From 9208d414975895f69e9aca49153060ddd31b18d0 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Thu, 21 Oct 2021 08:06:01 +0200
Subject: [PATCH 01/17] block: add a ->get_unique_id method

Add a method to query unique IDs from block devices.  It will be used to
remove code that deeply pokes into SCSI internals in the NFS server.
The implementation in the sd driver itself is also much nicer as it can
use the cached VPD page instead of always sending a command as the
current NFS code does.

For now the interface is kept very minimal but could be easily
extended when other users like a block-layer sysfs interface for
uniquue IDs shows up.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Link: https://lore.kernel.org/r/20211021060607.264371-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 include/linux/blkdev.h | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index f72ccb2829db..0d5826066e16 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1158,6 +1158,14 @@ static inline void blk_crypto_unregister(struct request_queue *q) { }
 
 #endif /* CONFIG_BLK_INLINE_ENCRYPTION */
 
+enum blk_unique_id {
+	/* these match the Designator Types specified in SPC */
+	BLK_UID_T10	= 1,
+	BLK_UID_EUI64	= 2,
+	BLK_UID_NAA	= 3,
+};
+
+#define NFL4_UFLG_MASK			0x0000003F
 
 struct block_device_operations {
 	void (*submit_bio)(struct bio *bio);
@@ -1176,6 +1184,9 @@ struct block_device_operations {
 	int (*report_zones)(struct gendisk *, sector_t sector,
 			unsigned int nr_zones, report_zones_cb cb, void *data);
 	char *(*devnode)(struct gendisk *disk, umode_t *mode);
+	/* returns the length of the identifier or a negative errno: */
+	int (*get_unique_id)(struct gendisk *disk, u8 id[16],
+			enum blk_unique_id id_type);
 	struct module *owner;
 	const struct pr_ops *pr_ops;
 

From b83ce214af3885437ff223b3a0c8ec6072a84167 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Thu, 21 Oct 2021 08:06:02 +0200
Subject: [PATCH 02/17] sd: implement ->get_unique_id

Add the method to query for a uniqueue ID of a given type by looking
it up in the cached device identification VPD page.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Link: https://lore.kernel.org/r/20211021060607.264371-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 drivers/scsi/sd.c | 39 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 39 insertions(+)

diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index d8f6add416c0..9b386ca39a91 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -1757,6 +1757,44 @@ static void sd_rescan(struct device *dev)
 	sd_revalidate_disk(sdkp->disk);
 }
 
+static int sd_get_unique_id(struct gendisk *disk, u8 id[16],
+		enum blk_unique_id type)
+{
+	struct scsi_device *sdev = scsi_disk(disk)->device;
+	const struct scsi_vpd *vpd;
+	const unsigned char *d;
+	int ret = -ENXIO, len;
+
+	rcu_read_lock();
+	vpd = rcu_dereference(sdev->vpd_pg83);
+	if (!vpd)
+		goto out_unlock;
+
+	ret = -EINVAL;
+	for (d = vpd->data + 4; d < vpd->data + vpd->len; d += d[3] + 4) {
+		/* we only care about designators with LU association */
+		if (((d[1] >> 4) & 0x3) != 0x00)
+			continue;
+		if ((d[1] & 0xf) != type)
+			continue;
+
+		/*
+		 * Only exit early if a 16-byte descriptor was found.  Otherwise
+		 * keep looking as one with more entropy might still show up.
+		 */
+		len = d[3];
+		if (len != 8 && len != 12 && len != 16)
+			continue;
+		ret = len;
+		memcpy(id, d + 4, len);
+		if (len == 16)
+			break;
+	}
+out_unlock:
+	rcu_read_unlock();
+	return ret;
+}
+
 static char sd_pr_type(enum pr_type type)
 {
 	switch (type) {
@@ -1861,6 +1899,7 @@ static const struct block_device_operations sd_fops = {
 	.check_events		= sd_check_events,
 	.unlock_native_capacity	= sd_unlock_native_capacity,
 	.report_zones		= sd_zbc_report_zones,
+	.get_unique_id		= sd_get_unique_id,
 	.pr_ops			= &sd_pr_ops,
 };
 

From 8c6aabd1c72bc241c55f5b71a86cea5ef28bceca Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Thu, 21 Oct 2021 08:06:03 +0200
Subject: [PATCH 03/17] nfsd/blocklayout: use ->get_unique_id instead of
 sending SCSI commands

Call the ->get_unique_id method to query the SCSI identifiers.  This can
use the cached VPD page in the sd driver instead of sending a command
on every LAYOUTGET.  It will also allow to support NVMe based volumes
if the draft for that ever takes off.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: J. Bruce Fields <bfields@redhat.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Link: https://lore.kernel.org/r/20211021060607.264371-4-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 fs/nfsd/Kconfig       |   1 -
 fs/nfsd/blocklayout.c | 158 +++++++++++-------------------------------
 fs/nfsd/nfs4layouts.c |   5 +-
 3 files changed, 44 insertions(+), 120 deletions(-)

diff --git a/fs/nfsd/Kconfig b/fs/nfsd/Kconfig
index 6e9ea4ee0f73..3d1d17256a91 100644
--- a/fs/nfsd/Kconfig
+++ b/fs/nfsd/Kconfig
@@ -109,7 +109,6 @@ config NFSD_SCSILAYOUT
 	depends on NFSD_V4 && BLOCK
 	select NFSD_PNFS
 	select EXPORTFS_BLOCK_OPS
-	select SCSI_COMMON
 	help
 	  This option enables support for the exporting pNFS SCSI layouts
 	  in the kernel's NFS server. The pNFS SCSI layout enables NFS
diff --git a/fs/nfsd/blocklayout.c b/fs/nfsd/blocklayout.c
index c99dee99a3c1..e5c0982a381d 100644
--- a/fs/nfsd/blocklayout.c
+++ b/fs/nfsd/blocklayout.c
@@ -9,9 +9,6 @@
 #include <linux/pr.h>
 
 #include <linux/nfsd/debug.h>
-#include <scsi/scsi_proto.h>
-#include <scsi/scsi_common.h>
-#include <scsi/scsi_request.h>
 
 #include "blocklayoutxdr.h"
 #include "pnfs.h"
@@ -211,109 +208,6 @@ const struct nfsd4_layout_ops bl_layout_ops = {
 #endif /* CONFIG_NFSD_BLOCKLAYOUT */
 
 #ifdef CONFIG_NFSD_SCSILAYOUT
-static int nfsd4_scsi_identify_device(struct block_device *bdev,
-		struct pnfs_block_volume *b)
-{
-	struct request_queue *q = bdev->bd_disk->queue;
-	struct request *rq;
-	struct scsi_request *req;
-	/*
-	 * The allocation length (passed in bytes 3 and 4 of the INQUIRY
-	 * command descriptor block) specifies the number of bytes that have
-	 * been allocated for the data-in buffer.
-	 * 252 is the highest one-byte value that is a multiple of 4.
-	 * 65532 is the highest two-byte value that is a multiple of 4.
-	 */
-	size_t bufflen = 252, maxlen = 65532, len, id_len;
-	u8 *buf, *d, type, assoc;
-	int retries = 1, error;
-
-	if (WARN_ON_ONCE(!blk_queue_scsi_passthrough(q)))
-		return -EINVAL;
-
-again:
-	buf = kzalloc(bufflen, GFP_KERNEL);
-	if (!buf)
-		return -ENOMEM;
-
-	rq = blk_get_request(q, REQ_OP_DRV_IN, 0);
-	if (IS_ERR(rq)) {
-		error = -ENOMEM;
-		goto out_free_buf;
-	}
-	req = scsi_req(rq);
-
-	error = blk_rq_map_kern(q, rq, buf, bufflen, GFP_KERNEL);
-	if (error)
-		goto out_put_request;
-
-	req->cmd[0] = INQUIRY;
-	req->cmd[1] = 1;
-	req->cmd[2] = 0x83;
-	req->cmd[3] = bufflen >> 8;
-	req->cmd[4] = bufflen & 0xff;
-	req->cmd_len = COMMAND_SIZE(INQUIRY);
-
-	blk_execute_rq(NULL, rq, 1);
-	if (req->result) {
-		pr_err("pNFS: INQUIRY 0x83 failed with: %x\n",
-			req->result);
-		error = -EIO;
-		goto out_put_request;
-	}
-
-	len = (buf[2] << 8) + buf[3] + 4;
-	if (len > bufflen) {
-		if (len <= maxlen && retries--) {
-			blk_put_request(rq);
-			kfree(buf);
-			bufflen = len;
-			goto again;
-		}
-		pr_err("pNFS: INQUIRY 0x83 response invalid (len = %zd)\n",
-			len);
-		goto out_put_request;
-	}
-
-	d = buf + 4;
-	for (d = buf + 4; d < buf + len; d += id_len + 4) {
-		id_len = d[3];
-		type = d[1] & 0xf;
-		assoc = (d[1] >> 4) & 0x3;
-
-		/*
-		 * We only care about a EUI-64 and NAA designator types
-		 * with LU association.
-		 */
-		if (assoc != 0x00)
-			continue;
-		if (type != 0x02 && type != 0x03)
-			continue;
-		if (id_len != 8 && id_len != 12 && id_len != 16)
-			continue;
-
-		b->scsi.code_set = PS_CODE_SET_BINARY;
-		b->scsi.designator_type = type == 0x02 ?
-			PS_DESIGNATOR_EUI64 : PS_DESIGNATOR_NAA;
-		b->scsi.designator_len = id_len;
-		memcpy(b->scsi.designator, d + 4, id_len);
-
-		/*
-		 * If we found a 8 or 12 byte descriptor continue on to
-		 * see if a 16 byte one is available.  If we find a
-		 * 16 byte descriptor we're done.
-		 */
-		if (id_len == 16)
-			break;
-	}
-
-out_put_request:
-	blk_put_request(rq);
-out_free_buf:
-	kfree(buf);
-	return error;
-}
-
 #define NFSD_MDS_PR_KEY		0x0100000000000000ULL
 
 /*
@@ -325,6 +219,31 @@ static u64 nfsd4_scsi_pr_key(struct nfs4_client *clp)
 	return ((u64)clp->cl_clientid.cl_boot << 32) | clp->cl_clientid.cl_id;
 }
 
+static const u8 designator_types[] = {
+	PS_DESIGNATOR_EUI64,
+	PS_DESIGNATOR_NAA,
+};
+
+static int
+nfsd4_block_get_unique_id(struct gendisk *disk, struct pnfs_block_volume *b)
+{
+	int ret, i;
+
+	for (i = 0; i < ARRAY_SIZE(designator_types); i++) {
+		u8 type = designator_types[i];
+
+		ret = disk->fops->get_unique_id(disk, b->scsi.designator, type);
+		if (ret > 0) {
+			b->scsi.code_set = PS_CODE_SET_BINARY;
+			b->scsi.designator_type = type;
+			b->scsi.designator_len = ret;
+			return 0;
+		}
+	}
+
+	return -EINVAL;
+}
+
 static int
 nfsd4_block_get_device_info_scsi(struct super_block *sb,
 		struct nfs4_client *clp,
@@ -333,7 +252,7 @@ nfsd4_block_get_device_info_scsi(struct super_block *sb,
 	struct pnfs_block_deviceaddr *dev;
 	struct pnfs_block_volume *b;
 	const struct pr_ops *ops;
-	int error;
+	int ret;
 
 	dev = kzalloc(sizeof(struct pnfs_block_deviceaddr) +
 		      sizeof(struct pnfs_block_volume), GFP_KERNEL);
@@ -347,33 +266,38 @@ nfsd4_block_get_device_info_scsi(struct super_block *sb,
 	b->type = PNFS_BLOCK_VOLUME_SCSI;
 	b->scsi.pr_key = nfsd4_scsi_pr_key(clp);
 
-	error = nfsd4_scsi_identify_device(sb->s_bdev, b);
-	if (error)
-		return error;
+	ret = nfsd4_block_get_unique_id(sb->s_bdev->bd_disk, b);
+	if (ret < 0)
+		goto out_free_dev;
 
+	ret = -EINVAL;
 	ops = sb->s_bdev->bd_disk->fops->pr_ops;
 	if (!ops) {
 		pr_err("pNFS: device %s does not support PRs.\n",
 			sb->s_id);
-		return -EINVAL;
+		goto out_free_dev;
 	}
 
-	error = ops->pr_register(sb->s_bdev, 0, NFSD_MDS_PR_KEY, true);
-	if (error) {
+	ret = ops->pr_register(sb->s_bdev, 0, NFSD_MDS_PR_KEY, true);
+	if (ret) {
 		pr_err("pNFS: failed to register key for device %s.\n",
 			sb->s_id);
-		return -EINVAL;
+		goto out_free_dev;
 	}
 
-	error = ops->pr_reserve(sb->s_bdev, NFSD_MDS_PR_KEY,
+	ret = ops->pr_reserve(sb->s_bdev, NFSD_MDS_PR_KEY,
 			PR_EXCLUSIVE_ACCESS_REG_ONLY, 0);
-	if (error) {
+	if (ret) {
 		pr_err("pNFS: failed to reserve device %s.\n",
 			sb->s_id);
-		return -EINVAL;
+		goto out_free_dev;
 	}
 
 	return 0;
+
+out_free_dev:
+	kfree(dev);
+	return ret;
 }
 
 static __be32
diff --git a/fs/nfsd/nfs4layouts.c b/fs/nfsd/nfs4layouts.c
index a97873f2d22b..6d1b5bb051c5 100644
--- a/fs/nfsd/nfs4layouts.c
+++ b/fs/nfsd/nfs4layouts.c
@@ -145,8 +145,9 @@ void nfsd4_setup_layout_type(struct svc_export *exp)
 #ifdef CONFIG_NFSD_SCSILAYOUT
 	if (sb->s_export_op->map_blocks &&
 	    sb->s_export_op->commit_blocks &&
-	    sb->s_bdev && sb->s_bdev->bd_disk->fops->pr_ops &&
-		blk_queue_scsi_passthrough(sb->s_bdev->bd_disk->queue))
+	    sb->s_bdev &&
+	    sb->s_bdev->bd_disk->fops->pr_ops &&
+	    sb->s_bdev->bd_disk->fops->get_unique_id)
 		exp->ex_layout_types |= 1 << LAYOUT_SCSI;
 #endif
 }

From 237ea1602fb4cd14cd31b745a56fd0639c58eea3 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Thu, 21 Oct 2021 08:06:04 +0200
Subject: [PATCH 04/17] bsg-lib: initialize the bsg_job in
 bsg_transport_sg_io_fn

Directly initialize the bsg_job structure instead of relying on the
->.initialize_rq_fn indirection.  This also removes the superflous
initialization of the second request used for BIDI requests.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Link: https://lore.kernel.org/r/20211021060607.264371-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/bsg-lib.c | 32 +++++++++++++-------------------
 1 file changed, 13 insertions(+), 19 deletions(-)

diff --git a/block/bsg-lib.c b/block/bsg-lib.c
index ccb98276c964..10aa378702fa 100644
--- a/block/bsg-lib.c
+++ b/block/bsg-lib.c
@@ -31,6 +31,7 @@ static int bsg_transport_sg_io_fn(struct request_queue *q, struct sg_io_v4 *hdr,
 	struct bsg_job *job;
 	struct request *rq;
 	struct bio *bio;
+	void *reply;
 	int ret;
 
 	if (hdr->protocol != BSG_PROTOCOL_SCSI  ||
@@ -39,22 +40,28 @@ static int bsg_transport_sg_io_fn(struct request_queue *q, struct sg_io_v4 *hdr,
 	if (!capable(CAP_SYS_RAWIO))
 		return -EPERM;
 
-	rq = blk_get_request(q, hdr->dout_xfer_len ?
+	rq = blk_mq_alloc_request(q, hdr->dout_xfer_len ?
 			     REQ_OP_DRV_OUT : REQ_OP_DRV_IN, 0);
 	if (IS_ERR(rq))
 		return PTR_ERR(rq);
 	rq->timeout = timeout;
 
 	job = blk_mq_rq_to_pdu(rq);
+	reply = job->reply;
+	memset(job, 0, sizeof(*job));
+	job->reply = reply;
+	job->reply_len = SCSI_SENSE_BUFFERSIZE;
+	job->dd_data = job + 1;
+
 	job->request_len = hdr->request_len;
 	job->request = memdup_user(uptr64(hdr->request), hdr->request_len);
 	if (IS_ERR(job->request)) {
 		ret = PTR_ERR(job->request);
-		goto out_put_request;
+		goto out_free_rq;
 	}
 
 	if (hdr->dout_xfer_len && hdr->din_xfer_len) {
-		job->bidi_rq = blk_get_request(rq->q, REQ_OP_DRV_IN, 0);
+		job->bidi_rq = blk_mq_alloc_request(rq->q, REQ_OP_DRV_IN, 0);
 		if (IS_ERR(job->bidi_rq)) {
 			ret = PTR_ERR(job->bidi_rq);
 			goto out_free_job_request;
@@ -134,11 +141,11 @@ out_unmap_bidi_rq:
 		blk_rq_unmap_user(job->bidi_bio);
 out_free_bidi_rq:
 	if (job->bidi_rq)
-		blk_put_request(job->bidi_rq);
+		blk_mq_free_request(job->bidi_rq);
 out_free_job_request:
 	kfree(job->request);
-out_put_request:
-	blk_put_request(rq);
+out_free_rq:
+	blk_mq_free_request(rq);
 	return ret;
 }
 
@@ -302,18 +309,6 @@ static int bsg_init_rq(struct blk_mq_tag_set *set, struct request *req,
 	return 0;
 }
 
-/* called right before the request is given to the request_queue user */
-static void bsg_initialize_rq(struct request *req)
-{
-	struct bsg_job *job = blk_mq_rq_to_pdu(req);
-	void *reply = job->reply;
-
-	memset(job, 0, sizeof(*job));
-	job->reply = reply;
-	job->reply_len = SCSI_SENSE_BUFFERSIZE;
-	job->dd_data = job + 1;
-}
-
 static void bsg_exit_rq(struct blk_mq_tag_set *set, struct request *req,
 		       unsigned int hctx_idx)
 {
@@ -350,7 +345,6 @@ static const struct blk_mq_ops bsg_mq_ops = {
 	.queue_rq		= bsg_queue_rq,
 	.init_request		= bsg_init_rq,
 	.exit_request		= bsg_exit_rq,
-	.initialize_rq_fn	= bsg_initialize_rq,
 	.complete		= bsg_complete,
 	.timeout		= bsg_timeout,
 };

From 68ec3b819a5d600a4ede8b596761dccac9f39ebc Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Thu, 21 Oct 2021 08:06:05 +0200
Subject: [PATCH 05/17] scsi: add a scsi_alloc_request helper

Add a new helper that calls blk_get_request and initializes the
scsi_request to avoid the indirect call through ->.initialize_rq_fn.

Note that this makes the pktcdvd driver depend on the SCSI core, but
given that only SCSI devices support SCSI passthrough requests that
is not a functional change.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Link: https://lore.kernel.org/r/20211021060607.264371-6-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 drivers/block/Kconfig              |  2 +-
 drivers/block/pktcdvd.c            |  2 +-
 drivers/scsi/scsi_bsg.c            |  4 ++--
 drivers/scsi/scsi_error.c          |  2 +-
 drivers/scsi/scsi_ioctl.c          |  4 ++--
 drivers/scsi/scsi_lib.c            | 19 +++++++++++++------
 drivers/scsi/sg.c                  |  4 ++--
 drivers/scsi/sr.c                  |  2 +-
 drivers/scsi/st.c                  |  2 +-
 drivers/target/target_core_pscsi.c |  3 +--
 include/scsi/scsi_cmnd.h           |  3 +++
 11 files changed, 28 insertions(+), 19 deletions(-)

diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig
index ab3e37aa1830..9151e8ffba1c 100644
--- a/drivers/block/Kconfig
+++ b/drivers/block/Kconfig
@@ -304,8 +304,8 @@ config BLK_DEV_RAM_SIZE
 config CDROM_PKTCDVD
 	tristate "Packet writing on CD/DVD media (DEPRECATED)"
 	depends on !UML
+	depends on SCSI
 	select CDROM
-	select SCSI_COMMON
 	help
 	  Note: This driver is deprecated and will be removed from the
 	  kernel in the near future!
diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index cb52cce6fb03..ea2262ec76d2 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -703,7 +703,7 @@ static int pkt_generic_packet(struct pktcdvd_device *pd, struct packet_command *
 	struct request *rq;
 	int ret = 0;
 
-	rq = blk_get_request(q, (cgc->data_direction == CGC_DATA_WRITE) ?
+	rq = scsi_alloc_request(q, (cgc->data_direction == CGC_DATA_WRITE) ?
 			     REQ_OP_DRV_OUT : REQ_OP_DRV_IN, 0);
 	if (IS_ERR(rq))
 		return PTR_ERR(rq);
diff --git a/drivers/scsi/scsi_bsg.c b/drivers/scsi/scsi_bsg.c
index 81c3853a2a80..551727a6f694 100644
--- a/drivers/scsi/scsi_bsg.c
+++ b/drivers/scsi/scsi_bsg.c
@@ -25,8 +25,8 @@ static int scsi_bsg_sg_io_fn(struct request_queue *q, struct sg_io_v4 *hdr,
 		return -EOPNOTSUPP;
 	}
 
-	rq = blk_get_request(q, hdr->dout_xfer_len ?
-			     REQ_OP_DRV_OUT : REQ_OP_DRV_IN, 0);
+	rq = scsi_alloc_request(q, hdr->dout_xfer_len ?
+				REQ_OP_DRV_OUT : REQ_OP_DRV_IN, 0);
 	if (IS_ERR(rq))
 		return PTR_ERR(rq);
 	rq->timeout = timeout;
diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index b6c86cce57bf..71d027b94be4 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -1998,7 +1998,7 @@ static void scsi_eh_lock_door(struct scsi_device *sdev)
 	struct request *req;
 	struct scsi_request *rq;
 
-	req = blk_get_request(sdev->request_queue, REQ_OP_DRV_IN, 0);
+	req = scsi_alloc_request(sdev->request_queue, REQ_OP_DRV_IN, 0);
 	if (IS_ERR(req))
 		return;
 	rq = scsi_req(req);
diff --git a/drivers/scsi/scsi_ioctl.c b/drivers/scsi/scsi_ioctl.c
index 6ff2207bd45a..0078975e3c07 100644
--- a/drivers/scsi/scsi_ioctl.c
+++ b/drivers/scsi/scsi_ioctl.c
@@ -438,7 +438,7 @@ static int sg_io(struct scsi_device *sdev, struct gendisk *disk,
 		at_head = 1;
 
 	ret = -ENOMEM;
-	rq = blk_get_request(sdev->request_queue, writing ?
+	rq = scsi_alloc_request(sdev->request_queue, writing ?
 			     REQ_OP_DRV_OUT : REQ_OP_DRV_IN, 0);
 	if (IS_ERR(rq))
 		return PTR_ERR(rq);
@@ -561,7 +561,7 @@ static int sg_scsi_ioctl(struct request_queue *q, struct gendisk *disk,
 
 	}
 
-	rq = blk_get_request(q, in_len ? REQ_OP_DRV_OUT : REQ_OP_DRV_IN, 0);
+	rq = scsi_alloc_request(q, in_len ? REQ_OP_DRV_OUT : REQ_OP_DRV_IN, 0);
 	if (IS_ERR(rq)) {
 		err = PTR_ERR(rq);
 		goto error_free_buffer;
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 30f7d0b4eb73..a0f801fc8943 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -216,7 +216,7 @@ int __scsi_execute(struct scsi_device *sdev, const unsigned char *cmd,
 	struct scsi_request *rq;
 	int ret;
 
-	req = blk_get_request(sdev->request_queue,
+	req = scsi_alloc_request(sdev->request_queue,
 			data_direction == DMA_TO_DEVICE ?
 			REQ_OP_DRV_OUT : REQ_OP_DRV_IN,
 			rq_flags & RQF_PM ? BLK_MQ_REQ_PM : 0);
@@ -1079,9 +1079,6 @@ EXPORT_SYMBOL(scsi_alloc_sgtables);
  * This function initializes the members of struct scsi_cmnd that must be
  * initialized before request processing starts and that won't be
  * reinitialized if a SCSI command is requeued.
- *
- * Called from inside blk_get_request() for pass-through requests and from
- * inside scsi_init_command() for filesystem requests.
  */
 static void scsi_initialize_rq(struct request *rq)
 {
@@ -1098,6 +1095,18 @@ static void scsi_initialize_rq(struct request *rq)
 	cmd->retries = 0;
 }
 
+struct request *scsi_alloc_request(struct request_queue *q,
+		unsigned int op, blk_mq_req_flags_t flags)
+{
+	struct request *rq;
+
+	rq = blk_get_request(q, op, flags);
+	if (!IS_ERR(rq))
+		scsi_initialize_rq(rq);
+	return rq;
+}
+EXPORT_SYMBOL_GPL(scsi_alloc_request);
+
 /*
  * Only called when the request isn't completed by SCSI, and not freed by
  * SCSI
@@ -1864,7 +1873,6 @@ static const struct blk_mq_ops scsi_mq_ops_no_commit = {
 #endif
 	.init_request	= scsi_mq_init_request,
 	.exit_request	= scsi_mq_exit_request,
-	.initialize_rq_fn = scsi_initialize_rq,
 	.cleanup_rq	= scsi_cleanup_rq,
 	.busy		= scsi_mq_lld_busy,
 	.map_queues	= scsi_map_queues,
@@ -1894,7 +1902,6 @@ static const struct blk_mq_ops scsi_mq_ops = {
 #endif
 	.init_request	= scsi_mq_init_request,
 	.exit_request	= scsi_mq_exit_request,
-	.initialize_rq_fn = scsi_initialize_rq,
 	.cleanup_rq	= scsi_cleanup_rq,
 	.busy		= scsi_mq_lld_busy,
 	.map_queues	= scsi_map_queues,
diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c
index 3c98f08dc25d..85f57ac0b844 100644
--- a/drivers/scsi/sg.c
+++ b/drivers/scsi/sg.c
@@ -1718,13 +1718,13 @@ sg_start_req(Sg_request *srp, unsigned char *cmd)
 	 *
 	 * With scsi-mq enabled, there are a fixed number of preallocated
 	 * requests equal in number to shost->can_queue.  If all of the
-	 * preallocated requests are already in use, then blk_get_request()
+	 * preallocated requests are already in use, then scsi_alloc_request()
 	 * will sleep until an active command completes, freeing up a request.
 	 * Although waiting in an asynchronous interface is less than ideal, we
 	 * do not want to use BLK_MQ_REQ_NOWAIT here because userspace might
 	 * not expect an EWOULDBLOCK from this condition.
 	 */
-	rq = blk_get_request(q, hp->dxfer_direction == SG_DXFER_TO_DEV ?
+	rq = scsi_alloc_request(q, hp->dxfer_direction == SG_DXFER_TO_DEV ?
 			REQ_OP_DRV_OUT : REQ_OP_DRV_IN, 0);
 	if (IS_ERR(rq)) {
 		kfree(long_cmdp);
diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c
index 115f7ef7a5de..7c4d9a964799 100644
--- a/drivers/scsi/sr.c
+++ b/drivers/scsi/sr.c
@@ -967,7 +967,7 @@ static int sr_read_cdda_bpc(struct cdrom_device_info *cdi, void __user *ubuf,
 	struct bio *bio;
 	int ret;
 
-	rq = blk_get_request(disk->queue, REQ_OP_DRV_IN, 0);
+	rq = scsi_alloc_request(disk->queue, REQ_OP_DRV_IN, 0);
 	if (IS_ERR(rq))
 		return PTR_ERR(rq);
 	req = scsi_req(rq);
diff --git a/drivers/scsi/st.c b/drivers/scsi/st.c
index 9933722acfd9..1275299f6159 100644
--- a/drivers/scsi/st.c
+++ b/drivers/scsi/st.c
@@ -543,7 +543,7 @@ static int st_scsi_execute(struct st_request *SRpnt, const unsigned char *cmd,
 	int err = 0;
 	struct scsi_tape *STp = SRpnt->stp;
 
-	req = blk_get_request(SRpnt->stp->device->request_queue,
+	req = scsi_alloc_request(SRpnt->stp->device->request_queue,
 			data_direction == DMA_TO_DEVICE ?
 			REQ_OP_DRV_OUT : REQ_OP_DRV_IN, 0);
 	if (IS_ERR(req))
diff --git a/drivers/target/target_core_pscsi.c b/drivers/target/target_core_pscsi.c
index 75ef52f008ff..b5705a2bd761 100644
--- a/drivers/target/target_core_pscsi.c
+++ b/drivers/target/target_core_pscsi.c
@@ -980,11 +980,10 @@ pscsi_execute_cmd(struct se_cmd *cmd)
 	memcpy(pt->pscsi_cdb, cmd->t_task_cdb,
 		scsi_command_size(cmd->t_task_cdb));
 
-	req = blk_get_request(pdv->pdv_sd->request_queue,
+	req = scsi_alloc_request(pdv->pdv_sd->request_queue,
 			cmd->data_direction == DMA_TO_DEVICE ?
 			REQ_OP_DRV_OUT : REQ_OP_DRV_IN, 0);
 	if (IS_ERR(req)) {
-		pr_err("PSCSI: blk_get_request() failed\n");
 		ret = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
 		goto fail;
 	}
diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h
index eaf04c9a1dfc..31078063afac 100644
--- a/include/scsi/scsi_cmnd.h
+++ b/include/scsi/scsi_cmnd.h
@@ -396,4 +396,7 @@ static inline unsigned scsi_transfer_length(struct scsi_cmnd *scmd)
 extern void scsi_build_sense(struct scsi_cmnd *scmd, int desc,
 			     u8 key, u8 asc, u8 ascq);
 
+struct request *scsi_alloc_request(struct request_queue *q,
+		unsigned int op, blk_mq_req_flags_t flags);
+
 #endif /* _SCSI_SCSI_CMND_H */

From 4abafdc4360d993104c2b2f85943938a0c6ad025 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Thu, 21 Oct 2021 08:06:06 +0200
Subject: [PATCH 06/17] block: remove the initialize_rq_fn blk_mq_ops method

Entirely unused now.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Link: https://lore.kernel.org/r/20211021060607.264371-7-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/blk-core.c       | 9 +--------
 include/linux/blk-mq.h | 5 -----
 2 files changed, 1 insertion(+), 13 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index fd389a16013c..5ffe05b1d17c 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -606,16 +606,9 @@ EXPORT_SYMBOL(blk_get_queue);
 struct request *blk_get_request(struct request_queue *q, unsigned int op,
 				blk_mq_req_flags_t flags)
 {
-	struct request *req;
-
 	WARN_ON_ONCE(op & REQ_NOWAIT);
 	WARN_ON_ONCE(flags & ~(BLK_MQ_REQ_NOWAIT | BLK_MQ_REQ_PM));
-
-	req = blk_mq_alloc_request(q, op, flags);
-	if (!IS_ERR(req) && q->mq_ops->initialize_rq_fn)
-		q->mq_ops->initialize_rq_fn(req);
-
-	return req;
+	return blk_mq_alloc_request(q, op, flags);
 }
 EXPORT_SYMBOL(blk_get_request);
 
diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h
index b4039fdf1b04..ebc45cf0450b 100644
--- a/include/linux/blk-mq.h
+++ b/include/linux/blk-mq.h
@@ -566,11 +566,6 @@ struct blk_mq_ops {
 	void (*exit_request)(struct blk_mq_tag_set *set, struct request *,
 			     unsigned int);
 
-	/**
-	 * @initialize_rq_fn: Called from inside blk_get_request().
-	 */
-	void (*initialize_rq_fn)(struct request *rq);
-
 	/**
 	 * @cleanup_rq: Called before freeing one request which isn't completed
 	 * yet, and usually for freeing the driver private data.

From 4845012eb5b4e56cadb5f484cb55dd4fd9d1df80 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Thu, 21 Oct 2021 08:06:07 +0200
Subject: [PATCH 07/17] block: remove QUEUE_FLAG_SCSI_PASSTHROUGH

Export scsi_device_from_queue for use with pktcdvd and use that instead
of the otherwise unused QUEUE_FLAG_SCSI_PASSTHROUGH queue flag.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Link: https://lore.kernel.org/r/20211021060607.264371-8-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/blk-mq-debugfs.c   | 1 -
 drivers/block/pktcdvd.c  | 5 ++++-
 drivers/scsi/scsi_lib.c  | 8 ++++++++
 drivers/scsi/scsi_scan.c | 1 -
 include/linux/blkdev.h   | 3 ---
 5 files changed, 12 insertions(+), 6 deletions(-)

diff --git a/block/blk-mq-debugfs.c b/block/blk-mq-debugfs.c
index 68ca5d21cda7..a317f05de466 100644
--- a/block/blk-mq-debugfs.c
+++ b/block/blk-mq-debugfs.c
@@ -124,7 +124,6 @@ static const char *const blk_queue_flag_name[] = {
 	QUEUE_FLAG_NAME(STATS),
 	QUEUE_FLAG_NAME(POLL_STATS),
 	QUEUE_FLAG_NAME(REGISTERED),
-	QUEUE_FLAG_NAME(SCSI_PASSTHROUGH),
 	QUEUE_FLAG_NAME(QUIESCED),
 	QUEUE_FLAG_NAME(PCI_P2PDMA),
 	QUEUE_FLAG_NAME(ZONE_RESETALL),
diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index ea2262ec76d2..cacf64eedad8 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -2536,6 +2536,7 @@ static int pkt_new_dev(struct pktcdvd_device *pd, dev_t dev)
 	int i;
 	char b[BDEVNAME_SIZE];
 	struct block_device *bdev;
+	struct scsi_device *sdev;
 
 	if (pd->pkt_dev == dev) {
 		pkt_err(pd, "recursive setup not allowed\n");
@@ -2559,10 +2560,12 @@ static int pkt_new_dev(struct pktcdvd_device *pd, dev_t dev)
 	bdev = blkdev_get_by_dev(dev, FMODE_READ | FMODE_NDELAY, NULL);
 	if (IS_ERR(bdev))
 		return PTR_ERR(bdev);
-	if (!blk_queue_scsi_passthrough(bdev_get_queue(bdev))) {
+	sdev = scsi_device_from_queue(bdev->bd_disk->queue);
+	if (!sdev) {
 		blkdev_put(bdev, FMODE_READ | FMODE_NDELAY);
 		return -EINVAL;
 	}
+	put_device(&sdev->sdev_gendev);
 
 	/* This is safe, since we have a reference from open(). */
 	__module_get(THIS_MODULE);
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index a0f801fc8943..9823b65d1536 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1967,6 +1967,14 @@ struct scsi_device *scsi_device_from_queue(struct request_queue *q)
 
 	return sdev;
 }
+/*
+ * pktcdvd should have been integrated into the SCSI layers, but for historical
+ * reasons like the old IDE driver it isn't.  This export allows it to safely
+ * probe if a given device is a SCSI one and only attach to that.
+ */
+#ifdef CONFIG_CDROM_PKTCDVD_MODULE
+EXPORT_SYMBOL_GPL(scsi_device_from_queue);
+#endif
 
 /**
  * scsi_block_requests - Utility function used by low-level drivers to prevent
diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c
index fe22191522a3..2808c0cb5711 100644
--- a/drivers/scsi/scsi_scan.c
+++ b/drivers/scsi/scsi_scan.c
@@ -280,7 +280,6 @@ static struct scsi_device *scsi_alloc_sdev(struct scsi_target *starget,
 	sdev->request_queue = q;
 	q->queuedata = sdev;
 	__scsi_init_queue(sdev->host, q);
-	blk_queue_flag_set(QUEUE_FLAG_SCSI_PASSTHROUGH, q);
 	WARN_ON_ONCE(!blk_get_queue(q));
 
 	depth = sdev->host->cmd_per_lun ?: 1;
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 0d5826066e16..1ad30f85d30e 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -357,7 +357,6 @@ struct request_queue {
 #define QUEUE_FLAG_STATS	20	/* track IO start and completion times */
 #define QUEUE_FLAG_POLL_STATS	21	/* collecting stats for hybrid polling */
 #define QUEUE_FLAG_REGISTERED	22	/* queue has been registered to a disk */
-#define QUEUE_FLAG_SCSI_PASSTHROUGH 23	/* queue supports SCSI commands */
 #define QUEUE_FLAG_QUIESCED	24	/* queue has been quiesced */
 #define QUEUE_FLAG_PCI_P2PDMA	25	/* device supports PCI p2p requests */
 #define QUEUE_FLAG_ZONE_RESETALL 26	/* supports Zone Reset All */
@@ -391,8 +390,6 @@ bool blk_queue_flag_test_and_set(unsigned int flag, struct request_queue *q);
 #define blk_queue_secure_erase(q) \
 	(test_bit(QUEUE_FLAG_SECERASE, &(q)->queue_flags))
 #define blk_queue_dax(q)	test_bit(QUEUE_FLAG_DAX, &(q)->queue_flags)
-#define blk_queue_scsi_passthrough(q)	\
-	test_bit(QUEUE_FLAG_SCSI_PASSTHROUGH, &(q)->queue_flags)
 #define blk_queue_pci_p2pdma(q)	\
 	test_bit(QUEUE_FLAG_PCI_P2PDMA, &(q)->queue_flags)
 #ifdef CONFIG_BLK_RQ_ALLOC_TIME

From 9a208ba5c9afa62c7b1e9c6f5e783066e84e2d3c Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Tue, 19 Oct 2021 08:25:24 +0200
Subject: [PATCH 08/17] fs: remove __sync_filesystem

There is no clear benefit in having this helper vs just open coding it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Link: https://lore.kernel.org/r/20211019062530.2174626-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 fs/sync.c | 38 +++++++++++++++++---------------------
 1 file changed, 17 insertions(+), 21 deletions(-)

diff --git a/fs/sync.c b/fs/sync.c
index 1373a610dc78..0d6cdc507cb9 100644
--- a/fs/sync.c
+++ b/fs/sync.c
@@ -21,25 +21,6 @@
 #define VALID_FLAGS (SYNC_FILE_RANGE_WAIT_BEFORE|SYNC_FILE_RANGE_WRITE| \
 			SYNC_FILE_RANGE_WAIT_AFTER)
 
-/*
- * Do the filesystem syncing work. For simple filesystems
- * writeback_inodes_sb(sb) just dirties buffers with inodes so we have to
- * submit IO for these buffers via __sync_blockdev(). This also speeds up the
- * wait == 1 case since in that case write_inode() functions do
- * sync_dirty_buffer() and thus effectively write one block at a time.
- */
-static int __sync_filesystem(struct super_block *sb, int wait)
-{
-	if (wait)
-		sync_inodes_sb(sb);
-	else
-		writeback_inodes_sb(sb, WB_REASON_SYNC);
-
-	if (sb->s_op->sync_fs)
-		sb->s_op->sync_fs(sb, wait);
-	return __sync_blockdev(sb->s_bdev, wait);
-}
-
 /*
  * Write out and wait upon all dirty data associated with this
  * superblock.  Filesystem data as well as the underlying block
@@ -61,10 +42,25 @@ int sync_filesystem(struct super_block *sb)
 	if (sb_rdonly(sb))
 		return 0;
 
-	ret = __sync_filesystem(sb, 0);
+	/*
+	 * Do the filesystem syncing work.  For simple filesystems
+	 * writeback_inodes_sb(sb) just dirties buffers with inodes so we have
+	 * to submit I/O for these buffers via __sync_blockdev().  This also
+	 * speeds up the wait == 1 case since in that case write_inode()
+	 * methods call sync_dirty_buffer() and thus effectively write one block
+	 * at a time.
+	 */
+	writeback_inodes_sb(sb, WB_REASON_SYNC);
+	if (sb->s_op->sync_fs)
+		sb->s_op->sync_fs(sb, 0);
+	ret = __sync_blockdev(sb->s_bdev, 0);
 	if (ret < 0)
 		return ret;
-	return __sync_filesystem(sb, 1);
+
+	sync_inodes_sb(sb);
+	if (sb->s_op->sync_fs)
+		sb->s_op->sync_fs(sb, 1);
+	return __sync_blockdev(sb->s_bdev, 1);
 }
 EXPORT_SYMBOL(sync_filesystem);
 

From 70164eb6ccb76ab679b016b4b60123bf4ec6c162 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Tue, 19 Oct 2021 08:25:25 +0200
Subject: [PATCH 09/17] block: remove __sync_blockdev

Instead offer a new sync_blockdev_nowait helper for the !wait case.
This new helper is exported as it will grow modular callers in a bit.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211019062530.2174626-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/bdev.c           | 11 ++++++-----
 fs/internal.h          |  5 -----
 fs/sync.c              |  7 ++++---
 include/linux/blkdev.h |  5 +++++
 4 files changed, 15 insertions(+), 13 deletions(-)

diff --git a/block/bdev.c b/block/bdev.c
index 7e6156203a71..9a33c414f450 100644
--- a/block/bdev.c
+++ b/block/bdev.c
@@ -185,14 +185,13 @@ int sb_min_blocksize(struct super_block *sb, int size)
 
 EXPORT_SYMBOL(sb_min_blocksize);
 
-int __sync_blockdev(struct block_device *bdev, int wait)
+int sync_blockdev_nowait(struct block_device *bdev)
 {
 	if (!bdev)
 		return 0;
-	if (!wait)
-		return filemap_flush(bdev->bd_inode->i_mapping);
-	return filemap_write_and_wait(bdev->bd_inode->i_mapping);
+	return filemap_flush(bdev->bd_inode->i_mapping);
 }
+EXPORT_SYMBOL_GPL(sync_blockdev_nowait);
 
 /*
  * Write out and wait upon all the dirty data associated with a block
@@ -200,7 +199,9 @@ int __sync_blockdev(struct block_device *bdev, int wait)
  */
 int sync_blockdev(struct block_device *bdev)
 {
-	return __sync_blockdev(bdev, 1);
+	if (!bdev)
+		return 0;
+	return filemap_write_and_wait(bdev->bd_inode->i_mapping);
 }
 EXPORT_SYMBOL(sync_blockdev);
 
diff --git a/fs/internal.h b/fs/internal.h
index 3cd065c8a66b..b5caa16f4645 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -23,7 +23,6 @@ struct pipe_inode_info;
 #ifdef CONFIG_BLOCK
 extern void __init bdev_cache_init(void);
 
-extern int __sync_blockdev(struct block_device *bdev, int wait);
 void iterate_bdevs(void (*)(struct block_device *, void *), void *);
 void emergency_thaw_bdev(struct super_block *sb);
 #else
@@ -31,10 +30,6 @@ static inline void bdev_cache_init(void)
 {
 }
 
-static inline int __sync_blockdev(struct block_device *bdev, int wait)
-{
-	return 0;
-}
 static inline void iterate_bdevs(void (*f)(struct block_device *, void *),
 		void *arg)
 {
diff --git a/fs/sync.c b/fs/sync.c
index 0d6cdc507cb9..a621089eb07e 100644
--- a/fs/sync.c
+++ b/fs/sync.c
@@ -3,6 +3,7 @@
  * High-level sync()-related operations
  */
 
+#include <linux/blkdev.h>
 #include <linux/kernel.h>
 #include <linux/file.h>
 #include <linux/fs.h>
@@ -45,7 +46,7 @@ int sync_filesystem(struct super_block *sb)
 	/*
 	 * Do the filesystem syncing work.  For simple filesystems
 	 * writeback_inodes_sb(sb) just dirties buffers with inodes so we have
-	 * to submit I/O for these buffers via __sync_blockdev().  This also
+	 * to submit I/O for these buffers via sync_blockdev().  This also
 	 * speeds up the wait == 1 case since in that case write_inode()
 	 * methods call sync_dirty_buffer() and thus effectively write one block
 	 * at a time.
@@ -53,14 +54,14 @@ int sync_filesystem(struct super_block *sb)
 	writeback_inodes_sb(sb, WB_REASON_SYNC);
 	if (sb->s_op->sync_fs)
 		sb->s_op->sync_fs(sb, 0);
-	ret = __sync_blockdev(sb->s_bdev, 0);
+	ret = sync_blockdev_nowait(sb->s_bdev);
 	if (ret < 0)
 		return ret;
 
 	sync_inodes_sb(sb);
 	if (sb->s_op->sync_fs)
 		sb->s_op->sync_fs(sb, 1);
-	return __sync_blockdev(sb->s_bdev, 1);
+	return sync_blockdev(sb->s_bdev);
 }
 EXPORT_SYMBOL(sync_filesystem);
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index f72ccb2829db..4b5a6bbdacd0 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1266,6 +1266,7 @@ int truncate_bdev_range(struct block_device *bdev, fmode_t mode, loff_t lstart,
 #ifdef CONFIG_BLOCK
 void invalidate_bdev(struct block_device *bdev);
 int sync_blockdev(struct block_device *bdev);
+int sync_blockdev_nowait(struct block_device *bdev);
 #else
 static inline void invalidate_bdev(struct block_device *bdev)
 {
@@ -1274,6 +1275,10 @@ static inline int sync_blockdev(struct block_device *bdev)
 {
 	return 0;
 }
+static inline int sync_blockdev_nowait(struct block_device *bdev)
+{
+	return 0;
+}
 #endif
 int fsync_bdev(struct block_device *bdev);
 

From d39b0a2fae366793e707bbd576eea055b7e9db9f Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Tue, 19 Oct 2021 08:25:26 +0200
Subject: [PATCH 10/17] xen-blkback: use sync_blockdev

Use sync_blockdev instead of opencoding it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Link: https://lore.kernel.org/r/20211019062530.2174626-4-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 drivers/block/xen-blkback/xenbus.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/block/xen-blkback/xenbus.c b/drivers/block/xen-blkback/xenbus.c
index 33eba3df4dd9..914587aabca0 100644
--- a/drivers/block/xen-blkback/xenbus.c
+++ b/drivers/block/xen-blkback/xenbus.c
@@ -98,7 +98,7 @@ static void xen_update_blkif_status(struct xen_blkif *blkif)
 		return;
 	}
 
-	err = filemap_write_and_wait(blkif->vbd.bdev->bd_inode->i_mapping);
+	err = sync_blockdev(blkif->vbd.bdev);
 	if (err) {
 		xenbus_dev_error(blkif->be->dev, err, "block flush");
 		return;

From 1226dfff572f8878fdfc4ac2fd84a59c7a75f489 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Tue, 19 Oct 2021 08:25:27 +0200
Subject: [PATCH 11/17] btrfs: use sync_blockdev

Use sync_blockdev instead of opencoding it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Acked-by: David Sterba <dsterba@suse.com>
Link: https://lore.kernel.org/r/20211019062530.2174626-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 fs/btrfs/volumes.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 2ec3b8ac8fa3..b51e4b464103 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -508,7 +508,7 @@ btrfs_get_bdev_and_sb(const char *device_path, fmode_t flags, void *holder,
 	}
 
 	if (flush)
-		filemap_write_and_wait((*bdev)->bd_inode->i_mapping);
+		sync_blockdev(*bdev);
 	ret = set_blocksize(*bdev, BTRFS_BDEV_BLOCKSIZE);
 	if (ret) {
 		blkdev_put(*bdev, flags);

From cb9568ee755c8289855a7a5ce0e1ee0082eca576 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Tue, 19 Oct 2021 08:25:28 +0200
Subject: [PATCH 12/17] fat: use sync_blockdev_nowait

Use sync_blockdev_nowait instead of opencoding it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Link: https://lore.kernel.org/r/20211019062530.2174626-6-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 fs/fat/inode.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index de0c9b013a85..2fd5bfddb695 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -1943,10 +1943,8 @@ int fat_flush_inodes(struct super_block *sb, struct inode *i1, struct inode *i2)
 		ret = writeback_inode(i1);
 	if (!ret && i2)
 		ret = writeback_inode(i2);
-	if (!ret) {
-		struct address_space *mapping = sb->s_bdev->bd_inode->i_mapping;
-		ret = filemap_flush(mapping);
-	}
+	if (!ret)
+		ret = sync_blockdev_nowait(sb->s_bdev);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(fat_flush_inodes);

From 680e667bc2e4e458a373bbfd367b7174e4972eb5 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Tue, 19 Oct 2021 08:25:29 +0200
Subject: [PATCH 13/17] ntfs3: use sync_blockdev_nowait

Use sync_blockdev_nowait instead of opencoding it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Link: https://lore.kernel.org/r/20211019062530.2174626-7-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 fs/ntfs3/inode.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/ntfs3/inode.c b/fs/ntfs3/inode.c
index 859951d785cb..a87ab3ad3cd3 100644
--- a/fs/ntfs3/inode.c
+++ b/fs/ntfs3/inode.c
@@ -1046,7 +1046,7 @@ int ntfs_flush_inodes(struct super_block *sb, struct inode *i1,
 	if (!ret && i2)
 		ret = writeback_inode(i2);
 	if (!ret)
-		ret = filemap_flush(sb->s_bdev->bd_inode->i_mapping);
+		ret = sync_blockdev_nowait(sb->s_bdev);
 	return ret;
 }
 

From 1e03a36bdff4709c1bbf0f57f60ae3f776d51adf Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Tue, 19 Oct 2021 08:25:30 +0200
Subject: [PATCH 14/17] block: simplify the block device syncing code

Get rid of the indirections and just provide a sync_bdevs
helper for the generic sync code.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20211019062530.2174626-8-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/bdev.c           | 17 ++++++++++++++---
 fs/internal.h          |  6 ------
 fs/sync.c              | 23 ++++-------------------
 include/linux/blkdev.h |  4 ++++
 4 files changed, 22 insertions(+), 28 deletions(-)

diff --git a/block/bdev.c b/block/bdev.c
index 9a33c414f450..b4dab2fb6a74 100644
--- a/block/bdev.c
+++ b/block/bdev.c
@@ -1021,7 +1021,7 @@ int __invalidate_device(struct block_device *bdev, bool kill_dirty)
 }
 EXPORT_SYMBOL(__invalidate_device);
 
-void iterate_bdevs(void (*func)(struct block_device *, void *), void *arg)
+void sync_bdevs(bool wait)
 {
 	struct inode *inode, *old_inode = NULL;
 
@@ -1052,8 +1052,19 @@ void iterate_bdevs(void (*func)(struct block_device *, void *), void *arg)
 		bdev = I_BDEV(inode);
 
 		mutex_lock(&bdev->bd_disk->open_mutex);
-		if (bdev->bd_openers)
-			func(bdev, arg);
+		if (!bdev->bd_openers) {
+			; /* skip */
+		} else if (wait) {
+			/*
+			 * We keep the error status of individual mapping so
+			 * that applications can catch the writeback error using
+			 * fsync(2). See filemap_fdatawait_keep_errors() for
+			 * details.
+			 */
+			filemap_fdatawait_keep_errors(inode->i_mapping);
+		} else {
+			filemap_fdatawrite(inode->i_mapping);
+		}
 		mutex_unlock(&bdev->bd_disk->open_mutex);
 
 		spin_lock(&blockdev_superblock->s_inode_list_lock);
diff --git a/fs/internal.h b/fs/internal.h
index b5caa16f4645..cdd83d4899bb 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -23,17 +23,11 @@ struct pipe_inode_info;
 #ifdef CONFIG_BLOCK
 extern void __init bdev_cache_init(void);
 
-void iterate_bdevs(void (*)(struct block_device *, void *), void *);
 void emergency_thaw_bdev(struct super_block *sb);
 #else
 static inline void bdev_cache_init(void)
 {
 }
-
-static inline void iterate_bdevs(void (*f)(struct block_device *, void *),
-		void *arg)
-{
-}
 static inline int emergency_thaw_bdev(struct super_block *sb)
 {
 	return 0;
diff --git a/fs/sync.c b/fs/sync.c
index a621089eb07e..3ce8e2137f31 100644
--- a/fs/sync.c
+++ b/fs/sync.c
@@ -78,21 +78,6 @@ static void sync_fs_one_sb(struct super_block *sb, void *arg)
 		sb->s_op->sync_fs(sb, *(int *)arg);
 }
 
-static void fdatawrite_one_bdev(struct block_device *bdev, void *arg)
-{
-	filemap_fdatawrite(bdev->bd_inode->i_mapping);
-}
-
-static void fdatawait_one_bdev(struct block_device *bdev, void *arg)
-{
-	/*
-	 * We keep the error status of individual mapping so that
-	 * applications can catch the writeback error using fsync(2).
-	 * See filemap_fdatawait_keep_errors() for details.
-	 */
-	filemap_fdatawait_keep_errors(bdev->bd_inode->i_mapping);
-}
-
 /*
  * Sync everything. We start by waking flusher threads so that most of
  * writeback runs on all devices in parallel. Then we sync all inodes reliably
@@ -111,8 +96,8 @@ void ksys_sync(void)
 	iterate_supers(sync_inodes_one_sb, NULL);
 	iterate_supers(sync_fs_one_sb, &nowait);
 	iterate_supers(sync_fs_one_sb, &wait);
-	iterate_bdevs(fdatawrite_one_bdev, NULL);
-	iterate_bdevs(fdatawait_one_bdev, NULL);
+	sync_bdevs(false);
+	sync_bdevs(true);
 	if (unlikely(laptop_mode))
 		laptop_sync_completion();
 }
@@ -133,10 +118,10 @@ static void do_sync_work(struct work_struct *work)
 	 */
 	iterate_supers(sync_inodes_one_sb, &nowait);
 	iterate_supers(sync_fs_one_sb, &nowait);
-	iterate_bdevs(fdatawrite_one_bdev, NULL);
+	sync_bdevs(false);
 	iterate_supers(sync_inodes_one_sb, &nowait);
 	iterate_supers(sync_fs_one_sb, &nowait);
-	iterate_bdevs(fdatawrite_one_bdev, NULL);
+	sync_bdevs(false);
 	printk("Emergency Sync complete\n");
 	kfree(work);
 }
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 4b5a6bbdacd0..09fdf7018b7f 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1267,6 +1267,7 @@ int truncate_bdev_range(struct block_device *bdev, fmode_t mode, loff_t lstart,
 void invalidate_bdev(struct block_device *bdev);
 int sync_blockdev(struct block_device *bdev);
 int sync_blockdev_nowait(struct block_device *bdev);
+void sync_bdevs(bool wait);
 #else
 static inline void invalidate_bdev(struct block_device *bdev)
 {
@@ -1279,6 +1280,9 @@ static inline int sync_blockdev_nowait(struct block_device *bdev)
 {
 	return 0;
 }
+static inline void sync_bdevs(bool wait)
+{
+}
 #endif
 int fsync_bdev(struct block_device *bdev);
 

From 4c3d8accdce26afa8089609593aa92965b500f72 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@kernel.dk>
Date: Sat, 23 Oct 2021 07:52:23 -0600
Subject: [PATCH 15/17] usb: remove res2 argument from gadget code completions

The USB gadget code is the only code that every tried to utilize the
2nd argument of the aio completions, but there are strong suspicions
that it was never actually used by anything on the userspace side.

Out of the 3 cases that touch it, two of them just pass in the same
as res, and the last one passes in error/transfer in res like any
other normal use case.

Remove the 2nd argument, pass 0 like the rest of the in-kernel users
of kiocb based IO.

Link: https://lore.kernel.org/linux-block/20211021174021.273c82b1.john@metanate.com/
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 drivers/usb/gadget/function/f_fs.c | 2 +-
 drivers/usb/gadget/legacy/inode.c  | 6 ++----
 2 files changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c
index 8260f38025b7..7bd22398d52f 100644
--- a/drivers/usb/gadget/function/f_fs.c
+++ b/drivers/usb/gadget/function/f_fs.c
@@ -831,7 +831,7 @@ static void ffs_user_copy_worker(struct work_struct *work)
 		kthread_unuse_mm(io_data->mm);
 	}
 
-	io_data->kiocb->ki_complete(io_data->kiocb, ret, ret);
+	io_data->kiocb->ki_complete(io_data->kiocb, ret, 0);
 
 	if (io_data->ffs->ffs_eventfd && !kiocb_has_eventfd)
 		eventfd_signal(io_data->ffs->ffs_eventfd, 1);
diff --git a/drivers/usb/gadget/legacy/inode.c b/drivers/usb/gadget/legacy/inode.c
index 539220d7f5b6..28d3d4e71182 100644
--- a/drivers/usb/gadget/legacy/inode.c
+++ b/drivers/usb/gadget/legacy/inode.c
@@ -469,7 +469,7 @@ static void ep_user_copy_worker(struct work_struct *work)
 		ret = -EFAULT;
 
 	/* completing the iocb can drop the ctx and mm, don't touch mm after */
-	iocb->ki_complete(iocb, ret, ret);
+	iocb->ki_complete(iocb, ret, 0);
 
 	kfree(priv->buf);
 	kfree(priv->to_free);
@@ -496,11 +496,9 @@ static void ep_aio_complete(struct usb_ep *ep, struct usb_request *req)
 		kfree(priv->to_free);
 		kfree(priv);
 		iocb->private = NULL;
-		/* aio_complete() reports bytes-transferred _and_ faults */
-
 		iocb->ki_complete(iocb,
 				req->actual ? req->actual : (long)req->status,
-				req->status);
+				0);
 	} else {
 		/* ep_copy_to_user() won't report both; we hide some faults */
 		if (unlikely(0 != req->status))

From 6b19b766e8f077f29cdb47da5003469a85bbfb9c Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@kernel.dk>
Date: Thu, 21 Oct 2021 09:22:35 -0600
Subject: [PATCH 16/17] fs: get rid of the res2 iocb->ki_complete argument

The second argument was only used by the USB gadget code, yet everyone
pays the overhead of passing a zero to be passed into aio, where it
ends up being part of the aio res2 value.

Now that everybody is passing in zero, kill off the extra argument.

Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/fops.c                       |  2 +-
 crypto/af_alg.c                    |  2 +-
 drivers/block/loop.c               |  4 ++--
 drivers/nvme/target/io-cmd-file.c  |  4 ++--
 drivers/target/target_core_file.c  |  4 ++--
 drivers/usb/gadget/function/f_fs.c |  2 +-
 drivers/usb/gadget/legacy/inode.c  |  5 ++---
 fs/aio.c                           |  6 +++---
 fs/cachefiles/io.c                 | 12 ++++++------
 fs/ceph/file.c                     |  2 +-
 fs/cifs/file.c                     |  4 ++--
 fs/direct-io.c                     |  2 +-
 fs/fuse/file.c                     |  2 +-
 fs/io_uring.c                      |  6 +++---
 fs/iomap/direct-io.c               |  2 +-
 fs/nfs/direct.c                    |  2 +-
 fs/overlayfs/file.c                |  4 ++--
 include/linux/fs.h                 |  2 +-
 18 files changed, 33 insertions(+), 34 deletions(-)

diff --git a/block/fops.c b/block/fops.c
index 396537598e3e..d86ebda73e8c 100644
--- a/block/fops.c
+++ b/block/fops.c
@@ -164,7 +164,7 @@ static void blkdev_bio_end_io(struct bio *bio)
 				ret = blk_status_to_errno(dio->bio.bi_status);
 			}
 
-			dio->iocb->ki_complete(iocb, ret, 0);
+			dio->iocb->ki_complete(iocb, ret);
 			if (dio->flags & DIO_MULTI_BIO)
 				bio_put(&dio->bio);
 		} else {
diff --git a/crypto/af_alg.c b/crypto/af_alg.c
index 8bd288d2b089..3dd5a773c320 100644
--- a/crypto/af_alg.c
+++ b/crypto/af_alg.c
@@ -1076,7 +1076,7 @@ void af_alg_async_cb(struct crypto_async_request *_req, int err)
 	af_alg_free_resources(areq);
 	sock_put(sk);
 
-	iocb->ki_complete(iocb, err ? err : (int)resultlen, 0);
+	iocb->ki_complete(iocb, err ? err : (int)resultlen);
 }
 EXPORT_SYMBOL_GPL(af_alg_async_cb);
 
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 7bf4686af774..469d87e24fd4 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -554,7 +554,7 @@ static void lo_rw_aio_do_completion(struct loop_cmd *cmd)
 		blk_mq_complete_request(rq);
 }
 
-static void lo_rw_aio_complete(struct kiocb *iocb, long ret, long ret2)
+static void lo_rw_aio_complete(struct kiocb *iocb, long ret)
 {
 	struct loop_cmd *cmd = container_of(iocb, struct loop_cmd, iocb);
 
@@ -627,7 +627,7 @@ static int lo_rw_aio(struct loop_device *lo, struct loop_cmd *cmd,
 	lo_rw_aio_do_completion(cmd);
 
 	if (ret != -EIOCBQUEUED)
-		cmd->iocb.ki_complete(&cmd->iocb, ret, 0);
+		lo_rw_aio_complete(&cmd->iocb, ret);
 	return 0;
 }
 
diff --git a/drivers/nvme/target/io-cmd-file.c b/drivers/nvme/target/io-cmd-file.c
index 1dd1a0fe2e81..6aa30f30b572 100644
--- a/drivers/nvme/target/io-cmd-file.c
+++ b/drivers/nvme/target/io-cmd-file.c
@@ -125,7 +125,7 @@ static ssize_t nvmet_file_submit_bvec(struct nvmet_req *req, loff_t pos,
 	return call_iter(iocb, &iter);
 }
 
-static void nvmet_file_io_done(struct kiocb *iocb, long ret, long ret2)
+static void nvmet_file_io_done(struct kiocb *iocb, long ret)
 {
 	struct nvmet_req *req = container_of(iocb, struct nvmet_req, f.iocb);
 	u16 status = NVME_SC_SUCCESS;
@@ -222,7 +222,7 @@ static bool nvmet_file_execute_io(struct nvmet_req *req, int ki_flags)
 	}
 
 complete:
-	nvmet_file_io_done(&req->f.iocb, ret, 0);
+	nvmet_file_io_done(&req->f.iocb, ret);
 	return true;
 }
 
diff --git a/drivers/target/target_core_file.c b/drivers/target/target_core_file.c
index 02f64453b4c5..8190b840065f 100644
--- a/drivers/target/target_core_file.c
+++ b/drivers/target/target_core_file.c
@@ -245,7 +245,7 @@ struct target_core_file_cmd {
 	struct bio_vec	bvecs[];
 };
 
-static void cmd_rw_aio_complete(struct kiocb *iocb, long ret, long ret2)
+static void cmd_rw_aio_complete(struct kiocb *iocb, long ret)
 {
 	struct target_core_file_cmd *cmd;
 
@@ -303,7 +303,7 @@ fd_execute_rw_aio(struct se_cmd *cmd, struct scatterlist *sgl, u32 sgl_nents,
 		ret = call_read_iter(file, &aio_cmd->iocb, &iter);
 
 	if (ret != -EIOCBQUEUED)
-		cmd_rw_aio_complete(&aio_cmd->iocb, ret, 0);
+		cmd_rw_aio_complete(&aio_cmd->iocb, ret);
 
 	return 0;
 }
diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c
index 7bd22398d52f..e20c19a0f106 100644
--- a/drivers/usb/gadget/function/f_fs.c
+++ b/drivers/usb/gadget/function/f_fs.c
@@ -831,7 +831,7 @@ static void ffs_user_copy_worker(struct work_struct *work)
 		kthread_unuse_mm(io_data->mm);
 	}
 
-	io_data->kiocb->ki_complete(io_data->kiocb, ret, 0);
+	io_data->kiocb->ki_complete(io_data->kiocb, ret);
 
 	if (io_data->ffs->ffs_eventfd && !kiocb_has_eventfd)
 		eventfd_signal(io_data->ffs->ffs_eventfd, 1);
diff --git a/drivers/usb/gadget/legacy/inode.c b/drivers/usb/gadget/legacy/inode.c
index 28d3d4e71182..78be94750232 100644
--- a/drivers/usb/gadget/legacy/inode.c
+++ b/drivers/usb/gadget/legacy/inode.c
@@ -469,7 +469,7 @@ static void ep_user_copy_worker(struct work_struct *work)
 		ret = -EFAULT;
 
 	/* completing the iocb can drop the ctx and mm, don't touch mm after */
-	iocb->ki_complete(iocb, ret, 0);
+	iocb->ki_complete(iocb, ret);
 
 	kfree(priv->buf);
 	kfree(priv->to_free);
@@ -497,8 +497,7 @@ static void ep_aio_complete(struct usb_ep *ep, struct usb_request *req)
 		kfree(priv);
 		iocb->private = NULL;
 		iocb->ki_complete(iocb,
-				req->actual ? req->actual : (long)req->status,
-				0);
+				req->actual ? req->actual : (long)req->status);
 	} else {
 		/* ep_copy_to_user() won't report both; we hide some faults */
 		if (unlikely(0 != req->status))
diff --git a/fs/aio.c b/fs/aio.c
index 51b08ab01dff..836dc7e48db7 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -1417,7 +1417,7 @@ static void aio_remove_iocb(struct aio_kiocb *iocb)
 	spin_unlock_irqrestore(&ctx->ctx_lock, flags);
 }
 
-static void aio_complete_rw(struct kiocb *kiocb, long res, long res2)
+static void aio_complete_rw(struct kiocb *kiocb, long res)
 {
 	struct aio_kiocb *iocb = container_of(kiocb, struct aio_kiocb, rw);
 
@@ -1437,7 +1437,7 @@ static void aio_complete_rw(struct kiocb *kiocb, long res, long res2)
 	}
 
 	iocb->ki_res.res = res;
-	iocb->ki_res.res2 = res2;
+	iocb->ki_res.res2 = 0;
 	iocb_put(iocb);
 }
 
@@ -1508,7 +1508,7 @@ static inline void aio_rw_done(struct kiocb *req, ssize_t ret)
 		ret = -EINTR;
 		fallthrough;
 	default:
-		req->ki_complete(req, ret, 0);
+		req->ki_complete(req, ret);
 	}
 }
 
diff --git a/fs/cachefiles/io.c b/fs/cachefiles/io.c
index fac2e8e7b533..effe37ef8629 100644
--- a/fs/cachefiles/io.c
+++ b/fs/cachefiles/io.c
@@ -37,11 +37,11 @@ static inline void cachefiles_put_kiocb(struct cachefiles_kiocb *ki)
 /*
  * Handle completion of a read from the cache.
  */
-static void cachefiles_read_complete(struct kiocb *iocb, long ret, long ret2)
+static void cachefiles_read_complete(struct kiocb *iocb, long ret)
 {
 	struct cachefiles_kiocb *ki = container_of(iocb, struct cachefiles_kiocb, iocb);
 
-	_enter("%ld,%ld", ret, ret2);
+	_enter("%ld", ret);
 
 	if (ki->term_func) {
 		if (ret >= 0)
@@ -139,7 +139,7 @@ static int cachefiles_read(struct netfs_cache_resources *cres,
 		fallthrough;
 	default:
 		ki->was_async = false;
-		cachefiles_read_complete(&ki->iocb, ret, 0);
+		cachefiles_read_complete(&ki->iocb, ret);
 		if (ret > 0)
 			ret = 0;
 		break;
@@ -159,12 +159,12 @@ presubmission_error:
 /*
  * Handle completion of a write to the cache.
  */
-static void cachefiles_write_complete(struct kiocb *iocb, long ret, long ret2)
+static void cachefiles_write_complete(struct kiocb *iocb, long ret)
 {
 	struct cachefiles_kiocb *ki = container_of(iocb, struct cachefiles_kiocb, iocb);
 	struct inode *inode = file_inode(ki->iocb.ki_filp);
 
-	_enter("%ld,%ld", ret, ret2);
+	_enter("%ld", ret);
 
 	/* Tell lockdep we inherited freeze protection from submission thread */
 	__sb_writers_acquired(inode->i_sb, SB_FREEZE_WRITE);
@@ -244,7 +244,7 @@ static int cachefiles_write(struct netfs_cache_resources *cres,
 		fallthrough;
 	default:
 		ki->was_async = false;
-		cachefiles_write_complete(&ki->iocb, ret, 0);
+		cachefiles_write_complete(&ki->iocb, ret);
 		if (ret > 0)
 			ret = 0;
 		break;
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index d16fd2d5fd42..e10a80c1b581 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -1023,7 +1023,7 @@ static void ceph_aio_complete(struct inode *inode,
 	ceph_put_cap_refs(ci, (aio_req->write ? CEPH_CAP_FILE_WR :
 						CEPH_CAP_FILE_RD));
 
-	aio_req->iocb->ki_complete(aio_req->iocb, ret, 0);
+	aio_req->iocb->ki_complete(aio_req->iocb, ret);
 
 	ceph_free_cap_flush(aio_req->prealloc_cf);
 	kfree(aio_req);
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index 13f3182cf796..1b855fcb179e 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -3184,7 +3184,7 @@ restart_loop:
 	mutex_unlock(&ctx->aio_mutex);
 
 	if (ctx->iocb && ctx->iocb->ki_complete)
-		ctx->iocb->ki_complete(ctx->iocb, ctx->rc, 0);
+		ctx->iocb->ki_complete(ctx->iocb, ctx->rc);
 	else
 		complete(&ctx->done);
 }
@@ -3917,7 +3917,7 @@ again:
 	mutex_unlock(&ctx->aio_mutex);
 
 	if (ctx->iocb && ctx->iocb->ki_complete)
-		ctx->iocb->ki_complete(ctx->iocb, ctx->rc, 0);
+		ctx->iocb->ki_complete(ctx->iocb, ctx->rc);
 	else
 		complete(&ctx->done);
 }
diff --git a/fs/direct-io.c b/fs/direct-io.c
index 453dcff0e7f5..654443558047 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -307,7 +307,7 @@ static ssize_t dio_complete(struct dio *dio, ssize_t ret, unsigned int flags)
 
 		if (ret > 0 && dio->op == REQ_OP_WRITE)
 			ret = generic_write_sync(dio->iocb, ret);
-		dio->iocb->ki_complete(dio->iocb, ret, 0);
+		dio->iocb->ki_complete(dio->iocb, ret);
 	}
 
 	kmem_cache_free(dio_cache, dio);
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 11404f8c21c7..e6039f22311b 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -687,7 +687,7 @@ static void fuse_aio_complete(struct fuse_io_priv *io, int err, ssize_t pos)
 			spin_unlock(&fi->lock);
 		}
 
-		io->iocb->ki_complete(io->iocb, res, 0);
+		io->iocb->ki_complete(io->iocb, res);
 	}
 
 	kref_put(&io->refcnt, fuse_io_release);
diff --git a/fs/io_uring.c b/fs/io_uring.c
index d4631a55a692..edf29406ba8b 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -2689,7 +2689,7 @@ static void __io_complete_rw(struct io_kiocb *req, long res, long res2,
 	__io_req_complete(req, issue_flags, req->result, io_put_rw_kbuf(req));
 }
 
-static void io_complete_rw(struct kiocb *kiocb, long res, long res2)
+static void io_complete_rw(struct kiocb *kiocb, long res)
 {
 	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb);
 
@@ -2700,7 +2700,7 @@ static void io_complete_rw(struct kiocb *kiocb, long res, long res2)
 	io_req_task_work_add(req);
 }
 
-static void io_complete_rw_iopoll(struct kiocb *kiocb, long res, long res2)
+static void io_complete_rw_iopoll(struct kiocb *kiocb, long res)
 {
 	struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb);
 
@@ -2913,7 +2913,7 @@ static inline void io_rw_done(struct kiocb *kiocb, ssize_t ret)
 		ret = -EINTR;
 		fallthrough;
 	default:
-		kiocb->ki_complete(kiocb, ret, 0);
+		kiocb->ki_complete(kiocb, ret);
 	}
 }
 
diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index 83ecfba53abe..811c898125a5 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -125,7 +125,7 @@ static void iomap_dio_complete_work(struct work_struct *work)
 	struct iomap_dio *dio = container_of(work, struct iomap_dio, aio.work);
 	struct kiocb *iocb = dio->iocb;
 
-	iocb->ki_complete(iocb, iomap_dio_complete(dio), 0);
+	iocb->ki_complete(iocb, iomap_dio_complete(dio));
 }
 
 /*
diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 2e894fec036b..7a5f287c5391 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -275,7 +275,7 @@ static void nfs_direct_complete(struct nfs_direct_req *dreq)
 			res = (long) dreq->count;
 			WARN_ON_ONCE(dreq->count < 0);
 		}
-		dreq->iocb->ki_complete(dreq->iocb, res, 0);
+		dreq->iocb->ki_complete(dreq->iocb, res);
 	}
 
 	complete(&dreq->completion);
diff --git a/fs/overlayfs/file.c b/fs/overlayfs/file.c
index c88ac571593d..ac461a499882 100644
--- a/fs/overlayfs/file.c
+++ b/fs/overlayfs/file.c
@@ -272,14 +272,14 @@ static void ovl_aio_cleanup_handler(struct ovl_aio_req *aio_req)
 	kmem_cache_free(ovl_aio_request_cachep, aio_req);
 }
 
-static void ovl_aio_rw_complete(struct kiocb *iocb, long res, long res2)
+static void ovl_aio_rw_complete(struct kiocb *iocb, long res)
 {
 	struct ovl_aio_req *aio_req = container_of(iocb,
 						   struct ovl_aio_req, iocb);
 	struct kiocb *orig_iocb = aio_req->orig_iocb;
 
 	ovl_aio_cleanup_handler(aio_req);
-	orig_iocb->ki_complete(orig_iocb, res, res2);
+	orig_iocb->ki_complete(orig_iocb, res);
 }
 
 static ssize_t ovl_read_iter(struct kiocb *iocb, struct iov_iter *iter)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 31029a91f440..0dcb9020a7b3 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -330,7 +330,7 @@ struct kiocb {
 	randomized_struct_fields_start
 
 	loff_t			ki_pos;
-	void (*ki_complete)(struct kiocb *iocb, long ret, long ret2);
+	void (*ki_complete)(struct kiocb *iocb, long ret);
 	void			*private;
 	int			ki_flags;
 	u16			ki_hint;

From 0bf6d96cb8294094ce1e44cbe8cf65b0899d0a3a Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Mon, 25 Oct 2021 09:05:07 +0200
Subject: [PATCH 17/17] block: remove blk_{get,put}_request

These are now pointless wrappers around blk_mq_{alloc,free}_request,
so remove them.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Link: https://lore.kernel.org/r/20211025070517.1548584-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 block/blk-core.c                   | 21 ---------------------
 drivers/block/paride/pd.c          |  4 ++--
 drivers/block/pktcdvd.c            |  2 +-
 drivers/block/virtio_blk.c         |  4 ++--
 drivers/md/dm-mpath.c              |  4 ++--
 drivers/mmc/core/block.c           | 20 ++++++++++----------
 drivers/scsi/scsi_bsg.c            |  2 +-
 drivers/scsi/scsi_error.c          |  2 +-
 drivers/scsi/scsi_ioctl.c          |  4 ++--
 drivers/scsi/scsi_lib.c            |  4 ++--
 drivers/scsi/sg.c                  |  6 +++---
 drivers/scsi/sr.c                  |  2 +-
 drivers/scsi/st.c                  |  4 ++--
 drivers/scsi/ufs/ufshcd.c          | 20 ++++++++++----------
 drivers/scsi/ufs/ufshpb.c          |  8 ++++----
 drivers/target/target_core_pscsi.c |  4 ++--
 include/linux/blk-mq.h             |  3 ---
 17 files changed, 45 insertions(+), 69 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index 5ffe05b1d17c..ac1de7d73a45 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -597,27 +597,6 @@ bool blk_get_queue(struct request_queue *q)
 }
 EXPORT_SYMBOL(blk_get_queue);
 
-/**
- * blk_get_request - allocate a request
- * @q: request queue to allocate a request for
- * @op: operation (REQ_OP_*) and REQ_* flags, e.g. REQ_SYNC.
- * @flags: BLK_MQ_REQ_* flags, e.g. BLK_MQ_REQ_NOWAIT.
- */
-struct request *blk_get_request(struct request_queue *q, unsigned int op,
-				blk_mq_req_flags_t flags)
-{
-	WARN_ON_ONCE(op & REQ_NOWAIT);
-	WARN_ON_ONCE(flags & ~(BLK_MQ_REQ_NOWAIT | BLK_MQ_REQ_PM));
-	return blk_mq_alloc_request(q, op, flags);
-}
-EXPORT_SYMBOL(blk_get_request);
-
-void blk_put_request(struct request *req)
-{
-	blk_mq_free_request(req);
-}
-EXPORT_SYMBOL(blk_put_request);
-
 static void handle_bad_sector(struct bio *bio, sector_t maxsector)
 {
 	char b[BDEVNAME_SIZE];
diff --git a/drivers/block/paride/pd.c b/drivers/block/paride/pd.c
index 675327df6aff..9cd0bd509b88 100644
--- a/drivers/block/paride/pd.c
+++ b/drivers/block/paride/pd.c
@@ -775,14 +775,14 @@ static int pd_special_command(struct pd_unit *disk,
 	struct request *rq;
 	struct pd_req *req;
 
-	rq = blk_get_request(disk->gd->queue, REQ_OP_DRV_IN, 0);
+	rq = blk_mq_alloc_request(disk->gd->queue, REQ_OP_DRV_IN, 0);
 	if (IS_ERR(rq))
 		return PTR_ERR(rq);
 	req = blk_mq_rq_to_pdu(rq);
 
 	req->func = func;
 	blk_execute_rq(disk->gd, rq, 0);
-	blk_put_request(rq);
+	blk_mq_free_request(rq);
 	return 0;
 }
 
diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index cacf64eedad8..40e7a45e3347 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -726,7 +726,7 @@ static int pkt_generic_packet(struct pktcdvd_device *pd, struct packet_command *
 	if (scsi_req(rq)->result)
 		ret = -EIO;
 out:
-	blk_put_request(rq);
+	blk_mq_free_request(rq);
 	return ret;
 }
 
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index 303caf2d17d0..f81a768943e1 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -312,7 +312,7 @@ static int virtblk_get_id(struct gendisk *disk, char *id_str)
 	struct request *req;
 	int err;
 
-	req = blk_get_request(q, REQ_OP_DRV_IN, 0);
+	req = blk_mq_alloc_request(q, REQ_OP_DRV_IN, 0);
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 
@@ -323,7 +323,7 @@ static int virtblk_get_id(struct gendisk *disk, char *id_str)
 	blk_execute_rq(vblk->disk, req, false);
 	err = blk_status_to_errno(virtblk_result(blk_mq_rq_to_pdu(req)));
 out:
-	blk_put_request(req);
+	blk_mq_free_request(req);
 	return err;
 }
 
diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index 694aaca4eea2..510f6c3ab98d 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -530,7 +530,7 @@ static int multipath_clone_and_map(struct dm_target *ti, struct request *rq,
 
 	bdev = pgpath->path.dev->bdev;
 	q = bdev_get_queue(bdev);
-	clone = blk_get_request(q, rq->cmd_flags | REQ_NOMERGE,
+	clone = blk_mq_alloc_request(q, rq->cmd_flags | REQ_NOMERGE,
 			BLK_MQ_REQ_NOWAIT);
 	if (IS_ERR(clone)) {
 		/* EBUSY, ENODEV or EWOULDBLOCK: requeue */
@@ -579,7 +579,7 @@ static void multipath_release_clone(struct request *clone,
 						    clone->io_start_time_ns);
 	}
 
-	blk_put_request(clone);
+	blk_mq_free_request(clone);
 }
 
 /*
diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c
index 431af5e8be2f..74882fa0f86d 100644
--- a/drivers/mmc/core/block.c
+++ b/drivers/mmc/core/block.c
@@ -258,7 +258,7 @@ static ssize_t power_ro_lock_store(struct device *dev,
 	mq = &md->queue;
 
 	/* Dispatch locking to the block layer */
-	req = blk_get_request(mq->queue, REQ_OP_DRV_OUT, 0);
+	req = blk_mq_alloc_request(mq->queue, REQ_OP_DRV_OUT, 0);
 	if (IS_ERR(req)) {
 		count = PTR_ERR(req);
 		goto out_put;
@@ -266,7 +266,7 @@ static ssize_t power_ro_lock_store(struct device *dev,
 	req_to_mmc_queue_req(req)->drv_op = MMC_DRV_OP_BOOT_WP;
 	blk_execute_rq(NULL, req, 0);
 	ret = req_to_mmc_queue_req(req)->drv_op_result;
-	blk_put_request(req);
+	blk_mq_free_request(req);
 
 	if (!ret) {
 		pr_info("%s: Locking boot partition ro until next power on\n",
@@ -646,7 +646,7 @@ static int mmc_blk_ioctl_cmd(struct mmc_blk_data *md,
 	 * Dispatch the ioctl() into the block request queue.
 	 */
 	mq = &md->queue;
-	req = blk_get_request(mq->queue,
+	req = blk_mq_alloc_request(mq->queue,
 		idata->ic.write_flag ? REQ_OP_DRV_OUT : REQ_OP_DRV_IN, 0);
 	if (IS_ERR(req)) {
 		err = PTR_ERR(req);
@@ -660,7 +660,7 @@ static int mmc_blk_ioctl_cmd(struct mmc_blk_data *md,
 	blk_execute_rq(NULL, req, 0);
 	ioc_err = req_to_mmc_queue_req(req)->drv_op_result;
 	err = mmc_blk_ioctl_copy_to_user(ic_ptr, idata);
-	blk_put_request(req);
+	blk_mq_free_request(req);
 
 cmd_done:
 	kfree(idata->buf);
@@ -716,7 +716,7 @@ static int mmc_blk_ioctl_multi_cmd(struct mmc_blk_data *md,
 	 * Dispatch the ioctl()s into the block request queue.
 	 */
 	mq = &md->queue;
-	req = blk_get_request(mq->queue,
+	req = blk_mq_alloc_request(mq->queue,
 		idata[0]->ic.write_flag ? REQ_OP_DRV_OUT : REQ_OP_DRV_IN, 0);
 	if (IS_ERR(req)) {
 		err = PTR_ERR(req);
@@ -733,7 +733,7 @@ static int mmc_blk_ioctl_multi_cmd(struct mmc_blk_data *md,
 	for (i = 0; i < num_of_cmds && !err; i++)
 		err = mmc_blk_ioctl_copy_to_user(&cmds[i], idata[i]);
 
-	blk_put_request(req);
+	blk_mq_free_request(req);
 
 cmd_err:
 	for (i = 0; i < num_of_cmds; i++) {
@@ -2730,7 +2730,7 @@ static int mmc_dbg_card_status_get(void *data, u64 *val)
 	int ret;
 
 	/* Ask the block layer about the card status */
-	req = blk_get_request(mq->queue, REQ_OP_DRV_IN, 0);
+	req = blk_mq_alloc_request(mq->queue, REQ_OP_DRV_IN, 0);
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 	req_to_mmc_queue_req(req)->drv_op = MMC_DRV_OP_GET_CARD_STATUS;
@@ -2740,7 +2740,7 @@ static int mmc_dbg_card_status_get(void *data, u64 *val)
 		*val = ret;
 		ret = 0;
 	}
-	blk_put_request(req);
+	blk_mq_free_request(req);
 
 	return ret;
 }
@@ -2766,7 +2766,7 @@ static int mmc_ext_csd_open(struct inode *inode, struct file *filp)
 		return -ENOMEM;
 
 	/* Ask the block layer for the EXT CSD */
-	req = blk_get_request(mq->queue, REQ_OP_DRV_IN, 0);
+	req = blk_mq_alloc_request(mq->queue, REQ_OP_DRV_IN, 0);
 	if (IS_ERR(req)) {
 		err = PTR_ERR(req);
 		goto out_free;
@@ -2775,7 +2775,7 @@ static int mmc_ext_csd_open(struct inode *inode, struct file *filp)
 	req_to_mmc_queue_req(req)->drv_op_data = &ext_csd;
 	blk_execute_rq(NULL, req, 0);
 	err = req_to_mmc_queue_req(req)->drv_op_result;
-	blk_put_request(req);
+	blk_mq_free_request(req);
 	if (err) {
 		pr_err("FAILED %d\n", err);
 		goto out_free;
diff --git a/drivers/scsi/scsi_bsg.c b/drivers/scsi/scsi_bsg.c
index 551727a6f694..081b84bb7985 100644
--- a/drivers/scsi/scsi_bsg.c
+++ b/drivers/scsi/scsi_bsg.c
@@ -95,7 +95,7 @@ static int scsi_bsg_sg_io_fn(struct request_queue *q, struct sg_io_v4 *hdr,
 out_free_cmd:
 	scsi_req_free_cmd(scsi_req(rq));
 out_put_request:
-	blk_put_request(rq);
+	blk_mq_free_request(rq);
 	return ret;
 }
 
diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index 71d027b94be4..36870b41c888 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -1979,7 +1979,7 @@ maybe_retry:
 
 static void eh_lock_door_done(struct request *req, blk_status_t status)
 {
-	blk_put_request(req);
+	blk_mq_free_request(req);
 }
 
 /**
diff --git a/drivers/scsi/scsi_ioctl.c b/drivers/scsi/scsi_ioctl.c
index 0078975e3c07..34412eac4566 100644
--- a/drivers/scsi/scsi_ioctl.c
+++ b/drivers/scsi/scsi_ioctl.c
@@ -490,7 +490,7 @@ static int sg_io(struct scsi_device *sdev, struct gendisk *disk,
 out_free_cdb:
 	scsi_req_free_cmd(req);
 out_put_request:
-	blk_put_request(rq);
+	blk_mq_free_request(rq);
 	return ret;
 }
 
@@ -634,7 +634,7 @@ static int sg_scsi_ioctl(struct request_queue *q, struct gendisk *disk,
 	}
 
 error:
-	blk_put_request(rq);
+	blk_mq_free_request(rq);
 
 error_free_buffer:
 	kfree(buffer);
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 9823b65d1536..9c2b99e12ce3 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -260,7 +260,7 @@ int __scsi_execute(struct scsi_device *sdev, const unsigned char *cmd,
 		scsi_normalize_sense(rq->sense, rq->sense_len, sshdr);
 	ret = rq->result;
  out:
-	blk_put_request(req);
+	blk_mq_free_request(req);
 
 	return ret;
 }
@@ -1100,7 +1100,7 @@ struct request *scsi_alloc_request(struct request_queue *q,
 {
 	struct request *rq;
 
-	rq = blk_get_request(q, op, flags);
+	rq = blk_mq_alloc_request(q, op, flags);
 	if (!IS_ERR(rq))
 		scsi_initialize_rq(rq);
 	return rq;
diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c
index 85f57ac0b844..141099ab9092 100644
--- a/drivers/scsi/sg.c
+++ b/drivers/scsi/sg.c
@@ -815,7 +815,7 @@ sg_common_write(Sg_fd * sfp, Sg_request * srp,
 	if (atomic_read(&sdp->detaching)) {
 		if (srp->bio) {
 			scsi_req_free_cmd(scsi_req(srp->rq));
-			blk_put_request(srp->rq);
+			blk_mq_free_request(srp->rq);
 			srp->rq = NULL;
 		}
 
@@ -1390,7 +1390,7 @@ sg_rq_end_io(struct request *rq, blk_status_t status)
 	 */
 	srp->rq = NULL;
 	scsi_req_free_cmd(scsi_req(rq));
-	blk_put_request(rq);
+	blk_mq_free_request(rq);
 
 	write_lock_irqsave(&sfp->rq_list_lock, iflags);
 	if (unlikely(srp->orphan)) {
@@ -1830,7 +1830,7 @@ sg_finish_rem_req(Sg_request *srp)
 
 	if (srp->rq) {
 		scsi_req_free_cmd(scsi_req(srp->rq));
-		blk_put_request(srp->rq);
+		blk_mq_free_request(srp->rq);
 	}
 
 	if (srp->res_used)
diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c
index 7c4d9a964799..3009b986d1d7 100644
--- a/drivers/scsi/sr.c
+++ b/drivers/scsi/sr.c
@@ -1003,7 +1003,7 @@ static int sr_read_cdda_bpc(struct cdrom_device_info *cdi, void __user *ubuf,
 	if (blk_rq_unmap_user(bio))
 		ret = -EFAULT;
 out_put_request:
-	blk_put_request(rq);
+	blk_mq_free_request(rq);
 	return ret;
 }
 
diff --git a/drivers/scsi/st.c b/drivers/scsi/st.c
index 1275299f6159..c2d5608f6b1a 100644
--- a/drivers/scsi/st.c
+++ b/drivers/scsi/st.c
@@ -530,7 +530,7 @@ static void st_scsi_execute_end(struct request *req, blk_status_t status)
 		complete(SRpnt->waiting);
 
 	blk_rq_unmap_user(tmp);
-	blk_put_request(req);
+	blk_mq_free_request(req);
 }
 
 static int st_scsi_execute(struct st_request *SRpnt, const unsigned char *cmd,
@@ -557,7 +557,7 @@ static int st_scsi_execute(struct st_request *SRpnt, const unsigned char *cmd,
 		err = blk_rq_map_user(req->q, req, mdata, NULL, bufflen,
 				      GFP_KERNEL);
 		if (err) {
-			blk_put_request(req);
+			blk_mq_free_request(req);
 			return err;
 		}
 	}
diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
index bf81da2ecf98..c85f540f6af4 100644
--- a/drivers/scsi/ufs/ufshcd.c
+++ b/drivers/scsi/ufs/ufshcd.c
@@ -2925,7 +2925,7 @@ static int ufshcd_exec_dev_cmd(struct ufs_hba *hba,
 	 * Even though we use wait_event() which sleeps indefinitely,
 	 * the maximum wait time is bounded by SCSI request timeout.
 	 */
-	req = blk_get_request(q, REQ_OP_DRV_OUT, 0);
+	req = blk_mq_alloc_request(q, REQ_OP_DRV_OUT, 0);
 	if (IS_ERR(req)) {
 		err = PTR_ERR(req);
 		goto out_unlock;
@@ -2952,7 +2952,7 @@ static int ufshcd_exec_dev_cmd(struct ufs_hba *hba,
 				    (struct utp_upiu_req *)lrbp->ucd_rsp_ptr);
 
 out:
-	blk_put_request(req);
+	blk_mq_free_request(req);
 out_unlock:
 	up_read(&hba->clk_scaling_lock);
 	return err;
@@ -6517,9 +6517,9 @@ static int __ufshcd_issue_tm_cmd(struct ufs_hba *hba,
 	int task_tag, err;
 
 	/*
-	 * blk_get_request() is used here only to get a free tag.
+	 * blk_mq_alloc_request() is used here only to get a free tag.
 	 */
-	req = blk_get_request(q, REQ_OP_DRV_OUT, 0);
+	req = blk_mq_alloc_request(q, REQ_OP_DRV_OUT, 0);
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 
@@ -6575,7 +6575,7 @@ static int __ufshcd_issue_tm_cmd(struct ufs_hba *hba,
 	spin_unlock_irqrestore(hba->host->host_lock, flags);
 
 	ufshcd_release(hba);
-	blk_put_request(req);
+	blk_mq_free_request(req);
 
 	return err;
 }
@@ -6660,7 +6660,7 @@ static int ufshcd_issue_devman_upiu_cmd(struct ufs_hba *hba,
 
 	down_read(&hba->clk_scaling_lock);
 
-	req = blk_get_request(q, REQ_OP_DRV_OUT, 0);
+	req = blk_mq_alloc_request(q, REQ_OP_DRV_OUT, 0);
 	if (IS_ERR(req)) {
 		err = PTR_ERR(req);
 		goto out_unlock;
@@ -6741,7 +6741,7 @@ static int ufshcd_issue_devman_upiu_cmd(struct ufs_hba *hba,
 				    (struct utp_upiu_req *)lrbp->ucd_rsp_ptr);
 
 out:
-	blk_put_request(req);
+	blk_mq_free_request(req);
 out_unlock:
 	up_read(&hba->clk_scaling_lock);
 	return err;
@@ -7912,7 +7912,7 @@ static void ufshcd_request_sense_done(struct request *rq, blk_status_t error)
 	if (error != BLK_STS_OK)
 		pr_err("%s: REQUEST SENSE failed (%d)\n", __func__, error);
 	kfree(rq->end_io_data);
-	blk_put_request(rq);
+	blk_mq_free_request(rq);
 }
 
 static int
@@ -7932,7 +7932,7 @@ ufshcd_request_sense_async(struct ufs_hba *hba, struct scsi_device *sdev)
 	if (!buffer)
 		return -ENOMEM;
 
-	req = blk_get_request(sdev->request_queue, REQ_OP_DRV_IN,
+	req = blk_mq_alloc_request(sdev->request_queue, REQ_OP_DRV_IN,
 			      /*flags=*/BLK_MQ_REQ_PM);
 	if (IS_ERR(req)) {
 		ret = PTR_ERR(req);
@@ -7957,7 +7957,7 @@ ufshcd_request_sense_async(struct ufs_hba *hba, struct scsi_device *sdev)
 	return 0;
 
 out_put:
-	blk_put_request(req);
+	blk_mq_free_request(req);
 out_free:
 	kfree(buffer);
 	return ret;
diff --git a/drivers/scsi/ufs/ufshpb.c b/drivers/scsi/ufs/ufshpb.c
index 589af5f6b940..a2a3080071e9 100644
--- a/drivers/scsi/ufs/ufshpb.c
+++ b/drivers/scsi/ufs/ufshpb.c
@@ -564,7 +564,7 @@ static int ufshpb_issue_pre_req(struct ufshpb_lu *hpb, struct scsi_cmnd *cmd,
 	int _read_id;
 	int ret = 0;
 
-	req = blk_get_request(cmd->device->request_queue,
+	req = blk_mq_alloc_request(cmd->device->request_queue,
 			      REQ_OP_DRV_OUT | REQ_SYNC, BLK_MQ_REQ_NOWAIT);
 	if (IS_ERR(req))
 		return -EAGAIN;
@@ -592,7 +592,7 @@ free_pre_req:
 	ufshpb_put_pre_req(hpb, pre_req);
 unlock_out:
 	spin_unlock_irqrestore(&hpb->rgn_state_lock, flags);
-	blk_put_request(req);
+	blk_mq_free_request(req);
 	return ret;
 }
 
@@ -721,7 +721,7 @@ static struct ufshpb_req *ufshpb_get_req(struct ufshpb_lu *hpb,
 		return NULL;
 
 retry:
-	req = blk_get_request(hpb->sdev_ufs_lu->request_queue, dir,
+	req = blk_mq_alloc_request(hpb->sdev_ufs_lu->request_queue, dir,
 			      BLK_MQ_REQ_NOWAIT);
 
 	if (!atomic && (PTR_ERR(req) == -EWOULDBLOCK) && (--retries > 0)) {
@@ -745,7 +745,7 @@ free_rq:
 
 static void ufshpb_put_req(struct ufshpb_lu *hpb, struct ufshpb_req *rq)
 {
-	blk_put_request(rq->req);
+	blk_mq_free_request(rq->req);
 	kmem_cache_free(hpb->map_req_cache, rq);
 }
 
diff --git a/drivers/target/target_core_pscsi.c b/drivers/target/target_core_pscsi.c
index b5705a2bd761..7fa57fb57bf2 100644
--- a/drivers/target/target_core_pscsi.c
+++ b/drivers/target/target_core_pscsi.c
@@ -1011,7 +1011,7 @@ pscsi_execute_cmd(struct se_cmd *cmd)
 	return 0;
 
 fail_put_request:
-	blk_put_request(req);
+	blk_mq_free_request(req);
 fail:
 	kfree(pt);
 	return ret;
@@ -1066,7 +1066,7 @@ static void pscsi_req_done(struct request *req, blk_status_t status)
 		break;
 	}
 
-	blk_put_request(req);
+	blk_mq_free_request(req);
 	kfree(pt);
 }
 
diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h
index ebc45cf0450b..8682663e7368 100644
--- a/include/linux/blk-mq.h
+++ b/include/linux/blk-mq.h
@@ -892,9 +892,6 @@ static inline bool rq_is_sync(struct request *rq)
 }
 
 void blk_rq_init(struct request_queue *q, struct request *rq);
-void blk_put_request(struct request *rq);
-struct request *blk_get_request(struct request_queue *q, unsigned int op,
-		blk_mq_req_flags_t flags);
 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);