[PATCH] block: return ELEVATOR_DISCARD_MERGE if possible

[PATCH] block: return ELEVATOR_DISCARD_MERGE if possible

[Gwendal Grignou]

From: Ming Lei <ming.lei@redhat.com>

When merging one bio to request, if they are discard IO and the queue
supports multi-range discard, we need to return ELEVATOR_DISCARD_MERGE
because both block core and related drivers(nvme, virtio-blk) doesn't
handle mixed discard io merge(traditional IO merge together with
discard merge) well.

Fix the issue by returning ELEVATOR_DISCARD_MERGE in this situation,
so both blk-mq and drivers just need to handle multi-range discard.

Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Fixes: 2705dfb20947 ("block: fix discard request merge")
Link: https://lore.kernel.org/r/20210729034226.1591070-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

commit 866663b7b52d2 upstream.

Similar to commit 87aa69aa10b42 ("block: return ELEVATOR_DISCARD_MERGE if possible")
in 5.10 kernel.

Conflicts:
   block/blk-merge.c: function at a different place.
   block/mq-deadline-main.c: not in 5.4, use mq-deadline.c instead.

Cc: <stable@vger.kernel.org> # 5.4.y
Signed-off-by: Gwendal Grignou <gwendal@chromium.org>
---
 block/bfq-iosched.c    |  3 +++
 block/blk-merge.c      | 15 ---------------
 block/elevator.c       |  3 +++
 block/mq-deadline.c    |  2 ++
 include/linux/blkdev.h | 16 ++++++++++++++++
 5 files changed, 24 insertions(+), 15 deletions(-)

diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index 1d443d17cf7c5..d46806182b051 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -2251,6 +2251,9 @@ static int bfq_request_merge(struct request_queue *q, struct request **req,
 	__rq = bfq_find_rq_fmerge(bfqd, bio, q);
 	if (__rq && elv_bio_merge_ok(__rq, bio)) {
 		*req = __rq;
+
+		if (blk_discard_mergable(__rq))
+			return ELEVATOR_DISCARD_MERGE;
 		return ELEVATOR_FRONT_MERGE;
 	}
 
diff --git a/block/blk-merge.c b/block/blk-merge.c
index a62692d135660..5219064cd72bb 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -721,21 +721,6 @@ static void blk_account_io_merge(struct request *req)
 		part_stat_unlock();
 	}
 }
-/*
- * Two cases of handling DISCARD merge:
- * If max_discard_segments > 1, the driver takes every bio
- * as a range and send them to controller together. The ranges
- * needn't to be contiguous.
- * Otherwise, the bios/requests will be handled as same as
- * others which should be contiguous.
- */
-static inline bool blk_discard_mergable(struct request *req)
-{
-	if (req_op(req) == REQ_OP_DISCARD &&
-	    queue_max_discard_segments(req->q) > 1)
-		return true;
-	return false;
-}
 
 static enum elv_merge blk_try_req_merge(struct request *req,
 					struct request *next)
diff --git a/block/elevator.c b/block/elevator.c
index 78805c74ea8a4..3ba826230c578 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -337,6 +337,9 @@ enum elv_merge elv_merge(struct request_queue *q, struct request **req,
 	__rq = elv_rqhash_find(q, bio->bi_iter.bi_sector);
 	if (__rq && elv_bio_merge_ok(__rq, bio)) {
 		*req = __rq;
+
+		if (blk_discard_mergable(__rq))
+			return ELEVATOR_DISCARD_MERGE;
 		return ELEVATOR_BACK_MERGE;
 	}
 
diff --git a/block/mq-deadline.c b/block/mq-deadline.c
index 19c6922e85f1b..6d6dda5cfffa3 100644
--- a/block/mq-deadline.c
+++ b/block/mq-deadline.c
@@ -452,6 +452,8 @@ static int dd_request_merge(struct request_queue *q, struct request **rq,
 
 		if (elv_bio_merge_ok(__rq, bio)) {
 			*rq = __rq;
+			if (blk_discard_mergable(__rq))
+				return ELEVATOR_DISCARD_MERGE;
 			return ELEVATOR_FRONT_MERGE;
 		}
 	}
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 8cc766743270f..308c2d8cdca19 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1409,6 +1409,22 @@ static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector
 	return offset << SECTOR_SHIFT;
 }
 
+/*
+ * Two cases of handling DISCARD merge:
+ * If max_discard_segments > 1, the driver takes every bio
+ * as a range and send them to controller together. The ranges
+ * needn't to be contiguous.
+ * Otherwise, the bios/requests will be handled as same as
+ * others which should be contiguous.
+ */
+static inline bool blk_discard_mergable(struct request *req)
+{
+	if (req_op(req) == REQ_OP_DISCARD &&
+	    queue_max_discard_segments(req->q) > 1)
+		return true;
+	return false;
+}
+
 static inline int bdev_discard_alignment(struct block_device *bdev)
 {
 	struct request_queue *q = bdev_get_queue(bdev);
-- 
2.36.1.124.g0e6072fb45-goog

Re: [PATCH] block: return ELEVATOR_DISCARD_MERGE if possible

[Greg KH]

On Sat, May 21, 2022 at 11:56:26AM -0700, Gwendal Grignou wrote:
> From: Ming Lei <ming.lei@redhat.com>
> 
> When merging one bio to request, if they are discard IO and the queue
> supports multi-range discard, we need to return ELEVATOR_DISCARD_MERGE
> because both block core and related drivers(nvme, virtio-blk) doesn't
> handle mixed discard io merge(traditional IO merge together with
> discard merge) well.
> 
> Fix the issue by returning ELEVATOR_DISCARD_MERGE in this situation,
> so both blk-mq and drivers just need to handle multi-range discard.
> 
> Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name>
> Signed-off-by: Ming Lei <ming.lei@redhat.com>
> Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
> Fixes: 2705dfb20947 ("block: fix discard request merge")
> Link: https://lore.kernel.org/r/20210729034226.1591070-1-ming.lei@redhat.com
> Signed-off-by: Jens Axboe <axboe@kernel.dk>
> 
> commit 866663b7b52d2 upstream.
> 
> Similar to commit 87aa69aa10b42 ("block: return ELEVATOR_DISCARD_MERGE if possible")
> in 5.10 kernel.
> 
> Conflicts:
>    block/blk-merge.c: function at a different place.
>    block/mq-deadline-main.c: not in 5.4, use mq-deadline.c instead.
> 
> Cc: <stable@vger.kernel.org> # 5.4.y
> Signed-off-by: Gwendal Grignou <gwendal@chromium.org>

Now queued up, thanks.

greg k-h