICOSS: Failed to load firmware %s\n", file); return ret; } rcvucodelen = *(u32 *)(fw->data + index); index += 4; switch (adapter->devid) { case SLIC_2GB_DEVICE_ID: if (rcvucodelen != OasisRcvUCodeLen) { release_firmware(fw); return -EINVAL; } break; case SLIC_1GB_DEVICE_ID: if (rcvucodelen != GBRcvUCodeLen) { release_firmware(fw); return -EINVAL; } break; } /* start download */ slic_reg32_write(&slic_regs->slic_rcv_wcs, SLIC_RCVWCS_BEGIN, FLUSH); /* download the rcv sequencer ucode */ for (codeaddr = 0; codeaddr < rcvucodelen; codeaddr++) { /* write out instruction address */ slic_reg32_write(&slic_regs->slic_rcv_wcs, codeaddr, FLUSH); instruction = *(u32 *)(fw->data + index); index += 4; /* write out the instruction data low addr */ slic_reg32_write(&slic_regs->slic_rcv_wcs, instruction, FLUSH); instruction = *(u8 *)(fw->data + index); index++; /* write out the instruction data high addr */ slic_reg32_write(&slic_regs->slic_rcv_wcs, (u8)instruction, FLUSH); } /* download finished */ release_firmware(fw); slic_reg32_write(&slic_regs->slic_rcv_wcs, SLIC_RCVWCS_FINISH, FLUSH); return 0; } MODULE_FIRMWARE("slicoss/oasisrcvucode.sys"); MODULE_FIRMWARE("slicoss/gbrcvucode.sys"); static int slic_card_download(struct adapter *adapter) { const struct firmware *fw; const char *file = ""; int ret; u32 section; int thissectionsize; int codeaddr; __iomem struct slic_regs *slic_regs = adapter->slic_regs; u32 instruction; u32 baseaddress; u32 i; u32 numsects = 0; u32 sectsize[3]; u32 sectstart[3]; int ucode_start, index = 0; switch (adapter->devid) { case SLIC_2GB_DEVICE_ID: file = "slicoss/oasisdownload.sys"; break; case SLIC_1GB_DEVICE_ID: file = "slicoss/gbdownload.sys"; break; default: return -ENOENT; } ret = request_firmware(&fw, file, &adapter->pcidev->dev); if (ret) { dev_err(&adapter->pcidev->dev, "SLICOSS: Failed to load firmware %s\n", file); return ret; } numsects = *(u32 *)(fw->data + index); index += 4; for (i = 0; i < numsects; i++) { sectsize[i] = *(u32 *)(fw->data + index); index += 4; } for (i = 0; i < numsects; i++) { sectstart[i] = *(u32 *)(fw->data + index); index += 4; } ucode_start = index; instruction = *(u32 *)(fw->data + index); index += 4; for (section = 0; section < numsects; section++) { baseaddress = sectstart[section]; thissectionsize = sectsize[section] >> 3; for (codeaddr = 0; codeaddr < thissectionsize; codeaddr++) { /* Write out instruction address */ slic_reg32_write(&slic_regs->slic_wcs, baseaddress + codeaddr, FLUSH); /* Write out instruction to low addr */ slic_reg32_write(&slic_regs->slic_wcs, instruction, FLUSH); instruction = *(u32 *)(fw->data + index); index += 4; /* Write out instruction to high addr */ slic_reg32_write(&slic_regs->slic_wcs, instruction, FLUSH); instruction = *(u32 *)(fw->data + index); index += 4; } } index = ucode_start; for (section = 0; section < numsects; section++) { instruction = *(u32 *)(fw->data + index); baseaddress = sectstart[section]; if (baseaddress < 0x8000) continue; thissectionsize = sectsize[section] >> 3; for (codeaddr = 0; codeaddr < thissectionsize; codeaddr++) { /* Write out instruction address */ slic_reg32_write(&slic_regs->slic_wcs, SLIC_WCS_COMPARE | (baseaddress + codeaddr), FLUSH); /* Write out instruction to low addr */ slic_reg32_write(&slic_regs->slic_wcs, instruction, FLUSH); instruction = *(u32 *)(fw->data + index); index += 4; /* Write out instruction to high addr */ slic_reg32_write(&slic_regs->slic_wcs, instruction, FLUSH); instruction = *(u32 *)(fw->data + index); index += 4; /* Check SRAM location zero. If it is non-zero. Abort.*/ /* failure = readl((u32 __iomem *)&slic_regs->slic_reset); if (failure) { release_firmware(fw); return -EIO; }*/ } } release_firmware(fw); /* Everything OK, kick off the card */ mdelay(10); slic_reg32_write(&slic_regs->slic_wcs, SLIC_WCS_START, FLUSH); /* stall for 20 ms, long enough for ucode to init card and reach mainloop */ mdelay(20); return 0; } MODULE_FIRMWARE("slicoss/oasisdownload.sys"); MODULE_FIRMWARE("slicoss/gbdownload.sys"); static void slic_adapter_set_hwaddr(struct adapter *adapter) { struct sliccard *card = adapter->card; if ((adapter->card) && (card->config_set)) { memcpy(adapter->macaddr, card->config.MacInfo[adapter->functionnumber].macaddrA, sizeof(struct slic_config_mac)); if (is_zero_ether_addr(adapter->currmacaddr)) memcpy(adapter->currmacaddr, adapter->macaddr, ETH_ALEN); if (adapter->netdev) memcpy(adapter->netdev->dev_addr, adapter->currmacaddr, ETH_ALEN); } } static void slic_intagg_set(struct adapter *adapter, u32 value) { slic_reg32_write(&adapter->slic_regs->slic_intagg, value, FLUSH); adapter->card->loadlevel_current = value; } static void slic_soft_reset(struct adapter *adapter) { if (adapter->card->state == CARD_UP) { slic_reg32_write(&adapter->slic_regs->slic_quiesce, 0, FLUSH); mdelay(1); } slic_reg32_write(&adapter->slic_regs->slic_reset, SLIC_RESET_MAGIC, FLUSH); mdelay(1); } static void slic_mac_address_config(struct adapter *adapter) { u32 value; u32 value2; __iomem struct slic_regs *slic_regs = adapter->slic_regs; value = *(u32 *) &adapter->currmacaddr[2]; value = ntohl(value); slic_reg32_write(&slic_regs->slic_wraddral, value, FLUSH); slic_reg32_write(&slic_regs->slic_wraddrbl, value, FLUSH); value2 = (u32) ((adapter->currmacaddr[0] << 8 | adapter->currmacaddr[1]) & 0xFFFF); slic_reg32_write(&slic_regs->slic_wraddrah, value2, FLUSH); slic_reg32_write(&slic_regs->slic_wraddrbh, value2, FLUSH); /* Write our multicast mask out to the card. This is done */ /* here in addition to the slic_mcast_addr_set routine */ /* because ALL_MCAST may have been enabled or disabled */ slic_mcast_set_mask(adapter); } static void slic_mac_config(struct adapter *adapter) { u32 value; __iomem struct slic_regs *slic_regs = adapter->slic_regs; /* Setup GMAC gaps */ if (adapter->linkspeed == LINK_1000MB) { value = ((GMCR_GAPBB_1000 << GMCR_GAPBB_SHIFT) | (GMCR_GAPR1_1000 << GMCR_GAPR1_SHIFT) | (GMCR_GAPR2_1000 << GMCR_GAPR2_SHIFT)); } else { value = ((GMCR_GAPBB_100 << GMCR_GAPBB_SHIFT) | (GMCR_GAPR1_100 << GMCR_GAPR1_SHIFT) | (GMCR_GAPR2_100 << GMCR_GAPR2_SHIFT)); } /* enable GMII */ if (adapter->linkspeed == LINK_1000MB) value |= GMCR_GBIT; /* enable fullduplex */ if ((adapter->linkduplex == LINK_FULLD) || (adapter->macopts & MAC_LOOPBACK)) { value |= GMCR_FULLD; } /* write mac config */ slic_reg32_write(&slic_regs->slic_wmcfg, value, FLUSH); /* setup mac addresses */ slic_mac_address_config(adapter); } static void slic_config_set(struct adapter *adapter, bool linkchange) { u32 value; u32 RcrReset; __iomem struct slic_regs *slic_regs = adapter->slic_regs; if (linkchange) { /* Setup MAC */ slic_mac_config(adapter); RcrReset = GRCR_RESET; } else { slic_mac_address_config(adapter); RcrReset = 0; } if (adapter->linkduplex == LINK_FULLD) { /* setup xmtcfg */ value = (GXCR_RESET | /* Always reset */ GXCR_XMTEN | /* Enable transmit */ GXCR_PAUSEEN); /* Enable pause */ slic_reg32_write(&slic_regs->slic_wxcfg, value, FLUSH); /* Setup rcvcfg last */ value = (RcrReset | /* Reset, if linkchange */ GRCR_CTLEN | /* Enable CTL frames */ GRCR_ADDRAEN | /* Address A enable */ GRCR_RCVBAD | /* Rcv bad frames */ (GRCR_HASHSIZE << GRCR_HASHSIZE_SHIFT)); } else { /* setup xmtcfg */ value = (GXCR_RESET | /* Always reset */ GXCR_XMTEN); /* Enable transmit */ slic_reg32_write(&slic_regs->slic_wxcfg, value, FLUSH); /* Setup rcvcfg last */ value = (RcrReset | /* Reset, if linkchange */ GRCR_ADDRAEN | /* Address A enable */ GRCR_RCVBAD | /* Rcv bad frames */ (GRCR_HASHSIZE << GRCR_HASHSIZE_SHIFT)); } if (adapter->state != ADAPT_DOWN) { /* Only enable receive if we are restarting or running */ value |= GRCR_RCVEN; } if (adapter->macopts & MAC_PROMISC) value |= GRCR_RCVALL; slic_reg32_write(&slic_regs->slic_wrcfg, value, FLUSH); } /* * Turn off RCV and XMT, power down PHY */ static void slic_config_clear(struct adapter *adapter) { u32 value; u32 phy_config; __iomem struct slic_regs *slic_regs = adapter->slic_regs; /* Setup xmtcfg */ value = (GXCR_RESET | /* Always reset */ GXCR_PAUSEEN); /* Enable pause */ slic_reg32_write(&slic_regs->slic_wxcfg, value, FLUSH); value = (GRCR_RESET | /* Always reset */ GRCR_CTLEN | /* Enable CTL frames */ GRCR_ADDRAEN | /* Address A enable */ (GRCR_HASHSIZE << GRCR_HASHSIZE_SHIFT)); slic_reg32_write(&slic_regs->slic_wrcfg, value, FLUSH); /* power down phy */ phy_config = (MIICR_REG_PCR | (PCR_POWERDOWN)); slic_reg32_write(&slic_regs->slic_wphy, phy_config, FLUSH); } static bool slic_mac_filter(struct adapter *adapter, struct ether_header *ether_frame) { struct net_device *netdev = adapter->netdev; u32 opts = adapter->macopts; if (opts & MAC_PROMISC) return true; if (is_broadcast_ether_addr(ether_frame->ether_dhost)) { if (opts & MAC_BCAST) { adapter->rcv_broadcasts++; return true; } else { return false; } } if (is_multicast_ether_addr(ether_frame->ether_dhost)) { if (opts & MAC_ALLMCAST) { adapter->rcv_multicasts++; netdev->stats.multicast++; return true; } if (opts & MAC_MCAST) { struct mcast_address *mcaddr = adapter->mcastaddrs; while (mcaddr) { if (ether_addr_equal(mcaddr->address, ether_frame->ether_dhost)) { adapter->rcv_multicasts++; netdev->stats.multicast++; return true; } mcaddr = mcaddr->next; } return false; } else { return false; } } if (opts & MAC_DIRECTED) { adapter->rcv_unicasts++; return true; } return false; } static int slic_mac_set_address(struct net_device *dev, void *ptr) { struct adapter *adapter = netdev_priv(dev); struct sockaddr *addr = ptr; if (netif_running(dev)) return -EBUSY; if (!adapter) return -EBUSY; if (!is_valid_ether_addr(addr->sa_data)) return -EINVAL; memcpy(dev->dev_addr, addr->sa_data, dev->addr_len); memcpy(adapter->currmacaddr, addr->sa_data, dev->addr_len); slic_config_set(adapter, true); return 0; } static void slic_timer_load_check(ulong cardaddr) { struct sliccard *card = (struct sliccard *)cardaddr; struct adapter *adapter = card->master; u32 __iomem *intagg; u32 load = card->events; u32 level = 0; if ((adapter) && (adapter->state == ADAPT_UP) && (card->state == CARD_UP) && (slic_global.dynamic_intagg)) { intagg = &adapter->slic_regs->slic_intagg; if (adapter->devid == SLIC_1GB_DEVICE_ID) { if (adapter->linkspeed == LINK_1000MB) level = 100; else { if (load > SLIC_LOAD_5) level = SLIC_INTAGG_5; else if (load > SLIC_LOAD_4) level = SLIC_INTAGG_4; else if (load > SLIC_LOAD_3) level = SLIC_INTAGG_3; else if (load > SLIC_LOAD_2) level = SLIC_INTAGG_2; else if (load > SLIC_LOAD_1) level = SLIC_INTAGG_1; else level = SLIC_INTAGG_0; } if (card->loadlevel_current != level) { card->loadlevel_current = level; slic_reg32_write(intagg, level, FLUSH); } } else { if (load > SLIC_LOAD_5) level = SLIC_INTAGG_5; else if (load > SLIC_LOAD_4) level = SLIC_INTAGG_4; else if (load > SLIC_LOAD_3) level = SLIC_INTAGG_3; else if (load > SLIC_LOAD_2) level = SLIC_INTAGG_2; else if (load > SLIC_LOAD_1) level = SLIC_INTAGG_1; else level = SLIC_INTAGG_0; if (card->loadlevel_current != level) { card->loadlevel_current = level; slic_reg32_write(intagg, level, FLUSH); } } } card->events = 0; card->loadtimer.expires = jiffies + (SLIC_LOADTIMER_PERIOD * HZ); add_timer(&card->loadtimer); } static int slic_upr_queue_request(struct adapter *adapter, u32 upr_request, u32 upr_data, u32 upr_data_h, u32 upr_buffer, u32 upr_buffer_h) { struct slic_upr *upr; struct slic_upr *uprqueue; upr = kmalloc(sizeof(struct slic_upr), GFP_ATOMIC); if (!upr) return -ENOMEM; upr->adapter = adapter->port; upr->upr_request = upr_request; upr->upr_data = upr_data; upr->upr_buffer = upr_buffer; upr->upr_data_h = upr_data_h; upr->upr_buffer_h = upr_buffer_h; upr->next = NULL; if (adapter->upr_list) { uprqueue = adapter->upr_list; while (uprqueue->next) uprqueue = uprqueue->next; uprqueue->next = upr; } else { adapter->upr_list = upr; } return 0; } static void slic_upr_start(struct adapter *adapter) { struct slic_upr *upr; __iomem struct slic_regs *slic_regs = adapter->slic_regs; /* char * ptr1; char * ptr2; uint cmdoffset; */ upr = adapter->upr_list; if (!upr) return; if (adapter->upr_busy) return; adapter->upr_busy = 1; switch (upr->upr_request) { case SLIC_UPR_STATS: if (upr->upr_data_h == 0) { slic_reg32_write(&slic_regs->slic_stats, upr->upr_data, FLUSH); } else { slic_reg64_write(adapter, &slic_regs->slic_stats64, upr->upr_data, &slic_regs->slic_addr_upper, upr->upr_data_h, FLUSH); } break; case SLIC_UPR_RLSR: slic_reg64_write(adapter, &slic_regs->slic_rlsr, upr->upr_data, &slic_regs->slic_addr_upper, upr->upr_data_h, FLUSH); break; case SLIC_UPR_RCONFIG: slic_reg64_write(adapter, &slic_regs->slic_rconfig, upr->upr_data, &slic_regs->slic_addr_upper, upr->upr_data_h, FLUSH); break; case SLIC_UPR_PING: slic_reg32_write(&slic_regs->slic_ping, 1, FLUSH); break; } } static int slic_upr_request(struct adapter *adapter, u32 upr_request, u32 upr_data, u32 upr_data_h, u32 upr_buffer, u32 upr_buffer_h) { int rc; spin_lock_irqsave(&adapter->upr_lock.lock, adapter->upr_lock.flags); rc = slic_upr_queue_request(adapter, upr_request, upr_data, upr_data_h, upr_buffer, upr_buffer_h); if (rc) goto err_unlock_irq; slic_upr_start(adapter); err_unlock_irq: spin_unlock_irqrestore(&adapter->upr_lock.lock, adapter->upr_lock.flags); return rc; } static void slic_link_upr_complete(struct adapter *adapter, u32 isr) { u32 linkstatus = adapter->pshmem->linkstatus; uint linkup; unsigned char linkspeed; unsigned char linkduplex; if ((isr & ISR_UPCERR) || (isr & ISR_UPCBSY)) { struct slic_shmem *pshmem; pshmem = (struct slic_shmem *)(unsigned long) adapter->phys_shmem; #if BITS_PER_LONG == 64 slic_upr_queue_request(adapter, SLIC_UPR_RLSR, SLIC_GET_ADDR_LOW(&pshmem->linkstatus), SLIC_GET_ADDR_HIGH(&pshmem->linkstatus), 0, 0); #else slic_upr_queue_request(adapter, SLIC_UPR_RLSR, (u32) &pshmem->linkstatus, SLIC_GET_ADDR_HIGH(pshmem), 0, 0); #endif return; } if (adapter->state != ADAPT_UP) return; linkup = linkstatus & GIG_LINKUP ? LINK_UP : LINK_DOWN; if (linkstatus & GIG_SPEED_1000) linkspeed = LINK_1000MB; else if (linkstatus & GIG_SPEED_100) linkspeed = LINK_100MB; else linkspeed = LINK_10MB; if (linkstatus & GIG_FULLDUPLEX) linkduplex = LINK_FULLD; else linkduplex = LINK_HALFD; if ((adapter->linkstate == LINK_DOWN) && (linkup == LINK_DOWN)) return; /* link up event, but nothing has changed */ if ((adapter->linkstate == LINK_UP) && (linkup == LINK_UP) && (adapter->linkspeed == linkspeed) && (adapter->linkduplex == linkduplex)) return; /* link has changed at this point */ /* link has gone from up to down */ if (linkup == LINK_DOWN) { adapter->linkstate = LINK_DOWN; return; } /* link has gone from down to up */ adapter->linkspeed = linkspeed; adapter->linkduplex = linkduplex; if (adapter->linkstate != LINK_UP) { /* setup the mac */ slic_config_set(adapter, true); adapter->linkstate = LINK_UP; netif_start_queue(adapter->netdev); } } static void slic_upr_request_complete(struct adapter *adapter, u32 isr) { struct sliccard *card = adapter->card; struct slic_upr *upr; spin_lock_irqsave(&adapter->upr_lock.lock, adapter->upr_lock.flags); upr = adapter->upr_list; if (!upr) { spin_unlock_irqrestore(&adapter->upr_lock.lock, adapter->upr_lock.flags); return; } adapter->upr_list = upr->next; upr->next = NULL; adapter->upr_busy = 0; switch (upr->upr_request) { case SLIC_UPR_STATS: { struct slic_stats *slicstats = (struct slic_stats *) &adapter->pshmem->inicstats; struct slic_stats *newstats = slicstats; struct slic_stats *old = &adapter->inicstats_prev; struct slicnet_stats *stst = &adapter->slic_stats; if (isr & ISR_UPCERR) { dev_err(&adapter->netdev->dev, "SLIC_UPR_STATS command failed isr[%x]\n", isr); break; } UPDATE_STATS_GB(stst->tcp.xmit_tcp_segs, newstats->xmit_tcp_segs_gb, old->xmit_tcp_segs_gb); UPDATE_STATS_GB(stst->tcp.xmit_tcp_bytes, newstats->xmit_tcp_bytes_gb, old->xmit_tcp_bytes_gb); UPDATE_STATS_GB(stst->tcp.rcv_tcp_segs, newstats->rcv_tcp_segs_gb, old->rcv_tcp_segs_gb); UPDATE_STATS_GB(stst->tcp.rcv_tcp_bytes, newstats->rcv_tcp_bytes_gb, old->rcv_tcp_bytes_gb); UPDATE_STATS_GB(stst->iface.xmt_bytes, newstats->xmit_bytes_gb, old->xmit_bytes_gb); UPDATE_STATS_GB(stst->iface.xmt_ucast, newstats->xmit_unicasts_gb, old->xmit_unicasts_gb); UPDATE_STATS_GB(stst->iface.rcv_bytes, newstats->rcv_bytes_gb, old->rcv_bytes_gb); UPDATE_STATS_GB(stst->iface.rcv_ucast, newstats->rcv_unicasts_gb, old->rcv_unicasts_gb); UPDATE_STATS_GB(stst->iface.xmt_errors, newstats->xmit_collisions_gb, old->xmit_collisions_gb); UPDATE_STATS_GB(stst->iface.xmt_errors, newstats->xmit_excess_collisions_gb, old->xmit_excess_collisions_gb); UPDATE_STATS_GB(stst->iface.xmt_errors, newstats->xmit_other_error_gb, old->xmit_other_error_gb); UPDATE_STATS_GB(stst->iface.rcv_errors, newstats->rcv_other_error_gb, old->rcv_other_error_gb); UPDATE_STATS_GB(stst->iface.rcv_discards, newstats->rcv_drops_gb, old->rcv_drops_gb); if (newstats->rcv_drops_gb > old->rcv_drops_gb) { adapter->rcv_drops += (newstats->rcv_drops_gb - old->rcv_drops_gb); } memcpy(old, newstats, sizeof(struct slic_stats)); break; } case SLIC_UPR_RLSR: slic_link_upr_complete(adapter, isr); break; case SLIC_UPR_RCONFIG: break; case SLIC_UPR_PING: card->pingstatus |= (isr & ISR_PINGDSMASK); break; } kfree(upr); slic_upr_start(adapter); spin_unlock_irqrestore(&adapter->upr_lock.lock, adapter->upr_lock.flags); } static int slic_config_get(struct adapter *adapter, u32 config, u32 config_h) { return slic_upr_request(adapter, SLIC_UPR_RCONFIG, config, config_h, 0, 0); } /* * Compute a checksum of the EEPROM according to RFC 1071. */ static u16 slic_eeprom_cksum(void *eeprom, unsigned len) { u16 *wp = eeprom; u32 checksum = 0; while (len > 1) { checksum += *(wp++); len -= 2; } if (len > 0) checksum += *(u8 *) wp; while (checksum >> 16) checksum = (checksum & 0xFFFF) + ((checksum >> 16) & 0xFFFF); return ~checksum; } static void slic_rspqueue_free(struct adapter *adapter) { int i; struct slic_rspqueue *rspq = &adapter->rspqueue; for (i = 0; i < rspq->num_pages; i++) { if (rspq->vaddr[i]) { pci_free_consistent(adapter->pcidev, PAGE_SIZE, rspq->vaddr[i], rspq->paddr[i]); } rspq->vaddr[i] = NULL; rspq->paddr[i] = 0; } rspq->offset = 0; rspq->pageindex = 0; rspq->rspbuf = NULL; } static int slic_rspqueue_init(struct adapter *adapter) { int i; struct slic_rspqueue *rspq = &adapter->rspqueue; __iomem struct slic_regs *slic_regs = adapter->slic_regs; u32 paddrh = 0; memset(rspq, 0, sizeof(struct slic_rspqueue)); rspq->num_pages = SLIC_RSPQ_PAGES_GB; for (i = 0; i < rspq->num_pages; i++) { rspq->vaddr[i] = pci_alloc_consistent(adapter->pcidev, PAGE_SIZE, &rspq->paddr[i]); if (!rspq->vaddr[i]) { dev_err(&adapter->pcidev->dev, "pci_alloc_consistent failed\n"); slic_rspqueue_free(adapter); return -ENOMEM; } /* FIXME: * do we really need this assertions (4K PAGE_SIZE aligned addr)? */ memset(rspq->vaddr[i], 0, PAGE_SIZE); if (paddrh == 0) { slic_reg32_write(&slic_regs->slic_rbar, (rspq->paddr[i] | SLIC_RSPQ_BUFSINPAGE), DONT_FLUSH); } else { slic_reg64_write(adapter, &slic_regs->slic_rbar64, (rspq->paddr[i] | SLIC_RSPQ_BUFSINPAGE), &slic_regs->slic_addr_upper, paddrh, DONT_FLUSH); } } rspq->offset = 0; rspq->pageindex = 0; rspq->rspbuf = (struct slic_rspbuf *)rspq->vaddr[0]; return 0; } static struct slic_rspbuf *slic_rspqueue_getnext(struct adapter *adapter) { struct slic_rspqueue *rspq = &adapter->rspqueue; struct slic_rspbuf *buf; if (!(rspq->rspbuf->status)) return NULL; buf = rspq->rspbuf; if (++rspq->offset < SLIC_RSPQ_BUFSINPAGE) { rspq->rspbuf++; } else { slic_reg64_write(adapter, &adapter->slic_regs->slic_rbar64, (rspq->paddr[rspq->pageindex] | SLIC_RSPQ_BUFSINPAGE), &adapter->slic_regs->slic_addr_upper, 0, DONT_FLUSH); rspq->pageindex = (rspq->pageindex + 1) % rspq->num_pages; rspq->offset = 0; rspq->rspbuf = (struct slic_rspbuf *) rspq->vaddr[rspq->pageindex]; } return buf; } static void slic_cmdqmem_free(struct adapter *adapter) { struct slic_cmdqmem *cmdqmem = &adapter->cmdqmem; int i; for (i = 0; i < SLIC_CMDQ_MAXPAGES; i++) { if (cmdqmem->pages[i]) { pci_free_consistent(adapter->pcidev, PAGE_SIZE, (void *) cmdqmem->pages[i], cmdqmem->dma_pages[i]); } } memset(cmdqmem, 0, sizeof(struct slic_cmdqmem)); } static u32 *slic_cmdqmem_addpage(struct adapter *adapter) { struct slic_cmdqmem *cmdqmem = &adapter->cmdqmem; u32 *pageaddr; if (cmdqmem->pagecnt >= SLIC_CMDQ_MAXPAGES) return NULL; pageaddr = pci_alloc_consistent(adapter->pcidev, PAGE_SIZE, &cmdqmem->dma_pages[cmdqmem->pagecnt]); if (!pageaddr) return NULL; cmdqmem->pages[cmdqmem->pagecnt] = pageaddr; cmdqmem->pagecnt++; return pageaddr; } static void slic_cmdq_free(struct adapter *adapter) { struct slic_hostcmd *cmd; cmd = adapter->cmdq_all.head; while (cmd) { if (cmd->busy) { struct sk_buff *tempskb; tempskb = cmd->skb; if (tempskb) { cmd->skb = NULL; dev_kfree_skb_irq(tempskb); } } cmd = cmd->next_all; } memset(&adapter->cmdq_all, 0, sizeof(struct slic_cmdqueue)); memset(&adapter->cmdq_free, 0, sizeof(struct slic_cmdqueue)); memset(&adapter->cmdq_done, 0, sizeof(struct slic_cmdqueue)); slic_cmdqmem_free(adapter); } static void slic_cmdq_addcmdpage(struct adapter *adapter, u32 *page) { struct slic_hostcmd *cmd; struct slic_hostcmd *prev; struct slic_hostcmd *tail; struct slic_cmdqueue *cmdq; int cmdcnt; void *cmdaddr; ulong phys_addr; u32 phys_addrl; u32 phys_addrh; struct slic_handle *pslic_handle; cmdaddr = page; cmd = (struct slic_hostcmd *)cmdaddr; cmdcnt = 0; phys_addr = virt_to_bus((void *)page); phys_addrl = SLIC_GET_ADDR_LOW(phys_addr); phys_addrh = SLIC_GET_ADDR_HIGH(phys_addr); prev = NULL; tail = cmd; while ((cmdcnt < SLIC_CMDQ_CMDSINPAGE) && (adapter->slic_handle_ix < 256)) { /* Allocate and initialize a SLIC_HANDLE for this command */ spin_lock_irqsave(&adapter->handle_lock.lock, adapter->handle_lock.flags); pslic_handle = adapter->pfree_slic_handles; adapter->pfree_slic_handles = pslic_handle->next; spin_unlock_irqrestore(&adapter->handle_lock.lock, adapter->handle_lock.flags); pslic_handle->type = SLIC_HANDLE_CMD; pslic_handle->address = (void *) cmd; pslic_handle->offset = (ushort) adapter->slic_handle_ix++; pslic_handle->other_handle = NULL; pslic_handle->next = NULL; cmd->pslic_handle = pslic_handle; cmd->cmd64.hosthandle = pslic_handle->token.handle_token; cmd->busy = false; cmd->paddrl = phys_addrl; cmd->paddrh = phys_addrh; cmd->next_all = prev; cmd->next = prev; prev = cmd; phys_addrl += SLIC_HOSTCMD_SIZE; cmdaddr += SLIC_HOSTCMD_SIZE; cmd = (struct slic_hostcmd *)cmdaddr; cmdcnt++; } cmdq = &adapter->cmdq_all; cmdq->count += cmdcnt; /* SLIC_CMDQ_CMDSINPAGE; mooktodo */ tail->next_all = cmdq->head; cmdq->head = prev; cmdq = &adapter->cmdq_free; spin_lock_irqsave(&cmdq->lock.lock, cmdq->lock.flags); cmdq->count += cmdcnt; /* SLIC_CMDQ_CMDSINPAGE; mooktodo */ tail->next = cmdq->head; cmdq->head = prev; spin_unlock_irqrestore(&cmdq->lock.lock, cmdq->lock.flags); } static int slic_cmdq_init(struct adapter *adapter) { int i; u32 *pageaddr; memset(&adapter->cmdq_all, 0, sizeof(struct slic_cmdqueue)); memset(&adapter->cmdq_free, 0, sizeof(struct slic_cmdqueue)); memset(&adapter->cmdq_done, 0, sizeof(struct slic_cmdqueue)); spin_lock_init(&adapter->cmdq_all.lock.lock); spin_lock_init(&adapter->cmdq_free.lock.lock); spin_lock_init(&adapter->cmdq_done.lock.lock); memset(&adapter->cmdqmem, 0, sizeof(struct slic_cmdqmem)); adapter->slic_handle_ix = 1; for (i = 0; i < SLIC_CMDQ_INITPAGES; i++) { pageaddr = slic_cmdqmem_addpage(adapter); if (!pageaddr) { slic_cmdq_free(adapter); return -ENOMEM; } slic_cmdq_addcmdpage(adapter, pageaddr); } adapter->slic_handle_ix = 1; return 0; } static void slic_cmdq_reset(struct adapter *adapter) { struct slic_hostcmd *hcmd; struct sk_buff *skb; u32 outstanding; spin_lock_irqsave(&adapter->cmdq_free.lock.lock, adapter->cmdq_free.lock.flags); spin_lock_irqsave(&adapter->cmdq_done.lock.lock, adapter->cmdq_done.lock.flags); outstanding = adapter->cmdq_all.count - adapter->cmdq_done.count; outstanding -= adapter->cmdq_free.count; hcmd = adapter->cmdq_all.head; while (hcmd) { if (hcmd->busy) { skb = hcmd->skb; hcmd->busy = 0; hcmd->skb = NULL; dev_kfree_skb_irq(skb); } hcmd = hcmd->next_all; } adapter->cmdq_free.count = 0; adapter->cmdq_free.head = NULL; adapter->cmdq_free.tail = NULL; adapter->cmdq_done.count = 0; adapter->cmdq_done.head = NULL; adapter->cmdq_done.tail = NULL; adapter->cmdq_free.head = adapter->cmdq_all.head; hcmd = adapter->cmdq_all.head; while (hcmd) { adapter->cmdq_free.count++; hcmd->next = hcmd->next_all; hcmd = hcmd->next_all; } if (adapter->cmdq_free.count != adapter->cmdq_all.count) { dev_err(&adapter->netdev->dev, "free_count %d != all count %d\n", adapter->cmdq_free.count, adapter->cmdq_all.count); } spin_unlock_irqrestore(&adapter->cmdq_done.lock.lock, adapter->cmdq_done.lock.flags); spin_unlock_irqrestore(&adapter->cmdq_free.lock.lock, adapter->cmdq_free.lock.flags); } static void slic_cmdq_getdone(struct adapter *adapter) { struct slic_cmdqueue *done_cmdq = &adapter->cmdq_done; struct slic_cmdqueue *free_cmdq = &adapter->cmdq_free; spin_lock_irqsave(&done_cmdq->lock.lock, done_cmdq->lock.flags); free_cmdq->head = done_cmdq->head; free_cmdq->count = done_cmdq->count; done_cmdq->head = NULL; done_cmdq->tail = NULL; done_cmdq->count = 0; spin_unlock_irqrestore(&done_cmdq->lock.lock, done_cmdq->lock.flags); } static struct slic_hostcmd *slic_cmdq_getfree(struct adapter *adapter) { struct slic_cmdqueue *cmdq = &adapter->cmdq_free; struct slic_hostcmd *cmd = NULL; lock_and_retry: spin_lock_irqsave(&cmdq->lock.lock, cmdq->lock.flags); retry: cmd = cmdq->head; if (cmd) { cmdq->head = cmd->next; cmdq->count--; spin_unlock_irqrestore(&cmdq->lock.lock, cmdq->lock.flags); } else { slic_cmdq_getdone(adapter); cmd = cmdq->head; if (cmd) { goto retry; } else { u32 *pageaddr; spin_unlock_irqrestore(&cmdq->lock.lock, cmdq->lock.flags); pageaddr = slic_cmdqmem_addpage(adapter); if (pageaddr) { slic_cmdq_addcmdpage(adapter, pageaddr); goto lock_and_retry; } } } return cmd; } static void slic_cmdq_putdone_irq(struct adapter *adapter, struct slic_hostcmd *cmd) { struct slic_cmdqueue *cmdq = &adapter->cmdq_done; spin_lock(&cmdq->lock.lock); cmd->busy = 0; cmd->next = cmdq->head; cmdq->head = cmd; cmdq->count++; if ((adapter->xmitq_full) && (cmdq->count > 10)) netif_wake_queue(adapter->netdev); spin_unlock(&cmdq->lock.lock); } static int slic_rcvqueue_fill(struct adapter *adapter) { void *paddr; u32 paddrl; u32 paddrh; struct slic_rcvqueue *rcvq = &adapter->rcvqueue; int i = 0; struct device *dev = &adapter->netdev->dev; while (i < SLIC_RCVQ_FILLENTRIES) { struct slic_rcvbuf *rcvbuf; struct sk_buff *skb; #ifdef KLUDGE_FOR_4GB_BOUNDARY retry_rcvqfill: #endif skb = alloc_skb(SLIC_RCVQ_RCVBUFSIZE, GFP_ATOMIC); if (skb) { paddr = (void *)(unsigned long) pci_map_single(adapter->pcidev, skb->data, SLIC_RCVQ_RCVBUFSIZE, PCI_DMA_FROMDEVICE); paddrl = SLIC_GET_ADDR_LOW(paddr); paddrh = SLIC_GET_ADDR_HIGH(paddr); skb->len = SLIC_RCVBUF_HEADSIZE; rcvbuf = (struct slic_rcvbuf *)skb->head; rcvbuf->status = 0; skb->next = NULL; #ifdef KLUDGE_FOR_4GB_BOUNDARY if (paddrl == 0) { dev_err(dev, "%s: LOW 32bits PHYSICAL ADDRESS == 0\n", __func__); dev_err(dev, "skb[%p] PROBLEM\n", skb); dev_err(dev, " skbdata[%p]\n", skb->data); dev_err(dev, " skblen[%x]\n", skb->len); dev_err(dev, " paddr[%p]\n", paddr); dev_err(dev, " paddrl[%x]\n", paddrl); dev_err(dev, " paddrh[%x]\n", paddrh); dev_err(dev, " rcvq->head[%p]\n", rcvq->head); dev_err(dev, " rcvq->tail[%p]\n", rcvq->tail); dev_err(dev, " rcvq->count[%x]\n", rcvq->count); dev_err(dev, "SKIP THIS SKB!!!!!!!!\n"); goto retry_rcvqfill; } #else if (paddrl == 0) { dev_err(dev, "%s: LOW 32bits PHYSICAL ADDRESS == 0\n", __func__); dev_err(dev, "skb[%p] PROBLEM\n", skb); dev_err(dev, " skbdata[%p]\n", skb->data); dev_err(dev, " skblen[%x]\n", skb->len); dev_err(dev, " paddr[%p]\n", paddr); dev_err(dev, " paddrl[%x]\n", paddrl); dev_err(dev, " paddrh[%x]\n", paddrh); dev_err(dev, " rcvq->head[%p]\n", rcvq->head); dev_err(dev, " rcvq->tail[%p]\n", rcvq->tail); dev_err(dev, " rcvq->count[%x]\n", rcvq->count); dev_err(dev, "GIVE TO CARD ANYWAY\n"); } #endif if (paddrh == 0) { slic_reg32_write(&adapter->slic_regs->slic_hbar, (u32)paddrl, DONT_FLUSH); } else { slic_reg64_write(adapter, &adapter->slic_regs->slic_hbar64, paddrl, &adapter->slic_regs->slic_addr_upper, paddrh, DONT_FLUSH); } if (rcvq->head) rcvq->tail->next = skb; else rcvq->head = skb; rcvq->tail = skb; rcvq->count++; i++; } else { dev_err(&adapter->netdev->dev, "slic_rcvqueue_fill could only get [%d] skbuffs\n", i); break; } } return i; } static void slic_rcvqueue_free(struct adapter *adapter) { struct slic_rcvqueue *rcvq = &adapter->rcvqueue; struct sk_buff *skb; while (rcvq->head) { skb = rcvq->head; rcvq->head = rcvq->head->next; dev_kfree_skb(skb); } rcvq->tail = NULL; rcvq->head = NULL; rcvq->count = 0; } static int slic_rcvqueue_init(struct adapter *adapter) { int i, count; struct slic_rcvqueue *rcvq = &adapter->rcvqueue; rcvq->tail = NULL; rcvq->head = NULL; rcvq->size = SLIC_RCVQ_ENTRIES; rcvq->errors = 0; rcvq->count = 0; i = (SLIC_RCVQ_ENTRIES / SLIC_RCVQ_FILLENTRIES); count = 0; while (i) { count += slic_rcvqueue_fill(adapter); i--; } if (rcvq->count < SLIC_RCVQ_MINENTRIES) { slic_rcvqueue_free(adapter); return -ENOMEM; } return 0; } static struct sk_buff *slic_rcvqueue_getnext(struct adapter *adapter) { struct slic_rcvqueue *rcvq = &adapter->rcvqueue; struct sk_buff *skb; struct slic_rcvbuf *rcvbuf; int count; if (rcvq->count) { skb = rcvq->head; rcvbuf = (struct slic_rcvbuf *)skb->head; if (rcvbuf->status & IRHDDR_SVALID) { rcvq->head = rcvq->head->next; skb->next = NULL; rcvq->count--; } else { skb = NULL; } } else { dev_err(&adapter->netdev->dev, "RcvQ Empty!! rcvq[%p] count[%x]\n", rcvq, rcvq->count); skb = NULL; } while (rcvq->count < SLIC_RCVQ_FILLTHRESH) { count = slic_rcvqueue_fill(adapter); if (!count) break; } if (skb) rcvq->errors = 0; return skb; } static u32 slic_rcvqueue_reinsert(struct adapter *adapter, struct sk_buff *skb) { struct slic_rcvqueue *rcvq = &adapter->rcvqueue; void *paddr; u32 paddrl; u32 paddrh; struct slic_rcvbuf *rcvbuf = (struct slic_rcvbuf *)skb->head; struct device *dev; paddr = (void *)(unsigned long) pci_map_single(adapter->pcidev, skb->head, SLIC_RCVQ_RCVBUFSIZE, PCI_DMA_FROMDEVICE); rcvbuf->status = 0; skb->next = NULL; paddrl = SLIC_GET_ADDR_LOW(paddr); paddrh = SLIC_GET_ADDR_HIGH(paddr); if (paddrl == 0) { dev = &adapter->netdev->dev; dev_err(dev, "%s: LOW 32bits PHYSICAL ADDRESS == 0\n", __func__); dev_err(dev, "skb[%p] PROBLEM\n", skb); dev_err(dev, " skbdata[%p]\n", skb->data); dev_err(dev, " skblen[%x]\n", skb->len); dev_err(dev, " paddr[%p]\n", paddr); dev_err(dev, " paddrl[%x]\n", paddrl); dev_err(dev, " paddrh[%x]\n", paddrh); dev_err(dev, " rcvq->head[%p]\n", rcvq->head); dev_err(dev, " rcvq->tail[%p]\n", rcvq->tail); dev_err(dev, " rcvq->count[%x]\n", rcvq->count); } if (paddrh == 0) { slic_reg32_write(&adapter->slic_regs->slic_hbar, (u32)paddrl, DONT_FLUSH); } else { slic_reg64_write(adapter, &adapter->slic_regs->slic_hbar64, paddrl, &adapter->slic_regs->slic_addr_upper, paddrh, DONT_FLUSH); } if (rcvq->head) rcvq->tail->next = skb; else rcvq->head = skb; rcvq->tail = skb; rcvq->count++; return rcvq->count; } /* * slic_link_event_handler - * * Initiate a link configuration sequence. The link configuration begins * by issuing a READ_LINK_STATUS command to the Utility Processor on the * SLIC. Since the command finishes asynchronously, the slic_upr_comlete * routine will follow it up witha UP configuration write command, which * will also complete asynchronously. * */ static void slic_link_event_handler(struct adapter *adapter) { int status; struct slic_shmem *pshmem; if (adapter->state != ADAPT_UP) { /* Adapter is not operational. Ignore. */ return; } pshmem = (struct slic_shmem *)(unsigned long)adapter->phys_shmem; #if BITS_PER_LONG == 64 status = slic_upr_request(adapter, SLIC_UPR_RLSR, SLIC_GET_ADDR_LOW(&pshmem->linkstatus), SLIC_GET_ADDR_HIGH(&pshmem->linkstatus), 0, 0); #else status = slic_upr_request(adapter, SLIC_UPR_RLSR, (u32) &pshmem->linkstatus, /* no 4GB wrap guaranteed */ 0, 0, 0); #endif } static void slic_init_cleanup(struct adapter *adapter) { if (adapter->intrregistered) { adapter->intrregistered = 0; free_irq(adapter->netdev->irq, adapter->netdev); } if (adapter->pshmem) { pci_free_consistent(adapter->pcidev, sizeof(struct slic_shmem), adapter->pshmem, adapter->phys_shmem); adapter->pshmem = NULL; adapter->phys_shmem = (dma_addr_t)(unsigned long)NULL; } if (adapter->pingtimerset) { adapter->pingtimerset = 0; del_timer(&adapter->pingtimer); } slic_rspqueue_free(adapter); slic_cmdq_free(adapter); slic_rcvqueue_free(adapter); } /* * Allocate a mcast_address structure to hold the multicast address. * Link it in. */ static int slic_mcast_add_list(struct adapter *adapter, char *address) { struct mcast_address *mcaddr, *mlist; /* Check to see if it already exists */ mlist = adapter->mcastaddrs; while (mlist) { if (ether_addr_equal(mlist->address, address)) return 0; mlist = mlist->next; } /* Doesn't already exist. Allocate a structure to hold it */ mcaddr = kmalloc(sizeof(struct mcast_address), GFP_ATOMIC); if (mcaddr == NULL) return 1; memcpy(mcaddr->address, address, ETH_ALEN); mcaddr->next = adapter->mcastaddrs; adapter->mcastaddrs = mcaddr; return 0; } static void slic_mcast_set_list(struct net_device *dev) { struct adapter *adapter = netdev_priv(dev); int status = 0; char *addresses; struct netdev_hw_addr *ha; netdev_for_each_mc_addr(ha, dev) { addresses = (char *) &ha->addr; status = slic_mcast_add_list(adapter, addresses); if (status != 0) break; slic_mcast_set_bit(adapter, addresses); } if (adapter->devflags_prev != dev->flags) { adapter->macopts = MAC_DIRECTED; if (dev->flags) { if (dev->flags & IFF_BROADCAST) adapter->macopts |= MAC_BCAST; if (dev->flags & IFF_PROMISC) adapter->macopts |= MAC_PROMISC; if (dev->flags & IFF_ALLMULTI) adapter->macopts |= MAC_ALLMCAST; if (dev->flags & IFF_MULTICAST) adapter->macopts |= MAC_MCAST; } adapter->devflags_prev = dev->flags; slic_config_set(adapter, true); } else { if (status == 0) slic_mcast_set_mask(adapter); } } #define XMIT_FAIL_LINK_STATE 1 #define XMIT_FAIL_ZERO_LENGTH 2 #define XMIT_FAIL_HOSTCMD_FAIL 3 static void slic_xmit_build_request(struct adapter *adapter, struct slic_hostcmd *hcmd, struct sk_buff *skb) { struct slic_host64_cmd *ihcmd; ulong phys_addr; ihcmd = &hcmd->cmd64; ihcmd->flags = (adapter->port << IHFLG_IFSHFT); ihcmd->command = IHCMD_XMT_REQ; ihcmd->u.slic_buffers.totlen = skb->len; phys_addr = pci_map_single(adapter->pcidev, skb->data, skb->len, PCI_DMA_TODEVICE); ihcmd->u.slic_buffers.bufs[0].paddrl = SLIC_GET_ADDR_LOW(phys_addr); ihcmd->u.slic_buffers.bufs[0].paddrh = SLIC_GET_ADDR_HIGH(phys_addr); ihcmd->u.slic_buffers.bufs[0].length = skb->len; #if BITS_PER_LONG == 64 hcmd->cmdsize = (u32) ((((u64)&ihcmd->u.slic_buffers.bufs[1] - (u64) hcmd) + 31) >> 5); #else hcmd->cmdsize = ((((u32) &ihcmd->u.slic_buffers.bufs[1] - (u32) hcmd) + 31) >> 5); #endif } static void slic_xmit_fail(struct adapter *adapter, struct sk_buff *skb, void *cmd, u32 skbtype, u32 status) { if (adapter->xmitq_full) netif_stop_queue(adapter->netdev); if ((cmd == NULL) && (status <= XMIT_FAIL_HOSTCMD_FAIL)) { switch (status) { case XMIT_FAIL_LINK_STATE: dev_err(&adapter->netdev->dev, "reject xmit skb[%p: %x] linkstate[%s] adapter[%s:%d] card[%s:%d]\n", skb, skb->pkt_type, SLIC_LINKSTATE(adapter->linkstate), SLIC_ADAPTER_STATE(adapter->state), adapter->state, SLIC_CARD_STATE(adapter->card->state), adapter->card->state); break; case XMIT_FAIL_ZERO_LENGTH: dev_err(&adapter->netdev->dev, "xmit_start skb->len == 0 skb[%p] type[%x]\n", skb, skb->pkt_type); break; case XMIT_FAIL_HOSTCMD_FAIL: dev_err(&adapter->netdev->dev, "xmit_start skb[%p] type[%x] No host commands available\n", skb, skb->pkt_type); break; } } dev_kfree_skb(skb); adapter->netdev->stats.tx_dropped++; } static void slic_rcv_handle_error(struct adapter *adapter, struct slic_rcvbuf *rcvbuf) { struct slic_hddr_wds *hdr = (struct slic_hddr_wds *)rcvbuf->data; struct net_device *netdev = adapter->netdev; if (adapter->devid != SLIC_1GB_DEVICE_ID) { if (hdr->frame_status14 & VRHSTAT_802OE) adapter->if_events.oflow802++; if (hdr->frame_status14 & VRHSTAT_TPOFLO) adapter->if_events.Tprtoflow++; if (hdr->frame_status_b14 & VRHSTATB_802UE) adapter->if_events.uflow802++; if (hdr->frame_status_b14 & VRHSTATB_RCVE) { adapter->if_events.rcvearly++; netdev->stats.rx_fifo_errors++; } if (hdr->frame_status_b14 & VRHSTATB_BUFF) { adapter->if_events.Bufov++; netdev->stats.rx_over_errors++; } if (hdr->frame_status_b14 & VRHSTATB_CARRE) { adapter->if_events.Carre++; netdev->stats.tx_carrier_errors++; } if (hdr->frame_status_b14 & VRHSTATB_LONGE) adapter->if_events.Longe++; if (hdr->frame_status_b14 & VRHSTATB_PREA) adapter->if_events.Invp++; if (hdr->frame_status_b14 & VRHSTATB_CRC) { adapter->if_events.Crc++; netdev->stats.rx_crc_errors++; } if (hdr->frame_status_b14 & VRHSTATB_DRBL) adapter->if_events.Drbl++; if (hdr->frame_status_b14 & VRHSTATB_CODE) adapter->if_events.Code++; if (hdr->frame_status_b14 & VRHSTATB_TPCSUM) adapter->if_events.TpCsum++; if (hdr->frame_status_b14 & VRHSTATB_TPHLEN) adapter->if_events.TpHlen++; if (hdr->frame_status_b14 & VRHSTATB_IPCSUM) adapter->if_events.IpCsum++; if (hdr->frame_status_b14 & VRHSTATB_IPLERR) adapter->if_events.IpLen++; if (hdr->frame_status_b14 & VRHSTATB_IPHERR) adapter->if_events.IpHlen++; } else { if (hdr->frame_statusGB & VGBSTAT_XPERR) { u32 xerr = hdr->frame_statusGB >> VGBSTAT_XERRSHFT; if (xerr == VGBSTAT_XCSERR) adapter->if_events.TpCsum++; if (xerr == VGBSTAT_XUFLOW) adapter->if_events.Tprtoflow++; if (xerr == VGBSTAT_XHLEN) adapter->if_events.TpHlen++; } if (hdr->frame_statusGB & VGBSTAT_NETERR) { u32 nerr = (hdr-> frame_statusGB >> VGBSTAT_NERRSHFT) & VGBSTAT_NERRMSK; if (nerr == VGBSTAT_NCSERR) adapter->if_events.IpCsum++; if (nerr == VGBSTAT_NUFLOW) adapter->if_events.IpLen++; if (nerr == VGBSTAT_NHLEN) adapter->if_events.IpHlen++; } if (hdr->frame_statusGB & VGBSTAT_LNKERR) { u32 lerr = hdr->frame_statusGB & VGBSTAT_LERRMSK; if (lerr == VGBSTAT_LDEARLY) adapter->if_events.rcvearly++; if (lerr == VGBSTAT_LBOFLO) adapter->if_events.Bufov++; if (lerr == VGBSTAT_LCODERR) adapter->if_events.Code++; if (lerr == VGBSTAT_LDBLNBL) adapter->if_events.Drbl++; if (lerr == VGBSTAT_LCRCERR) adapter->if_events.Crc++; if (lerr == VGBSTAT_LOFLO) adapter->if_events.oflow802++; if (lerr == VGBSTAT_LUFLO) adapter->if_events.uflow802++; } } return; } #define TCP_OFFLOAD_FRAME_PUSHFLAG 0x10000000 #define M_FAST_PATH 0x0040 static void slic_rcv_handler(struct adapter *adapter) { struct net_device *netdev = adapter->netdev; struct sk_buff *skb; struct slic_rcvbuf *rcvbuf; u32 frames = 0; while ((skb = slic_rcvqueue_getnext(adapter))) { u32 rx_bytes; rcvbuf = (struct slic_rcvbuf *)skb->head; adapter->card->events++; if (rcvbuf->status & IRHDDR_ERR) { adapter->rx_errors++; slic_rcv_handle_error(adapter, rcvbuf); slic_rcvqueue_reinsert(adapter, skb); continue; } if (!slic_mac_filter(adapter, (struct ether_header *) rcvbuf->data)) { slic_rcvqueue_reinsert(adapter, skb); continue; } skb_pull(skb, SLIC_RCVBUF_HEADSIZE); rx_bytes = (rcvbuf->length & IRHDDR_FLEN_MSK); skb_put(skb, rx_bytes); netdev->stats.rx_packets++; netdev->stats.rx_bytes += rx_bytes; #if SLIC_OFFLOAD_IP_CHECKSUM skb->ip_summed = CHECKSUM_UNNECESSARY; #endif skb->dev = adapter->netdev; skb->protocol = eth_type_trans(skb, skb->dev); netif_rx(skb); ++frames; #if SLIC_INTERRUPT_PROCESS_LIMIT if (frames >= SLIC_RCVQ_MAX_PROCESS_ISR) { adapter->rcv_interrupt_yields++; break; } #endif } adapter->max_isr_rcvs = max(adapter->max_isr_rcvs, frames); } static void slic_xmit_complete(struct adapter *adapter) { struct slic_hostcmd *hcmd; struct slic_rspbuf *rspbuf; u32 frames = 0; struct slic_handle_word slic_handle_word; do { rspbuf = slic_rspqueue_getnext(adapter); if (!rspbuf) break; adapter->xmit_completes++; adapter->card->events++; /* Get the complete host command buffer */ slic_handle_word.handle_token = rspbuf->hosthandle; hcmd = (struct slic_hostcmd *) adapter->slic_handles[slic_handle_word.handle_index]. address; /* hcmd = (struct slic_hostcmd *) rspbuf->hosthandle; */ if (hcmd->type == SLIC_CMD_DUMB) { if (hcmd->skb) dev_kfree_skb_irq(hcmd->skb); slic_cmdq_putdone_irq(adapter, hcmd); } rspbuf->status = 0; rspbuf->hosthandle = 0; frames++; } while (1); adapter->max_isr_xmits = max(adapter->max_isr_xmits, frames); } static void slic_interrupt_card_up(u32 isr, struct adapter *adapter, struct net_device *dev) { if (isr & ~ISR_IO) { if (isr & ISR_ERR) { adapter->error_interrupts++; if (isr & ISR_RMISS) { int count; int pre_count; int errors; struct slic_rcvqueue *rcvq = &adapter->rcvqueue; adapter->error_rmiss_interrupts++; if (!rcvq->errors) rcv_count = rcvq->count; pre_count = rcvq->count; errors = rcvq->errors; while (rcvq->count < SLIC_RCVQ_FILLTHRESH) { count = slic_rcvqueue_fill(adapter); if (!count) break; } } else if (isr & ISR_XDROP) { dev_err(&dev->dev, "isr & ISR_ERR [%x] ISR_XDROP\n", isr); } else { dev_err(&dev->dev, "isr & ISR_ERR [%x]\n", isr); } } if (isr & ISR_LEVENT) { adapter->linkevent_interrupts++; slic_link_event_handler(adapter); } if ((isr & ISR_UPC) || (isr & ISR_UPCERR) || (isr & ISR_UPCBSY)) { adapter->upr_interrupts++; slic_upr_request_complete(adapter, isr); } } if (isr & ISR_RCV) { adapter->rcv_interrupts++; slic_rcv_handler(adapter); } if (isr & ISR_CMD) { adapter->xmit_interrupts++; slic_xmit_complete(adapter); } } static irqreturn_t slic_interrupt(int irq, void *dev_id) { struct net_device *dev = (struct net_device *)dev_id; struct adapter *adapter = netdev_priv(dev); u32 isr; if ((adapter->pshmem) && (adapter->pshmem->isr)) { slic_reg32_write(&adapter->slic_regs->slic_icr, ICR_INT_MASK, FLUSH); isr = adapter->isrcopy = adapter->pshmem->isr; adapter->pshmem->isr = 0; adapter->num_isrs++; switch (adapter->card->state) { case CARD_UP: slic_interrupt_card_up(isr, adapter, dev); break; case CARD_DOWN: if ((isr & ISR_UPC) || (isr & ISR_UPCERR) || (isr & ISR_UPCBSY)) { adapter->upr_interrupts++; slic_upr_request_complete(adapter, isr); } break; } adapter->isrcopy = 0; adapter->all_reg_writes += 2; adapter->isr_reg_writes++; slic_reg32_write(&adapter->slic_regs->slic_isr, 0, FLUSH); } else { adapter->false_interrupts++; } return IRQ_HANDLED; } #define NORMAL_ETHFRAME 0 static netdev_tx_t slic_xmit_start(struct sk_buff *skb, struct net_device *dev) { struct sliccard *card; struct adapter *adapter = netdev_priv(dev); struct slic_hostcmd *hcmd = NULL; u32 status = 0; void *offloadcmd = NULL; card = adapter->card; if ((adapter->linkstate != LINK_UP) || (adapter->state != ADAPT_UP) || (card->state != CARD_UP)) { status = XMIT_FAIL_LINK_STATE; goto xmit_fail; } else if (skb->len == 0) { status = XMIT_FAIL_ZERO_LENGTH; goto xmit_fail; } hcmd = slic_cmdq_getfree(adapter); if (!hcmd) { adapter->xmitq_full = 1; status = XMIT_FAIL_HOSTCMD_FAIL; goto xmit_fail; } hcmd->skb = skb; hcmd->busy = 1; hcmd->type = SLIC_CMD_DUMB; slic_xmit_build_request(adapter, hcmd, skb); dev->stats.tx_packets++; dev->stats.tx_bytes += skb->len; #ifdef DEBUG_DUMP if (adapter->kill_card) { struct slic_host64_cmd ihcmd; ihcmd = &hcmd->cmd64; ihcmd->flags |= 0x40; adapter->kill_card = 0; /* only do this once */ } #endif if (hcmd->paddrh == 0) { slic_reg32_write(&adapter->slic_regs->slic_cbar, (hcmd->paddrl | hcmd->cmdsize), DONT_FLUSH); } else { slic_reg64_write(adapter, &adapter->slic_regs->slic_cbar64, (hcmd->paddrl | hcmd->cmdsize), &adapter->slic_regs->slic_addr_upper, hcmd->paddrh, DONT_FLUSH); } xmit_done: return NETDEV_TX_OK; xmit_fail: slic_xmit_fail(adapter, skb, offloadcmd, NORMAL_ETHFRAME, status); goto xmit_done; } static void slic_adapter_freeresources(struct adapter *adapter) { slic_init_cleanup(adapter); adapter->error_interrupts = 0; adapter->rcv_interrupts = 0; adapter->xmit_interrupts = 0; adapter->linkevent_interrupts = 0; adapter->upr_interrupts = 0; adapter->num_isrs = 0; adapter->xmit_completes = 0; adapter->rcv_broadcasts = 0; adapter->rcv_multicasts = 0; adapter->rcv_unicasts = 0; } static int slic_adapter_allocresources(struct adapter *adapter) { if (!adapter->intrregistered) { int retval; spin_unlock_irqrestore(&slic_global.driver_lock.lock, slic_global.driver_lock.flags); retval = request_irq(adapter->netdev->irq, &slic_interrupt, IRQF_SHARED, adapter->netdev->name, adapter->netdev); spin_lock_irqsave(&slic_global.driver_lock.lock, slic_global.driver_lock.flags); if (retval) { dev_err(&adapter->netdev->dev, "request_irq (%s) FAILED [%x]\n", adapter->netdev->name, retval); return retval; } adapter->intrregistered = 1; } return 0; } /* * slic_if_init * * Perform initialization of our slic interface. * */ static int slic_if_init(struct adapter *adapter) { struct sliccard *card = adapter->card; struct net_device *dev = adapter->netdev; __iomem struct slic_regs *slic_regs = adapter->slic_regs; struct slic_shmem *pshmem; int rc; /* adapter should be down at this point */ if (adapter->state != ADAPT_DOWN) { dev_err(&dev->dev, "%s: adapter->state != ADAPT_DOWN\n", __func__); rc = -EIO; goto err; } adapter->devflags_prev = dev->flags; adapter->macopts = MAC_DIRECTED; if (dev->flags) { if (dev->flags & IFF_BROADCAST) adapter->macopts |= MAC_BCAST; if (dev->flags & IFF_PROMISC) adapter->macopts |= MAC_PROMISC; if (dev->flags & IFF_ALLMULTI) adapter->macopts |= MAC_ALLMCAST; if (dev->flags & IFF_MULTICAST) adapter->macopts |= MAC_MCAST; } rc = slic_adapter_allocresources(adapter); if (rc) { dev_err(&dev->dev, "%s: slic_adapter_allocresources FAILED %x\n", __func__, rc); slic_adapter_freeresources(adapter); goto err; } if (!adapter->queues_initialized) { rc = slic_rspqueue_init(adapter); if (rc) goto err; rc = slic_cmdq_init(adapter); if (rc) goto err; rc = slic_rcvqueue_init(adapter); if (rc) goto err; adapter->queues_initialized = 1; } slic_reg32_write(&slic_regs->slic_icr, ICR_INT_OFF, FLUSH); mdelay(1); if (!adapter->isp_initialized) { pshmem = (struct slic_shmem *)(unsigned long) adapter->phys_shmem; spin_lock_irqsave(&adapter->bit64reglock.lock, adapter->bit64reglock.flags); #if BITS_PER_LONG == 64 slic_reg32_write(&slic_regs->slic_addr_upper, SLIC_GET_ADDR_HIGH(&pshmem->isr), DONT_FLUSH); slic_reg32_write(&slic_regs->slic_isp, SLIC_GET_ADDR_LOW(&pshmem->isr), FLUSH); #else slic_reg32_write(&slic_regs->slic_addr_upper, 0, DONT_FLUSH); slic_reg32_write(&slic_regs->slic_isp, (u32)&pshmem->isr, FLUSH); #endif spin_unlock_irqrestore(&adapter->bit64reglock.lock, adapter->bit64reglock.flags); adapter->isp_initialized = 1; } adapter->state = ADAPT_UP; if (!card->loadtimerset) { init_timer(&card->loadtimer); card->loadtimer.expires = jiffies + (SLIC_LOADTIMER_PERIOD * HZ); card->loadtimer.data = (ulong) card; card->loadtimer.function = &slic_timer_load_check; add_timer(&card->loadtimer); card->loadtimerset = 1; } if (!adapter->pingtimerset) { init_timer(&adapter->pingtimer); adapter->pingtimer.expires = jiffies + (PING_TIMER_INTERVAL * HZ); adapter->pingtimer.data = (ulong) dev; adapter->pingtimer.function = &slic_timer_ping; add_timer(&adapter->pingtimer); adapter->pingtimerset = 1; adapter->card->pingstatus = ISR_PINGMASK; } /* * clear any pending events, then enable interrupts */ adapter->isrcopy = 0; adapter->pshmem->isr = 0; slic_reg32_write(&slic_regs->slic_isr, 0, FLUSH); slic_reg32_write(&slic_regs->slic_icr, ICR_INT_ON, FLUSH); slic_link_config(adapter, LINK_AUTOSPEED, LINK_AUTOD); slic_link_event_handler(adapter); err: return rc; } static int slic_entry_open(struct net_device *dev) { struct adapter *adapter = netdev_priv(dev); struct sliccard *card = adapter->card; int status; netif_stop_queue(adapter->netdev); spin_lock_irqsave(&slic_global.driver_lock.lock, slic_global.driver_lock.flags); if (!adapter->activated) { card->adapters_activated++; slic_global.num_slic_ports_active++; adapter->activated = 1; } status = slic_if_init(adapter); if (status != 0) { if (adapter->activated) { card->adapters_activated--; slic_global.num_slic_ports_active--; adapter->activated = 0; } goto spin_unlock; } if (!card->master) card->master = adapter; spin_unlock: spin_unlock_irqrestore(&slic_global.driver_lock.lock, slic_global.driver_lock.flags); return status; } static void slic_card_cleanup(struct sliccard *card) { if (card->loadtimerset) { card->loadtimerset = 0; del_timer_sync(&card->loadtimer); } kfree(card); } static void slic_entry_remove(struct pci_dev *pcidev) { struct net_device *dev = pci_get_drvdata(pcidev); struct adapter *adapter = netdev_priv(dev); struct sliccard *card; struct mcast_address *mcaddr, *mlist; unregister_netdev(dev); slic_adapter_freeresources(adapter); slic_unmap_mmio_space(adapter); /* free multicast addresses */ mlist = adapter->mcastaddrs; while (mlist) { mcaddr = mlist; mlist = mlist->next; kfree(mcaddr); } card = adapter->card; card->adapters_allocated--; adapter->allocated = 0; if (!card->adapters_allocated) { struct sliccard *curr_card = slic_global.slic_card; if (curr_card == card) { slic_global.slic_card = card->next; } else { while (curr_card->next != card) curr_card = curr_card->next; curr_card->next = card->next; } slic_global.num_slic_cards--; slic_card_cleanup(card); } free_netdev(dev); pci_release_regions(pcidev); pci_disable_device(pcidev); } static int slic_entry_halt(struct net_device *dev) { struct adapter *adapter = netdev_priv(dev); struct sliccard *card = adapter->card; __iomem struct slic_regs *slic_regs = adapter->slic_regs; spin_lock_irqsave(&slic_global.driver_lock.lock, slic_global.driver_lock.flags); netif_stop_queue(adapter->netdev); adapter->state = ADAPT_DOWN; adapter->linkstate = LINK_DOWN; adapter->upr_list = NULL; adapter->upr_busy = 0; adapter->devflags_prev = 0; slic_reg32_write(&slic_regs->slic_icr, ICR_INT_OFF, FLUSH); adapter->all_reg_writes++; adapter->icr_reg_writes++; slic_config_clear(adapter); if (adapter->activated) { card->adapters_activated--; slic_global.num_slic_ports_active--; adapter->activated = 0; } #ifdef AUTOMATIC_RESET slic_reg32_write(&slic_regs->slic_reset_iface, 0, FLUSH); #endif /* * Reset the adapter's cmd queues */ slic_cmdq_reset(adapter); #ifdef AUTOMATIC_RESET if (!card->adapters_activated) slic_card_init(card, adapter); #endif spin_unlock_irqrestore(&slic_global.driver_lock.lock, slic_global.driver_lock.flags); return 0; } static struct net_device_stats *slic_get_stats(struct net_device *dev) { struct adapter *adapter = netdev_priv(dev); dev->stats.collisions = adapter->slic_stats.iface.xmit_collisions; dev->stats.rx_errors = adapter->slic_stats.iface.rcv_errors; dev->stats.tx_errors = adapter->slic_stats.iface.xmt_errors; dev->stats.rx_missed_errors = adapter->slic_stats.iface.rcv_discards; dev->stats.tx_heartbeat_errors = 0; dev->stats.tx_aborted_errors = 0; dev->stats.tx_window_errors = 0; dev->stats.tx_fifo_errors = 0; dev->stats.rx_frame_errors = 0; dev->stats.rx_length_errors = 0; return &dev->stats; } static int slic_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { struct adapter *adapter = netdev_priv(dev); struct ethtool_cmd edata; struct ethtool_cmd ecmd; u32 data[7]; u32 intagg; switch (cmd) { case SIOCSLICSETINTAGG: if (copy_from_user(data, rq->ifr_data, 28)) return -EFAULT; intagg = data[0]; dev_err(&dev->dev, "%s: set interrupt aggregation to %d\n", __func__, intagg); slic_intagg_set(adapter, intagg); return 0; #ifdef SLIC_TRACE_DUMP_ENABLED case SIOCSLICTRACEDUMP: { u32 value; DBG_IOCTL("slic_ioctl SIOCSLIC_TRACE_DUMP\n"); if (copy_from_user(data, rq->ifr_data, 28)) { PRINT_ERROR ("slic: copy_from_user FAILED getting initial simba param\n"); return -EFAULT; } value = data[0]; if (tracemon_request == SLIC_DUMP_DONE) { PRINT_ERROR ("ATK Diagnostic Trace Dump Requested\n"); tracemon_request = SLIC_DUMP_REQUESTED; tracemon_request_type = value; tracemon_timestamp = jiffies; } else if ((tracemon_request == SLIC_DUMP_REQUESTED) || (tracemon_request == SLIC_DUMP_IN_PROGRESS)) { PRINT_ERROR ("ATK Diagnostic Trace Dump Requested but already in progress... ignore\n"); } else { PRINT_ERROR ("ATK Diagnostic Trace Dump Requested\n"); tracemon_request = SLIC_DUMP_REQUESTED; tracemon_request_type = value; tracemon_timestamp = jiffies; } return 0; } #endif case SIOCETHTOOL: if (copy_from_user(&ecmd, rq->ifr_data, sizeof(ecmd))) return -EFAULT; if (ecmd.cmd == ETHTOOL_GSET) { memset(&edata, 0, sizeof(edata)); edata.supported = (SUPPORTED_10baseT_Half | SUPPORTED_10baseT_Full | SUPPORTED_100baseT_Half | SUPPORTED_100baseT_Full | SUPPORTED_Autoneg | SUPPORTED_MII); edata.port = PORT_MII; edata.transceiver = XCVR_INTERNAL; edata.phy_address = 0; if (adapter->linkspeed == LINK_100MB) edata.speed = SPEED_100; else if (adapter->linkspeed == LINK_10MB) edata.speed = SPEED_10; else edata.speed = 0; if (adapter->linkduplex == LINK_FULLD) edata.duplex = DUPLEX_FULL; else edata.duplex = DUPLEX_HALF; edata.autoneg = AUTONEG_ENABLE; edata.maxtxpkt = 1; edata.maxrxpkt = 1; if (copy_to_user(rq->ifr_data, &edata, sizeof(edata))) return -EFAULT; } else if (ecmd.cmd == ETHTOOL_SSET) { if (!capable(CAP_NET_ADMIN)) return -EPERM; if (adapter->linkspeed == LINK_100MB) edata.speed = SPEED_100; else if (adapter->linkspeed == LINK_10MB) edata.speed = SPEED_10; else edata.speed = 0; if (adapter->linkduplex == LINK_FULLD) edata.duplex = DUPLEX_FULL; else edata.duplex = DUPLEX_HALF; edata.autoneg = AUTONEG_ENABLE; edata.maxtxpkt = 1; edata.maxrxpkt = 1; if ((ecmd.speed != edata.speed) || (ecmd.duplex != edata.duplex)) { u32 speed; u32 duplex; if (ecmd.speed == SPEED_10) speed = 0; else speed = PCR_SPEED_100; if (ecmd.duplex == DUPLEX_FULL) duplex = PCR_DUPLEX_FULL; else duplex = 0; slic_link_config(adapter, speed, duplex); slic_link_event_handler(adapter); } } return 0; default: return -EOPNOTSUPP; } } static void slic_config_pci(struct pci_dev *pcidev) { u16 pci_command; u16 new_command; pci_read_config_word(pcidev, PCI_COMMAND, &pci_command); new_command = pci_command | PCI_COMMAND_MASTER | PCI_COMMAND_MEMORY | PCI_COMMAND_INVALIDATE | PCI_COMMAND_PARITY | PCI_COMMAND_SERR | PCI_COMMAND_FAST_BACK; if (pci_command != new_command) pci_write_config_word(pcidev, PCI_COMMAND, new_command); } static int slic_card_init(struct sliccard *card, struct adapter *adapter) { __iomem struct slic_regs *slic_regs = adapter->slic_regs; struct slic_eeprom *peeprom; struct oslic_eeprom *pOeeprom; dma_addr_t phys_config; u32 phys_configh; u32 phys_configl; u32 i = 0; struct slic_shmem *pshmem; int status; uint macaddrs = card->card_size; ushort eecodesize; ushort dramsize; ushort ee_chksum; ushort calc_chksum; struct slic_config_mac *pmac; unsigned char fruformat; unsigned char oemfruformat; struct atk_fru *patkfru; union oemfru *poemfru; /* Reset everything except PCI configuration space */ slic_soft_reset(adapter); /* Download the microcode */ status = slic_card_download(adapter); if (status != 0) { dev_err(&adapter->pcidev->dev, "download failed bus %d slot %d\n", adapter->busnumber, adapter->slotnumber); return status; } if (!card->config_set) { peeprom = pci_alloc_consistent(adapter->pcidev, sizeof(struct slic_eeprom), &phys_config); phys_configl = SLIC_GET_ADDR_LOW(phys_config); phys_configh = SLIC_GET_ADDR_HIGH(phys_config); if (!peeprom) { dev_err(&adapter->pcidev->dev, "eeprom read failed to get memory bus %d slot %d\n", adapter->busnumber, adapter->slotnumber); return -ENOMEM; } else { memset(peeprom, 0, sizeof(struct slic_eeprom)); } slic_reg32_write(&slic_regs->slic_icr, ICR_INT_OFF, FLUSH); mdelay(1); pshmem = (struct slic_shmem *)(unsigned long) adapter->phys_shmem; spin_lock_irqsave(&adapter->bit64reglock.lock, adapter->bit64reglock.flags); slic_reg32_write(&slic_regs->slic_addr_upper, SLIC_GET_ADDR_HIGH(&pshmem->isr), DONT_FLUSH); slic_reg32_write(&slic_regs->slic_isp, SLIC_GET_ADDR_LOW(&pshmem->isr), FLUSH); spin_unlock_irqrestore(&adapter->bit64reglock.lock, adapter->bit64reglock.flags); status = slic_config_get(adapter, phys_configl, phys_configh); if (status) { dev_err(&adapter->pcidev->dev, "Failed to fetch config data from device.\n"); goto card_init_err; } for (;;) { if (adapter->pshmem->isr) { if (adapter->pshmem->isr & ISR_UPC) { adapter->pshmem->isr = 0; slic_reg64_write(adapter, &slic_regs->slic_isp, 0, &slic_regs->slic_addr_upper, 0, FLUSH); slic_reg32_write(&slic_regs->slic_isr, 0, FLUSH); slic_upr_request_complete(adapter, 0); break; } else { adapter->pshmem->isr = 0; slic_reg32_write(&slic_regs->slic_isr, 0, FLUSH); } } else { mdelay(1); i++; if (i > 5000) { dev_err(&adapter->pcidev->dev, "%d config data fetch timed out!\n", adapter->port); slic_reg64_write(adapter, &slic_regs->slic_isp, 0, &slic_regs->slic_addr_upper, 0, FLUSH); status = -EINVAL; goto card_init_err; } } } switch (adapter->devid) { /* Oasis card */ case SLIC_2GB_DEVICE_ID: /* extract EEPROM data and pointers to EEPROM data */ pOeeprom = (struct oslic_eeprom *) peeprom; eecodesize = pOeeprom->EecodeSize; dramsize = pOeeprom->DramSize; pmac = pOeeprom->MacInfo; fruformat = pOeeprom->FruFormat; patkfru = &pOeeprom->AtkFru; oemfruformat = pOeeprom->OemFruFormat; poemfru = &pOeeprom->OemFru; macaddrs = 2; /* Minor kludge for Oasis card get 2 MAC addresses from the EEPROM to ensure that function 1 gets the Port 1 MAC address */ break; default: /* extract EEPROM data and pointers to EEPROM data */ eecodesize = peeprom->EecodeSize; dramsize = peeprom->DramSize; pmac = peeprom->u2.mac.MacInfo; fruformat = peeprom->FruFormat; patkfru = &peeprom->AtkFru; oemfruformat = peeprom->OemFruFormat; poemfru = &peeprom->OemFru; break; } card->config.EepromValid = false; /* see if the EEPROM is valid by checking it's checksum */ if ((eecodesize <= MAX_EECODE_SIZE) && (eecodesize >= MIN_EECODE_SIZE)) { ee_chksum = *(u16 *) ((char *) peeprom + (eecodesize - 2)); /* calculate the EEPROM checksum */ calc_chksum = slic_eeprom_cksum(peeprom, eecodesize - 2); /* if the ucdoe chksum flag bit worked, we wouldn't need this */ if (ee_chksum == calc_chksum) card->config.EepromValid = true; } /* copy in the DRAM size */ card->config.DramSize = dramsize; /* copy in the MAC address(es) */ for (i = 0; i < macaddrs; i++) { memcpy(&card->config.MacInfo[i], &pmac[i], sizeof(struct slic_config_mac)); } /* copy the Alacritech FRU information */ card->config.FruFormat = fruformat; memcpy(&card->config.AtkFru, patkfru, sizeof(struct atk_fru)); pci_free_consistent(adapter->pcidev, sizeof(struct slic_eeprom), peeprom, phys_config); if (!card->config.EepromValid) { slic_reg64_write(adapter, &slic_regs->slic_isp, 0, &slic_regs->slic_addr_upper, 0, FLUSH); dev_err(&adapter->pcidev->dev, "unsupported CONFIGURATION EEPROM invalid\n"); return -EINVAL; } card->config_set = 1; } if (slic_card_download_gbrcv(adapter)) { dev_err(&adapter->pcidev->dev, "unable to download GB receive microcode\n"); return -EINVAL; } if (slic_global.dynamic_intagg) slic_intagg_set(adapter, 0); else slic_intagg_set(adapter, intagg_delay); /* * Initialize ping status to "ok" */ card->pingstatus = ISR_PINGMASK; /* * Lastly, mark our card state as up and return success */ card->state = CARD_UP; card->reset_in_progress = 0; return 0; card_init_err: pci_free_consistent(adapter->pcidev, sizeof(struct slic_eeprom), peeprom, phys_config); return status; } static void slic_init_driver(void) { if (slic_first_init) { slic_first_init = 0; spin_lock_init(&slic_global.driver_lock.lock); } } static void slic_init_adapter(struct net_device *netdev, struct pci_dev *pcidev, const struct pci_device_id *pci_tbl_entry, void __iomem *memaddr, int chip_idx) { ushort index; struct slic_handle *pslic_handle; struct adapter *adapter = netdev_priv(netdev); /* adapter->pcidev = pcidev;*/ adapter->vendid = pci_tbl_entry->vendor; adapter->devid = pci_tbl_entry->device; adapter->subsysid = pci_tbl_entry->subdevice; adapter->busnumber = pcidev->bus->number; adapter->slotnumber = ((pcidev->devfn >> 3) & 0x1F); adapter->functionnumber = (pcidev->devfn & 0x7); adapter->slic_regs = (__iomem struct slic_regs *)memaddr; adapter->irq = pcidev->irq; /* adapter->netdev = netdev;*/ adapter->next_netdevice = head_netdevice; head_netdevice = netdev; adapter->chipid = chip_idx; adapter->port = 0; /*adapter->functionnumber;*/ adapter->cardindex = adapter->port; spin_lock_init(&adapter->upr_lock.lock); spin_lock_init(&adapter->bit64reglock.lock); spin_lock_init(&adapter->adapter_lock.lock); spin_lock_init(&adapter->reset_lock.lock); spin_lock_init(&adapter->handle_lock.lock); adapter->card_size = 1; /* Initialize slic_handle array */ /* Start with 1. 0 is an invalid host handle. */ for (index = 1, pslic_handle = &adapter->slic_handles[1]; index < SLIC_CMDQ_MAXCMDS; index++, pslic_handle++) { pslic_handle->token.handle_index = index; pslic_handle->type = SLIC_HANDLE_FREE; pslic_handle->next = adapter->pfree_slic_handles; adapter->pfree_slic_handles = pslic_handle; } adapter->pshmem = (struct slic_shmem *) pci_alloc_consistent(adapter->pcidev, sizeof(struct slic_shmem), &adapter-> phys_shmem); if (adapter->pshmem) memset(adapter->pshmem, 0, sizeof(struct slic_shmem)); } static const struct net_device_ops slic_netdev_ops = { .ndo_open = slic_entry_open, .ndo_stop = slic_entry_halt, .ndo_start_xmit = slic_xmit_start, .ndo_do_ioctl = slic_ioctl, .ndo_set_mac_address = slic_mac_set_address, .ndo_get_stats = slic_get_stats, .ndo_set_rx_mode = slic_mcast_set_list, .ndo_validate_addr = eth_validate_addr, .ndo_change_mtu = eth_change_mtu, }; static u32 slic_card_locate(struct adapter *adapter) { struct sliccard *card = slic_global.slic_card; struct physcard *physcard = slic_global.phys_card; ushort card_hostid; u16 __iomem *hostid_reg; uint i; uint rdhostid_offset = 0; switch (adapter->devid) { case SLIC_2GB_DEVICE_ID: rdhostid_offset = SLIC_RDHOSTID_2GB; break; case SLIC_1GB_DEVICE_ID: rdhostid_offset = SLIC_RDHOSTID_1GB; break; default: return -ENODEV; } hostid_reg = (u16 __iomem *) (((u8 __iomem *) (adapter->slic_regs)) + rdhostid_offset); /* read the 16 bit hostid from SRAM */ card_hostid = (ushort) readw(hostid_reg); /* Initialize a new card structure if need be */ if (card_hostid == SLIC_HOSTID_DEFAULT) { card = kzalloc(sizeof(struct sliccard), GFP_KERNEL); if (card == NULL) return -ENOMEM; card->next = slic_global.slic_card; slic_global.slic_card = card; card->busnumber = adapter->busnumber; card->slotnumber = adapter->slotnumber; /* Find an available cardnum */ for (i = 0; i < SLIC_MAX_CARDS; i++) { if (slic_global.cardnuminuse[i] == 0) { slic_global.cardnuminuse[i] = 1; card->cardnum = i; break; } } slic_global.num_slic_cards++; } else { /* Card exists, find the card this adapter belongs to */ while (card) { if (card->cardnum == card_hostid) break; card = card->next; } } if (!card) return -ENXIO; /* Put the adapter in the card's adapter list */ if (!card->adapter[adapter->port]) { card->adapter[adapter->port] = adapter; adapter->card = card; } card->card_size = 1; /* one port per *logical* card */ while (physcard) { for (i = 0; i < SLIC_MAX_PORTS; i++) { if (physcard->adapter[i]) break; } if (i == SLIC_MAX_PORTS) break; if (physcard->adapter[i]->slotnumber == adapter->slotnumber) break; physcard = physcard->next; } if (!physcard) { /* no structure allocated for this physical card yet */ physcard = kzalloc(sizeof(struct physcard), GFP_ATOMIC); if (!physcard) { if (card_hostid == SLIC_HOSTID_DEFAULT) kfree(card); return -ENOMEM; } physcard->next = slic_global.phys_card; slic_global.phys_card = physcard; physcard->adapters_allocd = 1; } else { physcard->adapters_allocd++; } /* Note - this is ZERO relative */ adapter->physport = physcard->adapters_allocd - 1; physcard->adapter[adapter->physport] = adapter; adapter->physcard = physcard; return 0; } static int slic_entry_probe(struct pci_dev *pcidev, const struct pci_device_id *pci_tbl_entry) { static int cards_found; static int did_version; int err = -ENODEV; struct net_device *netdev; struct adapter *adapter; void __iomem *memmapped_ioaddr = NULL; ulong mmio_start = 0; ulong mmio_len = 0; struct sliccard *card = NULL; int pci_using_dac = 0; slic_global.dynamic_intagg = dynamic_intagg; err = pci_enable_device(pcidev); if (err) return err; if (did_version++ == 0) { dev_info(&pcidev->dev, "%s\n", slic_banner); dev_info(&pcidev->dev, "%s\n", slic_proc_version); } if (!pci_set_dma_mask(pcidev, DMA_BIT_MASK(64))) { pci_using_dac = 1; err = pci_set_consistent_dma_mask(pcidev, DMA_BIT_MASK(64)); if (err) { dev_err(&pcidev->dev, "unable to obtain 64-bit DMA for consistent allocations\n"); goto err_out_disable_pci; } } else { err = pci_set_dma_mask(pcidev, DMA_BIT_MASK(32)); if (err) { dev_err(&pcidev->dev, "no usable DMA configuration\n"); goto err_out_disable_pci; } pci_using_dac = 0; pci_set_consistent_dma_mask(pcidev, DMA_BIT_MASK(32)); } err = pci_request_regions(pcidev, DRV_NAME); if (err) { dev_err(&pcidev->dev, "can't obtain PCI resources\n"); goto err_out_disable_pci; } pci_set_master(pcidev); netdev = alloc_etherdev(sizeof(struct adapter)); if (!netdev) { err = -ENOMEM; goto err_out_exit_slic_probe; } SET_NETDEV_DEV(netdev, &pcidev->dev); pci_set_drvdata(pcidev, netdev); adapter = netdev_priv(netdev); adapter->netdev = netdev; adapter->pcidev = pcidev; if (pci_using_dac) netdev->features |= NETIF_F_HIGHDMA; mmio_start = pci_resource_start(pcidev, 0); mmio_len = pci_resource_len(pcidev, 0); /* memmapped_ioaddr = (u32)ioremap_nocache(mmio_start, mmio_len);*/ memmapped_ioaddr = ioremap(mmio_start, mmio_len); if (!memmapped_ioaddr) { dev_err(&pcidev->dev, "cannot remap MMIO region %lx @ %lx\n", mmio_len, mmio_start); err = -ENOMEM; goto err_out_free_netdev; } slic_config_pci(pcidev); slic_init_driver(); slic_init_adapter(netdev, pcidev, pci_tbl_entry, memmapped_ioaddr, cards_found); err = slic_card_locate(adapter); if (err) { dev_err(&pcidev->dev, "cannot locate card\n"); goto err_out_unmap; } card = adapter->card; if (!adapter->allocated) { card->adapters_allocated++; adapter->allocated = 1; } err = slic_card_init(card, adapter); if (err) goto err_out_unmap; slic_adapter_set_hwaddr(adapter); netdev->base_addr = (unsigned long) memmapped_ioaddr; netdev->irq = adapter->irq; netdev->netdev_ops = &slic_netdev_ops; strcpy(netdev->name, "eth%d"); err = register_netdev(netdev); if (err) { dev_err(&pcidev->dev, "Cannot register net device, aborting.\n"); goto err_out_unmap; } cards_found++; return 0; err_out_unmap: iounmap(memmapped_ioaddr); err_out_free_netdev: free_netdev(netdev); err_out_exit_slic_probe: pci_release_regions(pcidev); err_out_disable_pci: pci_disable_device(pcidev); return err; } static struct pci_driver slic_driver = { .name = DRV_NAME, .id_table = slic_pci_tbl, .probe = slic_entry_probe, .remove = slic_entry_remove, }; static int __init slic_module_init(void) { slic_init_driver(); return pci_register_driver(&slic_driver); } static void __exit slic_module_cleanup(void) { pci_unregister_driver(&slic_driver); } module_init(slic_module_init); module_exit(slic_module_cleanup);