From 4f8c9510ba71bb54477841bebb90154ef140860f Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Fri, 17 Apr 2015 22:37:16 +0200
Subject: block: rename REQ_TYPE_SPECIAL to REQ_TYPE_DRV_PRIV

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 include/linux/blkdev.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux/blkdev.h')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 7f9a516f24de..98c90272443b 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -79,10 +79,10 @@ enum rq_cmd_type_bits {
 	REQ_TYPE_PM_SUSPEND,		/* suspend request */
 	REQ_TYPE_PM_RESUME,		/* resume request */
 	REQ_TYPE_PM_SHUTDOWN,		/* shutdown request */
-	REQ_TYPE_SPECIAL,		/* driver defined type */
+	REQ_TYPE_DRV_PRIV,		/* driver defined type */
 	/*
 	 * for ATA/ATAPI devices. this really doesn't belong here, ide should
-	 * use REQ_TYPE_SPECIAL and use rq->cmd[0] with the range of driver
+	 * use REQ_TYPE_DRV_PRIV and use rq->cmd[0] with the range of driver
 	 * private REQ_LB opcodes to differentiate what type of request this is
 	 */
 	REQ_TYPE_ATA_TASKFILE,
-- 
cgit v1.2.3-71-gd317


From b42171ef7d938a66fa52e66a3d911ed63770b5ca Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Fri, 17 Apr 2015 22:37:17 +0200
Subject: block: move REQ_TYPE_ATA_TASKFILE and REQ_TYPE_ATA_PC to ide.h

These values are only used by the IDE driver, so move them into it
by allowing drivers to take cmd_type values after the first private
one.  Note that we have to turn cmd_type into a plain unsigned integer
so that gcc doesn't complain about mismatching enum types.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 include/linux/blkdev.h | 11 ++---------
 include/linux/ide.h    |  7 +++++++
 2 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include/linux/blkdev.h')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 98c90272443b..9cb4d80a4987 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -79,14 +79,7 @@ enum rq_cmd_type_bits {
 	REQ_TYPE_PM_SUSPEND,		/* suspend request */
 	REQ_TYPE_PM_RESUME,		/* resume request */
 	REQ_TYPE_PM_SHUTDOWN,		/* shutdown request */
-	REQ_TYPE_DRV_PRIV,		/* driver defined type */
-	/*
-	 * for ATA/ATAPI devices. this really doesn't belong here, ide should
-	 * use REQ_TYPE_DRV_PRIV and use rq->cmd[0] with the range of driver
-	 * private REQ_LB opcodes to differentiate what type of request this is
-	 */
-	REQ_TYPE_ATA_TASKFILE,
-	REQ_TYPE_ATA_PC,
+	REQ_TYPE_DRV_PRIV,		/* driver defined types from here */
 };
 
 #define BLK_MAX_CDB	16
@@ -108,7 +101,7 @@ struct request {
 	struct blk_mq_ctx *mq_ctx;
 
 	u64 cmd_flags;
-	enum rq_cmd_type_bits cmd_type;
+	unsigned cmd_type;
 	unsigned long atomic_flags;
 
 	int cpu;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 93b5ca754b5b..62ac399144a6 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -39,6 +39,12 @@
 
 struct device;
 
+/* IDE-specific values for req->cmd_type */
+enum ata_cmd_type_bits {
+	REQ_TYPE_ATA_TASKFILE = REQ_TYPE_DRV_PRIV + 1,
+	REQ_TYPE_ATA_PC,
+};
+
 /* Error codes returned in rq->errors to the higher part of the driver. */
 enum {
 	IDE_DRV_ERROR_GENERAL	= 101,
@@ -1551,4 +1557,5 @@ static inline void ide_set_drivedata(ide_drive_t *drive, void *data)
 #define ide_host_for_each_port(i, port, host) \
 	for ((i) = 0; ((port) = (host)->ports[i]) || (i) < MAX_HOST_PORTS; (i)++)
 
+
 #endif /* _IDE_H */
-- 
cgit v1.2.3-71-gd317


From b0b93b48a30e809240ddd7449a6ad60a5ddf7b4d Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Fri, 17 Apr 2015 22:37:18 +0200
Subject: block: move REQ_TYPE_SENSE to the ide driver

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 drivers/ide/ide-atapi.c  | 6 +++---
 drivers/ide/ide-cd.c     | 8 ++++----
 drivers/ide/ide-floppy.c | 2 +-
 drivers/ide/ide-tape.c   | 2 +-
 include/linux/blkdev.h   | 1 -
 include/linux/ide.h      | 1 +
 6 files changed, 10 insertions(+), 10 deletions(-)

(limited to 'include/linux/blkdev.h')

diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index b367300ce479..1362ad80a76c 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -191,7 +191,7 @@ void ide_prep_sense(ide_drive_t *drive, struct request *rq)
 
 	BUG_ON(sense_len > sizeof(*sense));
 
-	if (rq->cmd_type == REQ_TYPE_SENSE || drive->sense_rq_armed)
+	if (rq->cmd_type == REQ_TYPE_ATA_SENSE || drive->sense_rq_armed)
 		return;
 
 	memset(sense, 0, sizeof(*sense));
@@ -210,7 +210,7 @@ void ide_prep_sense(ide_drive_t *drive, struct request *rq)
 	sense_rq->rq_disk = rq->rq_disk;
 	sense_rq->cmd[0] = GPCMD_REQUEST_SENSE;
 	sense_rq->cmd[4] = cmd_len;
-	sense_rq->cmd_type = REQ_TYPE_SENSE;
+	sense_rq->cmd_type = REQ_TYPE_ATA_SENSE;
 	sense_rq->cmd_flags |= REQ_PREEMPT;
 
 	if (drive->media == ide_tape)
@@ -310,7 +310,7 @@ int ide_cd_get_xferlen(struct request *rq)
 	switch (rq->cmd_type) {
 	case REQ_TYPE_FS:
 		return 32768;
-	case REQ_TYPE_SENSE:
+	case REQ_TYPE_ATA_SENSE:
 	case REQ_TYPE_BLOCK_PC:
 	case REQ_TYPE_ATA_PC:
 		return blk_rq_bytes(rq);
diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 9a32603c6c74..64a6b827b3dd 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -210,7 +210,7 @@ static void cdrom_analyze_sense_data(ide_drive_t *drive,
 static void ide_cd_complete_failed_rq(ide_drive_t *drive, struct request *rq)
 {
 	/*
-	 * For REQ_TYPE_SENSE, "rq->special" points to the original
+	 * For REQ_TYPE_ATA_SENSE, "rq->special" points to the original
 	 * failed request.  Also, the sense data should be read
 	 * directly from rq which might be different from the original
 	 * sense buffer if it got copied during mapping.
@@ -285,7 +285,7 @@ static int cdrom_decode_status(ide_drive_t *drive, u8 stat)
 				  "stat 0x%x",
 				  rq->cmd[0], rq->cmd_type, err, stat);
 
-	if (rq->cmd_type == REQ_TYPE_SENSE) {
+	if (rq->cmd_type == REQ_TYPE_ATA_SENSE) {
 		/*
 		 * We got an error trying to get sense info from the drive
 		 * (probably while trying to recover from a former error).
@@ -526,7 +526,7 @@ static ide_startstop_t cdrom_newpc_intr(ide_drive_t *drive)
 	ide_expiry_t *expiry = NULL;
 	int dma_error = 0, dma, thislen, uptodate = 0;
 	int write = (rq_data_dir(rq) == WRITE) ? 1 : 0, rc = 0;
-	int sense = (rq->cmd_type == REQ_TYPE_SENSE);
+	int sense = (rq->cmd_type == REQ_TYPE_ATA_SENSE);
 	unsigned int timeout;
 	u16 len;
 	u8 ireason, stat;
@@ -791,7 +791,7 @@ static ide_startstop_t ide_cd_do_request(ide_drive_t *drive, struct request *rq,
 		if (cdrom_start_rw(drive, rq) == ide_stopped)
 			goto out_end;
 		break;
-	case REQ_TYPE_SENSE:
+	case REQ_TYPE_ATA_SENSE:
 	case REQ_TYPE_BLOCK_PC:
 	case REQ_TYPE_ATA_PC:
 		if (!rq->timeout)
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 3dbfa5b6db6a..2fb5350c5410 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -266,7 +266,7 @@ static ide_startstop_t ide_floppy_do_request(ide_drive_t *drive,
 		idefloppy_create_rw_cmd(drive, pc, rq, (unsigned long)block);
 		break;
 	case REQ_TYPE_DRV_PRIV:
-	case REQ_TYPE_SENSE:
+	case REQ_TYPE_ATA_SENSE:
 		pc = (struct ide_atapi_pc *)rq->special;
 		break;
 	case REQ_TYPE_BLOCK_PC:
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 2a5d543db9f5..f5d51d1d09ee 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -577,7 +577,7 @@ static ide_startstop_t idetape_do_request(ide_drive_t *drive,
 		      blk_rq_sectors(rq));
 
 	BUG_ON(!(rq->cmd_type == REQ_TYPE_DRV_PRIV ||
-		 rq->cmd_type == REQ_TYPE_SENSE));
+		 rq->cmd_type == REQ_TYPE_ATA_SENSE));
 
 	/* Retry a failed packet command */
 	if (drive->failed_pc && drive->pc->c[0] == REQUEST_SENSE) {
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 9cb4d80a4987..6076b9e18dcb 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -75,7 +75,6 @@ struct request_list {
 enum rq_cmd_type_bits {
 	REQ_TYPE_FS		= 1,	/* fs request */
 	REQ_TYPE_BLOCK_PC,		/* scsi command */
-	REQ_TYPE_SENSE,			/* sense request */
 	REQ_TYPE_PM_SUSPEND,		/* suspend request */
 	REQ_TYPE_PM_RESUME,		/* resume request */
 	REQ_TYPE_PM_SHUTDOWN,		/* shutdown request */
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 62ac399144a6..9856b7d455d9 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -43,6 +43,7 @@ struct device;
 enum ata_cmd_type_bits {
 	REQ_TYPE_ATA_TASKFILE = REQ_TYPE_DRV_PRIV + 1,
 	REQ_TYPE_ATA_PC,
+	REQ_TYPE_ATA_SENSE,	/* sense request */
 };
 
 /* Error codes returned in rq->errors to the higher part of the driver. */
-- 
cgit v1.2.3-71-gd317


From ac7cdff00a33d48d27217560fa3b16d802e5f535 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Fri, 17 Apr 2015 22:37:19 +0200
Subject: block: remove REQ_TYPE_PM_SHUTDOWN

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 include/linux/blkdev.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include/linux/blkdev.h')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 6076b9e18dcb..c2829ba5e738 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -77,7 +77,6 @@ enum rq_cmd_type_bits {
 	REQ_TYPE_BLOCK_PC,		/* scsi command */
 	REQ_TYPE_PM_SUSPEND,		/* suspend request */
 	REQ_TYPE_PM_RESUME,		/* resume request */
-	REQ_TYPE_PM_SHUTDOWN,		/* shutdown request */
 	REQ_TYPE_DRV_PRIV,		/* driver defined types from here */
 };
 
-- 
cgit v1.2.3-71-gd317


From a7928c1578c550bd6f4dec62d65132e6db226c57 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Fri, 17 Apr 2015 22:37:20 +0200
Subject: block: move PM request support to IDE

This removes the request types and hacks from the block code and into the
old IDE driver.  There is a small amunt of code duplication due to this,
but it's not too bad.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 block/blk-core.c           |  1 +
 block/blk-exec.c           | 10 ---------
 block/blk.h                |  2 --
 drivers/ide/ide-eh.c       |  2 +-
 drivers/ide/ide-io.c       |  8 +++----
 drivers/ide/ide-pm.c       | 56 +++++++++++++++++++++++++++++++++++-----------
 drivers/ide/ide-taskfile.c |  2 +-
 include/linux/blkdev.h     | 21 +----------------
 include/linux/ide.h        | 19 ++++++++++++++++
 9 files changed, 70 insertions(+), 51 deletions(-)

(limited to 'include/linux/blkdev.h')

diff --git a/block/blk-core.c b/block/blk-core.c
index fd154b94447a..2e5020f37d55 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -285,6 +285,7 @@ inline void __blk_run_queue_uncond(struct request_queue *q)
 	q->request_fn(q);
 	q->request_fn_active--;
 }
+EXPORT_SYMBOL_GPL(__blk_run_queue_uncond);
 
 /**
  * __blk_run_queue - run a single device queue
diff --git a/block/blk-exec.c b/block/blk-exec.c
index 9924725fa50d..3fec8a29d0fa 100644
--- a/block/blk-exec.c
+++ b/block/blk-exec.c
@@ -53,7 +53,6 @@ void blk_execute_rq_nowait(struct request_queue *q, struct gendisk *bd_disk,
 			   rq_end_io_fn *done)
 {
 	int where = at_head ? ELEVATOR_INSERT_FRONT : ELEVATOR_INSERT_BACK;
-	bool is_pm_resume;
 
 	WARN_ON(irqs_disabled());
 	WARN_ON(rq->cmd_type == REQ_TYPE_FS);
@@ -70,12 +69,6 @@ void blk_execute_rq_nowait(struct request_queue *q, struct gendisk *bd_disk,
 		return;
 	}
 
-	/*
-	 * need to check this before __blk_run_queue(), because rq can
-	 * be freed before that returns.
-	 */
-	is_pm_resume = rq->cmd_type == REQ_TYPE_PM_RESUME;
-
 	spin_lock_irq(q->queue_lock);
 
 	if (unlikely(blk_queue_dying(q))) {
@@ -88,9 +81,6 @@ void blk_execute_rq_nowait(struct request_queue *q, struct gendisk *bd_disk,
 
 	__elv_add_request(q, rq, where);
 	__blk_run_queue(q);
-	/* the queue is stopped so it won't be run */
-	if (is_pm_resume)
-		__blk_run_queue_uncond(q);
 	spin_unlock_irq(q->queue_lock);
 }
 EXPORT_SYMBOL_GPL(blk_execute_rq_nowait);
diff --git a/block/blk.h b/block/blk.h
index 43b036185712..4b48d55e588e 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -193,8 +193,6 @@ int blk_try_merge(struct request *rq, struct bio *bio);
 
 void blk_queue_congestion_threshold(struct request_queue *q);
 
-void __blk_run_queue_uncond(struct request_queue *q);
-
 int blk_dev_init(void);
 
 
diff --git a/drivers/ide/ide-eh.c b/drivers/ide/ide-eh.c
index 19d809c48a8d..d6da011299f5 100644
--- a/drivers/ide/ide-eh.c
+++ b/drivers/ide/ide-eh.c
@@ -129,7 +129,7 @@ ide_startstop_t ide_error(ide_drive_t *drive, const char *msg, u8 stat)
 
 			if (cmd)
 				ide_complete_cmd(drive, cmd, stat, err);
-		} else if (blk_pm_request(rq)) {
+		} else if (ata_pm_request(rq)) {
 			rq->errors = 1;
 			ide_complete_pm_rq(drive, rq);
 			return ide_stopped;
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 8e55abd03a24..669ea1e45795 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -320,7 +320,7 @@ static ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 		goto kill_rq;
 	}
 
-	if (blk_pm_request(rq))
+	if (ata_pm_request(rq))
 		ide_check_pm_state(drive, rq);
 
 	drive->hwif->tp_ops->dev_select(drive);
@@ -342,8 +342,8 @@ static ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 
 		if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE)
 			return execute_drive_cmd(drive, rq);
-		else if (blk_pm_request(rq)) {
-			struct request_pm_state *pm = rq->special;
+		else if (ata_pm_request(rq)) {
+			struct ide_pm_state *pm = rq->special;
 #ifdef DEBUG_PM
 			printk("%s: start_power_step(step: %d)\n",
 				drive->name, pm->pm_step);
@@ -538,7 +538,7 @@ repeat:
 		 * state machine.
 		 */
 		if ((drive->dev_flags & IDE_DFLAG_BLOCKED) &&
-		    blk_pm_request(rq) == 0 &&
+		    ata_pm_request(rq) == 0 &&
 		    (rq->cmd_flags & REQ_PREEMPT) == 0) {
 			/* there should be no pending command at this point */
 			ide_unlock_port(hwif);
diff --git a/drivers/ide/ide-pm.c b/drivers/ide/ide-pm.c
index 8d1e32d7cd97..081e43458d50 100644
--- a/drivers/ide/ide-pm.c
+++ b/drivers/ide/ide-pm.c
@@ -8,7 +8,7 @@ int generic_ide_suspend(struct device *dev, pm_message_t mesg)
 	ide_drive_t *pair = ide_get_pair_dev(drive);
 	ide_hwif_t *hwif = drive->hwif;
 	struct request *rq;
-	struct request_pm_state rqpm;
+	struct ide_pm_state rqpm;
 	int ret;
 
 	if (ide_port_acpi(hwif)) {
@@ -19,7 +19,7 @@ int generic_ide_suspend(struct device *dev, pm_message_t mesg)
 
 	memset(&rqpm, 0, sizeof(rqpm));
 	rq = blk_get_request(drive->queue, READ, __GFP_WAIT);
-	rq->cmd_type = REQ_TYPE_PM_SUSPEND;
+	rq->cmd_type = REQ_TYPE_ATA_PM_SUSPEND;
 	rq->special = &rqpm;
 	rqpm.pm_step = IDE_PM_START_SUSPEND;
 	if (mesg.event == PM_EVENT_PRETHAW)
@@ -38,13 +38,43 @@ int generic_ide_suspend(struct device *dev, pm_message_t mesg)
 	return ret;
 }
 
+static void ide_end_sync_rq(struct request *rq, int error)
+{
+	complete(rq->end_io_data);
+}
+
+static int ide_pm_execute_rq(struct request *rq)
+{
+	struct request_queue *q = rq->q;
+	DECLARE_COMPLETION_ONSTACK(wait);
+
+	rq->end_io_data = &wait;
+	rq->end_io = ide_end_sync_rq;
+
+	spin_lock_irq(q->queue_lock);
+	if (unlikely(blk_queue_dying(q))) {
+		rq->cmd_flags |= REQ_QUIET;
+		rq->errors = -ENXIO;
+		__blk_end_request_all(rq, rq->errors);
+		spin_unlock_irq(q->queue_lock);
+		return -ENXIO;
+	}
+	__elv_add_request(q, rq, ELEVATOR_INSERT_FRONT);
+	__blk_run_queue_uncond(q);
+	spin_unlock_irq(q->queue_lock);
+
+	wait_for_completion_io(&wait);
+
+	return rq->errors ? -EIO : 0;
+}
+
 int generic_ide_resume(struct device *dev)
 {
 	ide_drive_t *drive = to_ide_device(dev);
 	ide_drive_t *pair = ide_get_pair_dev(drive);
 	ide_hwif_t *hwif = drive->hwif;
 	struct request *rq;
-	struct request_pm_state rqpm;
+	struct ide_pm_state rqpm;
 	int err;
 
 	if (ide_port_acpi(hwif)) {
@@ -59,13 +89,13 @@ int generic_ide_resume(struct device *dev)
 
 	memset(&rqpm, 0, sizeof(rqpm));
 	rq = blk_get_request(drive->queue, READ, __GFP_WAIT);
-	rq->cmd_type = REQ_TYPE_PM_RESUME;
+	rq->cmd_type = REQ_TYPE_ATA_PM_RESUME;
 	rq->cmd_flags |= REQ_PREEMPT;
 	rq->special = &rqpm;
 	rqpm.pm_step = IDE_PM_START_RESUME;
 	rqpm.pm_state = PM_EVENT_ON;
 
-	err = blk_execute_rq(drive->queue, NULL, rq, 1);
+	err = ide_pm_execute_rq(rq);
 	blk_put_request(rq);
 
 	if (err == 0 && dev->driver) {
@@ -80,7 +110,7 @@ int generic_ide_resume(struct device *dev)
 
 void ide_complete_power_step(ide_drive_t *drive, struct request *rq)
 {
-	struct request_pm_state *pm = rq->special;
+	struct ide_pm_state *pm = rq->special;
 
 #ifdef DEBUG_PM
 	printk(KERN_INFO "%s: complete_power_step(step: %d)\n",
@@ -110,7 +140,7 @@ void ide_complete_power_step(ide_drive_t *drive, struct request *rq)
 
 ide_startstop_t ide_start_power_step(ide_drive_t *drive, struct request *rq)
 {
-	struct request_pm_state *pm = rq->special;
+	struct ide_pm_state *pm = rq->special;
 	struct ide_cmd cmd = { };
 
 	switch (pm->pm_step) {
@@ -182,7 +212,7 @@ out_do_tf:
 void ide_complete_pm_rq(ide_drive_t *drive, struct request *rq)
 {
 	struct request_queue *q = drive->queue;
-	struct request_pm_state *pm = rq->special;
+	struct ide_pm_state *pm = rq->special;
 	unsigned long flags;
 
 	ide_complete_power_step(drive, rq);
@@ -191,10 +221,10 @@ void ide_complete_pm_rq(ide_drive_t *drive, struct request *rq)
 
 #ifdef DEBUG_PM
 	printk("%s: completing PM request, %s\n", drive->name,
-	       (rq->cmd_type == REQ_TYPE_PM_SUSPEND) ? "suspend" : "resume");
+	       (rq->cmd_type == REQ_TYPE_ATA_PM_SUSPEND) ? "suspend" : "resume");
 #endif
 	spin_lock_irqsave(q->queue_lock, flags);
-	if (rq->cmd_type == REQ_TYPE_PM_SUSPEND)
+	if (rq->cmd_type == REQ_TYPE_ATA_PM_SUSPEND)
 		blk_stop_queue(q);
 	else
 		drive->dev_flags &= ~IDE_DFLAG_BLOCKED;
@@ -208,13 +238,13 @@ void ide_complete_pm_rq(ide_drive_t *drive, struct request *rq)
 
 void ide_check_pm_state(ide_drive_t *drive, struct request *rq)
 {
-	struct request_pm_state *pm = rq->special;
+	struct ide_pm_state *pm = rq->special;
 
-	if (rq->cmd_type == REQ_TYPE_PM_SUSPEND &&
+	if (rq->cmd_type == REQ_TYPE_ATA_PM_SUSPEND &&
 	    pm->pm_step == IDE_PM_START_SUSPEND)
 		/* Mark drive blocked when starting the suspend sequence. */
 		drive->dev_flags |= IDE_DFLAG_BLOCKED;
-	else if (rq->cmd_type == REQ_TYPE_PM_RESUME &&
+	else if (rq->cmd_type == REQ_TYPE_ATA_PM_RESUME &&
 		 pm->pm_step == IDE_PM_START_RESUME) {
 		/*
 		 * The first thing we do on wakeup is to wait for BSY bit to
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index dabb88b1cbec..0979e126fff1 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -186,7 +186,7 @@ static ide_startstop_t task_no_data_intr(ide_drive_t *drive)
 	    tf->command == ATA_CMD_CHK_POWER) {
 		struct request *rq = hwif->rq;
 
-		if (blk_pm_request(rq))
+		if (ata_pm_request(rq))
 			ide_complete_pm_rq(drive, rq);
 		else
 			ide_finish_cmd(drive, cmd, stat);
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index c2829ba5e738..2da818a48097 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -30,7 +30,6 @@ struct scsi_ioctl_command;
 
 struct request_queue;
 struct elevator_queue;
-struct request_pm_state;
 struct blk_trace;
 struct request;
 struct sg_io_hdr;
@@ -75,8 +74,6 @@ struct request_list {
 enum rq_cmd_type_bits {
 	REQ_TYPE_FS		= 1,	/* fs request */
 	REQ_TYPE_BLOCK_PC,		/* scsi command */
-	REQ_TYPE_PM_SUSPEND,		/* suspend request */
-	REQ_TYPE_PM_RESUME,		/* resume request */
 	REQ_TYPE_DRV_PRIV,		/* driver defined types from here */
 };
 
@@ -207,19 +204,6 @@ static inline unsigned short req_get_ioprio(struct request *req)
 	return req->ioprio;
 }
 
-/*
- * State information carried for REQ_TYPE_PM_SUSPEND and REQ_TYPE_PM_RESUME
- * requests. Some step values could eventually be made generic.
- */
-struct request_pm_state
-{
-	/* PM state machine step value, currently driver specific */
-	int	pm_step;
-	/* requested PM state value (S1, S2, S3, S4, ...) */
-	u32	pm_state;
-	void*	data;		/* for driver use */
-};
-
 #include <linux/elevator.h>
 
 struct blk_queue_ctx;
@@ -601,10 +585,6 @@ static inline void queue_flag_clear(unsigned int flag, struct request_queue *q)
 	(((rq)->cmd_flags & REQ_STARTED) && \
 	 ((rq)->cmd_type == REQ_TYPE_FS))
 
-#define blk_pm_request(rq)	\
-	((rq)->cmd_type == REQ_TYPE_PM_SUSPEND || \
-	 (rq)->cmd_type == REQ_TYPE_PM_RESUME)
-
 #define blk_rq_cpu_valid(rq)	((rq)->cpu != -1)
 #define blk_bidi_rq(rq)		((rq)->next_rq != NULL)
 /* rq->queuelist of dequeued request must be list_empty() */
@@ -838,6 +818,7 @@ extern void blk_stop_queue(struct request_queue *q);
 extern void blk_sync_queue(struct request_queue *q);
 extern void __blk_stop_queue(struct request_queue *q);
 extern void __blk_run_queue(struct request_queue *q);
+extern void __blk_run_queue_uncond(struct request_queue *q);
 extern void blk_run_queue(struct request_queue *);
 extern void blk_run_queue_async(struct request_queue *q);
 extern int blk_rq_map_user(struct request_queue *, struct request *,
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 9856b7d455d9..a633898f36ac 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -44,8 +44,14 @@ enum ata_cmd_type_bits {
 	REQ_TYPE_ATA_TASKFILE = REQ_TYPE_DRV_PRIV + 1,
 	REQ_TYPE_ATA_PC,
 	REQ_TYPE_ATA_SENSE,	/* sense request */
+	REQ_TYPE_ATA_PM_SUSPEND,/* suspend request */
+	REQ_TYPE_ATA_PM_RESUME,	/* resume request */
 };
 
+#define ata_pm_request(rq)	\
+	((rq)->cmd_type == REQ_TYPE_ATA_PM_SUSPEND || \
+	 (rq)->cmd_type == REQ_TYPE_ATA_PM_RESUME)
+
 /* Error codes returned in rq->errors to the higher part of the driver. */
 enum {
 	IDE_DRV_ERROR_GENERAL	= 101,
@@ -1321,6 +1327,19 @@ struct ide_port_info {
 	u8			udma_mask;
 };
 
+/*
+ * State information carried for REQ_TYPE_ATA_PM_SUSPEND and REQ_TYPE_ATA_PM_RESUME
+ * requests.
+ */
+struct ide_pm_state {
+	/* PM state machine step value, currently driver specific */
+	int	pm_step;
+	/* requested PM state value (S1, S2, S3, S4, ...) */
+	u32	pm_state;
+	void*	data;		/* for driver use */
+};
+
+
 int ide_pci_init_one(struct pci_dev *, const struct ide_port_info *, void *);
 int ide_pci_init_two(struct pci_dev *, struct pci_dev *,
 		     const struct ide_port_info *, void *);
-- 
cgit v1.2.3-71-gd317


From 4ecd4fef3a074c8bb43c391a57742c422469ebbd Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Thu, 7 May 2015 09:38:13 +0200
Subject: block: use an atomic_t for mq_freeze_depth

lockdep gets unhappy about the not disabling irqs when using the queue_lock
around it.  Instead of trying to fix that up just switch to an atomic_t
and get rid of the lock.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 block/blk-mq.c         | 24 ++++++++++--------------
 include/linux/blkdev.h |  2 +-
 2 files changed, 11 insertions(+), 15 deletions(-)

(limited to 'include/linux/blkdev.h')

diff --git a/block/blk-mq.c b/block/blk-mq.c
index 31df47443699..c382a34fe5ac 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -89,7 +89,8 @@ static int blk_mq_queue_enter(struct request_queue *q, gfp_t gfp)
 			return -EBUSY;
 
 		ret = wait_event_interruptible(q->mq_freeze_wq,
-				!q->mq_freeze_depth || blk_queue_dying(q));
+				!atomic_read(&q->mq_freeze_depth) ||
+				blk_queue_dying(q));
 		if (blk_queue_dying(q))
 			return -ENODEV;
 		if (ret)
@@ -112,13 +113,10 @@ static void blk_mq_usage_counter_release(struct percpu_ref *ref)
 
 void blk_mq_freeze_queue_start(struct request_queue *q)
 {
-	bool freeze;
+	int freeze_depth;
 
-	spin_lock_irq(q->queue_lock);
-	freeze = !q->mq_freeze_depth++;
-	spin_unlock_irq(q->queue_lock);
-
-	if (freeze) {
+	freeze_depth = atomic_inc_return(&q->mq_freeze_depth);
+	if (freeze_depth == 1) {
 		percpu_ref_kill(&q->mq_usage_counter);
 		blk_mq_run_hw_queues(q, false);
 	}
@@ -143,13 +141,11 @@ EXPORT_SYMBOL_GPL(blk_mq_freeze_queue);
 
 void blk_mq_unfreeze_queue(struct request_queue *q)
 {
-	bool wake;
+	int freeze_depth;
 
-	spin_lock_irq(q->queue_lock);
-	wake = !--q->mq_freeze_depth;
-	WARN_ON_ONCE(q->mq_freeze_depth < 0);
-	spin_unlock_irq(q->queue_lock);
-	if (wake) {
+	freeze_depth = atomic_dec_return(&q->mq_freeze_depth);
+	WARN_ON_ONCE(freeze_depth < 0);
+	if (!freeze_depth) {
 		percpu_ref_reinit(&q->mq_usage_counter);
 		wake_up_all(&q->mq_freeze_wq);
 	}
@@ -2081,7 +2077,7 @@ void blk_mq_free_queue(struct request_queue *q)
 /* Basically redo blk_mq_init_queue with queue frozen */
 static void blk_mq_queue_reinit(struct request_queue *q)
 {
-	WARN_ON_ONCE(!q->mq_freeze_depth);
+	WARN_ON_ONCE(!atomic_read(&q->mq_freeze_depth));
 
 	blk_mq_sysfs_unregister(q);
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 2da818a48097..bc917956a6d0 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -444,7 +444,7 @@ struct request_queue {
 	struct mutex		sysfs_lock;
 
 	int			bypass_depth;
-	int			mq_freeze_depth;
+	atomic_t		mq_freeze_depth;
 
 #if defined(CONFIG_BLK_DEV_BSG)
 	bsg_job_fn		*bsg_job_fn;
-- 
cgit v1.2.3-71-gd317


From 5f1b670d0bef508a5554d92525f5f6d00d640b38 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Fri, 22 May 2015 09:14:04 -0400
Subject: block, dm: don't copy bios for request clones

Currently dm-multipath has to clone the bios for every request sent
to the lower devices, which wastes cpu cycles and ties down memory.

This patch instead adds a new REQ_CLONE flag that instructs req_bio_endio
to not complete bios attached to a request, which we set on clone
requests similar to bios in a flush sequence.  With this change I/O
errors on a path failure only get propagated to dm-multipath, which
can then either resubmit the I/O or complete the bios on the original
request.

I've done some basic testing of this on a Linux target with ALUA support,
and it survives path failures during I/O nicely.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
---
 block/blk-core.c          |  94 +++----------------------
 drivers/md/dm-table.c     |  25 ++++---
 drivers/md/dm.c           | 171 +++++++++++-----------------------------------
 drivers/md/dm.h           |   5 +-
 include/linux/blk_types.h |   2 +
 include/linux/blkdev.h    |   6 +-
 6 files changed, 73 insertions(+), 230 deletions(-)

(limited to 'include/linux/blkdev.h')

diff --git a/block/blk-core.c b/block/blk-core.c
index de474b5dee2b..aa819a58ea24 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -117,7 +117,7 @@ EXPORT_SYMBOL(blk_rq_init);
 static void req_bio_endio(struct request *rq, struct bio *bio,
 			  unsigned int nbytes, int error)
 {
-	if (error)
+	if (error && !(rq->cmd_flags & REQ_CLONE))
 		clear_bit(BIO_UPTODATE, &bio->bi_flags);
 	else if (!test_bit(BIO_UPTODATE, &bio->bi_flags))
 		error = -EIO;
@@ -128,7 +128,8 @@ static void req_bio_endio(struct request *rq, struct bio *bio,
 	bio_advance(bio, nbytes);
 
 	/* don't actually finish bio if it's part of flush sequence */
-	if (bio->bi_iter.bi_size == 0 && !(rq->cmd_flags & REQ_FLUSH_SEQ))
+	if (bio->bi_iter.bi_size == 0 &&
+	    !(rq->cmd_flags & (REQ_FLUSH_SEQ|REQ_CLONE)))
 		bio_endio(bio, error);
 }
 
@@ -2909,95 +2910,22 @@ int blk_lld_busy(struct request_queue *q)
 }
 EXPORT_SYMBOL_GPL(blk_lld_busy);
 
-/**
- * blk_rq_unprep_clone - Helper function to free all bios in a cloned request
- * @rq: the clone request to be cleaned up
- *
- * Description:
- *     Free all bios in @rq for a cloned request.
- */
-void blk_rq_unprep_clone(struct request *rq)
-{
-	struct bio *bio;
-
-	while ((bio = rq->bio) != NULL) {
-		rq->bio = bio->bi_next;
-
-		bio_put(bio);
-	}
-}
-EXPORT_SYMBOL_GPL(blk_rq_unprep_clone);
-
-/*
- * Copy attributes of the original request to the clone request.
- * The actual data parts (e.g. ->cmd, ->sense) are not copied.
- */
-static void __blk_rq_prep_clone(struct request *dst, struct request *src)
+void blk_rq_prep_clone(struct request *dst, struct request *src)
 {
 	dst->cpu = src->cpu;
-	dst->cmd_flags |= (src->cmd_flags & REQ_CLONE_MASK) | REQ_NOMERGE;
+	dst->cmd_flags |= (src->cmd_flags & REQ_CLONE_MASK);
+	dst->cmd_flags |= REQ_NOMERGE | REQ_CLONE;
 	dst->cmd_type = src->cmd_type;
 	dst->__sector = blk_rq_pos(src);
 	dst->__data_len = blk_rq_bytes(src);
 	dst->nr_phys_segments = src->nr_phys_segments;
 	dst->ioprio = src->ioprio;
 	dst->extra_len = src->extra_len;
-}
-
-/**
- * blk_rq_prep_clone - Helper function to setup clone request
- * @rq: the request to be setup
- * @rq_src: original request to be cloned
- * @bs: bio_set that bios for clone are allocated from
- * @gfp_mask: memory allocation mask for bio
- * @bio_ctr: setup function to be called for each clone bio.
- *           Returns %0 for success, non %0 for failure.
- * @data: private data to be passed to @bio_ctr
- *
- * Description:
- *     Clones bios in @rq_src to @rq, and copies attributes of @rq_src to @rq.
- *     The actual data parts of @rq_src (e.g. ->cmd, ->sense)
- *     are not copied, and copying such parts is the caller's responsibility.
- *     Also, pages which the original bios are pointing to are not copied
- *     and the cloned bios just point same pages.
- *     So cloned bios must be completed before original bios, which means
- *     the caller must complete @rq before @rq_src.
- */
-int blk_rq_prep_clone(struct request *rq, struct request *rq_src,
-		      struct bio_set *bs, gfp_t gfp_mask,
-		      int (*bio_ctr)(struct bio *, struct bio *, void *),
-		      void *data)
-{
-	struct bio *bio, *bio_src;
-
-	if (!bs)
-		bs = fs_bio_set;
-
-	__rq_for_each_bio(bio_src, rq_src) {
-		bio = bio_clone_fast(bio_src, gfp_mask, bs);
-		if (!bio)
-			goto free_and_out;
-
-		if (bio_ctr && bio_ctr(bio, bio_src, data))
-			goto free_and_out;
-
-		if (rq->bio) {
-			rq->biotail->bi_next = bio;
-			rq->biotail = bio;
-		} else
-			rq->bio = rq->biotail = bio;
-	}
-
-	__blk_rq_prep_clone(rq, rq_src);
-
-	return 0;
-
-free_and_out:
-	if (bio)
-		bio_put(bio);
-	blk_rq_unprep_clone(rq);
-
-	return -ENOMEM;
+	dst->bio = src->bio;
+	dst->biotail = src->biotail;
+	dst->cmd = src->cmd;
+	dst->cmd_len = src->cmd_len;
+	dst->sense = src->sense;
 }
 EXPORT_SYMBOL_GPL(blk_rq_prep_clone);
 
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index d9b00b8565c6..3662b2e49b8d 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -940,21 +940,28 @@ static int dm_table_alloc_md_mempools(struct dm_table *t, struct mapped_device *
 {
 	unsigned type = dm_table_get_type(t);
 	unsigned per_bio_data_size = 0;
-	struct dm_target *tgt;
 	unsigned i;
 
-	if (unlikely(type == DM_TYPE_NONE)) {
+	switch (type) {
+	case DM_TYPE_BIO_BASED:
+		for (i = 0; i < t->num_targets; i++) {
+			struct dm_target *tgt = t->targets + i;
+
+			per_bio_data_size = max(per_bio_data_size,
+						tgt->per_bio_data_size);
+		}
+		t->mempools = dm_alloc_bio_mempools(t->integrity_supported,
+						    per_bio_data_size);
+		break;
+	case DM_TYPE_REQUEST_BASED:
+	case DM_TYPE_MQ_REQUEST_BASED:
+		t->mempools = dm_alloc_rq_mempools(md, type);
+		break;
+	default:
 		DMWARN("no table type is set, can't allocate mempools");
 		return -EINVAL;
 	}
 
-	if (type == DM_TYPE_BIO_BASED)
-		for (i = 0; i < t->num_targets; i++) {
-			tgt = t->targets + i;
-			per_bio_data_size = max(per_bio_data_size, tgt->per_bio_data_size);
-		}
-
-	t->mempools = dm_alloc_md_mempools(md, type, t->integrity_supported, per_bio_data_size);
 	if (!t->mempools)
 		return -ENOMEM;
 
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index a930b72314ac..38837f8ea327 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -990,57 +990,6 @@ static void clone_endio(struct bio *bio, int error)
 	dec_pending(io, error);
 }
 
-/*
- * Partial completion handling for request-based dm
- */
-static void end_clone_bio(struct bio *clone, int error)
-{
-	struct dm_rq_clone_bio_info *info =
-		container_of(clone, struct dm_rq_clone_bio_info, clone);
-	struct dm_rq_target_io *tio = info->tio;
-	struct bio *bio = info->orig;
-	unsigned int nr_bytes = info->orig->bi_iter.bi_size;
-
-	bio_put(clone);
-
-	if (tio->error)
-		/*
-		 * An error has already been detected on the request.
-		 * Once error occurred, just let clone->end_io() handle
-		 * the remainder.
-		 */
-		return;
-	else if (error) {
-		/*
-		 * Don't notice the error to the upper layer yet.
-		 * The error handling decision is made by the target driver,
-		 * when the request is completed.
-		 */
-		tio->error = error;
-		return;
-	}
-
-	/*
-	 * I/O for the bio successfully completed.
-	 * Notice the data completion to the upper layer.
-	 */
-
-	/*
-	 * bios are processed from the head of the list.
-	 * So the completing bio should always be rq->bio.
-	 * If it's not, something wrong is happening.
-	 */
-	if (tio->orig->bio != bio)
-		DMERR("bio completion is going in the middle of the request");
-
-	/*
-	 * Update the original request.
-	 * Do not use blk_end_request() here, because it may complete
-	 * the original request before the clone, and break the ordering.
-	 */
-	blk_update_request(tio->orig, 0, nr_bytes);
-}
-
 static struct dm_rq_target_io *tio_from_request(struct request *rq)
 {
 	return (rq->q->mq_ops ? blk_mq_rq_to_pdu(rq) : rq->special);
@@ -1089,8 +1038,6 @@ static void free_rq_clone(struct request *clone, bool must_be_mapped)
 
 	WARN_ON_ONCE(must_be_mapped && !clone->q);
 
-	blk_rq_unprep_clone(clone);
-
 	if (md->type == DM_TYPE_MQ_REQUEST_BASED)
 		/* stacked on blk-mq queue(s) */
 		tio->ti->type->release_clone_rq(clone);
@@ -1821,39 +1768,13 @@ static void dm_dispatch_clone_request(struct request *clone, struct request *rq)
 		dm_complete_request(rq, r);
 }
 
-static int dm_rq_bio_constructor(struct bio *bio, struct bio *bio_orig,
-				 void *data)
+static void setup_clone(struct request *clone, struct request *rq,
+		        struct dm_rq_target_io *tio)
 {
-	struct dm_rq_target_io *tio = data;
-	struct dm_rq_clone_bio_info *info =
-		container_of(bio, struct dm_rq_clone_bio_info, clone);
-
-	info->orig = bio_orig;
-	info->tio = tio;
-	bio->bi_end_io = end_clone_bio;
-
-	return 0;
-}
-
-static int setup_clone(struct request *clone, struct request *rq,
-		       struct dm_rq_target_io *tio, gfp_t gfp_mask)
-{
-	int r;
-
-	r = blk_rq_prep_clone(clone, rq, tio->md->bs, gfp_mask,
-			      dm_rq_bio_constructor, tio);
-	if (r)
-		return r;
-
-	clone->cmd = rq->cmd;
-	clone->cmd_len = rq->cmd_len;
-	clone->sense = rq->sense;
+	blk_rq_prep_clone(clone, rq);
 	clone->end_io = end_clone_request;
 	clone->end_io_data = tio;
-
 	tio->clone = clone;
-
-	return 0;
 }
 
 static struct request *clone_rq(struct request *rq, struct mapped_device *md,
@@ -1874,12 +1795,7 @@ static struct request *clone_rq(struct request *rq, struct mapped_device *md,
 		clone = tio->clone;
 
 	blk_rq_init(NULL, clone);
-	if (setup_clone(clone, rq, tio, gfp_mask)) {
-		/* -ENOMEM */
-		if (alloc_clone)
-			free_clone_request(md, clone);
-		return NULL;
-	}
+	setup_clone(clone, rq, tio);
 
 	return clone;
 }
@@ -1973,11 +1889,7 @@ static int map_request(struct dm_rq_target_io *tio, struct request *rq,
 		}
 		if (IS_ERR(clone))
 			return DM_MAPIO_REQUEUE;
-		if (setup_clone(clone, rq, tio, GFP_ATOMIC)) {
-			/* -ENOMEM */
-			ti->type->release_clone_rq(clone);
-			return DM_MAPIO_REQUEUE;
-		}
+		setup_clone(clone, rq, tio);
 	}
 
 	switch (r) {
@@ -2431,8 +2343,6 @@ static void __bind_mempools(struct mapped_device *md, struct dm_table *t)
 		goto out;
 	}
 
-	BUG_ON(!p || md->io_pool || md->rq_pool || md->bs);
-
 	md->io_pool = p->io_pool;
 	p->io_pool = NULL;
 	md->rq_pool = p->rq_pool;
@@ -3536,48 +3446,23 @@ int dm_noflush_suspending(struct dm_target *ti)
 }
 EXPORT_SYMBOL_GPL(dm_noflush_suspending);
 
-struct dm_md_mempools *dm_alloc_md_mempools(struct mapped_device *md, unsigned type,
-					    unsigned integrity, unsigned per_bio_data_size)
+struct dm_md_mempools *dm_alloc_bio_mempools(unsigned integrity,
+					     unsigned per_bio_data_size)
 {
-	struct dm_md_mempools *pools = kzalloc(sizeof(*pools), GFP_KERNEL);
-	struct kmem_cache *cachep = NULL;
-	unsigned int pool_size = 0;
+	struct dm_md_mempools *pools;
+	unsigned int pool_size = dm_get_reserved_bio_based_ios();
 	unsigned int front_pad;
 
+	pools = kzalloc(sizeof(*pools), GFP_KERNEL);
 	if (!pools)
 		return NULL;
 
-	type = filter_md_type(type, md);
+	front_pad = roundup(per_bio_data_size, __alignof__(struct dm_target_io)) +
+		offsetof(struct dm_target_io, clone);
 
-	switch (type) {
-	case DM_TYPE_BIO_BASED:
-		cachep = _io_cache;
-		pool_size = dm_get_reserved_bio_based_ios();
-		front_pad = roundup(per_bio_data_size, __alignof__(struct dm_target_io)) + offsetof(struct dm_target_io, clone);
-		break;
-	case DM_TYPE_REQUEST_BASED:
-		cachep = _rq_tio_cache;
-		pool_size = dm_get_reserved_rq_based_ios();
-		pools->rq_pool = mempool_create_slab_pool(pool_size, _rq_cache);
-		if (!pools->rq_pool)
-			goto out;
-		/* fall through to setup remaining rq-based pools */
-	case DM_TYPE_MQ_REQUEST_BASED:
-		if (!pool_size)
-			pool_size = dm_get_reserved_rq_based_ios();
-		front_pad = offsetof(struct dm_rq_clone_bio_info, clone);
-		/* per_bio_data_size is not used. See __bind_mempools(). */
-		WARN_ON(per_bio_data_size != 0);
-		break;
-	default:
-		BUG();
-	}
-
-	if (cachep) {
-		pools->io_pool = mempool_create_slab_pool(pool_size, cachep);
-		if (!pools->io_pool)
-			goto out;
-	}
+	pools->io_pool = mempool_create_slab_pool(pool_size, _io_cache);
+	if (!pools->io_pool)
+		goto out;
 
 	pools->bs = bioset_create_nobvec(pool_size, front_pad);
 	if (!pools->bs)
@@ -3587,10 +3472,34 @@ struct dm_md_mempools *dm_alloc_md_mempools(struct mapped_device *md, unsigned t
 		goto out;
 
 	return pools;
-
 out:
 	dm_free_md_mempools(pools);
+	return NULL;
+}
+
+struct dm_md_mempools *dm_alloc_rq_mempools(struct mapped_device *md,
+					    unsigned type)
+{
+	unsigned int pool_size = dm_get_reserved_rq_based_ios();
+	struct dm_md_mempools *pools;
 
+	pools = kzalloc(sizeof(*pools), GFP_KERNEL);
+	if (!pools)
+		return NULL;
+
+	if (filter_md_type(type, md) == DM_TYPE_REQUEST_BASED) {
+		pools->rq_pool = mempool_create_slab_pool(pool_size, _rq_cache);
+		if (!pools->rq_pool)
+			goto out;
+	}
+
+	pools->io_pool = mempool_create_slab_pool(pool_size, _rq_tio_cache);
+	if (!pools->io_pool)
+		goto out;
+
+	return pools;
+out:
+	dm_free_md_mempools(pools);
 	return NULL;
 }
 
diff --git a/drivers/md/dm.h b/drivers/md/dm.h
index 6123c2bf9150..e6e66d087b26 100644
--- a/drivers/md/dm.h
+++ b/drivers/md/dm.h
@@ -222,8 +222,9 @@ void dm_kcopyd_exit(void);
 /*
  * Mempool operations
  */
-struct dm_md_mempools *dm_alloc_md_mempools(struct mapped_device *md, unsigned type,
-					    unsigned integrity, unsigned per_bio_data_size);
+struct dm_md_mempools *dm_alloc_bio_mempools(unsigned integrity,
+					     unsigned per_bio_data_size);
+struct dm_md_mempools *dm_alloc_rq_mempools(struct mapped_device *md, unsigned type);
 void dm_free_md_mempools(struct dm_md_mempools *pools);
 
 /*
diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h
index 3f4ded0b1a34..45a6be89957c 100644
--- a/include/linux/blk_types.h
+++ b/include/linux/blk_types.h
@@ -192,6 +192,7 @@ enum rq_flag_bits {
 	__REQ_HASHED,		/* on IO scheduler merge hash */
 	__REQ_MQ_INFLIGHT,	/* track inflight for MQ */
 	__REQ_NO_TIMEOUT,	/* requests may never expire */
+	__REQ_CLONE,		/* cloned bios */
 	__REQ_NR_BITS,		/* stops here */
 };
 
@@ -246,5 +247,6 @@ enum rq_flag_bits {
 #define REQ_HASHED		(1ULL << __REQ_HASHED)
 #define REQ_MQ_INFLIGHT		(1ULL << __REQ_MQ_INFLIGHT)
 #define REQ_NO_TIMEOUT		(1ULL << __REQ_NO_TIMEOUT)
+#define REQ_CLONE		(1ULL << __REQ_CLONE)
 
 #endif /* __LINUX_BLK_TYPES_H */
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index bc917956a6d0..9ded80da2c16 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -775,11 +775,7 @@ extern void blk_add_request_payload(struct request *rq, struct page *page,
 		unsigned int len);
 extern int blk_rq_check_limits(struct request_queue *q, struct request *rq);
 extern int blk_lld_busy(struct request_queue *q);
-extern int blk_rq_prep_clone(struct request *rq, struct request *rq_src,
-			     struct bio_set *bs, gfp_t gfp_mask,
-			     int (*bio_ctr)(struct bio *, struct bio *, void *),
-			     void *data);
-extern void blk_rq_unprep_clone(struct request *rq);
+extern void blk_rq_prep_clone(struct request *rq, struct request *rq_src);
 extern int blk_insert_cloned_request(struct request_queue *q,
 				     struct request *rq);
 extern void blk_delay_queue(struct request_queue *, unsigned long);
-- 
cgit v1.2.3-71-gd317


From e548ca4ee4595f65b262661d166310ad8a149bec Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@fb.com>
Date: Fri, 29 May 2015 13:11:32 -0600
Subject: block: don't honor chunk sizes for data-less IO

We don't need to honor chunk sizes for IO that doesn't carry any
data.

Signed-off-by: Jens Axboe <axboe@fb.com>
---
 include/linux/blkdev.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux/blkdev.h')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 9ded80da2c16..ccaa9aecd593 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -903,7 +903,7 @@ static inline unsigned int blk_rq_get_max_sectors(struct request *rq)
 	if (unlikely(rq->cmd_type == REQ_TYPE_BLOCK_PC))
 		return q->limits.max_hw_sectors;
 
-	if (!q->limits.chunk_sectors)
+	if (!q->limits.chunk_sectors || (rq->cmd_flags & REQ_DISCARD))
 		return blk_queue_get_max_sectors(q, rq->cmd_flags);
 
 	return min(blk_max_size_offset(q, blk_rq_pos(rq)),
-- 
cgit v1.2.3-71-gd317


From 3f21c265cd5f7ae867cc0e86a1f6d5093f1963cc Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@fb.com>
Date: Fri, 5 Jun 2015 10:57:37 -0600
Subject: block: add blk_set_queue_dying() to blkdev.h

We export this function and NVMe wants to use it, but for some reason
it was never added to the block header. Do that.

Signed-off-by: Jens Axboe <axboe@fb.com>
---
 include/linux/blkdev.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux/blkdev.h')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index ccaa9aecd593..a31380c35918 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1024,6 +1024,7 @@ bool __must_check blk_get_queue(struct request_queue *);
 struct request_queue *blk_alloc_queue(gfp_t);
 struct request_queue *blk_alloc_queue_node(gfp_t, int);
 extern void blk_put_queue(struct request_queue *);
+extern void blk_set_queue_dying(struct request_queue *);
 
 /*
  * block layer runtime pm functions
-- 
cgit v1.2.3-71-gd317