[PATCH v20 00/13] Improve write performance for zoned UFS devices

[PATCH v20 00/13] Improve write performance for zoned UFS devices

[Bart Van Assche]

Hi Jens,

This patch series improves small write IOPS by a factor of two for zoned UFS
devices on my test setup. The changes included in this patch series are as
follows:
 - A new request queue limits flag is introduced that allows block drivers to
   declare whether or not they preserve the request order per hardware queue.
 - The order of zoned writes is preserved in the block layer by submitting all
   zoned writes from the same CPU core as long as any zoned writes are pending.
   A new member 'from_cpu' is introduced in the per-zone data structure
   'blk_zone_wplug' to track from which CPU to submit zoned writes. This data
   member is reset to -1 after all pending zoned writes for a zone have
   completed.
 - The retry count for zoned writes is increased in the SCSI core to deal with
   reordering caused by unit attention conditions or the SCSI error handler.
 - New functionality is added in the null_blk and scsi_debug drivers to
   make it easier to test the changes introduced by this patch series.

Please consider this patch series for the next merge window.

Thanks,

Bart.

Changes compared to v19:
 - Dropped patch 2/11 "block: Support allocating from a specific software queue"
 - Implemented Damien's proposal to always add pipelined bios to the plug list
   and to submit all pipelined bios from the bio work for a zone.
 - Added three refactoring patches to make this patch series easier to review.

Changes compared to v18:
 - Dropped patch 2/12 "block: Rework request allocation in blk_mq_submit_bio()".
 - Improved patch descriptions.

Changes compared to v17:
 - Rebased the patch series on top of kernel v6.16-rc1.
 - Dropped support for UFSHCI 3.0 controllers because the UFSHCI 3.0 auto-
   hibernation mechanism causes request reordering. UFSHCI 4.0 controllers
   remain supported.
 - Removed the error handling and write pointer tracking mechanisms again
   from block/blk-zoned.c.
 - Dropped the dm-linear patch from this patch series since I'm not aware of
   any use cases for write pipelining and dm-linear.

Changes compared to v16:
 - Rebased the entire patch series on top of Jens' for-next branch. Compared
   to when v16 of this series was posted, the BLK_ZONE_WPLUG_NEED_WP_UPDATE
   flag has been introduced and support for REQ_NOWAIT has been fixed.
 - The behavior for SMR disks is preserved: if .driver_preserves_write_order
   has not been set, BLK_ZONE_WPLUG_NEED_WP_UPDATE is still set if a write
   error has been encountered. If .driver_preserves_write_order has not been
   set, the write pointer is restored and the failed zoned writes are retried.
 - The superfluous "disk->zone_wplugs_hash_bits != 0" tests have been removed.

Changes compared to v15:
 - Reworked this patch series on top of the zone write plugging approach.
 - Moved support for requeuing requests from the SCSI core into the block
   layer core.
 - In the UFS driver, instead of disabling write pipelining if
   auto-hibernation is enabled, rely on the requeuing mechanism to handle
   reordering caused by resuming from auto-hibernation.

Changes compared to v14:
 - Removed the drivers/scsi/Kconfig.kunit and drivers/scsi/Makefile.kunit
   files. Instead, modified drivers/scsi/Kconfig and added #include "*_test.c"
   directives in the appropriate .c files. Removed the EXPORT_SYMBOL()
   directives that were added to make the unit tests link.
 - Fixed a double free in a unit test.

Changes compared to v13:
 - Reworked patch "block: Preserve the order of requeued zoned writes".
 - Addressed a performance concern by removing the eh_needs_prepare_resubmit
   SCSI driver callback and by introducing the SCSI host template flag
   .needs_prepare_resubmit instead.
 - Added a patch that adds a 'host' argument to scsi_eh_flush_done_q().
 - Made the code in unit tests less repetitive.

Changes compared to v12:
 - Added two new patches: "block: Preserve the order of requeued zoned writes"
   and "scsi: sd: Add a unit test for sd_cmp_sector()"
 - Restricted the number of zoned write retries. To my surprise I had to add
   "&& scmd->retries <= scmd->allowed" in the SCSI error handler to limit the
   number of retries.
 - In patch "scsi: ufs: Inform the block layer about write ordering", only set
   ELEVATOR_F_ZBD_SEQ_WRITE for zoned block devices.

Changes compared to v11:
 - Fixed a NULL pointer dereference that happened when booting from an ATA
   device by adding an scmd->device != NULL check in scsi_needs_preparation().
 - Updated Reviewed-by tags.

Changes compared to v10:
 - Dropped the UFS MediaTek and HiSilicon patches because these are not correct
   and because it is safe to drop these patches.
 - Updated Acked-by / Reviewed-by tags.

Changes compared to v9:
 - Introduced an additional scsi_driver callback: .eh_needs_prepare_resubmit().
 - Renamed the scsi_debug kernel module parameter 'no_zone_write_lock' into
   'preserves_write_order'.
 - Fixed an out-of-bounds access in the unit scsi_call_prepare_resubmit() unit
   test.
 - Wrapped ufshcd_auto_hibern8_update() calls in UFS host drivers with
   WARN_ON_ONCE() such that a kernel stack appears in case an error code is
   returned.
 - Elaborated a comment in the UFSHCI driver.

Changes compared to v8:
 - Fixed handling of 'driver_preserves_write_order' and 'use_zone_write_lock'
   in blk_stack_limits().
 - Added a comment in disk_set_zoned().
 - Modified blk_req_needs_zone_write_lock() such that it returns false if
   q->limits.use_zone_write_lock is false.
 - Modified disk_clear_zone_settings() such that it clears
   q->limits.use_zone_write_lock.
 - Left out one change from the mq-deadline patch that became superfluous due to
   the blk_req_needs_zone_write_lock() change.
 - Modified scsi_call_prepare_resubmit() such that it only calls list_sort() if
   zoned writes have to be resubmitted for which zone write locking is disabled.
 - Added an additional unit test for scsi_call_prepare_resubmit().
 - Modified the sorting code in the sd driver such that only those SCSI commands
   are sorted for which write locking is disabled.
 - Modified sd_zbc.c such that ELEVATOR_F_ZBD_SEQ_WRITE is only set if the
   write order is not preserved.
 - Included three patches for UFS host drivers that rework code that wrote
   directly to the auto-hibernation controller register.
 - Modified the UFS driver such that enabling auto-hibernation is not allowed
   if a zoned logical unit is present and if the controller operates in legacy
   mode.
 - Also in the UFS driver, simplified ufshcd_auto_hibern8_update().

Changes compared to v7:
 - Split the queue_limits member variable `use_zone_write_lock' into two member
   variables: `use_zone_write_lock' (set by disk_set_zoned()) and
   `driver_preserves_write_order' (set by the block driver or SCSI LLD). This
   should clear up the confusion about the purpose of this variable.
 - Moved the code for sorting SCSI commands by LBA from the SCSI error handler
   into the SCSI disk (sd) driver as requested by Christoph.
   
Changes compared to v6:
 - Removed QUEUE_FLAG_NO_ZONE_WRITE_LOCK and instead introduced a flag in
   the request queue limits data structure.

Changes compared to v5:
 - Renamed scsi_cmp_lba() into scsi_cmp_sector().
 - Improved several source code comments.

Changes compared to v4:
 - Dropped the patch that introduces the REQ_NO_ZONE_WRITE_LOCK flag.
 - Dropped the null_blk patch and added two scsi_debug patches instead.
 - Dropped the f2fs patch.
 - Split the patch for the UFS driver into two patches.
 - Modified several patch descriptions and source code comments.
 - Renamed dd_use_write_locking() into dd_use_zone_write_locking().
 - Moved the list_sort() call from scsi_unjam_host() into scsi_eh_flush_done_q()
   such that sorting happens just before reinserting.
 - Removed the scsi_cmd_retry_allowed() call from scsi_check_sense() to make
   sure that the retry counter is adjusted once per retry instead of twice.

Changes compared to v3:
 - Restored the patch that introduces QUEUE_FLAG_NO_ZONE_WRITE_LOCK. That patch
   had accidentally been left out from v2.
 - In patch "block: Introduce the flag REQ_NO_ZONE_WRITE_LOCK", improved the
   patch description and added the function blk_no_zone_write_lock().
 - In patch "block/mq-deadline: Only use zone locking if necessary", moved the
   blk_queue_is_zoned() call into dd_use_write_locking().
 - In patch "fs/f2fs: Disable zone write locking", set REQ_NO_ZONE_WRITE_LOCK
   from inside __bio_alloc() instead of in f2fs_submit_write_bio().

Changes compared to v2:
 - Renamed the request queue flag for disabling zone write locking.
 - Introduced a new request flag for disabling zone write locking.
 - Modified the mq-deadline scheduler such that zone write locking is only
   disabled if both flags are set.
 - Added an F2FS patch that sets the request flag for disabling zone write
   locking.
 - Only disable zone write locking in the UFS driver if auto-hibernation is
   disabled.

Changes compared to v1:
 - Left out the patches that are already upstream.
 - Switched the approach in patch "scsi: Retry unaligned zoned writes" from
   retrying immediately to sending unaligned write commands to the SCSI error
   handler.

Bart Van Assche (13):
  block: Support block drivers that preserve the order of write requests
  blk-mq: Restore the zone write order when requeuing
  blk-zoned: Add an argument to blk_zone_plug_bio()
  blk-zoned: Split an if-statement
  blk-zoned: Move code from disk_zone_wplug_add_bio() into its caller
  blk-zoned: Split blk_zone_wplug_bio_work()
  blk-zoned: Support pipelining of zoned writes
  null_blk: Add the preserves_write_order attribute
  scsi: core: Retry unaligned zoned writes
  scsi: sd: Increase retry count for zoned writes
  scsi: scsi_debug: Add the preserves_write_order module parameter
  scsi: scsi_debug: Support injecting unaligned write errors
  scsi: ufs: Inform the block layer about write ordering

 block/bfq-iosched.c               |   2 +
 block/blk-mq.c                    |  34 +++++++--
 block/blk-mq.h                    |   2 +
 block/blk-settings.c              |   2 +
 block/blk-zoned.c                 | 119 ++++++++++++++++++++----------
 block/kyber-iosched.c             |   2 +
 block/mq-deadline.c               |   7 +-
 drivers/block/null_blk/main.c     |   3 +
 drivers/block/null_blk/null_blk.h |   1 +
 drivers/md/dm.c                   |   5 +-
 drivers/scsi/scsi_debug.c         |  21 +++++-
 drivers/scsi/scsi_error.c         |  16 ++++
 drivers/scsi/sd.c                 |   7 ++
 drivers/ufs/core/ufshcd.c         |   6 ++
 include/linux/blk-mq.h            |  13 +++-
 include/linux/blkdev.h            |  10 ++-
 16 files changed, 196 insertions(+), 54 deletions(-)

[PATCH v20 01/13] block: Support block drivers that preserve the order of write requests

[Bart Van Assche]

Some storage controllers preserve the request order per hardware queue.
Some but not all device mapper drivers preserve the bio order. Introduce
the request queue limit member variable 'driver_preserves_write_order' to
allow block drivers and device mapper drivers to indicate that the order
of write commands is preserved per hardware queue and hence that
serialization of writes per zone is not required if all pending writes are
submitted to the same hardware queue.

Cc: Damien Le Moal <dlemoal@kernel.org>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Nitesh Shetty <nj.shetty@samsung.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
 block/blk-settings.c   | 2 ++
 include/linux/blkdev.h | 5 +++++
 2 files changed, 7 insertions(+)

diff --git a/block/blk-settings.c b/block/blk-settings.c
index a000daafbfb4..bceb9a9cb5ba 100644
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@@ -814,6 +814,8 @@ int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
 	}
 	t->max_secure_erase_sectors = min_not_zero(t->max_secure_erase_sectors,
 						   b->max_secure_erase_sectors);
+	t->driver_preserves_write_order = t->driver_preserves_write_order &&
+		b->driver_preserves_write_order;
 	t->zone_write_granularity = max(t->zone_write_granularity,
 					b->zone_write_granularity);
 	if (!(t->features & BLK_FEAT_ZONED)) {
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 5f14c20c8bc0..4dec1d91b7f2 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -413,6 +413,11 @@ struct queue_limits {
 
 	unsigned int		max_open_zones;
 	unsigned int		max_active_zones;
+	/*
+	 * Whether or not the block driver preserves the order of write
+	 * requests per hardware queue. Set by the block driver.
+	 */
+	bool			driver_preserves_write_order;
 
 	/*
 	 * Drivers that set dma_alignment to less than 511 must be prepared to

[PATCH v20 02/13] blk-mq: Restore the zone write order when requeuing

[Bart Van Assche]

Zoned writes may be requeued. This happens if a block driver returns
BLK_STS_RESOURCE, to handle SCSI unit attentions or by the SCSI error
handler after error handling has finished. A later patch enables write
pipelining and increases the number of pending writes per zone. If
multiple writes are pending per zone, write requests may be requeued in
another order than submitted. Restore the request order if requests are
requeued. Add RQF_DONTPREP to RQF_NOMERGE_FLAGS because this patch may
cause RQF_DONTPREP requests to be sent to the code that checks whether
a request can be merged and RQF_DONTPREP requests must not be merged.

Cc: Christoph Hellwig <hch@lst.de>
Cc: Damien Le Moal <dlemoal@kernel.org>
Cc: Yu Kuai <yukuai3@huawei.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
 block/bfq-iosched.c    |  2 ++
 block/blk-mq.c         | 20 +++++++++++++++++++-
 block/blk-mq.h         |  2 ++
 block/kyber-iosched.c  |  2 ++
 block/mq-deadline.c    |  7 ++++++-
 include/linux/blk-mq.h | 13 ++++++++++++-
 6 files changed, 43 insertions(+), 3 deletions(-)

diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index 0cb1e9873aab..1bd3afe5d779 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -6276,6 +6276,8 @@ static void bfq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq,
 
 	if (flags & BLK_MQ_INSERT_AT_HEAD) {
 		list_add(&rq->queuelist, &bfqd->dispatch);
+	} else if (flags & BLK_MQ_INSERT_ORDERED) {
+		blk_mq_insert_ordered(rq, &bfqd->dispatch);
 	} else if (!bfqq) {
 		list_add_tail(&rq->queuelist, &bfqd->dispatch);
 	} else {
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 0c61492724d2..aa81526ad531 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -1557,7 +1557,9 @@ static void blk_mq_requeue_work(struct work_struct *work)
 		 * already.  Insert it into the hctx dispatch list to avoid
 		 * block layer merges for the request.
 		 */
-		if (rq->rq_flags & RQF_DONTPREP)
+		if (blk_rq_is_seq_zoned_write(rq))
+			blk_mq_insert_request(rq, BLK_MQ_INSERT_ORDERED);
+		else if (rq->rq_flags & RQF_DONTPREP)
 			blk_mq_request_bypass_insert(rq, 0);
 		else
 			blk_mq_insert_request(rq, BLK_MQ_INSERT_AT_HEAD);
@@ -2590,6 +2592,20 @@ static void blk_mq_insert_requests(struct blk_mq_hw_ctx *hctx,
 	blk_mq_run_hw_queue(hctx, run_queue_async);
 }
 
+void blk_mq_insert_ordered(struct request *rq, struct list_head *list)
+{
+	struct request_queue *q = rq->q;
+	struct request *rq2;
+
+	list_for_each_entry(rq2, list, queuelist)
+		if (rq2->q == q && blk_rq_pos(rq2) > blk_rq_pos(rq))
+			break;
+
+	/* Insert rq before rq2. If rq2 is the list head, append at the end. */
+	list_add_tail(&rq->queuelist, &rq2->queuelist);
+}
+EXPORT_SYMBOL_GPL(blk_mq_insert_ordered);
+
 static void blk_mq_insert_request(struct request *rq, blk_insert_t flags)
 {
 	struct request_queue *q = rq->q;
@@ -2644,6 +2660,8 @@ static void blk_mq_insert_request(struct request *rq, blk_insert_t flags)
 		spin_lock(&ctx->lock);
 		if (flags & BLK_MQ_INSERT_AT_HEAD)
 			list_add(&rq->queuelist, &ctx->rq_lists[hctx->type]);
+		else if (flags & BLK_MQ_INSERT_ORDERED)
+			blk_mq_insert_ordered(rq, &ctx->rq_lists[hctx->type]);
 		else
 			list_add_tail(&rq->queuelist,
 				      &ctx->rq_lists[hctx->type]);
diff --git a/block/blk-mq.h b/block/blk-mq.h
index affb2e14b56e..393660311a56 100644
--- a/block/blk-mq.h
+++ b/block/blk-mq.h
@@ -40,8 +40,10 @@ enum {
 
 typedef unsigned int __bitwise blk_insert_t;
 #define BLK_MQ_INSERT_AT_HEAD		((__force blk_insert_t)0x01)
+#define BLK_MQ_INSERT_ORDERED		((__force blk_insert_t)0x02)
 
 void blk_mq_submit_bio(struct bio *bio);
+void blk_mq_insert_ordered(struct request *rq, struct list_head *list);
 int blk_mq_poll(struct request_queue *q, blk_qc_t cookie, struct io_comp_batch *iob,
 		unsigned int flags);
 void blk_mq_exit_queue(struct request_queue *q);
diff --git a/block/kyber-iosched.c b/block/kyber-iosched.c
index 4dba8405bd01..051c05ceafd7 100644
--- a/block/kyber-iosched.c
+++ b/block/kyber-iosched.c
@@ -603,6 +603,8 @@ static void kyber_insert_requests(struct blk_mq_hw_ctx *hctx,
 		trace_block_rq_insert(rq);
 		if (flags & BLK_MQ_INSERT_AT_HEAD)
 			list_move(&rq->queuelist, head);
+		else if (flags & BLK_MQ_INSERT_ORDERED)
+			blk_mq_insert_ordered(rq, head);
 		else
 			list_move_tail(&rq->queuelist, head);
 		sbitmap_set_bit(&khd->kcq_map[sched_domain],
diff --git a/block/mq-deadline.c b/block/mq-deadline.c
index 2edf1cac06d5..110fef65b829 100644
--- a/block/mq-deadline.c
+++ b/block/mq-deadline.c
@@ -710,7 +710,12 @@ static void dd_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq,
 		 * set expire time and add to fifo list
 		 */
 		rq->fifo_time = jiffies + dd->fifo_expire[data_dir];
-		list_add_tail(&rq->queuelist, &per_prio->fifo_list[data_dir]);
+		if (flags & BLK_MQ_INSERT_ORDERED)
+			blk_mq_insert_ordered(rq,
+					      &per_prio->fifo_list[data_dir]);
+		else
+			list_add_tail(&rq->queuelist,
+				      &per_prio->fifo_list[data_dir]);
 	}
 }
 
diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h
index 2a5a828f19a0..1c516151fff0 100644
--- a/include/linux/blk-mq.h
+++ b/include/linux/blk-mq.h
@@ -86,7 +86,7 @@ enum rqf_flags {
 
 /* flags that prevent us from merging requests: */
 #define RQF_NOMERGE_FLAGS \
-	(RQF_STARTED | RQF_FLUSH_SEQ | RQF_SPECIAL_PAYLOAD)
+	(RQF_STARTED | RQF_FLUSH_SEQ | RQF_DONTPREP | RQF_SPECIAL_PAYLOAD)
 
 enum mq_rq_state {
 	MQ_RQ_IDLE		= 0,
@@ -1191,4 +1191,15 @@ static inline int blk_rq_map_sg(struct request *rq, struct scatterlist *sglist)
 }
 void blk_dump_rq_flags(struct request *, char *);
 
+static inline bool blk_rq_is_seq_zoned_write(struct request *rq)
+{
+	switch (req_op(rq)) {
+	case REQ_OP_WRITE:
+	case REQ_OP_WRITE_ZEROES:
+		return bdev_zone_is_seq(rq->q->disk->part0, blk_rq_pos(rq));
+	default:
+		return false;
+	}
+}
+
 #endif /* BLK_MQ_H */

[PATCH v20 03/13] blk-zoned: Add an argument to blk_zone_plug_bio()

[Bart Van Assche]

Software that submits zoned writes, e.g. a filesystem, may submit zoned
writes from multiple CPUs as long as the zoned writes are serialized per
zone. Submitting bios from different CPUs may cause bio reordering if
e.g. different bios reach the storage device through different queues.
Prepare for preserving the order of pipelined zoned writes per zone by
adding the 'rq_cpu` argument to blk_zone_plug_bio(). This argument tells
blk_zone_plug_bio() from which CPU a cached request has been allocated.
The cached request will only be used if it matches the CPU from which
zoned writes are being submitted for the zone associated with the bio.

Cc: Christoph Hellwig <hch@lst.de>
Cc: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
 block/blk-mq.c         | 10 ++++++----
 block/blk-zoned.c      |  5 ++++-
 drivers/md/dm.c        |  5 ++---
 include/linux/blkdev.h |  5 +++--
 4 files changed, 15 insertions(+), 10 deletions(-)

diff --git a/block/blk-mq.c b/block/blk-mq.c
index aa81526ad531..c02cf4140717 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -3133,11 +3133,14 @@ void blk_mq_submit_bio(struct bio *bio)
 	unsigned int nr_segs;
 	struct request *rq;
 	blk_status_t ret;
+	int rq_cpu = -1;
 
 	/*
 	 * If the plug has a cached request for this queue, try to use it.
 	 */
 	rq = blk_mq_peek_cached_request(plug, q, bio->bi_opf);
+	if (rq)
+		rq_cpu = rq->mq_ctx->cpu;
 
 	/*
 	 * A BIO that was released from a zone write plug has already been
@@ -3187,10 +3190,9 @@ void blk_mq_submit_bio(struct bio *bio)
 	if (blk_mq_attempt_bio_merge(q, bio, nr_segs))
 		goto queue_exit;
 
-	if (bio_needs_zone_write_plugging(bio)) {
-		if (blk_zone_plug_bio(bio, nr_segs))
-			goto queue_exit;
-	}
+	if (bio_needs_zone_write_plugging(bio) &&
+	    blk_zone_plug_bio(bio, nr_segs, rq_cpu))
+		goto queue_exit;
 
 new_request:
 	if (rq) {
diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index efe71b1a1da1..3f285e372460 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -1104,6 +1104,9 @@ static void blk_zone_wplug_handle_native_zone_append(struct bio *bio)
  * blk_zone_plug_bio - Handle a zone write BIO with zone write plugging
  * @bio: The BIO being submitted
  * @nr_segs: The number of physical segments of @bio
+ * @rq_cpu: software queue onto which a request will be queued. -1 if the caller
+ *	has not yet decided onto which software queue to queue the request or if
+ *	the bio won't be converted into a request.
  *
  * Handle write, write zeroes and zone append operations requiring emulation
  * using zone write plugging.
@@ -1112,7 +1115,7 @@ static void blk_zone_wplug_handle_native_zone_append(struct bio *bio)
  * write plug. Otherwise, return false to let the submission path process
  * @bio normally.
  */
-bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs)
+bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs, int rq_cpu)
 {
 	struct block_device *bdev = bio->bi_bdev;
 
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index ca889328fdfe..5033af6d687c 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1814,9 +1814,8 @@ static inline bool dm_zone_bio_needs_split(struct mapped_device *md,
 
 static inline bool dm_zone_plug_bio(struct mapped_device *md, struct bio *bio)
 {
-	if (!bio_needs_zone_write_plugging(bio))
-		return false;
-	return blk_zone_plug_bio(bio, 0);
+	return bio_needs_zone_write_plugging(bio) &&
+		blk_zone_plug_bio(bio, 0, -1);
 }
 
 static blk_status_t __send_zone_reset_all_emulated(struct clone_info *ci,
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 4dec1d91b7f2..3cdb4f7c602d 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -896,7 +896,7 @@ static inline bool bio_needs_zone_write_plugging(struct bio *bio)
 	}
 }
 
-bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs);
+bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs, int rq_cpu);
 
 /**
  * disk_zone_capacity - returns the zone capacity of zone containing @sector
@@ -931,7 +931,8 @@ static inline bool bio_needs_zone_write_plugging(struct bio *bio)
 	return false;
 }
 
-static inline bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs)
+static inline bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs,
+				     int rq_cpu)
 {
 	return false;
 }

[PATCH v20 04/13] blk-zoned: Split an if-statement

[Bart Van Assche]

Prepare for moving code from disk_zone_wplug_add_bio() into its caller.
Split an if-statement and also the comment above that if-statement. No
functionality has been changed.

Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
 block/blk-zoned.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index 3f285e372460..089f106fd82e 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -1029,14 +1029,14 @@ static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs)
 	/* Indicate that this BIO is being handled using zone write plugging. */
 	bio_set_flag(bio, BIO_ZONE_WRITE_PLUGGING);
 
+	/* If the zone is already plugged, add the BIO to the plug BIO list. */
+	if (zwplug->flags & BLK_ZONE_WPLUG_PLUGGED)
+		goto plug;
 	/*
-	 * If the zone is already plugged, add the BIO to the plug BIO list.
 	 * Do the same for REQ_NOWAIT BIOs to ensure that we will not see a
 	 * BLK_STS_AGAIN failure if we let the BIO execute.
-	 * Otherwise, plug and let the BIO execute.
 	 */
-	if ((zwplug->flags & BLK_ZONE_WPLUG_PLUGGED) ||
-	    (bio->bi_opf & REQ_NOWAIT))
+	if (bio->bi_opf & REQ_NOWAIT)
 		goto plug;
 
 	if (!blk_zone_wplug_prepare_bio(zwplug, bio)) {
@@ -1045,6 +1045,7 @@ static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs)
 		return true;
 	}
 
+	/* Otherwise, plug and submit the BIO. */
 	zwplug->flags |= BLK_ZONE_WPLUG_PLUGGED;
 
 	spin_unlock_irqrestore(&zwplug->lock, flags);

[PATCH v20 05/13] blk-zoned: Move code from disk_zone_wplug_add_bio() into its caller

[Bart Van Assche]

Move the following code into the only caller of disk_zone_wplug_add_bio():
 - bio->bi_opf &= ~REQ_NOWAIT
 - wplug->flags |= BLK_ZONE_WPLUG_PLUGGED
 - The disk_zone_wplug_schedule_bio_work() call.

No functionality has been changed.

This patch prepares for zoned write pipelining by removing the code from
disk_zone_wplug_add_bio() that does not apply to all zoned write pipelining
bio processing cases.

Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
 block/blk-zoned.c | 34 ++++++++++++++--------------------
 1 file changed, 14 insertions(+), 20 deletions(-)

diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index 089f106fd82e..2a85e3b7b081 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -778,8 +778,6 @@ static inline void disk_zone_wplug_add_bio(struct gendisk *disk,
 				struct blk_zone_wplug *zwplug,
 				struct bio *bio, unsigned int nr_segs)
 {
-	bool schedule_bio_work = false;
-
 	/*
 	 * Grab an extra reference on the BIO request queue usage counter.
 	 * This reference will be reused to submit a request for the BIO for
@@ -795,16 +793,6 @@ static inline void disk_zone_wplug_add_bio(struct gendisk *disk,
 	 */
 	bio_clear_polled(bio);
 
-	/*
-	 * REQ_NOWAIT BIOs are always handled using the zone write plug BIO
-	 * work, which can block. So clear the REQ_NOWAIT flag and schedule the
-	 * work if this is the first BIO we are plugging.
-	 */
-	if (bio->bi_opf & REQ_NOWAIT) {
-		schedule_bio_work = !(zwplug->flags & BLK_ZONE_WPLUG_PLUGGED);
-		bio->bi_opf &= ~REQ_NOWAIT;
-	}
-
 	/*
 	 * Reuse the poll cookie field to store the number of segments when
 	 * split to the hardware limits.
@@ -818,11 +806,6 @@ static inline void disk_zone_wplug_add_bio(struct gendisk *disk,
 	 * at the tail of the list to preserve the sequential write order.
 	 */
 	bio_list_add(&zwplug->bio_list, bio);
-
-	zwplug->flags |= BLK_ZONE_WPLUG_PLUGGED;
-
-	if (schedule_bio_work)
-		disk_zone_wplug_schedule_bio_work(disk, zwplug);
 }
 
 /*
@@ -987,6 +970,7 @@ static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs)
 {
 	struct gendisk *disk = bio->bi_bdev->bd_disk;
 	sector_t sector = bio->bi_iter.bi_sector;
+	bool schedule_bio_work = false;
 	struct blk_zone_wplug *zwplug;
 	gfp_t gfp_mask = GFP_NOIO;
 	unsigned long flags;
@@ -1031,13 +1015,17 @@ static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs)
 
 	/* If the zone is already plugged, add the BIO to the plug BIO list. */
 	if (zwplug->flags & BLK_ZONE_WPLUG_PLUGGED)
-		goto plug;
+		goto queue_bio;
 	/*
 	 * Do the same for REQ_NOWAIT BIOs to ensure that we will not see a
 	 * BLK_STS_AGAIN failure if we let the BIO execute.
 	 */
-	if (bio->bi_opf & REQ_NOWAIT)
-		goto plug;
+	if (bio->bi_opf & REQ_NOWAIT) {
+		bio->bi_opf &= ~REQ_NOWAIT;
+		if (!(zwplug->flags & BLK_ZONE_WPLUG_PLUGGED))
+			goto plug;
+		goto queue_bio;
+	}
 
 	if (!blk_zone_wplug_prepare_bio(zwplug, bio)) {
 		spin_unlock_irqrestore(&zwplug->lock, flags);
@@ -1053,7 +1041,13 @@ static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs)
 	return false;
 
 plug:
+	zwplug->flags |= BLK_ZONE_WPLUG_PLUGGED;
+	schedule_bio_work = true;
+
+queue_bio:
 	disk_zone_wplug_add_bio(disk, zwplug, bio, nr_segs);
+	if (schedule_bio_work)
+		disk_zone_wplug_schedule_bio_work(disk, zwplug);
 
 	spin_unlock_irqrestore(&zwplug->lock, flags);
 

[PATCH v20 06/13] blk-zoned: Split blk_zone_wplug_bio_work()

[Bart Van Assche]

Prepare for submitting multiple bios from inside a single
blk_zone_wplug_bio_work() call. No functionality has been changed.

Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
 block/blk-zoned.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index 2a85e3b7b081..d2aa7671ddd6 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -1257,10 +1257,8 @@ void blk_zone_write_plug_finish_request(struct request *req)
 	disk_put_zone_wplug(zwplug);
 }
 
-static void blk_zone_wplug_bio_work(struct work_struct *work)
+static void blk_zone_submit_one_bio(struct blk_zone_wplug *zwplug)
 {
-	struct blk_zone_wplug *zwplug =
-		container_of(work, struct blk_zone_wplug, bio_work);
 	struct block_device *bdev;
 	unsigned long flags;
 	struct bio *bio;
@@ -1276,7 +1274,7 @@ static void blk_zone_wplug_bio_work(struct work_struct *work)
 	if (!bio) {
 		zwplug->flags &= ~BLK_ZONE_WPLUG_PLUGGED;
 		spin_unlock_irqrestore(&zwplug->lock, flags);
-		goto put_zwplug;
+		return;
 	}
 
 	if (!blk_zone_wplug_prepare_bio(zwplug, bio)) {
@@ -1300,8 +1298,15 @@ static void blk_zone_wplug_bio_work(struct work_struct *work)
 	} else {
 		blk_mq_submit_bio(bio);
 	}
+}
+
+static void blk_zone_wplug_bio_work(struct work_struct *work)
+{
+	struct blk_zone_wplug *zwplug =
+		container_of(work, struct blk_zone_wplug, bio_work);
+
+	blk_zone_submit_one_bio(zwplug);
 
-put_zwplug:
 	/* Drop the reference we took in disk_zone_wplug_schedule_bio_work(). */
 	disk_put_zone_wplug(zwplug);
 }

[PATCH v20 07/13] blk-zoned: Support pipelining of zoned writes

[Bart Van Assche]

Support pipelining of zoned writes if the block driver preserves the write
order per hardware queue. Track per zone to which software queue writes
have been queued. If zoned writes are pipelined, submit new writes to the
same software queue as the writes that are already in progress. This
prevents reordering by submitting requests for the same zone to different
software or hardware queues.

Cc: Christoph Hellwig <hch@lst.de>
Cc: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
 block/blk-mq.c    |  4 +--
 block/blk-zoned.c | 68 ++++++++++++++++++++++++++++++++++++-----------
 2 files changed, 55 insertions(+), 17 deletions(-)

diff --git a/block/blk-mq.c b/block/blk-mq.c
index c02cf4140717..1b45c98c7e0c 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -3145,8 +3145,8 @@ void blk_mq_submit_bio(struct bio *bio)
 	/*
 	 * A BIO that was released from a zone write plug has already been
 	 * through the preparation in this function, already holds a reference
-	 * on the queue usage counter, and is the only write BIO in-flight for
-	 * the target zone. Go straight to preparing a request for it.
+	 * on the queue usage counter. Go straight to preparing a request for
+	 * it.
 	 */
 	if (bio_zone_write_plugging(bio)) {
 		nr_segs = bio->__bi_nr_segments;
diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index d2aa7671ddd6..7c99e64647d0 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -51,6 +51,8 @@ static const char *const zone_cond_name[] = {
  * @zone_no: The number of the zone the plug is managing.
  * @wp_offset: The zone write pointer location relative to the start of the zone
  *             as a number of 512B sectors.
+ * @from_cpu: Software queue to submit writes from for drivers that preserve
+ *	the write order.
  * @bio_list: The list of BIOs that are currently plugged.
  * @bio_work: Work struct to handle issuing of plugged BIOs
  * @rcu_head: RCU head to free zone write plugs with an RCU grace period.
@@ -63,6 +65,7 @@ struct blk_zone_wplug {
 	unsigned int		flags;
 	unsigned int		zone_no;
 	unsigned int		wp_offset;
+	int			from_cpu;
 	struct bio_list		bio_list;
 	struct work_struct	bio_work;
 	struct rcu_head		rcu_head;
@@ -72,8 +75,7 @@ struct blk_zone_wplug {
 /*
  * Zone write plug flags bits:
  *  - BLK_ZONE_WPLUG_PLUGGED: Indicates that the zone write plug is plugged,
- *    that is, that write BIOs are being throttled due to a write BIO already
- *    being executed or the zone write plug bio list is not empty.
+ *    that is, that write BIOs are being throttled.
  *  - BLK_ZONE_WPLUG_NEED_WP_UPDATE: Indicates that we lost track of a zone
  *    write pointer offset and need to update it.
  *  - BLK_ZONE_WPLUG_UNHASHED: Indicates that the zone write plug was removed
@@ -568,6 +570,7 @@ static struct blk_zone_wplug *disk_get_and_lock_zone_wplug(struct gendisk *disk,
 	zwplug->flags = 0;
 	zwplug->zone_no = zno;
 	zwplug->wp_offset = bdev_offset_from_zone_start(disk->part0, sector);
+	zwplug->from_cpu = -1;
 	bio_list_init(&zwplug->bio_list);
 	INIT_WORK(&zwplug->bio_work, blk_zone_wplug_bio_work);
 	zwplug->disk = disk;
@@ -764,14 +767,18 @@ static bool blk_zone_wplug_handle_reset_all(struct bio *bio)
 static void disk_zone_wplug_schedule_bio_work(struct gendisk *disk,
 					      struct blk_zone_wplug *zwplug)
 {
+	int cpu;
+
+	lockdep_assert_held(&zwplug->lock);
+
 	/*
 	 * Take a reference on the zone write plug and schedule the submission
 	 * of the next plugged BIO. blk_zone_wplug_bio_work() will release the
 	 * reference we take here.
 	 */
-	WARN_ON_ONCE(!(zwplug->flags & BLK_ZONE_WPLUG_PLUGGED));
 	refcount_inc(&zwplug->ref);
-	queue_work(disk->zone_wplugs_wq, &zwplug->bio_work);
+	cpu = zwplug->from_cpu >= 0 ? zwplug->from_cpu : WORK_CPU_UNBOUND;
+	queue_work_on(cpu, disk->zone_wplugs_wq, &zwplug->bio_work);
 }
 
 static inline void disk_zone_wplug_add_bio(struct gendisk *disk,
@@ -932,7 +939,8 @@ static bool blk_zone_wplug_prepare_bio(struct blk_zone_wplug *zwplug,
 	 * We know such BIO will fail, and that would potentially overflow our
 	 * write pointer offset beyond the end of the zone.
 	 */
-	if (disk_zone_wplug_is_full(disk, zwplug))
+	if (!disk->queue->limits.driver_preserves_write_order
+	    && disk_zone_wplug_is_full(disk, zwplug))
 		return false;
 
 	if (bio_op(bio) == REQ_OP_ZONE_APPEND) {
@@ -956,7 +964,8 @@ static bool blk_zone_wplug_prepare_bio(struct blk_zone_wplug *zwplug,
 		 * with a start sector not unaligned to the zone write pointer
 		 * will fail.
 		 */
-		if (bio_offset_from_zone_start(bio) != zwplug->wp_offset)
+		if (!disk->queue->limits.driver_preserves_write_order
+		    && bio_offset_from_zone_start(bio) != zwplug->wp_offset)
 			return false;
 	}
 
@@ -966,9 +975,11 @@ static bool blk_zone_wplug_prepare_bio(struct blk_zone_wplug *zwplug,
 	return true;
 }
 
-static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs)
+static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs,
+					int from_cpu)
 {
 	struct gendisk *disk = bio->bi_bdev->bd_disk;
+	const bool dpwo = disk->queue->limits.driver_preserves_write_order;
 	sector_t sector = bio->bi_iter.bi_sector;
 	bool schedule_bio_work = false;
 	struct blk_zone_wplug *zwplug;
@@ -1033,8 +1044,23 @@ static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs)
 		return true;
 	}
 
-	/* Otherwise, plug and submit the BIO. */
-	zwplug->flags |= BLK_ZONE_WPLUG_PLUGGED;
+	if (dpwo && zwplug->from_cpu < 0) {
+		/* No zoned writes are in progress. Select the current CPU. */
+		zwplug->from_cpu = raw_smp_processor_id();
+		goto plug;
+	} else if (dpwo) {
+		/*
+		 * The block driver preserves the write order. Submit the bio
+		 * from zwplug->from_cpu.
+		 */
+		goto plug;
+	} else {
+		/*
+		 * The block driver does not preserve the write order. Plug and
+		 * submit the BIO.
+		 */
+		zwplug->flags |= BLK_ZONE_WPLUG_PLUGGED;
+	}
 
 	spin_unlock_irqrestore(&zwplug->lock, flags);
 
@@ -1143,7 +1169,7 @@ bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs, int rq_cpu)
 		fallthrough;
 	case REQ_OP_WRITE:
 	case REQ_OP_WRITE_ZEROES:
-		return blk_zone_wplug_handle_write(bio, nr_segs);
+		return blk_zone_wplug_handle_write(bio, nr_segs, rq_cpu);
 	case REQ_OP_ZONE_RESET:
 		return blk_zone_wplug_handle_reset_or_finish(bio, 0);
 	case REQ_OP_ZONE_FINISH:
@@ -1175,6 +1201,9 @@ static void disk_zone_wplug_unplug_bio(struct gendisk *disk,
 
 	zwplug->flags &= ~BLK_ZONE_WPLUG_PLUGGED;
 
+	if (refcount_read(&zwplug->ref) == 2)
+		zwplug->from_cpu = -1;
+
 	/*
 	 * If the zone is full (it was fully written or finished, or empty
 	 * (it was reset), remove its zone write plug from the hash table.
@@ -1257,9 +1286,10 @@ void blk_zone_write_plug_finish_request(struct request *req)
 	disk_put_zone_wplug(zwplug);
 }
 
-static void blk_zone_submit_one_bio(struct blk_zone_wplug *zwplug)
+static bool blk_zone_submit_one_bio(struct blk_zone_wplug *zwplug)
 {
 	struct block_device *bdev;
+	struct gendisk *disk;
 	unsigned long flags;
 	struct bio *bio;
 
@@ -1274,7 +1304,7 @@ static void blk_zone_submit_one_bio(struct blk_zone_wplug *zwplug)
 	if (!bio) {
 		zwplug->flags &= ~BLK_ZONE_WPLUG_PLUGGED;
 		spin_unlock_irqrestore(&zwplug->lock, flags);
-		return;
+		return false;
 	}
 
 	if (!blk_zone_wplug_prepare_bio(zwplug, bio)) {
@@ -1285,6 +1315,7 @@ static void blk_zone_submit_one_bio(struct blk_zone_wplug *zwplug)
 	spin_unlock_irqrestore(&zwplug->lock, flags);
 
 	bdev = bio->bi_bdev;
+	disk = bdev->bd_disk;
 
 	/*
 	 * blk-mq devices will reuse the extra reference on the request queue
@@ -1298,6 +1329,9 @@ static void blk_zone_submit_one_bio(struct blk_zone_wplug *zwplug)
 	} else {
 		blk_mq_submit_bio(bio);
 	}
+
+	return disk->queue->limits.driver_preserves_write_order &&
+		!need_resched();
 }
 
 static void blk_zone_wplug_bio_work(struct work_struct *work)
@@ -1305,7 +1339,8 @@ static void blk_zone_wplug_bio_work(struct work_struct *work)
 	struct blk_zone_wplug *zwplug =
 		container_of(work, struct blk_zone_wplug, bio_work);
 
-	blk_zone_submit_one_bio(zwplug);
+	while (blk_zone_submit_one_bio(zwplug))
+		;
 
 	/* Drop the reference we took in disk_zone_wplug_schedule_bio_work(). */
 	disk_put_zone_wplug(zwplug);
@@ -1831,6 +1866,7 @@ static void queue_zone_wplug_show(struct blk_zone_wplug *zwplug,
 	unsigned int zwp_zone_no, zwp_ref;
 	unsigned int zwp_bio_list_size;
 	unsigned long flags;
+	int from_cpu;
 
 	spin_lock_irqsave(&zwplug->lock, flags);
 	zwp_zone_no = zwplug->zone_no;
@@ -1838,10 +1874,12 @@ static void queue_zone_wplug_show(struct blk_zone_wplug *zwplug,
 	zwp_ref = refcount_read(&zwplug->ref);
 	zwp_wp_offset = zwplug->wp_offset;
 	zwp_bio_list_size = bio_list_size(&zwplug->bio_list);
+	from_cpu = zwplug->from_cpu;
 	spin_unlock_irqrestore(&zwplug->lock, flags);
 
-	seq_printf(m, "%u 0x%x %u %u %u\n", zwp_zone_no, zwp_flags, zwp_ref,
-		   zwp_wp_offset, zwp_bio_list_size);
+	seq_printf(m, "zone_no %u flags 0x%x ref %u wp_offset %u bio_list_size %u from_cpu %d\n",
+		   zwp_zone_no, zwp_flags, zwp_ref, zwp_wp_offset,
+		   zwp_bio_list_size, from_cpu);
 }
 
 int queue_zone_wplugs_show(void *data, struct seq_file *m)

[PATCH v20 08/13] null_blk: Add the preserves_write_order attribute

[Bart Van Assche]

Support configuring the preserves_write_order in the null_blk driver to
make it easier to test support for this attribute.

Cc: Christoph Hellwig <hch@lst.de>
Cc: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
 drivers/block/null_blk/main.c     | 3 +++
 drivers/block/null_blk/null_blk.h | 1 +
 2 files changed, 4 insertions(+)

diff --git a/drivers/block/null_blk/main.c b/drivers/block/null_blk/main.c
index aa163ae9b2aa..45eaa6f0e983 100644
--- a/drivers/block/null_blk/main.c
+++ b/drivers/block/null_blk/main.c
@@ -475,6 +475,7 @@ NULLB_DEVICE_ATTR(fua, bool, NULL);
 NULLB_DEVICE_ATTR(rotational, bool, NULL);
 NULLB_DEVICE_ATTR(badblocks_once, bool, NULL);
 NULLB_DEVICE_ATTR(badblocks_partial_io, bool, NULL);
+NULLB_DEVICE_ATTR(preserves_write_order, bool, NULL);
 
 static ssize_t nullb_device_power_show(struct config_item *item, char *page)
 {
@@ -613,6 +614,7 @@ static struct configfs_attribute *nullb_device_attrs[] = {
 	&nullb_device_attr_no_sched,
 	&nullb_device_attr_poll_queues,
 	&nullb_device_attr_power,
+	&nullb_device_attr_preserves_write_order,
 	&nullb_device_attr_queue_mode,
 	&nullb_device_attr_rotational,
 	&nullb_device_attr_shared_tag_bitmap,
@@ -1979,6 +1981,7 @@ static int null_add_dev(struct nullb_device *dev)
 	if (dev->virt_boundary)
 		lim.virt_boundary_mask = PAGE_SIZE - 1;
 	null_config_discard(nullb, &lim);
+	lim.driver_preserves_write_order = dev->preserves_write_order;
 	if (dev->zoned) {
 		rv = null_init_zoned_dev(dev, &lim);
 		if (rv)
diff --git a/drivers/block/null_blk/null_blk.h b/drivers/block/null_blk/null_blk.h
index 7bb6128dbaaf..08b732cf853f 100644
--- a/drivers/block/null_blk/null_blk.h
+++ b/drivers/block/null_blk/null_blk.h
@@ -110,6 +110,7 @@ struct nullb_device {
 	bool shared_tag_bitmap; /* use hostwide shared tags */
 	bool fua; /* Support FUA */
 	bool rotational; /* Fake rotational device */
+	bool preserves_write_order;
 };
 
 struct nullb {

[PATCH v20 09/13] scsi: core: Retry unaligned zoned writes

[Bart Van Assche]

If zoned writes (REQ_OP_WRITE) for a sequential write required zone have
a starting LBA that differs from the write pointer, e.g. because a prior
write triggered a unit attention condition, then the storage device will
respond with an UNALIGNED WRITE COMMAND error. Retry commands that failed
with an unaligned write error.

Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
 drivers/scsi/scsi_error.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index 746ff6a1f309..d742fa67fe0e 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -713,6 +713,22 @@ enum scsi_disposition scsi_check_sense(struct scsi_cmnd *scmd)
 		fallthrough;
 
 	case ILLEGAL_REQUEST:
+		/*
+		 * Unaligned write command. This may indicate that zoned writes
+		 * have been received by the device in the wrong order. If write
+		 * pipelining is enabled, retry.
+		 */
+		if (sshdr.asc == 0x21 && sshdr.ascq == 0x04 &&
+		    req->q->limits.driver_preserves_write_order &&
+		    blk_rq_is_seq_zoned_write(req) &&
+		    scsi_cmd_retry_allowed(scmd)) {
+			SCSI_LOG_ERROR_RECOVERY(1,
+				sdev_printk(KERN_WARNING, scmd->device,
+				"Retrying unaligned write at LBA %#llx.\n",
+				scsi_get_lba(scmd)));
+			return NEEDS_RETRY;
+		}
+
 		if (sshdr.asc == 0x20 || /* Invalid command operation code */
 		    sshdr.asc == 0x21 || /* Logical block address out of range */
 		    sshdr.asc == 0x22 || /* Invalid function */

[PATCH v20 10/13] scsi: sd: Increase retry count for zoned writes

[Bart Van Assche]

If the write order is preserved, increase the number of retries for
write commands sent to a sequential zone to the maximum number of
outstanding commands because in the worst case the number of times
reordered zoned writes have to be retried is (number of outstanding
writes per sequential zone) - 1.

Cc: Damien Le Moal <dlemoal@kernel.org>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
 drivers/scsi/sd.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index eeaa6af294b8..9a499ca9dc3c 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -1404,6 +1404,13 @@ static blk_status_t sd_setup_read_write_cmnd(struct scsi_cmnd *cmd)
 	cmd->transfersize = sdp->sector_size;
 	cmd->underflow = nr_blocks << 9;
 	cmd->allowed = sdkp->max_retries;
+	/*
+	 * Increase the number of allowed retries for zoned writes if the driver
+	 * preserves the command order.
+	 */
+	if (rq->q->limits.driver_preserves_write_order &&
+	    blk_rq_is_seq_zoned_write(rq))
+		cmd->allowed += rq->q->nr_requests;
 	cmd->sdb.length = nr_blocks * sdp->sector_size;
 
 	SCSI_LOG_HLQUEUE(1,

[PATCH v20 11/13] scsi: scsi_debug: Add the preserves_write_order module parameter

[Bart Van Assche]

Zone write locking is not used for zoned devices if the block driver
reports that it preserves the order of write commands. Make it easier to
test not using zone write locking by adding support for setting the
driver_preserves_write_order flag.

Acked-by: Douglas Gilbert <dgilbert@interlog.com>
Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
 drivers/scsi/scsi_debug.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/scsi/scsi_debug.c b/drivers/scsi/scsi_debug.c
index aef33d1e346a..5575bc8a833d 100644
--- a/drivers/scsi/scsi_debug.c
+++ b/drivers/scsi/scsi_debug.c
@@ -1004,6 +1004,7 @@ static int dix_reads;
 static int dif_errors;
 
 /* ZBC global data */
+static bool sdeb_preserves_write_order;
 static bool sdeb_zbc_in_use;	/* true for host-aware and host-managed disks */
 static int sdeb_zbc_zone_cap_mb;
 static int sdeb_zbc_zone_size_mb;
@@ -6607,10 +6608,14 @@ static struct sdebug_dev_info *find_build_dev_info(struct scsi_device *sdev)
 
 static int scsi_debug_sdev_init(struct scsi_device *sdp)
 {
+	struct request_queue *q = sdp->request_queue;
+
 	if (sdebug_verbose)
 		pr_info("sdev_init <%u %u %u %llu>\n",
 		       sdp->host->host_no, sdp->channel, sdp->id, sdp->lun);
 
+	q->limits.driver_preserves_write_order = sdeb_preserves_write_order;
+
 	return 0;
 }
 
@@ -7339,6 +7344,8 @@ module_param_named(statistics, sdebug_statistics, bool, S_IRUGO | S_IWUSR);
 module_param_named(strict, sdebug_strict, bool, S_IRUGO | S_IWUSR);
 module_param_named(submit_queues, submit_queues, int, S_IRUGO);
 module_param_named(poll_queues, poll_queues, int, S_IRUGO);
+module_param_named(preserves_write_order, sdeb_preserves_write_order, bool,
+		   S_IRUGO);
 module_param_named(tur_ms_to_ready, sdeb_tur_ms_to_ready, int, S_IRUGO);
 module_param_named(unmap_alignment, sdebug_unmap_alignment, int, S_IRUGO);
 module_param_named(unmap_granularity, sdebug_unmap_granularity, int, S_IRUGO);
@@ -7411,6 +7418,8 @@ MODULE_PARM_DESC(opts, "1->noise, 2->medium_err, 4->timeout, 8->recovered_err...
 MODULE_PARM_DESC(per_host_store, "If set, next positive add_host will get new store (def=0)");
 MODULE_PARM_DESC(physblk_exp, "physical block exponent (def=0)");
 MODULE_PARM_DESC(poll_queues, "support for iouring iopoll queues (1 to max(submit_queues - 1))");
+MODULE_PARM_DESC(preserves_write_order,
+		 "Whether or not to inform the block layer that this driver preserves the order of WRITE commands (def=0)");
 MODULE_PARM_DESC(ptype, "SCSI peripheral type(def=0[disk])");
 MODULE_PARM_DESC(random, "If set, uniformly randomize command duration between 0 and delay_in_ns");
 MODULE_PARM_DESC(removable, "claim to have removable media (def=0)");

[PATCH v20 12/13] scsi: scsi_debug: Support injecting unaligned write errors

[Bart Van Assche]

Allow user space software, e.g. a blktests test, to inject unaligned
write errors.

Acked-by: Douglas Gilbert <dgilbert@interlog.com>
Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
 drivers/scsi/scsi_debug.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/scsi/scsi_debug.c b/drivers/scsi/scsi_debug.c
index 5575bc8a833d..963246f5894a 100644
--- a/drivers/scsi/scsi_debug.c
+++ b/drivers/scsi/scsi_debug.c
@@ -230,6 +230,7 @@ struct tape_block {
 #define SDEBUG_OPT_NO_CDB_NOISE		0x4000
 #define SDEBUG_OPT_HOST_BUSY		0x8000
 #define SDEBUG_OPT_CMD_ABORT		0x10000
+#define SDEBUG_OPT_UNALIGNED_WRITE	0x20000
 #define SDEBUG_OPT_ALL_NOISE (SDEBUG_OPT_NOISE | SDEBUG_OPT_Q_NOISE | \
 			      SDEBUG_OPT_RESET_NOISE)
 #define SDEBUG_OPT_ALL_INJECTING (SDEBUG_OPT_RECOVERED_ERR | \
@@ -237,7 +238,8 @@ struct tape_block {
 				  SDEBUG_OPT_DIF_ERR | SDEBUG_OPT_DIX_ERR | \
 				  SDEBUG_OPT_SHORT_TRANSFER | \
 				  SDEBUG_OPT_HOST_BUSY | \
-				  SDEBUG_OPT_CMD_ABORT)
+				  SDEBUG_OPT_CMD_ABORT | \
+				  SDEBUG_OPT_UNALIGNED_WRITE)
 #define SDEBUG_OPT_RECOV_DIF_DIX (SDEBUG_OPT_RECOVERED_ERR | \
 				  SDEBUG_OPT_DIF_ERR | SDEBUG_OPT_DIX_ERR)
 
@@ -4915,6 +4917,14 @@ static int resp_write_dt0(struct scsi_cmnd *scp, struct sdebug_dev_info *devip)
 	u8 *cmd = scp->cmnd;
 	bool meta_data_locked = false;
 
+	if (unlikely(sdebug_opts & SDEBUG_OPT_UNALIGNED_WRITE &&
+		     atomic_read(&sdeb_inject_pending))) {
+		atomic_set(&sdeb_inject_pending, 0);
+		mk_sense_buffer(scp, ILLEGAL_REQUEST, LBA_OUT_OF_RANGE,
+				UNALIGNED_WRITE_ASCQ);
+		return check_condition_result;
+	}
+
 	switch (cmd[0]) {
 	case WRITE_16:
 		ei_lba = 0;

[PATCH v20 13/13] scsi: ufs: Inform the block layer about write ordering

[Bart Van Assche]

From the UFSHCI 4.0 specification, about the MCQ mode:
"Command Submission
1. Host SW writes an Entry to SQ
2. Host SW updates SQ doorbell tail pointer

Command Processing
3. After fetching the Entry, Host Controller updates SQ doorbell head
   pointer
4. Host controller sends COMMAND UPIU to UFS device"

In other words, in MCQ mode, UFS controllers are required to forward
commands to the UFS device in the order these commands have been
received from the host.

This patch improves performance as follows on a test setup with UFSHCI
4.0 controller:
- When not using an I/O scheduler: 2.3x more IOPS for small writes.
- With the mq-deadline scheduler: 2.0x more IOPS for small writes.

Reviewed-by: Avri Altman <avri.altman@wdc.com>
Cc: Bao D. Nguyen <quic_nguyenb@quicinc.com>
Cc: Can Guo <quic_cang@quicinc.com>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
 drivers/ufs/core/ufshcd.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c
index 50adfb8b335b..1f2fc47ca7b4 100644
--- a/drivers/ufs/core/ufshcd.c
+++ b/drivers/ufs/core/ufshcd.c
@@ -5281,6 +5281,12 @@ static int ufshcd_sdev_configure(struct scsi_device *sdev,
 	struct ufs_hba *hba = shost_priv(sdev->host);
 	struct request_queue *q = sdev->request_queue;
 
+	/*
+	 * The write order is preserved per MCQ. Without MCQ, auto-hibernation
+	 * may cause write reordering that results in unaligned write errors.
+	 */
+	lim->driver_preserves_write_order = hba->mcq_enabled;
+
 	lim->dma_pad_mask = PRDT_DATA_BYTE_COUNT_PAD - 1;
 
 	/*

Re: [PATCH v20 01/13] block: Support block drivers that preserve the order of write requests

[Damien Le Moal]

On 7/9/25 7:06 AM, Bart Van Assche wrote:
> Some storage controllers preserve the request order per hardware queue.
> Some but not all device mapper drivers preserve the bio order. Introduce
> the request queue limit member variable 'driver_preserves_write_order' to
> allow block drivers and device mapper drivers to indicate that the order
> of write commands is preserved per hardware queue and hence that
> serialization of writes per zone is not required if all pending writes are
> submitted to the same hardware queue.
> 
> Cc: Damien Le Moal <dlemoal@kernel.org>
> Cc: Hannes Reinecke <hare@suse.de>
> Cc: Nitesh Shetty <nj.shetty@samsung.com>
> Cc: Christoph Hellwig <hch@lst.de>
> Cc: Ming Lei <ming.lei@redhat.com>
> Signed-off-by: Bart Van Assche <bvanassche@acm.org>
> ---
>  block/blk-settings.c   | 2 ++
>  include/linux/blkdev.h | 5 +++++
>  2 files changed, 7 insertions(+)
> 
> diff --git a/block/blk-settings.c b/block/blk-settings.c
> index a000daafbfb4..bceb9a9cb5ba 100644
> --- a/block/blk-settings.c
> +++ b/block/blk-settings.c
> @@ -814,6 +814,8 @@ int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
>  	}
>  	t->max_secure_erase_sectors = min_not_zero(t->max_secure_erase_sectors,
>  						   b->max_secure_erase_sectors);
> +	t->driver_preserves_write_order = t->driver_preserves_write_order &&
> +		b->driver_preserves_write_order;

See blk_stack_limits() and the section:

/*
 * Some feaures need to be supported both by the stacking driver and all
 * underlying devices.  The stacking driver sets these flags before
 * stacking the limits, and this will clear the flags if any of the
 * underlying devices does not support it.
 */
if (!(b->features & BLK_FEAT_NOWAIT))
	t->features &= ~BLK_FEAT_NOWAIT;
if (!(b->features & BLK_FEAT_POLL))
	t->features &= ~BLK_FEAT_POLL;

And make driver_preserves_write_order a feature instead of treating it
specially. Also, the name "driver_preserves_write_order" is not great. The
driver may be preserving write order, but the hardware not (e.g. libata and AHCI).

>  	t->zone_write_granularity = max(t->zone_write_granularity,
>  					b->zone_write_granularity);
>  	if (!(t->features & BLK_FEAT_ZONED)) {
> diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
> index 5f14c20c8bc0..4dec1d91b7f2 100644
> --- a/include/linux/blkdev.h
> +++ b/include/linux/blkdev.h
> @@ -413,6 +413,11 @@ struct queue_limits {
>  
>  	unsigned int		max_open_zones;
>  	unsigned int		max_active_zones;
> +	/*
> +	 * Whether or not the block driver preserves the order of write
> +	 * requests per hardware queue. Set by the block driver.
> +	 */
> +	bool			driver_preserves_write_order;
>  
>  	/*
>  	 * Drivers that set dma_alignment to less than 511 must be prepared to


-- 
Damien Le Moal
Western Digital Research

Re: [PATCH v20 02/13] blk-mq: Restore the zone write order when requeuing

[Damien Le Moal]

On 7/9/25 7:06 AM, Bart Van Assche wrote:
> Zoned writes may be requeued. This happens if a block driver returns
> BLK_STS_RESOURCE, to handle SCSI unit attentions or by the SCSI error
> handler after error handling has finished. A later patch enables write
> pipelining and increases the number of pending writes per zone. If
> multiple writes are pending per zone, write requests may be requeued in
> another order than submitted. Restore the request order if requests are
> requeued. Add RQF_DONTPREP to RQF_NOMERGE_FLAGS because this patch may
> cause RQF_DONTPREP requests to be sent to the code that checks whether
> a request can be merged and RQF_DONTPREP requests must not be merged.
> 
> Cc: Christoph Hellwig <hch@lst.de>
> Cc: Damien Le Moal <dlemoal@kernel.org>
> Cc: Yu Kuai <yukuai3@huawei.com>
> Signed-off-by: Bart Van Assche <bvanassche@acm.org>

[...]

> diff --git a/block/blk-mq.c b/block/blk-mq.c
> index 0c61492724d2..aa81526ad531 100644
> --- a/block/blk-mq.c
> +++ b/block/blk-mq.c
> @@ -1557,7 +1557,9 @@ static void blk_mq_requeue_work(struct work_struct *work)
>  		 * already.  Insert it into the hctx dispatch list to avoid
>  		 * block layer merges for the request.
>  		 */
> -		if (rq->rq_flags & RQF_DONTPREP)
> +		if (blk_rq_is_seq_zoned_write(rq))
> +			blk_mq_insert_request(rq, BLK_MQ_INSERT_ORDERED);

Every zone write will go through this, even for devices that do not have
driver_preserves_write_order == true. Given that thi insert ordered function is
heavier than inserting at head, it would be nice to avoid this.

> +		else if (rq->rq_flags & RQF_DONTPREP)
>  			blk_mq_request_bypass_insert(rq, 0);
>  		else
>  			blk_mq_insert_request(rq, BLK_MQ_INSERT_AT_HEAD);



-- 
Damien Le Moal
Western Digital Research

Re: [PATCH v20 03/13] blk-zoned: Add an argument to blk_zone_plug_bio()

[Damien Le Moal]

On 7/9/25 7:07 AM, Bart Van Assche wrote:
> Software that submits zoned writes, e.g. a filesystem, may submit zoned
> writes from multiple CPUs as long as the zoned writes are serialized per
> zone. Submitting bios from different CPUs may cause bio reordering if
> e.g. different bios reach the storage device through different queues.
> Prepare for preserving the order of pipelined zoned writes per zone by
> adding the 'rq_cpu` argument to blk_zone_plug_bio(). This argument tells
> blk_zone_plug_bio() from which CPU a cached request has been allocated.
> The cached request will only be used if it matches the CPU from which
> zoned writes are being submitted for the zone associated with the bio.

This needs to be squashed in patch 7. Otherwise, this cannot be reviewed as we
have no idea what that new argument to blk_zone_plug_bio() will be used for.


-- 
Damien Le Moal
Western Digital Research

Re: [PATCH v20 05/13] blk-zoned: Move code from disk_zone_wplug_add_bio() into its caller

[Damien Le Moal]

On 7/9/25 7:07 AM, Bart Van Assche wrote:
> Move the following code into the only caller of disk_zone_wplug_add_bio():
>  - bio->bi_opf &= ~REQ_NOWAIT
>  - wplug->flags |= BLK_ZONE_WPLUG_PLUGGED
>  - The disk_zone_wplug_schedule_bio_work() call.

Please make sentences instead of copy-pasting code. We can see that in the
patch itself.

> 
> No functionality has been changed.
> 
> This patch prepares for zoned write pipelining by removing the code from
> disk_zone_wplug_add_bio() that does not apply to all zoned write pipelining
> bio processing cases.
> 
> Signed-off-by: Bart Van Assche <bvanassche@acm.org>
> ---
>  block/blk-zoned.c | 34 ++++++++++++++--------------------
>  1 file changed, 14 insertions(+), 20 deletions(-)
> 
> diff --git a/block/blk-zoned.c b/block/blk-zoned.c
> index 089f106fd82e..2a85e3b7b081 100644
> --- a/block/blk-zoned.c
> +++ b/block/blk-zoned.c
> @@ -778,8 +778,6 @@ static inline void disk_zone_wplug_add_bio(struct gendisk *disk,
>  				struct blk_zone_wplug *zwplug,
>  				struct bio *bio, unsigned int nr_segs)
>  {
> -	bool schedule_bio_work = false;
> -
>  	/*
>  	 * Grab an extra reference on the BIO request queue usage counter.
>  	 * This reference will be reused to submit a request for the BIO for
> @@ -795,16 +793,6 @@ static inline void disk_zone_wplug_add_bio(struct gendisk *disk,
>  	 */
>  	bio_clear_polled(bio);
>  
> -	/*
> -	 * REQ_NOWAIT BIOs are always handled using the zone write plug BIO
> -	 * work, which can block. So clear the REQ_NOWAIT flag and schedule the
> -	 * work if this is the first BIO we are plugging.
> -	 */
> -	if (bio->bi_opf & REQ_NOWAIT) {
> -		schedule_bio_work = !(zwplug->flags & BLK_ZONE_WPLUG_PLUGGED);
> -		bio->bi_opf &= ~REQ_NOWAIT;
> -	}
> -
>  	/*
>  	 * Reuse the poll cookie field to store the number of segments when
>  	 * split to the hardware limits.
> @@ -818,11 +806,6 @@ static inline void disk_zone_wplug_add_bio(struct gendisk *disk,
>  	 * at the tail of the list to preserve the sequential write order.
>  	 */
>  	bio_list_add(&zwplug->bio_list, bio);
> -
> -	zwplug->flags |= BLK_ZONE_WPLUG_PLUGGED;
> -
> -	if (schedule_bio_work)
> -		disk_zone_wplug_schedule_bio_work(disk, zwplug);
>  }
>  
>  /*
> @@ -987,6 +970,7 @@ static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs)
>  {
>  	struct gendisk *disk = bio->bi_bdev->bd_disk;
>  	sector_t sector = bio->bi_iter.bi_sector;
> +	bool schedule_bio_work = false;
>  	struct blk_zone_wplug *zwplug;
>  	gfp_t gfp_mask = GFP_NOIO;
>  	unsigned long flags;
> @@ -1031,13 +1015,17 @@ static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs)
>  
>  	/* If the zone is already plugged, add the BIO to the plug BIO list. */
>  	if (zwplug->flags & BLK_ZONE_WPLUG_PLUGGED)
> -		goto plug;
> +		goto queue_bio;

"queue_bio" is not a good name. Please make that "add_bio" or "plug_bio" to
match the call to disk_zone_wplug_add_bio() done.

>  	/*
>  	 * Do the same for REQ_NOWAIT BIOs to ensure that we will not see a
>  	 * BLK_STS_AGAIN failure if we let the BIO execute.
>  	 */
> -	if (bio->bi_opf & REQ_NOWAIT)
> -		goto plug;
> +	if (bio->bi_opf & REQ_NOWAIT) {
> +		bio->bi_opf &= ~REQ_NOWAIT;
> +		if (!(zwplug->flags & BLK_ZONE_WPLUG_PLUGGED))
> +			goto plug;
> +		goto queue_bio;
> +	}
>  
>  	if (!blk_zone_wplug_prepare_bio(zwplug, bio)) {
>  		spin_unlock_irqrestore(&zwplug->lock, flags);
> @@ -1053,7 +1041,13 @@ static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs)
>  	return false;
>  
>  plug:
> +	zwplug->flags |= BLK_ZONE_WPLUG_PLUGGED;
> +	schedule_bio_work = true;
> +
> +queue_bio:
>  	disk_zone_wplug_add_bio(disk, zwplug, bio, nr_segs);
> +	if (schedule_bio_work)
> +		disk_zone_wplug_schedule_bio_work(disk, zwplug);
>  
>  	spin_unlock_irqrestore(&zwplug->lock, flags);
>  


-- 
Damien Le Moal
Western Digital Research

Re: [PATCH v20 06/13] blk-zoned: Split blk_zone_wplug_bio_work()

[Damien Le Moal]

On 7/9/25 7:07 AM, Bart Van Assche wrote:
> Prepare for submitting multiple bios from inside a single
> blk_zone_wplug_bio_work() call. No functionality has been changed.

The patch title says "Split blk_zone_wplug_bio_work()" but you do not explain
what you split into what.

Other than this, looks OK to me.

-- 
Damien Le Moal
Western Digital Research

Re: [PATCH v20 07/13] blk-zoned: Support pipelining of zoned writes

[Damien Le Moal]

On 7/9/25 7:07 AM, Bart Van Assche wrote:
> Support pipelining of zoned writes if the block driver preserves the write
> order per hardware queue. Track per zone to which software queue writes
> have been queued. If zoned writes are pipelined, submit new writes to the
> same software queue as the writes that are already in progress. This
> prevents reordering by submitting requests for the same zone to different
> software or hardware queues.
> 
> Cc: Christoph Hellwig <hch@lst.de>
> Cc: Damien Le Moal <dlemoal@kernel.org>
> Signed-off-by: Bart Van Assche <bvanassche@acm.org>

[...]

> @@ -764,14 +767,18 @@ static bool blk_zone_wplug_handle_reset_all(struct bio *bio)
>  static void disk_zone_wplug_schedule_bio_work(struct gendisk *disk,
>  					      struct blk_zone_wplug *zwplug)
>  {
> +	int cpu;
> +
> +	lockdep_assert_held(&zwplug->lock);
> +
>  	/*
>  	 * Take a reference on the zone write plug and schedule the submission
>  	 * of the next plugged BIO. blk_zone_wplug_bio_work() will release the
>  	 * reference we take here.
>  	 */
> -	WARN_ON_ONCE(!(zwplug->flags & BLK_ZONE_WPLUG_PLUGGED));
>  	refcount_inc(&zwplug->ref);
> -	queue_work(disk->zone_wplugs_wq, &zwplug->bio_work);
> +	cpu = zwplug->from_cpu >= 0 ? zwplug->from_cpu : WORK_CPU_UNBOUND;
> +	queue_work_on(cpu, disk->zone_wplugs_wq, &zwplug->bio_work);

Please change this to a more readable form with an expanded "if". The local
variable "cpu" should probably be named "work_cpu" for clarity too.

>  }
>  
>  static inline void disk_zone_wplug_add_bio(struct gendisk *disk,
> @@ -932,7 +939,8 @@ static bool blk_zone_wplug_prepare_bio(struct blk_zone_wplug *zwplug,
>  	 * We know such BIO will fail, and that would potentially overflow our
>  	 * write pointer offset beyond the end of the zone.
>  	 */
> -	if (disk_zone_wplug_is_full(disk, zwplug))
> +	if (!disk->queue->limits.driver_preserves_write_order
> +	    && disk_zone_wplug_is_full(disk, zwplug))

Writing to a zone that is full is an error, pipelining or not. So why do you
change this ? This does not make sense.

>  		return false;
>  
>  	if (bio_op(bio) == REQ_OP_ZONE_APPEND) {
> @@ -956,7 +964,8 @@ static bool blk_zone_wplug_prepare_bio(struct blk_zone_wplug *zwplug,
>  		 * with a start sector not unaligned to the zone write pointer
>  		 * will fail.
>  		 */
> -		if (bio_offset_from_zone_start(bio) != zwplug->wp_offset)
> +		if (!disk->queue->limits.driver_preserves_write_order
> +		    && bio_offset_from_zone_start(bio) != zwplug->wp_offset)

Same here. This does not depend pipelining: write should have been received in
order and be aligned with the wp. So why change this condition ?

>  			return false;
>  	}
>  

> @@ -1033,8 +1044,23 @@ static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs)
>  		return true;
>  	}
>  
> -	/* Otherwise, plug and submit the BIO. */
> -	zwplug->flags |= BLK_ZONE_WPLUG_PLUGGED;
> +	if (dpwo && zwplug->from_cpu < 0) {
> +		/* No zoned writes are in progress. Select the current CPU. */
> +		zwplug->from_cpu = raw_smp_processor_id();
> +		goto plug;
> +	} else if (dpwo) {
> +		/*
> +		 * The block driver preserves the write order. Submit the bio
> +		 * from zwplug->from_cpu.
> +		 */
> +		goto plug;

Can you change this to:

	if (dpwo) {
		/*
		 * The block driver preserves the write order: if we do not
		 * have any writes in progress already, use the current CPU
		 * to submit the BIO. Otherwise, keep using zwplug->from_cpu.
		 */
		if (zwplug->from_cpu < 0)
			zwplug->from_cpu = raw_smp_processor_id();
		goto plug;
	}

	/*
	 * The block driver does not preserve the write order. Plug and
	 * submit the BIO.
	 */
	zwplug->flags |= BLK_ZONE_WPLUG_PLUGGED;

That is a lot simpler and easier to read.

> +	} else {
> +		/*
> +		 * The block driver does not preserve the write order. Plug and
> +		 * submit the BIO.
> +		 */
> +		zwplug->flags |= BLK_ZONE_WPLUG_PLUGGED;
> +	}
>  
>  	spin_unlock_irqrestore(&zwplug->lock, flags);
>  

> @@ -1298,6 +1329,9 @@ static void blk_zone_submit_one_bio(struct blk_zone_wplug *zwplug)
>  	} else {
>  		blk_mq_submit_bio(bio);
>  	}
> +
> +	return disk->queue->limits.driver_preserves_write_order &&
> +		!need_resched();

I think we really need a helper for that
"disk->queue->limits.driver_preserves_write_order". But if you make this a
feature, it will be better.

Also, here, the test using need_resched() really need a comment explaining why
you look at that. I do not get it personally.

>  }
>  
>  static void blk_zone_wplug_bio_work(struct work_struct *work)
> @@ -1305,7 +1339,8 @@ static void blk_zone_wplug_bio_work(struct work_struct *work)
>  	struct blk_zone_wplug *zwplug =
>  		container_of(work, struct blk_zone_wplug, bio_work);
>  
> -	blk_zone_submit_one_bio(zwplug);
> +	while (blk_zone_submit_one_bio(zwplug))
> +		;

So patch 6 split created that blk_zone_submit_one_bio() function, but all you
do here is this change. I really do not see how that is better. Let's not do
that and simply have the loop expanded here to something like:

	do {
		...
	} while (disk->queue->limits.driver_preserves_write_order &&
		 !need_resched());

>  
>  	/* Drop the reference we took in disk_zone_wplug_schedule_bio_work(). */
>  	disk_put_zone_wplug(zwplug);
> @@ -1831,6 +1866,7 @@ static void queue_zone_wplug_show(struct blk_zone_wplug *zwplug,
>  	unsigned int zwp_zone_no, zwp_ref;
>  	unsigned int zwp_bio_list_size;
>  	unsigned long flags;
> +	int from_cpu;
>  
>  	spin_lock_irqsave(&zwplug->lock, flags);
>  	zwp_zone_no = zwplug->zone_no;
> @@ -1838,10 +1874,12 @@ static void queue_zone_wplug_show(struct blk_zone_wplug *zwplug,
>  	zwp_ref = refcount_read(&zwplug->ref);
>  	zwp_wp_offset = zwplug->wp_offset;
>  	zwp_bio_list_size = bio_list_size(&zwplug->bio_list);
> +	from_cpu = zwplug->from_cpu;
>  	spin_unlock_irqrestore(&zwplug->lock, flags);
>  
> -	seq_printf(m, "%u 0x%x %u %u %u\n", zwp_zone_no, zwp_flags, zwp_ref,
> -		   zwp_wp_offset, zwp_bio_list_size);
> +	seq_printf(m, "zone_no %u flags 0x%x ref %u wp_offset %u bio_list_size %u from_cpu %d\n",
> +		   zwp_zone_no, zwp_flags, zwp_ref, zwp_wp_offset,
> +		   zwp_bio_list_size, from_cpu);
>  }
>  
>  int queue_zone_wplugs_show(void *data, struct seq_file *m)


-- 
Damien Le Moal
Western Digital Research

Re: [PATCH v20 01/13] block: Support block drivers that preserve the order of write requests

[Bart Van Assche]

On 7/9/25 9:28 PM, Damien Le Moal wrote:
> See blk_stack_limits() and the section:
> 
> /*
>   * Some feaures need to be supported both by the stacking driver and all
>   * underlying devices.  The stacking driver sets these flags before
>   * stacking the limits, and this will clear the flags if any of the
>   * underlying devices does not support it.
>   */
> if (!(b->features & BLK_FEAT_NOWAIT))
> 	t->features &= ~BLK_FEAT_NOWAIT;
> if (!(b->features & BLK_FEAT_POLL))
> 	t->features &= ~BLK_FEAT_POLL;
> 
> And make driver_preserves_write_order a feature instead of treating it
> specially. Also, the name "driver_preserves_write_order" is not great. The
> driver may be preserving write order, but the hardware not (e.g. libata and AHCI).

How about adding this in include/linux/blkdev.h?

/*
  * The request order is preserved per hardware queue by the block 
driver and
  * the block device.
  */
#define BLK_FEAT_ORDERED_HWQ		((__force blk_features_t)(1u << 14))

Thanks,

Bart.

Re: [PATCH v20 05/13] blk-zoned: Move code from disk_zone_wplug_add_bio() into its caller

[Bart Van Assche]

On 7/9/25 9:46 PM, Damien Le Moal wrote:
> On 7/9/25 7:07 AM, Bart Van Assche wrote:
>> Move the following code into the only caller of disk_zone_wplug_add_bio():
>>   - bio->bi_opf &= ~REQ_NOWAIT
>>   - wplug->flags |= BLK_ZONE_WPLUG_PLUGGED
>>   - The disk_zone_wplug_schedule_bio_work() call.
> 
> Please make sentences instead of copy-pasting code. We can see that in the
> patch itself.

Is this good enough?

Move the following code into the only caller of disk_zone_wplug_add_bio():
  - The code for clearing the REQ_NOWAIT flag.
  - The code that sets the BLK_ZONE_WPLUG_PLUGGED flag.
  - The disk_zone_wplug_schedule_bio_work() call.

>>   	/* If the zone is already plugged, add the BIO to the plug BIO list. */
>>   	if (zwplug->flags & BLK_ZONE_WPLUG_PLUGGED)
>> -		goto plug;
>> +		goto queue_bio;
> 
> "queue_bio" is not a good name. Please make that "add_bio" or "plug_bio" to
> match the call to disk_zone_wplug_add_bio() done.

bio_list_add() adds BIOs at the end of zwplug->bio_list and
bio_list_pop() removes BIOs from the front of zwplug->bio_list. Hence,
zwplug->bio_list is used as a queue, isn't it? This is why I chose the
name "queue_bio". Anyway, if that label name isn't considered
good enough, how about changing it into "add_to_bio_list"?

Thanks,

Bart.

Re: [PATCH v20 07/13] blk-zoned: Support pipelining of zoned writes

[Bart Van Assche]

On 7/9/25 10:09 PM, Damien Le Moal wrote:
> On 7/9/25 7:07 AM, Bart Van Assche wrote:
>> @@ -932,7 +939,8 @@ static bool blk_zone_wplug_prepare_bio(struct blk_zone_wplug *zwplug,
>>   	 * We know such BIO will fail, and that would potentially overflow our
>>   	 * write pointer offset beyond the end of the zone.
>>   	 */
>> -	if (disk_zone_wplug_is_full(disk, zwplug))
>> +	if (!disk->queue->limits.driver_preserves_write_order
>> +	    && disk_zone_wplug_is_full(disk, zwplug))
> 
> Writing to a zone that is full is an error, pipelining or not. So why do you
> change this ? This does not make sense.

Agreed. I will drop this change.

>> @@ -956,7 +964,8 @@ static bool blk_zone_wplug_prepare_bio(struct blk_zone_wplug *zwplug,
>>   		 * with a start sector not unaligned to the zone write pointer
>>   		 * will fail.
>>   		 */
>> -		if (bio_offset_from_zone_start(bio) != zwplug->wp_offset)
>> +		if (!disk->queue->limits.driver_preserves_write_order
>> +		    && bio_offset_from_zone_start(bio) != zwplug->wp_offset)
> 
> Same here. This does not depend pipelining: write should have been received in
> order and be aligned with the wp. So why change this condition ?

Agreed. I will drop this change too. The reason I hit this and the above
condition is because the blk_zone_wplug_prepare_bio() call in this patch
is misplaced: it should happen after all the 'goto plug' statements
instead of before.

>> @@ -1298,6 +1329,9 @@ static void blk_zone_submit_one_bio(struct blk_zone_wplug *zwplug)
>>   	} else {
>>   		blk_mq_submit_bio(bio);
>>   	}
>> +
>> +	return disk->queue->limits.driver_preserves_write_order &&
>> +		!need_resched();
> 
> I think we really need a helper for that
> "disk->queue->limits.driver_preserves_write_order". But if you make this a
> feature, it will be better.
> 
> Also, here, the test using need_resched() really need a comment explaining why
> you look at that. I do not get it personally.

The need_resched() call was added to break out of the loop if another
thread would be runnable. Anyway, since I'm not sure that call is needed
so I will drop it.

It would be appreciated if you could take a look at the reworked version
of this patch below before I repost the entire series. The changes
compared to the patch at the start of this email thread are as follows:
- The ternary operator in the queue_work_on() call has been converted
   into an if/then/else statement.
- The two blk_zone_wplug_prepare_bio() changes have been dropped.
- The behavior changes for zone append writes in
   blk_zone_wplug_handle_write() have been dropped.
- blk_zone_wplug_handle_write() has been changed such that
   blk_zone_wplug_prepare_bio() is only called once for each bio.
- blk_zone_wplug_handle_write() has been modified such that no
   plugging happens if a cached request has been allocated from the
   same cpu as zwplug->from_cpu.
- The function blk_zone_submit_one_bio() has been removed and has been
   inlined into its only caller, blk_zone_wplug_bio_work().

Thanks,

Bart.


[PATCH] blk-zoned: Support pipelining of zoned writes

Support pipelining of zoned writes if the block driver preserves the write
order per hardware queue. Track per zone to which software queue writes
have been queued. If zoned writes are pipelined, submit new writes to the
same software queue as the writes that are already in progress. This
prevents reordering by submitting requests for the same zone to different
software or hardware queues.

Cc: Christoph Hellwig <hch@lst.de>
Cc: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
  block/blk-mq.c    |  4 +--
  block/blk-zoned.c | 67 ++++++++++++++++++++++++++++++++++++++---------
  2 files changed, 56 insertions(+), 15 deletions(-)

diff --git a/block/blk-mq.c b/block/blk-mq.c
index 5d216940e9fe..76c45bf81b02 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -3148,8 +3148,8 @@ void blk_mq_submit_bio(struct bio *bio)
  	/*
  	 * A BIO that was released from a zone write plug has already been
  	 * through the preparation in this function, already holds a reference
-	 * on the queue usage counter, and is the only write BIO in-flight for
-	 * the target zone. Go straight to preparing a request for it.
+	 * on the queue usage counter. Go straight to preparing a request for
+	 * it.
  	 */
  	if (bio_zone_write_plugging(bio)) {
  		nr_segs = bio->__bi_nr_segments;
diff --git a/block/blk-zoned.c b/block/blk-zoned.c
index de3fb0cd6d37..d9d624d90693 100644
--- a/block/blk-zoned.c
+++ b/block/blk-zoned.c
@@ -53,6 +53,8 @@ static const char *const zone_cond_name[] = {
   * @zone_no: The number of the zone the plug is managing.
   * @wp_offset: The zone write pointer location relative to the start 
of the zone
   *             as a number of 512B sectors.
+ * @from_cpu: Software queue to submit writes from for drivers that 
preserve
+ *	the write order.
   * @bio_list: The list of BIOs that are currently plugged.
   * @bio_work: Work struct to handle issuing of plugged BIOs
   * @rcu_head: RCU head to free zone write plugs with an RCU grace period.
@@ -65,6 +67,7 @@ struct blk_zone_wplug {
  	unsigned int		flags;
  	unsigned int		zone_no;
  	unsigned int		wp_offset;
+	int			from_cpu;
  	struct bio_list		bio_list;
  	struct work_struct	bio_work;
  	struct rcu_head		rcu_head;
@@ -74,8 +77,7 @@ struct blk_zone_wplug {
  /*
   * Zone write plug flags bits:
   *  - BLK_ZONE_WPLUG_PLUGGED: Indicates that the zone write plug is 
plugged,
- *    that is, that write BIOs are being throttled due to a write BIO 
already
- *    being executed or the zone write plug bio list is not empty.
+ *    that is, that write BIOs are being throttled.
   *  - BLK_ZONE_WPLUG_NEED_WP_UPDATE: Indicates that we lost track of a 
zone
   *    write pointer offset and need to update it.
   *  - BLK_ZONE_WPLUG_UNHASHED: Indicates that the zone write plug was 
removed
@@ -572,6 +574,7 @@ static struct blk_zone_wplug 
*disk_get_and_lock_zone_wplug(struct gendisk *disk,
  	zwplug->flags = 0;
  	zwplug->zone_no = zno;
  	zwplug->wp_offset = bdev_offset_from_zone_start(disk->part0, sector);
+	zwplug->from_cpu = -1;
  	bio_list_init(&zwplug->bio_list);
  	INIT_WORK(&zwplug->bio_work, blk_zone_wplug_bio_work);
  	zwplug->disk = disk;
@@ -768,14 +771,19 @@ static bool blk_zone_wplug_handle_reset_all(struct 
bio *bio)
  static void disk_zone_wplug_schedule_bio_work(struct gendisk *disk,
  					      struct blk_zone_wplug *zwplug)
  {
+	lockdep_assert_held(&zwplug->lock);
+
  	/*
  	 * Take a reference on the zone write plug and schedule the submission
  	 * of the next plugged BIO. blk_zone_wplug_bio_work() will release the
  	 * reference we take here.
  	 */
-	WARN_ON_ONCE(!(zwplug->flags & BLK_ZONE_WPLUG_PLUGGED));
  	refcount_inc(&zwplug->ref);
-	queue_work(disk->zone_wplugs_wq, &zwplug->bio_work);
+	if (zwplug->from_cpu >= 0)
+		queue_work_on(zwplug->from_cpu, disk->zone_wplugs_wq,
+			      &zwplug->bio_work);
+	else
+		queue_work(disk->zone_wplugs_wq, &zwplug->bio_work);
  }

  static inline void disk_zone_wplug_add_bio(struct gendisk *disk,
@@ -972,9 +980,12 @@ static bool blk_zone_wplug_prepare_bio(struct 
blk_zone_wplug *zwplug,
  	return true;
  }

-static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int 
nr_segs)
+static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int 
nr_segs,
+					int from_cpu)
  {
  	struct gendisk *disk = bio->bi_bdev->bd_disk;
+	const bool ordered_hwq = bio_op(bio) != REQ_OP_ZONE_APPEND &&
+		(disk->queue->limits.features & BLK_FEAT_ORDERED_HWQ);
  	sector_t sector = bio->bi_iter.bi_sector;
  	bool schedule_bio_work = false;
  	struct blk_zone_wplug *zwplug;
@@ -1034,15 +1045,36 @@ static bool blk_zone_wplug_handle_write(struct 
bio *bio, unsigned int nr_segs)
  	if (zwplug->flags & BLK_ZONE_WPLUG_PLUGGED)
  		goto add_to_bio_list;

+	if (ordered_hwq && zwplug->from_cpu < 0) {
+		/* No zoned writes are in progress. Select the current CPU. */
+		zwplug->from_cpu = raw_smp_processor_id();
+	}
+	if (ordered_hwq && zwplug->from_cpu == from_cpu) {
+		/*
+		 * The block driver preserves the write order, zoned writes have
+		 * not been plugged and the zoned write will be submitted from
+		 * zwplug->from_cpu. Let the caller submit the bio.
+		 */
+	} else if (ordered_hwq) {
+		/*
+		 * The block driver preserves the write order. Submit the bio
+		 * from zwplug->from_cpu.
+		 */
+		goto plug;
+	} else {
+		/*
+		 * The block driver does not preserve the write order. Plug and
+		 * submit the BIO.
+		 */
+		zwplug->flags |= BLK_ZONE_WPLUG_PLUGGED;
+	}
+
  	if (!blk_zone_wplug_prepare_bio(zwplug, bio)) {
  		spin_unlock_irqrestore(&zwplug->lock, flags);
  		bio_io_error(bio);
  		return true;
  	}

-	/* Otherwise, plug and submit the BIO. */
-	zwplug->flags |= BLK_ZONE_WPLUG_PLUGGED;
-
  	spin_unlock_irqrestore(&zwplug->lock, flags);

  	return false;
@@ -1150,7 +1182,7 @@ bool blk_zone_plug_bio(struct bio *bio, unsigned 
int nr_segs, int rq_cpu)
  		fallthrough;
  	case REQ_OP_WRITE:
  	case REQ_OP_WRITE_ZEROES:
-		return blk_zone_wplug_handle_write(bio, nr_segs);
+		return blk_zone_wplug_handle_write(bio, nr_segs, rq_cpu);
  	case REQ_OP_ZONE_RESET:
  		return blk_zone_wplug_handle_reset_or_finish(bio, 0);
  	case REQ_OP_ZONE_FINISH:
@@ -1182,6 +1214,9 @@ static void disk_zone_wplug_unplug_bio(struct 
gendisk *disk,

  	zwplug->flags &= ~BLK_ZONE_WPLUG_PLUGGED;

+	if (refcount_read(&zwplug->ref) == 2)
+		zwplug->from_cpu = -1;
+
  	/*
  	 * If the zone is full (it was fully written or finished, or empty
  	 * (it was reset), remove its zone write plug from the hash table.
@@ -1283,6 +1318,7 @@ static void blk_zone_wplug_bio_work(struct 
work_struct *work)
  	struct blk_zone_wplug *zwplug =
  		container_of(work, struct blk_zone_wplug, bio_work);
  	struct block_device *bdev;
+	bool ordered_hwq;
  	struct bio *bio;

  	do {
@@ -1303,6 +1339,8 @@ static void blk_zone_wplug_bio_work(struct 
work_struct *work)
  						 bio->bi_iter.bi_sector,
  						 bio_sectors(bio));

+			ordered_hwq = zwplug->disk->queue->limits.features &
+				BLK_FEAT_ORDERED_HWQ;
  			if (!blk_zone_wplug_prepare_bio(zwplug, bio)) {
  				blk_zone_wplug_bio_io_error(zwplug, bio);
  				goto again;
@@ -1323,7 +1361,7 @@ static void blk_zone_wplug_bio_work(struct 
work_struct *work)
  		} else {
  			blk_mq_submit_bio(bio);
  		}
-	} while (0);
+	} while (ordered_hwq);

  put_zwplug:
  	/* Drop the reference we took in disk_zone_wplug_schedule_bio_work(). */
@@ -1850,6 +1888,7 @@ static void queue_zone_wplug_show(struct 
blk_zone_wplug *zwplug,
  	unsigned int zwp_zone_no, zwp_ref;
  	unsigned int zwp_bio_list_size;
  	unsigned long flags;
+	int from_cpu;

  	spin_lock_irqsave(&zwplug->lock, flags);
  	zwp_zone_no = zwplug->zone_no;
@@ -1857,10 +1896,12 @@ static void queue_zone_wplug_show(struct 
blk_zone_wplug *zwplug,
  	zwp_ref = refcount_read(&zwplug->ref);
  	zwp_wp_offset = zwplug->wp_offset;
  	zwp_bio_list_size = bio_list_size(&zwplug->bio_list);
+	from_cpu = zwplug->from_cpu;
  	spin_unlock_irqrestore(&zwplug->lock, flags);

-	seq_printf(m, "%u 0x%x %u %u %u\n", zwp_zone_no, zwp_flags, zwp_ref,
-		   zwp_wp_offset, zwp_bio_list_size);
+	seq_printf(m, "zone_no %u flags 0x%x ref %u wp_offset %u bio_list_size 
%u from_cpu %d\n",
+		   zwp_zone_no, zwp_flags, zwp_ref, zwp_wp_offset,
+		   zwp_bio_list_size, from_cpu);
  }

  int queue_zone_wplugs_show(void *data, struct seq_file *m)