From eb571eeade2598635f813b3284d02c13a380301e Mon Sep 17 00:00:00 2001
From: Joe Lawrence <joe.lawrence@stratus.com>
Date: Wed, 2 Jul 2014 15:35:16 -0400
Subject: block,scsi: verify return pointer from blk_get_request

The blk-core dead queue checks introduce an error scenario to
blk_get_request that returns NULL if the request queue has been
shutdown. This affects the behavior for __GFP_WAIT callers, who should
verify the return value before dereferencing.

Signed-off-by: Joe Lawrence <joe.lawrence@stratus.com>
Acked-by: Jiri Kosina <jkosina@suse.cz> [for pktdvd]
Reviewed-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/block/paride/pd.c | 2 ++
 drivers/block/pktcdvd.c   | 2 ++
 drivers/scsi/scsi_error.c | 2 ++
 3 files changed, 6 insertions(+)

(limited to 'drivers')

diff --git a/drivers/block/paride/pd.c b/drivers/block/paride/pd.c
index fea7e76a00de..ca831f741d89 100644
--- a/drivers/block/paride/pd.c
+++ b/drivers/block/paride/pd.c
@@ -722,6 +722,8 @@ static int pd_special_command(struct pd_unit *disk,
 	int err = 0;
 
 	rq = blk_get_request(disk->gd->queue, READ, __GFP_WAIT);
+	if (!rq)
+		return -ENODEV;
 
 	rq->cmd_type = REQ_TYPE_SPECIAL;
 	rq->special = func;
diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index 758ac442c5b5..7fa8c80e8982 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -704,6 +704,8 @@ static int pkt_generic_packet(struct pktcdvd_device *pd, struct packet_command *
 
 	rq = blk_get_request(q, (cgc->data_direction == CGC_DATA_WRITE) ?
 			     WRITE : READ, __GFP_WAIT);
+	if (!rq)
+		return -ENODEV;
 	blk_rq_set_block_pc(rq);
 
 	if (cgc->buflen) {
diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index 5db8454474ee..4c433bf47a06 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -1960,6 +1960,8 @@ static void scsi_eh_lock_door(struct scsi_device *sdev)
 	 * request becomes available
 	 */
 	req = blk_get_request(sdev->request_queue, READ, GFP_KERNEL);
+	if (!req)
+		return;
 
 	blk_rq_set_block_pc(req);
 
-- 
cgit v1.2.3


From a492f075450f3ba87de36e5ffe92a9d0c7af9723 Mon Sep 17 00:00:00 2001
From: Joe Lawrence <joe.lawrence@stratus.com>
Date: Thu, 28 Aug 2014 08:15:21 -0600
Subject: block,scsi: fixup blk_get_request dead queue scenarios

The blk_get_request function may fail in low-memory conditions or during
device removal (even if __GFP_WAIT is set). To distinguish between these
errors, modify the blk_get_request call stack to return the appropriate
ERR_PTR. Verify that all callers check the return status and consider
IS_ERR instead of a simple NULL pointer check.

For consistency, make a similar change to the blk_mq_alloc_request leg
of blk_get_request.  It may fail if the queue is dead, or the caller was
unwilling to wait.

Signed-off-by: Joe Lawrence <joe.lawrence@stratus.com>
Acked-by: Jiri Kosina <jkosina@suse.cz> [for pktdvd]
Acked-by: Boaz Harrosh <bharrosh@panasas.com> [for osd]
Reviewed-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/block/paride/pd.c                   | 4 ++--
 drivers/block/pktcdvd.c                     | 4 ++--
 drivers/block/sx8.c                         | 2 +-
 drivers/cdrom/cdrom.c                       | 4 ++--
 drivers/ide/ide-park.c                      | 2 +-
 drivers/scsi/device_handler/scsi_dh_alua.c  | 2 +-
 drivers/scsi/device_handler/scsi_dh_emc.c   | 2 +-
 drivers/scsi/device_handler/scsi_dh_hp_sw.c | 4 ++--
 drivers/scsi/device_handler/scsi_dh_rdac.c  | 2 +-
 drivers/scsi/osd/osd_initiator.c            | 4 ++--
 drivers/scsi/osst.c                         | 2 +-
 drivers/scsi/scsi_error.c                   | 2 +-
 drivers/scsi/scsi_lib.c                     | 2 +-
 drivers/scsi/sg.c                           | 4 ++--
 drivers/scsi/st.c                           | 2 +-
 drivers/target/target_core_pscsi.c          | 2 +-
 16 files changed, 22 insertions(+), 22 deletions(-)

(limited to 'drivers')

diff --git a/drivers/block/paride/pd.c b/drivers/block/paride/pd.c
index ca831f741d89..d48715b287e6 100644
--- a/drivers/block/paride/pd.c
+++ b/drivers/block/paride/pd.c
@@ -722,8 +722,8 @@ static int pd_special_command(struct pd_unit *disk,
 	int err = 0;
 
 	rq = blk_get_request(disk->gd->queue, READ, __GFP_WAIT);
-	if (!rq)
-		return -ENODEV;
+	if (IS_ERR(rq))
+		return PTR_ERR(rq);
 
 	rq->cmd_type = REQ_TYPE_SPECIAL;
 	rq->special = func;
diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index 7fa8c80e8982..09e628dafd9d 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -704,8 +704,8 @@ static int pkt_generic_packet(struct pktcdvd_device *pd, struct packet_command *
 
 	rq = blk_get_request(q, (cgc->data_direction == CGC_DATA_WRITE) ?
 			     WRITE : READ, __GFP_WAIT);
-	if (!rq)
-		return -ENODEV;
+	if (IS_ERR(rq))
+		return PTR_ERR(rq);
 	blk_rq_set_block_pc(rq);
 
 	if (cgc->buflen) {
diff --git a/drivers/block/sx8.c b/drivers/block/sx8.c
index d5e2d12b9d9e..5d552857de41 100644
--- a/drivers/block/sx8.c
+++ b/drivers/block/sx8.c
@@ -568,7 +568,7 @@ static struct carm_request *carm_get_special(struct carm_host *host)
 		return NULL;
 
 	rq = blk_get_request(host->oob_q, WRITE /* bogus */, GFP_KERNEL);
-	if (!rq) {
+	if (IS_ERR(rq)) {
 		spin_lock_irqsave(&host->lock, flags);
 		carm_put_request(host, crq);
 		spin_unlock_irqrestore(&host->lock, flags);
diff --git a/drivers/cdrom/cdrom.c b/drivers/cdrom/cdrom.c
index 898b84bba28a..5d28a45d2960 100644
--- a/drivers/cdrom/cdrom.c
+++ b/drivers/cdrom/cdrom.c
@@ -2180,8 +2180,8 @@ static int cdrom_read_cdda_bpc(struct cdrom_device_info *cdi, __u8 __user *ubuf,
 		len = nr * CD_FRAMESIZE_RAW;
 
 		rq = blk_get_request(q, READ, GFP_KERNEL);
-		if (!rq) {
-			ret = -ENOMEM;
+		if (IS_ERR(rq)) {
+			ret = PTR_ERR(rq);
 			break;
 		}
 		blk_rq_set_block_pc(rq);
diff --git a/drivers/ide/ide-park.c b/drivers/ide/ide-park.c
index f41558a0bcd1..ca958604cda2 100644
--- a/drivers/ide/ide-park.c
+++ b/drivers/ide/ide-park.c
@@ -46,7 +46,7 @@ static void issue_park_cmd(ide_drive_t *drive, unsigned long timeout)
 	 * timeout has expired, so power management will be reenabled.
 	 */
 	rq = blk_get_request(q, READ, GFP_NOWAIT);
-	if (unlikely(!rq))
+	if (IS_ERR(rq))
 		goto out;
 
 	rq->cmd[0] = REQ_UNPARK_HEADS;
diff --git a/drivers/scsi/device_handler/scsi_dh_alua.c b/drivers/scsi/device_handler/scsi_dh_alua.c
index 7bcf67eec921..e99507ed0e3c 100644
--- a/drivers/scsi/device_handler/scsi_dh_alua.c
+++ b/drivers/scsi/device_handler/scsi_dh_alua.c
@@ -115,7 +115,7 @@ static struct request *get_alua_req(struct scsi_device *sdev,
 
 	rq = blk_get_request(q, rw, GFP_NOIO);
 
-	if (!rq) {
+	if (IS_ERR(rq)) {
 		sdev_printk(KERN_INFO, sdev,
 			    "%s: blk_get_request failed\n", __func__);
 		return NULL;
diff --git a/drivers/scsi/device_handler/scsi_dh_emc.c b/drivers/scsi/device_handler/scsi_dh_emc.c
index 6f07f7fe3aa1..84765384c47c 100644
--- a/drivers/scsi/device_handler/scsi_dh_emc.c
+++ b/drivers/scsi/device_handler/scsi_dh_emc.c
@@ -275,7 +275,7 @@ static struct request *get_req(struct scsi_device *sdev, int cmd,
 
 	rq = blk_get_request(sdev->request_queue,
 			(cmd != INQUIRY) ? WRITE : READ, GFP_NOIO);
-	if (!rq) {
+	if (IS_ERR(rq)) {
 		sdev_printk(KERN_INFO, sdev, "get_req: blk_get_request failed");
 		return NULL;
 	}
diff --git a/drivers/scsi/device_handler/scsi_dh_hp_sw.c b/drivers/scsi/device_handler/scsi_dh_hp_sw.c
index e9d9fea9e272..4ee2759f5299 100644
--- a/drivers/scsi/device_handler/scsi_dh_hp_sw.c
+++ b/drivers/scsi/device_handler/scsi_dh_hp_sw.c
@@ -117,7 +117,7 @@ static int hp_sw_tur(struct scsi_device *sdev, struct hp_sw_dh_data *h)
 
 retry:
 	req = blk_get_request(sdev->request_queue, WRITE, GFP_NOIO);
-	if (!req)
+	if (IS_ERR(req))
 		return SCSI_DH_RES_TEMP_UNAVAIL;
 
 	blk_rq_set_block_pc(req);
@@ -247,7 +247,7 @@ static int hp_sw_start_stop(struct hp_sw_dh_data *h)
 	struct request *req;
 
 	req = blk_get_request(h->sdev->request_queue, WRITE, GFP_ATOMIC);
-	if (!req)
+	if (IS_ERR(req))
 		return SCSI_DH_RES_TEMP_UNAVAIL;
 
 	blk_rq_set_block_pc(req);
diff --git a/drivers/scsi/device_handler/scsi_dh_rdac.c b/drivers/scsi/device_handler/scsi_dh_rdac.c
index 826069db9848..1b5bc9293e37 100644
--- a/drivers/scsi/device_handler/scsi_dh_rdac.c
+++ b/drivers/scsi/device_handler/scsi_dh_rdac.c
@@ -274,7 +274,7 @@ static struct request *get_rdac_req(struct scsi_device *sdev,
 
 	rq = blk_get_request(q, rw, GFP_NOIO);
 
-	if (!rq) {
+	if (IS_ERR(rq)) {
 		sdev_printk(KERN_INFO, sdev,
 				"get_rdac_req: blk_get_request failed.\n");
 		return NULL;
diff --git a/drivers/scsi/osd/osd_initiator.c b/drivers/scsi/osd/osd_initiator.c
index 5f4cbf0c4759..fd19fd8468ac 100644
--- a/drivers/scsi/osd/osd_initiator.c
+++ b/drivers/scsi/osd/osd_initiator.c
@@ -1567,8 +1567,8 @@ static struct request *_make_request(struct request_queue *q, bool has_write,
 		struct request *req;
 
 		req = blk_get_request(q, has_write ? WRITE : READ, flags);
-		if (unlikely(!req))
-			return ERR_PTR(-ENOMEM);
+		if (IS_ERR(req))
+			return req;
 
 		blk_rq_set_block_pc(req);
 		return req;
diff --git a/drivers/scsi/osst.c b/drivers/scsi/osst.c
index 0727ea7cc387..dff37a250d79 100644
--- a/drivers/scsi/osst.c
+++ b/drivers/scsi/osst.c
@@ -362,7 +362,7 @@ static int osst_execute(struct osst_request *SRpnt, const unsigned char *cmd,
 	int write = (data_direction == DMA_TO_DEVICE);
 
 	req = blk_get_request(SRpnt->stp->device->request_queue, write, GFP_KERNEL);
-	if (!req)
+	if (IS_ERR(req))
 		return DRIVER_ERROR << 24;
 
 	blk_rq_set_block_pc(req);
diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index 4c433bf47a06..a2c3d3d255a1 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -1960,7 +1960,7 @@ static void scsi_eh_lock_door(struct scsi_device *sdev)
 	 * request becomes available
 	 */
 	req = blk_get_request(sdev->request_queue, READ, GFP_KERNEL);
-	if (!req)
+	if (IS_ERR(req))
 		return;
 
 	blk_rq_set_block_pc(req);
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index ce62e8798cc8..972d0a8adf2e 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -221,7 +221,7 @@ int scsi_execute(struct scsi_device *sdev, const unsigned char *cmd,
 	int ret = DRIVER_ERROR << 24;
 
 	req = blk_get_request(sdev->request_queue, write, __GFP_WAIT);
-	if (!req)
+	if (IS_ERR(req))
 		return ret;
 	blk_rq_set_block_pc(req);
 
diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c
index 01cf88888797..60354449d9ed 100644
--- a/drivers/scsi/sg.c
+++ b/drivers/scsi/sg.c
@@ -1711,9 +1711,9 @@ sg_start_req(Sg_request *srp, unsigned char *cmd)
 	}
 
 	rq = blk_get_request(q, rw, GFP_ATOMIC);
-	if (!rq) {
+	if (IS_ERR(rq)) {
 		kfree(long_cmdp);
-		return -ENOMEM;
+		return PTR_ERR(rq);
 	}
 
 	blk_rq_set_block_pc(rq);
diff --git a/drivers/scsi/st.c b/drivers/scsi/st.c
index aff9689de0f7..59db5bfc11db 100644
--- a/drivers/scsi/st.c
+++ b/drivers/scsi/st.c
@@ -490,7 +490,7 @@ static int st_scsi_execute(struct st_request *SRpnt, const unsigned char *cmd,
 
 	req = blk_get_request(SRpnt->stp->device->request_queue, write,
 			      GFP_KERNEL);
-	if (!req)
+	if (IS_ERR(req))
 		return DRIVER_ERROR << 24;
 
 	blk_rq_set_block_pc(req);
diff --git a/drivers/target/target_core_pscsi.c b/drivers/target/target_core_pscsi.c
index 943b1dbe859a..70d9f6dabba0 100644
--- a/drivers/target/target_core_pscsi.c
+++ b/drivers/target/target_core_pscsi.c
@@ -1050,7 +1050,7 @@ pscsi_execute_cmd(struct se_cmd *cmd)
 		req = blk_get_request(pdv->pdv_sd->request_queue,
 				(data_direction == DMA_TO_DEVICE),
 				GFP_KERNEL);
-		if (!req) {
+		if (IS_ERR(req)) {
 			pr_err("PSCSI: blk_get_request() failed\n");
 			ret = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
 			goto fail;
-- 
cgit v1.2.3


From bf57229745f849e500ba69ff91e35bc8160a7373 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Sat, 13 Sep 2014 16:40:08 -0700
Subject: blk-mq: remove REQ_END

Pass an explicit parameter for the last request in a batch to ->queue_rq
instead of using a request flag.  Besides being a cleaner and non-stateful
interface this is also required for the next patch, which fixes the blk-mq
I/O submission code to not start a time too early.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/block/mtip32xx/mtip32xx.c | 3 ++-
 drivers/block/null_blk.c          | 3 ++-
 drivers/block/virtio_blk.c        | 4 ++--
 drivers/scsi/scsi_lib.c           | 3 ++-
 4 files changed, 8 insertions(+), 5 deletions(-)

(limited to 'drivers')

diff --git a/drivers/block/mtip32xx/mtip32xx.c b/drivers/block/mtip32xx/mtip32xx.c
index 5c8e7fe07745..0e2084f37c67 100644
--- a/drivers/block/mtip32xx/mtip32xx.c
+++ b/drivers/block/mtip32xx/mtip32xx.c
@@ -3775,7 +3775,8 @@ static bool mtip_check_unal_depth(struct blk_mq_hw_ctx *hctx,
 	return false;
 }
 
-static int mtip_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *rq)
+static int mtip_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *rq,
+		bool last)
 {
 	int ret;
 
diff --git a/drivers/block/null_blk.c b/drivers/block/null_blk.c
index 00d469c7f9f7..c5b7315c2c13 100644
--- a/drivers/block/null_blk.c
+++ b/drivers/block/null_blk.c
@@ -313,7 +313,8 @@ static void null_request_fn(struct request_queue *q)
 	}
 }
 
-static int null_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *rq)
+static int null_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *rq,
+		bool last)
 {
 	struct nullb_cmd *cmd = blk_mq_rq_to_pdu(rq);
 
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index 0a581400de0f..13756e016797 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -164,14 +164,14 @@ static void virtblk_done(struct virtqueue *vq)
 	spin_unlock_irqrestore(&vblk->vqs[qid].lock, flags);
 }
 
-static int virtio_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *req)
+static int virtio_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *req,
+		bool last)
 {
 	struct virtio_blk *vblk = hctx->queue->queuedata;
 	struct virtblk_req *vbr = blk_mq_rq_to_pdu(req);
 	unsigned long flags;
 	unsigned int num;
 	int qid = hctx->queue_num;
-	const bool last = (req->cmd_flags & REQ_END) != 0;
 	int err;
 	bool notify = false;
 
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 1f2bae475cb7..f1df41168391 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1855,7 +1855,8 @@ static void scsi_mq_done(struct scsi_cmnd *cmd)
 	blk_mq_complete_request(cmd->request);
 }
 
-static int scsi_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *req)
+static int scsi_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *req,
+		bool last)
 {
 	struct request_queue *q = req->q;
 	struct scsi_device *sdev = q->queuedata;
-- 
cgit v1.2.3


From e2490073cd7c3d6f6ef6e029a208edd4d38efac4 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Sat, 13 Sep 2014 16:40:09 -0700
Subject: blk-mq: call blk_mq_start_request from ->queue_rq

When we call blk_mq_start_request from the core blk-mq code before calling into
->queue_rq there is a racy window where the timeout handler can hit before we've
fully set up the driver specific part of the command.

Move the call to blk_mq_start_request into the driver so the driver can start
the request only once it is fully set up.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/block/mtip32xx/mtip32xx.c | 2 ++
 drivers/block/null_blk.c          | 2 ++
 drivers/block/virtio_blk.c        | 2 ++
 drivers/scsi/scsi_lib.c           | 1 +
 4 files changed, 7 insertions(+)

(limited to 'drivers')

diff --git a/drivers/block/mtip32xx/mtip32xx.c b/drivers/block/mtip32xx/mtip32xx.c
index 0e2084f37c67..4042440a0470 100644
--- a/drivers/block/mtip32xx/mtip32xx.c
+++ b/drivers/block/mtip32xx/mtip32xx.c
@@ -3783,6 +3783,8 @@ static int mtip_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *rq,
 	if (unlikely(mtip_check_unal_depth(hctx, rq)))
 		return BLK_MQ_RQ_QUEUE_BUSY;
 
+	blk_mq_start_request(rq);
+
 	ret = mtip_submit_request(hctx, rq);
 	if (likely(!ret))
 		return BLK_MQ_RQ_QUEUE_OK;
diff --git a/drivers/block/null_blk.c b/drivers/block/null_blk.c
index c5b7315c2c13..332ce20d45da 100644
--- a/drivers/block/null_blk.c
+++ b/drivers/block/null_blk.c
@@ -321,6 +321,8 @@ static int null_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *rq,
 	cmd->rq = rq;
 	cmd->nq = hctx->driver_data;
 
+	blk_mq_start_request(rq);
+
 	null_handle_cmd(cmd);
 	return BLK_MQ_RQ_QUEUE_OK;
 }
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index 13756e016797..83816bf6882f 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -205,6 +205,8 @@ static int virtio_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *req,
 		}
 	}
 
+	blk_mq_start_request(req);
+
 	num = blk_rq_map_sg(hctx->queue, vbr->req, vbr->sg);
 	if (num) {
 		if (rq_data_dir(vbr->req) == WRITE)
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index f1df41168391..2dcd9078de48 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1890,6 +1890,7 @@ static int scsi_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *req,
 	scsi_init_cmd_errh(cmd);
 	cmd->scsi_done = scsi_mq_done;
 
+	blk_mq_start_request(req);
 	reason = scsi_dispatch_cmd(cmd);
 	if (reason) {
 		scsi_set_blocked(cmd, reason);
-- 
cgit v1.2.3


From c8a446ad695ada43a885ec12b38411dbd190a11b Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Sat, 13 Sep 2014 16:40:10 -0700
Subject: blk-mq: rename blk_mq_end_io to blk_mq_end_request

Now that we've changed the driver API on the submission side use the
opportunity to fix up the name on the completion side to fit into the
general scheme.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/block/mtip32xx/mtip32xx.c | 4 ++--
 drivers/block/null_blk.c          | 2 +-
 drivers/block/virtio_blk.c        | 2 +-
 drivers/scsi/scsi_lib.c           | 4 ++--
 4 files changed, 6 insertions(+), 6 deletions(-)

(limited to 'drivers')

diff --git a/drivers/block/mtip32xx/mtip32xx.c b/drivers/block/mtip32xx/mtip32xx.c
index 4042440a0470..6b7e8d0fba99 100644
--- a/drivers/block/mtip32xx/mtip32xx.c
+++ b/drivers/block/mtip32xx/mtip32xx.c
@@ -247,7 +247,7 @@ static void mtip_async_complete(struct mtip_port *port,
 	if (unlikely(cmd->unaligned))
 		up(&port->cmd_slot_unal);
 
-	blk_mq_end_io(rq, status ? -EIO : 0);
+	blk_mq_end_request(rq, status ? -EIO : 0);
 }
 
 /*
@@ -3739,7 +3739,7 @@ static int mtip_submit_request(struct blk_mq_hw_ctx *hctx, struct request *rq)
 		int err;
 
 		err = mtip_send_trim(dd, blk_rq_pos(rq), blk_rq_sectors(rq));
-		blk_mq_end_io(rq, err);
+		blk_mq_end_request(rq, err);
 		return 0;
 	}
 
diff --git a/drivers/block/null_blk.c b/drivers/block/null_blk.c
index 332ce20d45da..ac50a2931044 100644
--- a/drivers/block/null_blk.c
+++ b/drivers/block/null_blk.c
@@ -177,7 +177,7 @@ static void end_cmd(struct nullb_cmd *cmd)
 {
 	switch (queue_mode)  {
 	case NULL_Q_MQ:
-		blk_mq_end_io(cmd->rq, 0);
+		blk_mq_end_request(cmd->rq, 0);
 		return;
 	case NULL_Q_RQ:
 		INIT_LIST_HEAD(&cmd->rq->queuelist);
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index 83816bf6882f..f751fc392ba9 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -135,7 +135,7 @@ static inline void virtblk_request_done(struct request *req)
 		req->errors = (error != 0);
 	}
 
-	blk_mq_end_io(req, error);
+	blk_mq_end_request(req, error);
 }
 
 static void virtblk_done(struct virtqueue *vq)
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 2dcd9078de48..73ce7d27f5c8 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -713,7 +713,7 @@ static bool scsi_end_request(struct request *req, int error,
 
 	if (req->mq_ctx) {
 		/*
-		 * In the MQ case the command gets freed by __blk_mq_end_io,
+		 * In the MQ case the command gets freed by __blk_mq_end_request,
 		 * so we have to do all cleanup that depends on it earlier.
 		 *
 		 * We also can't kick the queues from irq context, so we
@@ -721,7 +721,7 @@ static bool scsi_end_request(struct request *req, int error,
 		 */
 		scsi_mq_uninit_cmd(cmd);
 
-		__blk_mq_end_io(req, error);
+		__blk_mq_end_request(req, error);
 
 		if (scsi_target(sdev)->single_lun ||
 		    !list_empty(&sdev->host->starved_list))
-- 
cgit v1.2.3


From 0152fb6b57c4fae769ee75ea2ae670f4ff39fba9 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Sat, 13 Sep 2014 16:40:13 -0700
Subject: blk-mq: pass a reserved argument to the timeout handler

Allow blk-mq to pass an argument to the timeout handler to indicate
if we're timing out a reserved or regular command.  For many drivers
those need to be handled different.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/scsi/scsi_lib.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

(limited to 'drivers')

diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 73ce7d27f5c8..86b1156edb82 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1932,6 +1932,14 @@ out:
 	return ret;
 }
 
+static enum blk_eh_timer_return scsi_timeout(struct request *req,
+		bool reserved)
+{
+	if (reserved)
+		return BLK_EH_RESET_TIMER;
+	return scsi_times_out(req);
+}
+
 static int scsi_init_request(void *data, struct request *rq,
 		unsigned int hctx_idx, unsigned int request_idx,
 		unsigned int numa_node)
@@ -2043,7 +2051,7 @@ static struct blk_mq_ops scsi_mq_ops = {
 	.map_queue	= blk_mq_map_queue,
 	.queue_rq	= scsi_queue_rq,
 	.complete	= scsi_softirq_done,
-	.timeout	= scsi_times_out,
+	.timeout	= scsi_timeout,
 	.init_request	= scsi_init_request,
 	.exit_request	= scsi_exit_request,
 };
-- 
cgit v1.2.3


From fe052529e465daff25225aac769828baa88b7252 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Mon, 22 Sep 2014 15:59:31 +0200
Subject: scsi: move blk_mq_start_request call earlier

Some ATA drivers need the dma drain size workaround, and thus need to
call blk_mq_start_request before the S/G mapping.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/scsi/scsi_lib.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

(limited to 'drivers')

diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 86b1156edb82..5c5617ec834f 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1846,6 +1846,8 @@ static int scsi_mq_prep_fn(struct request *req)
 		next_rq->special = bidi_sdb;
 	}
 
+	blk_mq_start_request(req);
+
 	return scsi_setup_cmnd(sdev, req);
 }
 
@@ -1880,17 +1882,19 @@ static int scsi_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *req,
 	if (!scsi_host_queue_ready(q, shost, sdev))
 		goto out_dec_target_busy;
 
+
 	if (!(req->cmd_flags & REQ_DONTPREP)) {
 		ret = prep_to_mq(scsi_mq_prep_fn(req));
 		if (ret)
 			goto out_dec_host_busy;
 		req->cmd_flags |= REQ_DONTPREP;
+	} else {
+		blk_mq_start_request(req);
 	}
 
 	scsi_init_cmd_errh(cmd);
 	cmd->scsi_done = scsi_mq_done;
 
-	blk_mq_start_request(req);
 	reason = scsi_dispatch_cmd(cmd);
 	if (reason) {
 		scsi_set_blocked(cmd, reason);
-- 
cgit v1.2.3


From 180b2f95dd331010a9930a65c8a18d6d81b94dc1 Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Fri, 26 Sep 2014 19:19:56 -0400
Subject: block: Replace bi_integrity with bi_special

For commands like REQ_COPY we need a way to pass extra information along
with each bio. Like integrity metadata this information must be
available at the bottom of the stack so bi_private does not suffice.

Rename the existing bi_integrity field to bi_special and make it a union
so we can have different bio extensions for each class of command.

We previously used bi_integrity != NULL as a way to identify whether a
bio had integrity metadata or not. Introduce a REQ_INTEGRITY to be the
indicator now that bi_special can contain different things.

In addition, bio_integrity(bio) will now return a pointer to the
integrity payload (when applicable).

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/scsi/sd_dif.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'drivers')

diff --git a/drivers/scsi/sd_dif.c b/drivers/scsi/sd_dif.c
index a7a691d0af7d..29f0477a8708 100644
--- a/drivers/scsi/sd_dif.c
+++ b/drivers/scsi/sd_dif.c
@@ -383,9 +383,9 @@ void sd_dif_prepare(struct request *rq, sector_t hw_sector,
 		if (bio_flagged(bio, BIO_MAPPED_INTEGRITY))
 			break;
 
-		virt = bio->bi_integrity->bip_iter.bi_sector & 0xffffffff;
+		virt = bio_integrity(bio)->bip_iter.bi_sector & 0xffffffff;
 
-		bip_for_each_vec(iv, bio->bi_integrity, iter) {
+		bip_for_each_vec(iv, bio_integrity(bio), iter) {
 			sdt = kmap_atomic(iv.bv_page)
 				+ iv.bv_offset;
 
@@ -434,9 +434,9 @@ void sd_dif_complete(struct scsi_cmnd *scmd, unsigned int good_bytes)
 		struct bio_vec iv;
 		struct bvec_iter iter;
 
-		virt = bio->bi_integrity->bip_iter.bi_sector & 0xffffffff;
+		virt = bio_integrity(bio)->bip_iter.bi_sector & 0xffffffff;
 
-		bip_for_each_vec(iv, bio->bi_integrity, iter) {
+		bip_for_each_vec(iv, bio_integrity(bio), iter) {
 			sdt = kmap_atomic(iv.bv_page)
 				+ iv.bv_offset;
 
-- 
cgit v1.2.3


From 8492b68bc4025e7bce1d57761bd7c047efda2f81 Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Fri, 26 Sep 2014 19:19:57 -0400
Subject: block: Remove integrity tagging functions

None of the filesystems appear interested in using the integrity tagging
feature. Potentially because very few storage devices actually permit
using the application tag space.

Remove the tagging functions.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/scsi/sd_dif.c | 65 ---------------------------------------------------
 1 file changed, 65 deletions(-)

(limited to 'drivers')

diff --git a/drivers/scsi/sd_dif.c b/drivers/scsi/sd_dif.c
index 29f0477a8708..38a7778631be 100644
--- a/drivers/scsi/sd_dif.c
+++ b/drivers/scsi/sd_dif.c
@@ -128,39 +128,10 @@ static int sd_dif_type1_verify_ip(struct blk_integrity_exchg *bix)
 	return sd_dif_type1_verify(bix, sd_dif_ip_fn);
 }
 
-/*
- * Functions for interleaving and deinterleaving application tags
- */
-static void sd_dif_type1_set_tag(void *prot, void *tag_buf, unsigned int sectors)
-{
-	struct sd_dif_tuple *sdt = prot;
-	u8 *tag = tag_buf;
-	unsigned int i, j;
-
-	for (i = 0, j = 0 ; i < sectors ; i++, j += 2, sdt++) {
-		sdt->app_tag = tag[j] << 8 | tag[j+1];
-		BUG_ON(sdt->app_tag == 0xffff);
-	}
-}
-
-static void sd_dif_type1_get_tag(void *prot, void *tag_buf, unsigned int sectors)
-{
-	struct sd_dif_tuple *sdt = prot;
-	u8 *tag = tag_buf;
-	unsigned int i, j;
-
-	for (i = 0, j = 0 ; i < sectors ; i++, j += 2, sdt++) {
-		tag[j] = (sdt->app_tag & 0xff00) >> 8;
-		tag[j+1] = sdt->app_tag & 0xff;
-	}
-}
-
 static struct blk_integrity dif_type1_integrity_crc = {
 	.name			= "T10-DIF-TYPE1-CRC",
 	.generate_fn		= sd_dif_type1_generate_crc,
 	.verify_fn		= sd_dif_type1_verify_crc,
-	.get_tag_fn		= sd_dif_type1_get_tag,
-	.set_tag_fn		= sd_dif_type1_set_tag,
 	.tuple_size		= sizeof(struct sd_dif_tuple),
 	.tag_size		= 0,
 };
@@ -169,8 +140,6 @@ static struct blk_integrity dif_type1_integrity_ip = {
 	.name			= "T10-DIF-TYPE1-IP",
 	.generate_fn		= sd_dif_type1_generate_ip,
 	.verify_fn		= sd_dif_type1_verify_ip,
-	.get_tag_fn		= sd_dif_type1_get_tag,
-	.set_tag_fn		= sd_dif_type1_set_tag,
 	.tuple_size		= sizeof(struct sd_dif_tuple),
 	.tag_size		= 0,
 };
@@ -245,42 +214,10 @@ static int sd_dif_type3_verify_ip(struct blk_integrity_exchg *bix)
 	return sd_dif_type3_verify(bix, sd_dif_ip_fn);
 }
 
-static void sd_dif_type3_set_tag(void *prot, void *tag_buf, unsigned int sectors)
-{
-	struct sd_dif_tuple *sdt = prot;
-	u8 *tag = tag_buf;
-	unsigned int i, j;
-
-	for (i = 0, j = 0 ; i < sectors ; i++, j += 6, sdt++) {
-		sdt->app_tag = tag[j] << 8 | tag[j+1];
-		sdt->ref_tag = tag[j+2] << 24 | tag[j+3] << 16 |
-			tag[j+4] << 8 | tag[j+5];
-	}
-}
-
-static void sd_dif_type3_get_tag(void *prot, void *tag_buf, unsigned int sectors)
-{
-	struct sd_dif_tuple *sdt = prot;
-	u8 *tag = tag_buf;
-	unsigned int i, j;
-
-	for (i = 0, j = 0 ; i < sectors ; i++, j += 2, sdt++) {
-		tag[j] = (sdt->app_tag & 0xff00) >> 8;
-		tag[j+1] = sdt->app_tag & 0xff;
-		tag[j+2] = (sdt->ref_tag & 0xff000000) >> 24;
-		tag[j+3] = (sdt->ref_tag & 0xff0000) >> 16;
-		tag[j+4] = (sdt->ref_tag & 0xff00) >> 8;
-		tag[j+5] = sdt->ref_tag & 0xff;
-		BUG_ON(sdt->app_tag == 0xffff || sdt->ref_tag == 0xffffffff);
-	}
-}
-
 static struct blk_integrity dif_type3_integrity_crc = {
 	.name			= "T10-DIF-TYPE3-CRC",
 	.generate_fn		= sd_dif_type3_generate_crc,
 	.verify_fn		= sd_dif_type3_verify_crc,
-	.get_tag_fn		= sd_dif_type3_get_tag,
-	.set_tag_fn		= sd_dif_type3_set_tag,
 	.tuple_size		= sizeof(struct sd_dif_tuple),
 	.tag_size		= 0,
 };
@@ -289,8 +226,6 @@ static struct blk_integrity dif_type3_integrity_ip = {
 	.name			= "T10-DIF-TYPE3-IP",
 	.generate_fn		= sd_dif_type3_generate_ip,
 	.verify_fn		= sd_dif_type3_verify_ip,
-	.get_tag_fn		= sd_dif_type3_get_tag,
-	.set_tag_fn		= sd_dif_type3_set_tag,
 	.tuple_size		= sizeof(struct sd_dif_tuple),
 	.tag_size		= 0,
 };
-- 
cgit v1.2.3


From 3be91c4a3d090bd700bd6ee5bf457c1bbf189a4f Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Fri, 26 Sep 2014 19:19:59 -0400
Subject: block: Deprecate the use of the term sector in the context of block
 integrity

The protection interval is not necessarily tied to the logical block
size of a block device. Stop using the terms "sector" and "sectors".

Going forward we will use the term "seed" to describe the initial
reference tag value for a given I/O. "Interval" will be used to describe
the portion of the data buffer that a given piece of protection
information is associated with.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/scsi/sd_dif.c | 46 +++++++++++++++++++++++-----------------------
 1 file changed, 23 insertions(+), 23 deletions(-)

(limited to 'drivers')

diff --git a/drivers/scsi/sd_dif.c b/drivers/scsi/sd_dif.c
index 38a7778631be..1600270a46e5 100644
--- a/drivers/scsi/sd_dif.c
+++ b/drivers/scsi/sd_dif.c
@@ -57,16 +57,16 @@ static void sd_dif_type1_generate(struct blk_integrity_exchg *bix, csum_fn *fn)
 {
 	void *buf = bix->data_buf;
 	struct sd_dif_tuple *sdt = bix->prot_buf;
-	sector_t sector = bix->sector;
+	sector_t seed = bix->seed;
 	unsigned int i;
 
-	for (i = 0 ; i < bix->data_size ; i += bix->sector_size, sdt++) {
-		sdt->guard_tag = fn(buf, bix->sector_size);
-		sdt->ref_tag = cpu_to_be32(sector & 0xffffffff);
+	for (i = 0 ; i < bix->data_size ; i += bix->interval, sdt++) {
+		sdt->guard_tag = fn(buf, bix->interval);
+		sdt->ref_tag = cpu_to_be32(seed & 0xffffffff);
 		sdt->app_tag = 0;
 
-		buf += bix->sector_size;
-		sector++;
+		buf += bix->interval;
+		seed++;
 	}
 }
 
@@ -84,35 +84,35 @@ static int sd_dif_type1_verify(struct blk_integrity_exchg *bix, csum_fn *fn)
 {
 	void *buf = bix->data_buf;
 	struct sd_dif_tuple *sdt = bix->prot_buf;
-	sector_t sector = bix->sector;
+	sector_t seed = bix->seed;
 	unsigned int i;
 	__u16 csum;
 
-	for (i = 0 ; i < bix->data_size ; i += bix->sector_size, sdt++) {
+	for (i = 0 ; i < bix->data_size ; i += bix->interval, sdt++) {
 		/* Unwritten sectors */
 		if (sdt->app_tag == 0xffff)
 			return 0;
 
-		if (be32_to_cpu(sdt->ref_tag) != (sector & 0xffffffff)) {
+		if (be32_to_cpu(sdt->ref_tag) != (seed & 0xffffffff)) {
 			printk(KERN_ERR
 			       "%s: ref tag error on sector %lu (rcvd %u)\n",
-			       bix->disk_name, (unsigned long)sector,
+			       bix->disk_name, (unsigned long)seed,
 			       be32_to_cpu(sdt->ref_tag));
 			return -EIO;
 		}
 
-		csum = fn(buf, bix->sector_size);
+		csum = fn(buf, bix->interval);
 
 		if (sdt->guard_tag != csum) {
 			printk(KERN_ERR "%s: guard tag error on sector %lu " \
 			       "(rcvd %04x, data %04x)\n", bix->disk_name,
-			       (unsigned long)sector,
+			       (unsigned long)seed,
 			       be16_to_cpu(sdt->guard_tag), be16_to_cpu(csum));
 			return -EIO;
 		}
 
-		buf += bix->sector_size;
-		sector++;
+		buf += bix->interval;
+		seed++;
 	}
 
 	return 0;
@@ -155,12 +155,12 @@ static void sd_dif_type3_generate(struct blk_integrity_exchg *bix, csum_fn *fn)
 	struct sd_dif_tuple *sdt = bix->prot_buf;
 	unsigned int i;
 
-	for (i = 0 ; i < bix->data_size ; i += bix->sector_size, sdt++) {
-		sdt->guard_tag = fn(buf, bix->sector_size);
+	for (i = 0 ; i < bix->data_size ; i += bix->interval, sdt++) {
+		sdt->guard_tag = fn(buf, bix->interval);
 		sdt->ref_tag = 0;
 		sdt->app_tag = 0;
 
-		buf += bix->sector_size;
+		buf += bix->interval;
 	}
 }
 
@@ -178,27 +178,27 @@ static int sd_dif_type3_verify(struct blk_integrity_exchg *bix, csum_fn *fn)
 {
 	void *buf = bix->data_buf;
 	struct sd_dif_tuple *sdt = bix->prot_buf;
-	sector_t sector = bix->sector;
+	sector_t seed = bix->seed;
 	unsigned int i;
 	__u16 csum;
 
-	for (i = 0 ; i < bix->data_size ; i += bix->sector_size, sdt++) {
+	for (i = 0 ; i < bix->data_size ; i += bix->interval, sdt++) {
 		/* Unwritten sectors */
 		if (sdt->app_tag == 0xffff && sdt->ref_tag == 0xffffffff)
 			return 0;
 
-		csum = fn(buf, bix->sector_size);
+		csum = fn(buf, bix->interval);
 
 		if (sdt->guard_tag != csum) {
 			printk(KERN_ERR "%s: guard tag error on sector %lu " \
 			       "(rcvd %04x, data %04x)\n", bix->disk_name,
-			       (unsigned long)sector,
+			       (unsigned long)seed,
 			       be16_to_cpu(sdt->guard_tag), be16_to_cpu(csum));
 			return -EIO;
 		}
 
-		buf += bix->sector_size;
-		sector++;
+		buf += bix->interval;
+		seed++;
 	}
 
 	return 0;
-- 
cgit v1.2.3


From 1859308853b19c4daf4afaab910d3d52ac1ec2ff Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Fri, 26 Sep 2014 19:20:01 -0400
Subject: block: Clean up the code used to generate and verify integrity
 metadata

Instead of the "operate" parameter we pass in a seed value and a pointer
to a function that can be used to process the integrity metadata. The
generation function is changed to have a return value to fit into this
scheme.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/scsi/sd_dif.c | 106 ++++++++++++++++++++++++++------------------------
 1 file changed, 56 insertions(+), 50 deletions(-)

(limited to 'drivers')

diff --git a/drivers/scsi/sd_dif.c b/drivers/scsi/sd_dif.c
index 1600270a46e5..801c41851a01 100644
--- a/drivers/scsi/sd_dif.c
+++ b/drivers/scsi/sd_dif.c
@@ -53,42 +53,44 @@ static __u16 sd_dif_ip_fn(void *data, unsigned int len)
  * Type 1 and Type 2 protection use the same format: 16 bit guard tag,
  * 16 bit app tag, 32 bit reference tag.
  */
-static void sd_dif_type1_generate(struct blk_integrity_exchg *bix, csum_fn *fn)
+static void sd_dif_type1_generate(struct blk_integrity_iter *iter, csum_fn *fn)
 {
-	void *buf = bix->data_buf;
-	struct sd_dif_tuple *sdt = bix->prot_buf;
-	sector_t seed = bix->seed;
+	void *buf = iter->data_buf;
+	struct sd_dif_tuple *sdt = iter->prot_buf;
+	sector_t seed = iter->seed;
 	unsigned int i;
 
-	for (i = 0 ; i < bix->data_size ; i += bix->interval, sdt++) {
-		sdt->guard_tag = fn(buf, bix->interval);
+	for (i = 0 ; i < iter->data_size ; i += iter->interval, sdt++) {
+		sdt->guard_tag = fn(buf, iter->interval);
 		sdt->ref_tag = cpu_to_be32(seed & 0xffffffff);
 		sdt->app_tag = 0;
 
-		buf += bix->interval;
+		buf += iter->interval;
 		seed++;
 	}
 }
 
-static void sd_dif_type1_generate_crc(struct blk_integrity_exchg *bix)
+static int sd_dif_type1_generate_crc(struct blk_integrity_iter *iter)
 {
-	sd_dif_type1_generate(bix, sd_dif_crc_fn);
+	sd_dif_type1_generate(iter, sd_dif_crc_fn);
+	return 0;
 }
 
-static void sd_dif_type1_generate_ip(struct blk_integrity_exchg *bix)
+static int sd_dif_type1_generate_ip(struct blk_integrity_iter *iter)
 {
-	sd_dif_type1_generate(bix, sd_dif_ip_fn);
+	sd_dif_type1_generate(iter, sd_dif_ip_fn);
+	return 0;
 }
 
-static int sd_dif_type1_verify(struct blk_integrity_exchg *bix, csum_fn *fn)
+static int sd_dif_type1_verify(struct blk_integrity_iter *iter, csum_fn *fn)
 {
-	void *buf = bix->data_buf;
-	struct sd_dif_tuple *sdt = bix->prot_buf;
-	sector_t seed = bix->seed;
+	void *buf = iter->data_buf;
+	struct sd_dif_tuple *sdt = iter->prot_buf;
+	sector_t seed = iter->seed;
 	unsigned int i;
 	__u16 csum;
 
-	for (i = 0 ; i < bix->data_size ; i += bix->interval, sdt++) {
+	for (i = 0 ; i < iter->data_size ; i += iter->interval, sdt++) {
 		/* Unwritten sectors */
 		if (sdt->app_tag == 0xffff)
 			return 0;
@@ -96,36 +98,36 @@ static int sd_dif_type1_verify(struct blk_integrity_exchg *bix, csum_fn *fn)
 		if (be32_to_cpu(sdt->ref_tag) != (seed & 0xffffffff)) {
 			printk(KERN_ERR
 			       "%s: ref tag error on sector %lu (rcvd %u)\n",
-			       bix->disk_name, (unsigned long)seed,
+			       iter->disk_name, (unsigned long)seed,
 			       be32_to_cpu(sdt->ref_tag));
 			return -EIO;
 		}
 
-		csum = fn(buf, bix->interval);
+		csum = fn(buf, iter->interval);
 
 		if (sdt->guard_tag != csum) {
 			printk(KERN_ERR "%s: guard tag error on sector %lu " \
-			       "(rcvd %04x, data %04x)\n", bix->disk_name,
+			       "(rcvd %04x, data %04x)\n", iter->disk_name,
 			       (unsigned long)seed,
 			       be16_to_cpu(sdt->guard_tag), be16_to_cpu(csum));
 			return -EIO;
 		}
 
-		buf += bix->interval;
+		buf += iter->interval;
 		seed++;
 	}
 
 	return 0;
 }
 
-static int sd_dif_type1_verify_crc(struct blk_integrity_exchg *bix)
+static int sd_dif_type1_verify_crc(struct blk_integrity_iter *iter)
 {
-	return sd_dif_type1_verify(bix, sd_dif_crc_fn);
+	return sd_dif_type1_verify(iter, sd_dif_crc_fn);
 }
 
-static int sd_dif_type1_verify_ip(struct blk_integrity_exchg *bix)
+static int sd_dif_type1_verify_ip(struct blk_integrity_iter *iter)
 {
-	return sd_dif_type1_verify(bix, sd_dif_ip_fn);
+	return sd_dif_type1_verify(iter, sd_dif_ip_fn);
 }
 
 static struct blk_integrity dif_type1_integrity_crc = {
@@ -149,69 +151,71 @@ static struct blk_integrity dif_type1_integrity_ip = {
  * Type 3 protection has a 16-bit guard tag and 16 + 32 bits of opaque
  * tag space.
  */
-static void sd_dif_type3_generate(struct blk_integrity_exchg *bix, csum_fn *fn)
+static void sd_dif_type3_generate(struct blk_integrity_iter *iter, csum_fn *fn)
 {
-	void *buf = bix->data_buf;
-	struct sd_dif_tuple *sdt = bix->prot_buf;
+	void *buf = iter->data_buf;
+	struct sd_dif_tuple *sdt = iter->prot_buf;
 	unsigned int i;
 
-	for (i = 0 ; i < bix->data_size ; i += bix->interval, sdt++) {
-		sdt->guard_tag = fn(buf, bix->interval);
+	for (i = 0 ; i < iter->data_size ; i += iter->interval, sdt++) {
+		sdt->guard_tag = fn(buf, iter->interval);
 		sdt->ref_tag = 0;
 		sdt->app_tag = 0;
 
-		buf += bix->interval;
+		buf += iter->interval;
 	}
 }
 
-static void sd_dif_type3_generate_crc(struct blk_integrity_exchg *bix)
+static int sd_dif_type3_generate_crc(struct blk_integrity_iter *iter)
 {
-	sd_dif_type3_generate(bix, sd_dif_crc_fn);
+	sd_dif_type3_generate(iter, sd_dif_crc_fn);
+	return 0;
 }
 
-static void sd_dif_type3_generate_ip(struct blk_integrity_exchg *bix)
+static int sd_dif_type3_generate_ip(struct blk_integrity_iter *iter)
 {
-	sd_dif_type3_generate(bix, sd_dif_ip_fn);
+	sd_dif_type3_generate(iter, sd_dif_ip_fn);
+	return 0;
 }
 
-static int sd_dif_type3_verify(struct blk_integrity_exchg *bix, csum_fn *fn)
+static int sd_dif_type3_verify(struct blk_integrity_iter *iter, csum_fn *fn)
 {
-	void *buf = bix->data_buf;
-	struct sd_dif_tuple *sdt = bix->prot_buf;
-	sector_t seed = bix->seed;
+	void *buf = iter->data_buf;
+	struct sd_dif_tuple *sdt = iter->prot_buf;
+	sector_t seed = iter->seed;
 	unsigned int i;
 	__u16 csum;
 
-	for (i = 0 ; i < bix->data_size ; i += bix->interval, sdt++) {
+	for (i = 0 ; i < iter->data_size ; i += iter->interval, sdt++) {
 		/* Unwritten sectors */
 		if (sdt->app_tag == 0xffff && sdt->ref_tag == 0xffffffff)
 			return 0;
 
-		csum = fn(buf, bix->interval);
+		csum = fn(buf, iter->interval);
 
 		if (sdt->guard_tag != csum) {
 			printk(KERN_ERR "%s: guard tag error on sector %lu " \
-			       "(rcvd %04x, data %04x)\n", bix->disk_name,
+			       "(rcvd %04x, data %04x)\n", iter->disk_name,
 			       (unsigned long)seed,
 			       be16_to_cpu(sdt->guard_tag), be16_to_cpu(csum));
 			return -EIO;
 		}
 
-		buf += bix->interval;
+		buf += iter->interval;
 		seed++;
 	}
 
 	return 0;
 }
 
-static int sd_dif_type3_verify_crc(struct blk_integrity_exchg *bix)
+static int sd_dif_type3_verify_crc(struct blk_integrity_iter *iter)
 {
-	return sd_dif_type3_verify(bix, sd_dif_crc_fn);
+	return sd_dif_type3_verify(iter, sd_dif_crc_fn);
 }
 
-static int sd_dif_type3_verify_ip(struct blk_integrity_exchg *bix)
+static int sd_dif_type3_verify_ip(struct blk_integrity_iter *iter)
 {
-	return sd_dif_type3_verify(bix, sd_dif_ip_fn);
+	return sd_dif_type3_verify(iter, sd_dif_ip_fn);
 }
 
 static struct blk_integrity dif_type3_integrity_crc = {
@@ -310,6 +314,7 @@ void sd_dif_prepare(struct request *rq, sector_t hw_sector,
 	phys = hw_sector & 0xffffffff;
 
 	__rq_for_each_bio(bio, rq) {
+		struct bio_integrity_payload *bip = bio_integrity(bio);
 		struct bio_vec iv;
 		struct bvec_iter iter;
 		unsigned int j;
@@ -318,9 +323,9 @@ void sd_dif_prepare(struct request *rq, sector_t hw_sector,
 		if (bio_flagged(bio, BIO_MAPPED_INTEGRITY))
 			break;
 
-		virt = bio_integrity(bio)->bip_iter.bi_sector & 0xffffffff;
+		virt = bip_get_seed(bip) & 0xffffffff;
 
-		bip_for_each_vec(iv, bio_integrity(bio), iter) {
+		bip_for_each_vec(iv, bip, iter) {
 			sdt = kmap_atomic(iv.bv_page)
 				+ iv.bv_offset;
 
@@ -366,12 +371,13 @@ void sd_dif_complete(struct scsi_cmnd *scmd, unsigned int good_bytes)
 		phys >>= 3;
 
 	__rq_for_each_bio(bio, scmd->request) {
+		struct bio_integrity_payload *bip = bio_integrity(bio);
 		struct bio_vec iv;
 		struct bvec_iter iter;
 
-		virt = bio_integrity(bio)->bip_iter.bi_sector & 0xffffffff;
+		virt = bip_get_seed(bip) & 0xffffffff;
 
-		bip_for_each_vec(iv, bio_integrity(bio), iter) {
+		bip_for_each_vec(iv, bip, iter) {
 			sdt = kmap_atomic(iv.bv_page)
 				+ iv.bv_offset;
 
-- 
cgit v1.2.3


From 3aec2f41a8baeb70aa77556a4e4dcec7d9d70b4d Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Fri, 26 Sep 2014 19:20:03 -0400
Subject: block: Add a disk flag to block integrity profile

So far we have relied on the app tag size to determine whether a disk
has been formatted with T10 protection information or not. However, not
all target devices provide application tag storage.

Add a flag to the block integrity profile that indicates whether the
disk has been formatted with protection information.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@dev.mellanox.co.il>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/scsi/sd_dif.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

(limited to 'drivers')

diff --git a/drivers/scsi/sd_dif.c b/drivers/scsi/sd_dif.c
index 801c41851a01..1e971c6f8c2b 100644
--- a/drivers/scsi/sd_dif.c
+++ b/drivers/scsi/sd_dif.c
@@ -270,7 +270,13 @@ void sd_dif_config_host(struct scsi_disk *sdkp)
 		  "Enabling DIX %s protection\n", disk->integrity->name);
 
 	/* Signal to block layer that we support sector tagging */
-	if (dif && type && sdkp->ATO) {
+	if (dif && type) {
+
+		disk->integrity->flags |= BLK_INTEGRITY_DEVICE_CAPABLE;
+
+		if (!sdkp)
+			return;
+
 		if (type == SD_DIF_TYPE3_PROTECTION)
 			disk->integrity->tag_size = sizeof(u16) + sizeof(u32);
 		else
-- 
cgit v1.2.3


From b1f01388574c9329922f760fc2a7335c2d14b08b Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Fri, 26 Sep 2014 19:20:04 -0400
Subject: block: Relocate bio integrity flags

Move flags affecting the integrity code out of the bio bi_flags and into
the block integrity payload.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/scsi/sd_dif.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'drivers')

diff --git a/drivers/scsi/sd_dif.c b/drivers/scsi/sd_dif.c
index 1e971c6f8c2b..4ce636fdc15f 100644
--- a/drivers/scsi/sd_dif.c
+++ b/drivers/scsi/sd_dif.c
@@ -326,7 +326,7 @@ void sd_dif_prepare(struct request *rq, sector_t hw_sector,
 		unsigned int j;
 
 		/* Already remapped? */
-		if (bio_flagged(bio, BIO_MAPPED_INTEGRITY))
+		if (bip->bip_flags & BIP_MAPPED_INTEGRITY)
 			break;
 
 		virt = bip_get_seed(bip) & 0xffffffff;
@@ -347,7 +347,7 @@ void sd_dif_prepare(struct request *rq, sector_t hw_sector,
 			kunmap_atomic(sdt);
 		}
 
-		bio->bi_flags |= (1 << BIO_MAPPED_INTEGRITY);
+		bip->bip_flags |= BIP_MAPPED_INTEGRITY;
 	}
 }
 
-- 
cgit v1.2.3


From aae7df50190a640e51bfe11c93f94741ac82ff0b Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Fri, 26 Sep 2014 19:20:05 -0400
Subject: block: Integrity checksum flag

Make the choice of checksum a per-I/O property by introducing a flag
that can be inspected by the SCSI layer. There are several reasons for
this:

 1. It allows us to switch choice of checksum without unloading and
    reloading the HBA driver.

 2. During error recovery we need to be able to tell the HBA that
    checksums read from disk should not be verified and converted to IP
    checksums.

 3. For error injection purposes we need to be able to write a bad guard
    tag to storage. Since the storage device only supports T10 CRC we
    need to be able to disable IP checksum conversion on the HBA.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/scsi/sd_dif.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

(limited to 'drivers')

diff --git a/drivers/scsi/sd_dif.c b/drivers/scsi/sd_dif.c
index 4ce636fdc15f..2198abee619e 100644
--- a/drivers/scsi/sd_dif.c
+++ b/drivers/scsi/sd_dif.c
@@ -255,12 +255,14 @@ void sd_dif_config_host(struct scsi_disk *sdkp)
 		return;
 
 	/* Enable DMA of protection information */
-	if (scsi_host_get_guard(sdkp->device->host) & SHOST_DIX_GUARD_IP)
+	if (scsi_host_get_guard(sdkp->device->host) & SHOST_DIX_GUARD_IP) {
 		if (type == SD_DIF_TYPE3_PROTECTION)
 			blk_integrity_register(disk, &dif_type3_integrity_ip);
 		else
 			blk_integrity_register(disk, &dif_type1_integrity_ip);
-	else
+
+		disk->integrity->flags |= BLK_INTEGRITY_IP_CHECKSUM;
+	} else
 		if (type == SD_DIF_TYPE3_PROTECTION)
 			blk_integrity_register(disk, &dif_type3_integrity_crc);
 		else
-- 
cgit v1.2.3


From 2341c2f8c33196d02cf5a721746eea4e3c06674a Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Fri, 26 Sep 2014 19:20:07 -0400
Subject: block: Add T10 Protection Information functions

The T10 Protection Information format is also used by some devices that
do not go through the SCSI layer (virtual block devices, NVMe). Relocate
the relevant functions to a block layer library that can be used without
involving SCSI.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/scsi/Kconfig  |   1 -
 drivers/scsi/sd_dif.c | 241 +++++---------------------------------------------
 2 files changed, 20 insertions(+), 222 deletions(-)

(limited to 'drivers')

diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
index 18a3358eb1d4..9ece13f922a6 100644
--- a/drivers/scsi/Kconfig
+++ b/drivers/scsi/Kconfig
@@ -62,7 +62,6 @@ comment "SCSI support type (disk, tape, CD-ROM)"
 config BLK_DEV_SD
 	tristate "SCSI disk support"
 	depends on SCSI
-	select CRC_T10DIF if BLK_DEV_INTEGRITY
 	---help---
 	  If you want to use SCSI hard disks, Fibre Channel disks,
 	  Serial ATA (SATA) or Parallel ATA (PATA) hard disks,
diff --git a/drivers/scsi/sd_dif.c b/drivers/scsi/sd_dif.c
index 2198abee619e..b7eaeadc18f9 100644
--- a/drivers/scsi/sd_dif.c
+++ b/drivers/scsi/sd_dif.c
@@ -21,7 +21,7 @@
  */
 
 #include <linux/blkdev.h>
-#include <linux/crc-t10dif.h>
+#include <linux/t10-pi.h>
 
 #include <scsi/scsi.h>
 #include <scsi/scsi_cmnd.h>
@@ -33,207 +33,8 @@
 #include <scsi/scsi_ioctl.h>
 #include <scsi/scsicam.h>
 
-#include <net/checksum.h>
-
 #include "sd.h"
 
-typedef __u16 (csum_fn) (void *, unsigned int);
-
-static __u16 sd_dif_crc_fn(void *data, unsigned int len)
-{
-	return cpu_to_be16(crc_t10dif(data, len));
-}
-
-static __u16 sd_dif_ip_fn(void *data, unsigned int len)
-{
-	return ip_compute_csum(data, len);
-}
-
-/*
- * Type 1 and Type 2 protection use the same format: 16 bit guard tag,
- * 16 bit app tag, 32 bit reference tag.
- */
-static void sd_dif_type1_generate(struct blk_integrity_iter *iter, csum_fn *fn)
-{
-	void *buf = iter->data_buf;
-	struct sd_dif_tuple *sdt = iter->prot_buf;
-	sector_t seed = iter->seed;
-	unsigned int i;
-
-	for (i = 0 ; i < iter->data_size ; i += iter->interval, sdt++) {
-		sdt->guard_tag = fn(buf, iter->interval);
-		sdt->ref_tag = cpu_to_be32(seed & 0xffffffff);
-		sdt->app_tag = 0;
-
-		buf += iter->interval;
-		seed++;
-	}
-}
-
-static int sd_dif_type1_generate_crc(struct blk_integrity_iter *iter)
-{
-	sd_dif_type1_generate(iter, sd_dif_crc_fn);
-	return 0;
-}
-
-static int sd_dif_type1_generate_ip(struct blk_integrity_iter *iter)
-{
-	sd_dif_type1_generate(iter, sd_dif_ip_fn);
-	return 0;
-}
-
-static int sd_dif_type1_verify(struct blk_integrity_iter *iter, csum_fn *fn)
-{
-	void *buf = iter->data_buf;
-	struct sd_dif_tuple *sdt = iter->prot_buf;
-	sector_t seed = iter->seed;
-	unsigned int i;
-	__u16 csum;
-
-	for (i = 0 ; i < iter->data_size ; i += iter->interval, sdt++) {
-		/* Unwritten sectors */
-		if (sdt->app_tag == 0xffff)
-			return 0;
-
-		if (be32_to_cpu(sdt->ref_tag) != (seed & 0xffffffff)) {
-			printk(KERN_ERR
-			       "%s: ref tag error on sector %lu (rcvd %u)\n",
-			       iter->disk_name, (unsigned long)seed,
-			       be32_to_cpu(sdt->ref_tag));
-			return -EIO;
-		}
-
-		csum = fn(buf, iter->interval);
-
-		if (sdt->guard_tag != csum) {
-			printk(KERN_ERR "%s: guard tag error on sector %lu " \
-			       "(rcvd %04x, data %04x)\n", iter->disk_name,
-			       (unsigned long)seed,
-			       be16_to_cpu(sdt->guard_tag), be16_to_cpu(csum));
-			return -EIO;
-		}
-
-		buf += iter->interval;
-		seed++;
-	}
-
-	return 0;
-}
-
-static int sd_dif_type1_verify_crc(struct blk_integrity_iter *iter)
-{
-	return sd_dif_type1_verify(iter, sd_dif_crc_fn);
-}
-
-static int sd_dif_type1_verify_ip(struct blk_integrity_iter *iter)
-{
-	return sd_dif_type1_verify(iter, sd_dif_ip_fn);
-}
-
-static struct blk_integrity dif_type1_integrity_crc = {
-	.name			= "T10-DIF-TYPE1-CRC",
-	.generate_fn		= sd_dif_type1_generate_crc,
-	.verify_fn		= sd_dif_type1_verify_crc,
-	.tuple_size		= sizeof(struct sd_dif_tuple),
-	.tag_size		= 0,
-};
-
-static struct blk_integrity dif_type1_integrity_ip = {
-	.name			= "T10-DIF-TYPE1-IP",
-	.generate_fn		= sd_dif_type1_generate_ip,
-	.verify_fn		= sd_dif_type1_verify_ip,
-	.tuple_size		= sizeof(struct sd_dif_tuple),
-	.tag_size		= 0,
-};
-
-
-/*
- * Type 3 protection has a 16-bit guard tag and 16 + 32 bits of opaque
- * tag space.
- */
-static void sd_dif_type3_generate(struct blk_integrity_iter *iter, csum_fn *fn)
-{
-	void *buf = iter->data_buf;
-	struct sd_dif_tuple *sdt = iter->prot_buf;
-	unsigned int i;
-
-	for (i = 0 ; i < iter->data_size ; i += iter->interval, sdt++) {
-		sdt->guard_tag = fn(buf, iter->interval);
-		sdt->ref_tag = 0;
-		sdt->app_tag = 0;
-
-		buf += iter->interval;
-	}
-}
-
-static int sd_dif_type3_generate_crc(struct blk_integrity_iter *iter)
-{
-	sd_dif_type3_generate(iter, sd_dif_crc_fn);
-	return 0;
-}
-
-static int sd_dif_type3_generate_ip(struct blk_integrity_iter *iter)
-{
-	sd_dif_type3_generate(iter, sd_dif_ip_fn);
-	return 0;
-}
-
-static int sd_dif_type3_verify(struct blk_integrity_iter *iter, csum_fn *fn)
-{
-	void *buf = iter->data_buf;
-	struct sd_dif_tuple *sdt = iter->prot_buf;
-	sector_t seed = iter->seed;
-	unsigned int i;
-	__u16 csum;
-
-	for (i = 0 ; i < iter->data_size ; i += iter->interval, sdt++) {
-		/* Unwritten sectors */
-		if (sdt->app_tag == 0xffff && sdt->ref_tag == 0xffffffff)
-			return 0;
-
-		csum = fn(buf, iter->interval);
-
-		if (sdt->guard_tag != csum) {
-			printk(KERN_ERR "%s: guard tag error on sector %lu " \
-			       "(rcvd %04x, data %04x)\n", iter->disk_name,
-			       (unsigned long)seed,
-			       be16_to_cpu(sdt->guard_tag), be16_to_cpu(csum));
-			return -EIO;
-		}
-
-		buf += iter->interval;
-		seed++;
-	}
-
-	return 0;
-}
-
-static int sd_dif_type3_verify_crc(struct blk_integrity_iter *iter)
-{
-	return sd_dif_type3_verify(iter, sd_dif_crc_fn);
-}
-
-static int sd_dif_type3_verify_ip(struct blk_integrity_iter *iter)
-{
-	return sd_dif_type3_verify(iter, sd_dif_ip_fn);
-}
-
-static struct blk_integrity dif_type3_integrity_crc = {
-	.name			= "T10-DIF-TYPE3-CRC",
-	.generate_fn		= sd_dif_type3_generate_crc,
-	.verify_fn		= sd_dif_type3_verify_crc,
-	.tuple_size		= sizeof(struct sd_dif_tuple),
-	.tag_size		= 0,
-};
-
-static struct blk_integrity dif_type3_integrity_ip = {
-	.name			= "T10-DIF-TYPE3-IP",
-	.generate_fn		= sd_dif_type3_generate_ip,
-	.verify_fn		= sd_dif_type3_verify_ip,
-	.tuple_size		= sizeof(struct sd_dif_tuple),
-	.tag_size		= 0,
-};
-
 /*
  * Configure exchange of protection information between OS and HBA.
  */
@@ -257,16 +58,16 @@ void sd_dif_config_host(struct scsi_disk *sdkp)
 	/* Enable DMA of protection information */
 	if (scsi_host_get_guard(sdkp->device->host) & SHOST_DIX_GUARD_IP) {
 		if (type == SD_DIF_TYPE3_PROTECTION)
-			blk_integrity_register(disk, &dif_type3_integrity_ip);
+			blk_integrity_register(disk, &t10_pi_type3_ip);
 		else
-			blk_integrity_register(disk, &dif_type1_integrity_ip);
+			blk_integrity_register(disk, &t10_pi_type1_ip);
 
 		disk->integrity->flags |= BLK_INTEGRITY_IP_CHECKSUM;
 	} else
 		if (type == SD_DIF_TYPE3_PROTECTION)
-			blk_integrity_register(disk, &dif_type3_integrity_crc);
+			blk_integrity_register(disk, &t10_pi_type3_crc);
 		else
-			blk_integrity_register(disk, &dif_type1_integrity_crc);
+			blk_integrity_register(disk, &t10_pi_type1_crc);
 
 	sd_printk(KERN_NOTICE, sdkp,
 		  "Enabling DIX %s protection\n", disk->integrity->name);
@@ -308,10 +109,10 @@ void sd_dif_config_host(struct scsi_disk *sdkp)
 void sd_dif_prepare(struct request *rq, sector_t hw_sector,
 		    unsigned int sector_sz)
 {
-	const int tuple_sz = sizeof(struct sd_dif_tuple);
+	const int tuple_sz = sizeof(struct t10_pi_tuple);
 	struct bio *bio;
 	struct scsi_disk *sdkp;
-	struct sd_dif_tuple *sdt;
+	struct t10_pi_tuple *pi;
 	u32 phys, virt;
 
 	sdkp = rq->bio->bi_bdev->bd_disk->private_data;
@@ -334,19 +135,18 @@ void sd_dif_prepare(struct request *rq, sector_t hw_sector,
 		virt = bip_get_seed(bip) & 0xffffffff;
 
 		bip_for_each_vec(iv, bip, iter) {
-			sdt = kmap_atomic(iv.bv_page)
-				+ iv.bv_offset;
+			pi = kmap_atomic(iv.bv_page) + iv.bv_offset;
 
-			for (j = 0; j < iv.bv_len; j += tuple_sz, sdt++) {
+			for (j = 0; j < iv.bv_len; j += tuple_sz, pi++) {
 
-				if (be32_to_cpu(sdt->ref_tag) == virt)
-					sdt->ref_tag = cpu_to_be32(phys);
+				if (be32_to_cpu(pi->ref_tag) == virt)
+					pi->ref_tag = cpu_to_be32(phys);
 
 				virt++;
 				phys++;
 			}
 
-			kunmap_atomic(sdt);
+			kunmap_atomic(pi);
 		}
 
 		bip->bip_flags |= BIP_MAPPED_INTEGRITY;
@@ -359,10 +159,10 @@ void sd_dif_prepare(struct request *rq, sector_t hw_sector,
  */
 void sd_dif_complete(struct scsi_cmnd *scmd, unsigned int good_bytes)
 {
-	const int tuple_sz = sizeof(struct sd_dif_tuple);
+	const int tuple_sz = sizeof(struct t10_pi_tuple);
 	struct scsi_disk *sdkp;
 	struct bio *bio;
-	struct sd_dif_tuple *sdt;
+	struct t10_pi_tuple *pi;
 	unsigned int j, sectors, sector_sz;
 	u32 phys, virt;
 
@@ -386,25 +186,24 @@ void sd_dif_complete(struct scsi_cmnd *scmd, unsigned int good_bytes)
 		virt = bip_get_seed(bip) & 0xffffffff;
 
 		bip_for_each_vec(iv, bip, iter) {
-			sdt = kmap_atomic(iv.bv_page)
-				+ iv.bv_offset;
+			pi = kmap_atomic(iv.bv_page) + iv.bv_offset;
 
-			for (j = 0; j < iv.bv_len; j += tuple_sz, sdt++) {
+			for (j = 0; j < iv.bv_len; j += tuple_sz, pi++) {
 
 				if (sectors == 0) {
-					kunmap_atomic(sdt);
+					kunmap_atomic(pi);
 					return;
 				}
 
-				if (be32_to_cpu(sdt->ref_tag) == phys)
-					sdt->ref_tag = cpu_to_be32(virt);
+				if (be32_to_cpu(pi->ref_tag) == phys)
+					pi->ref_tag = cpu_to_be32(virt);
 
 				virt++;
 				phys++;
 				sectors--;
 			}
 
-			kunmap_atomic(sdt);
+			kunmap_atomic(pi);
 		}
 	}
 }
-- 
cgit v1.2.3


From c611529e7cd3465ec0eada0f44200e8420c38908 Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Fri, 26 Sep 2014 19:20:08 -0400
Subject: sd: Honor block layer integrity handling flags

A set of flags introduced in the block layer enable better control over
how protection information is handled. These flags are useful for both
error injection and data recovery purposes. Checking can be enabled and
disabled for controller and disk, and the guard tag format is now a
per-I/O property.

Update sd_protect_op to communicate the relevant information to the
low-level device driver via a set of flags in scsi_cmnd.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/scsi/sd.c     | 73 ++++++++++++++++++++++++++++++---------------------
 drivers/scsi/sd.h     | 66 ++++++++++++++++++++++++++++++++++++++++++++--
 drivers/scsi/sd_dif.c | 23 +++++++---------
 3 files changed, 116 insertions(+), 46 deletions(-)

(limited to 'drivers')

diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index 2c2041ca4b70..9f7099f4b537 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -610,29 +610,44 @@ static void scsi_disk_put(struct scsi_disk *sdkp)
 	mutex_unlock(&sd_ref_mutex);
 }
 
-static void sd_prot_op(struct scsi_cmnd *scmd, unsigned int dif)
-{
-	unsigned int prot_op = SCSI_PROT_NORMAL;
-	unsigned int dix = scsi_prot_sg_count(scmd);
-
-	if (scmd->sc_data_direction == DMA_FROM_DEVICE) {
-		if (dif && dix)
-			prot_op = SCSI_PROT_READ_PASS;
-		else if (dif && !dix)
-			prot_op = SCSI_PROT_READ_STRIP;
-		else if (!dif && dix)
-			prot_op = SCSI_PROT_READ_INSERT;
-	} else {
-		if (dif && dix)
-			prot_op = SCSI_PROT_WRITE_PASS;
-		else if (dif && !dix)
-			prot_op = SCSI_PROT_WRITE_INSERT;
-		else if (!dif && dix)
-			prot_op = SCSI_PROT_WRITE_STRIP;
+
+
+static unsigned char sd_setup_protect_cmnd(struct scsi_cmnd *scmd,
+					   unsigned int dix, unsigned int dif)
+{
+	struct bio *bio = scmd->request->bio;
+	unsigned int prot_op = sd_prot_op(rq_data_dir(scmd->request), dix, dif);
+	unsigned int protect = 0;
+
+	if (dix) {				/* DIX Type 0, 1, 2, 3 */
+		if (bio_integrity_flagged(bio, BIP_IP_CHECKSUM))
+			scmd->prot_flags |= SCSI_PROT_IP_CHECKSUM;
+
+		if (bio_integrity_flagged(bio, BIP_CTRL_NOCHECK) == false)
+			scmd->prot_flags |= SCSI_PROT_GUARD_CHECK;
+	}
+
+	if (dif != SD_DIF_TYPE3_PROTECTION) {	/* DIX/DIF Type 0, 1, 2 */
+		scmd->prot_flags |= SCSI_PROT_REF_INCREMENT;
+
+		if (bio_integrity_flagged(bio, BIP_CTRL_NOCHECK) == false)
+			scmd->prot_flags |= SCSI_PROT_REF_CHECK;
+	}
+
+	if (dif) {				/* DIX/DIF Type 1, 2, 3 */
+		scmd->prot_flags |= SCSI_PROT_TRANSFER_PI;
+
+		if (bio_integrity_flagged(bio, BIP_DISK_NOCHECK))
+			protect = 3 << 5;	/* Disable target PI checking */
+		else
+			protect = 1 << 5;	/* Enable target PI checking */
 	}
 
 	scsi_set_prot_op(scmd, prot_op);
 	scsi_set_prot_type(scmd, dif);
+	scmd->prot_flags &= sd_prot_flag_mask(prot_op);
+
+	return protect;
 }
 
 static void sd_config_discard(struct scsi_disk *sdkp, unsigned int mode)
@@ -893,7 +908,8 @@ static int sd_setup_read_write_cmnd(struct scsi_cmnd *SCpnt)
 	sector_t block = blk_rq_pos(rq);
 	sector_t threshold;
 	unsigned int this_count = blk_rq_sectors(rq);
-	int ret, host_dif;
+	unsigned int dif, dix;
+	int ret;
 	unsigned char protect;
 
 	ret = scsi_init_io(SCpnt, GFP_ATOMIC);
@@ -995,7 +1011,7 @@ static int sd_setup_read_write_cmnd(struct scsi_cmnd *SCpnt)
 		SCpnt->cmnd[0] = WRITE_6;
 
 		if (blk_integrity_rq(rq))
-			sd_dif_prepare(rq, block, sdp->sector_size);
+			sd_dif_prepare(SCpnt);
 
 	} else if (rq_data_dir(rq) == READ) {
 		SCpnt->cmnd[0] = READ_6;
@@ -1010,14 +1026,15 @@ static int sd_setup_read_write_cmnd(struct scsi_cmnd *SCpnt)
 					"writing" : "reading", this_count,
 					blk_rq_sectors(rq)));
 
-	/* Set RDPROTECT/WRPROTECT if disk is formatted with DIF */
-	host_dif = scsi_host_dif_capable(sdp->host, sdkp->protection_type);
-	if (host_dif)
-		protect = 1 << 5;
+	dix = scsi_prot_sg_count(SCpnt);
+	dif = scsi_host_dif_capable(SCpnt->device->host, sdkp->protection_type);
+
+	if (dif || dix)
+		protect = sd_setup_protect_cmnd(SCpnt, dix, dif);
 	else
 		protect = 0;
 
-	if (host_dif == SD_DIF_TYPE2_PROTECTION) {
+	if (protect && sdkp->protection_type == SD_DIF_TYPE2_PROTECTION) {
 		SCpnt->cmnd = mempool_alloc(sd_cdb_pool, GFP_ATOMIC);
 
 		if (unlikely(SCpnt->cmnd == NULL)) {
@@ -1102,10 +1119,6 @@ static int sd_setup_read_write_cmnd(struct scsi_cmnd *SCpnt)
 	}
 	SCpnt->sdb.length = this_count * sdp->sector_size;
 
-	/* If DIF or DIX is enabled, tell HBA how to handle request */
-	if (host_dif || scsi_prot_sg_count(SCpnt))
-		sd_prot_op(SCpnt, host_dif);
-
 	/*
 	 * We shouldn't disconnect in the middle of a sector, so with a dumb
 	 * host adapter, it's safe to assume that we can at least transfer
diff --git a/drivers/scsi/sd.h b/drivers/scsi/sd.h
index 4c3ab8377fd3..467377884b63 100644
--- a/drivers/scsi/sd.h
+++ b/drivers/scsi/sd.h
@@ -166,6 +166,68 @@ enum sd_dif_target_protection_types {
 	SD_DIF_TYPE3_PROTECTION = 0x3,
 };
 
+/*
+ * Look up the DIX operation based on whether the command is read or
+ * write and whether dix and dif are enabled.
+ */
+static inline unsigned int sd_prot_op(bool write, bool dix, bool dif)
+{
+	/* Lookup table: bit 2 (write), bit 1 (dix), bit 0 (dif) */
+	const unsigned int ops[] = {	/* wrt dix dif */
+		SCSI_PROT_NORMAL,	/*  0	0   0  */
+		SCSI_PROT_READ_STRIP,	/*  0	0   1  */
+		SCSI_PROT_READ_INSERT,	/*  0	1   0  */
+		SCSI_PROT_READ_PASS,	/*  0	1   1  */
+		SCSI_PROT_NORMAL,	/*  1	0   0  */
+		SCSI_PROT_WRITE_INSERT, /*  1	0   1  */
+		SCSI_PROT_WRITE_STRIP,	/*  1	1   0  */
+		SCSI_PROT_WRITE_PASS,	/*  1	1   1  */
+	};
+
+	return ops[write << 2 | dix << 1 | dif];
+}
+
+/*
+ * Returns a mask of the protection flags that are valid for a given DIX
+ * operation.
+ */
+static inline unsigned int sd_prot_flag_mask(unsigned int prot_op)
+{
+	const unsigned int flag_mask[] = {
+		[SCSI_PROT_NORMAL]		= 0,
+
+		[SCSI_PROT_READ_STRIP]		= SCSI_PROT_TRANSFER_PI |
+						  SCSI_PROT_GUARD_CHECK |
+						  SCSI_PROT_REF_CHECK |
+						  SCSI_PROT_REF_INCREMENT,
+
+		[SCSI_PROT_READ_INSERT]		= SCSI_PROT_REF_INCREMENT |
+						  SCSI_PROT_IP_CHECKSUM,
+
+		[SCSI_PROT_READ_PASS]		= SCSI_PROT_TRANSFER_PI |
+						  SCSI_PROT_GUARD_CHECK |
+						  SCSI_PROT_REF_CHECK |
+						  SCSI_PROT_REF_INCREMENT |
+						  SCSI_PROT_IP_CHECKSUM,
+
+		[SCSI_PROT_WRITE_INSERT]	= SCSI_PROT_TRANSFER_PI |
+						  SCSI_PROT_REF_INCREMENT,
+
+		[SCSI_PROT_WRITE_STRIP]		= SCSI_PROT_GUARD_CHECK |
+						  SCSI_PROT_REF_CHECK |
+						  SCSI_PROT_REF_INCREMENT |
+						  SCSI_PROT_IP_CHECKSUM,
+
+		[SCSI_PROT_WRITE_PASS]		= SCSI_PROT_TRANSFER_PI |
+						  SCSI_PROT_GUARD_CHECK |
+						  SCSI_PROT_REF_CHECK |
+						  SCSI_PROT_REF_INCREMENT |
+						  SCSI_PROT_IP_CHECKSUM,
+	};
+
+	return flag_mask[prot_op];
+}
+
 /*
  * Data Integrity Field tuple.
  */
@@ -178,7 +240,7 @@ struct sd_dif_tuple {
 #ifdef CONFIG_BLK_DEV_INTEGRITY
 
 extern void sd_dif_config_host(struct scsi_disk *);
-extern void sd_dif_prepare(struct request *rq, sector_t, unsigned int);
+extern void sd_dif_prepare(struct scsi_cmnd *scmd);
 extern void sd_dif_complete(struct scsi_cmnd *, unsigned int);
 
 #else /* CONFIG_BLK_DEV_INTEGRITY */
@@ -186,7 +248,7 @@ extern void sd_dif_complete(struct scsi_cmnd *, unsigned int);
 static inline void sd_dif_config_host(struct scsi_disk *disk)
 {
 }
-static inline int sd_dif_prepare(struct request *rq, sector_t s, unsigned int a)
+static inline int sd_dif_prepare(struct scsi_cmnd *scmd)
 {
 	return 0;
 }
diff --git a/drivers/scsi/sd_dif.c b/drivers/scsi/sd_dif.c
index b7eaeadc18f9..14c7d42a11c2 100644
--- a/drivers/scsi/sd_dif.c
+++ b/drivers/scsi/sd_dif.c
@@ -106,8 +106,7 @@ void sd_dif_config_host(struct scsi_disk *sdkp)
  *
  * Type 3 does not have a reference tag so no remapping is required.
  */
-void sd_dif_prepare(struct request *rq, sector_t hw_sector,
-		    unsigned int sector_sz)
+void sd_dif_prepare(struct scsi_cmnd *scmd)
 {
 	const int tuple_sz = sizeof(struct t10_pi_tuple);
 	struct bio *bio;
@@ -115,14 +114,14 @@ void sd_dif_prepare(struct request *rq, sector_t hw_sector,
 	struct t10_pi_tuple *pi;
 	u32 phys, virt;
 
-	sdkp = rq->bio->bi_bdev->bd_disk->private_data;
+	sdkp = scsi_disk(scmd->request->rq_disk);
 
 	if (sdkp->protection_type == SD_DIF_TYPE3_PROTECTION)
 		return;
 
-	phys = hw_sector & 0xffffffff;
+	phys = scsi_prot_ref_tag(scmd);
 
-	__rq_for_each_bio(bio, rq) {
+	__rq_for_each_bio(bio, scmd->request) {
 		struct bio_integrity_payload *bip = bio_integrity(bio);
 		struct bio_vec iv;
 		struct bvec_iter iter;
@@ -163,7 +162,7 @@ void sd_dif_complete(struct scsi_cmnd *scmd, unsigned int good_bytes)
 	struct scsi_disk *sdkp;
 	struct bio *bio;
 	struct t10_pi_tuple *pi;
-	unsigned int j, sectors, sector_sz;
+	unsigned int j, intervals;
 	u32 phys, virt;
 
 	sdkp = scsi_disk(scmd->request->rq_disk);
@@ -171,12 +170,8 @@ void sd_dif_complete(struct scsi_cmnd *scmd, unsigned int good_bytes)
 	if (sdkp->protection_type == SD_DIF_TYPE3_PROTECTION || good_bytes == 0)
 		return;
 
-	sector_sz = scmd->device->sector_size;
-	sectors = good_bytes / sector_sz;
-
-	phys = blk_rq_pos(scmd->request) & 0xffffffff;
-	if (sector_sz == 4096)
-		phys >>= 3;
+	intervals = good_bytes / scsi_prot_interval(scmd);
+	phys = scsi_prot_ref_tag(scmd);
 
 	__rq_for_each_bio(bio, scmd->request) {
 		struct bio_integrity_payload *bip = bio_integrity(bio);
@@ -190,7 +185,7 @@ void sd_dif_complete(struct scsi_cmnd *scmd, unsigned int good_bytes)
 
 			for (j = 0; j < iv.bv_len; j += tuple_sz, pi++) {
 
-				if (sectors == 0) {
+				if (intervals == 0) {
 					kunmap_atomic(pi);
 					return;
 				}
@@ -200,7 +195,7 @@ void sd_dif_complete(struct scsi_cmnd *scmd, unsigned int good_bytes)
 
 				virt++;
 				phys++;
-				sectors--;
+				intervals--;
 			}
 
 			kunmap_atomic(pi);
-- 
cgit v1.2.3