[PATCH AUTOSEL 4.19 41/65] block: cover another queue enter recursion via BIO_QUEUE_ENTERED

[PATCH AUTOSEL 4.19 41/65] block: cover another queue enter recursion via BIO_QUEUE_ENTERED

[Sasha Levin]

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

[ Upstream commit 698cef173983b086977e633e46476e0f925ca01e ]

Except for blk_queue_split(), bio_split() is used for splitting bio too,
then the remained bio is often resubmit to queue via generic_make_request().
So the same queue enter recursion exits in this case too. Unfortunatley
commit cd4a4ae4683dc2 doesn't help this case.

This patch covers the above case by setting BIO_QUEUE_ENTERED before calling
q->make_request_fn.

In theory the per-bio flag is used to simulate one stack variable, it is
just fine to clear it after q->make_request_fn is returned. Especially
the same bio can't be submitted from another context.

Fixes: cd4a4ae4683dc2 ("block: don't use blocking queue entered for recursive bio submits")
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: NeilBrown <neilb@suse.com>
Reviewed-by:  Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 block/blk-core.c  | 11 +++++++++++
 block/blk-merge.c | 10 ----------
 2 files changed, 11 insertions(+), 10 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index eb8b52241453..a9fde80606d7 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -2454,7 +2454,18 @@ blk_qc_t generic_make_request(struct bio *bio)
 			/* Create a fresh bio_list for all subordinate requests */
 			bio_list_on_stack[1] = bio_list_on_stack[0];
 			bio_list_init(&bio_list_on_stack[0]);
+
+			/*
+			 * Since we're recursing into make_request here, ensure
+			 * that we mark this bio as already having entered the queue.
+			 * If not, and the queue is going away, we can get stuck
+			 * forever on waiting for the queue reference to drop. But
+			 * that will never happen, as we're already holding a
+			 * reference to it.
+			 */
+			bio_set_flag(bio, BIO_QUEUE_ENTERED);
 			ret = q->make_request_fn(q, bio);
+			bio_clear_flag(bio, BIO_QUEUE_ENTERED);
 
 			/* sort new bios into those for a lower level
 			 * and those for the same level
diff --git a/block/blk-merge.c b/block/blk-merge.c
index 2e042190a4f1..ba7f732c519e 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -211,16 +211,6 @@ void blk_queue_split(struct request_queue *q, struct bio **bio)
 		/* there isn't chance to merge the splitted bio */
 		split->bi_opf |= REQ_NOMERGE;
 
-		/*
-		 * Since we're recursing into make_request here, ensure
-		 * that we mark this bio as already having entered the queue.
-		 * If not, and the queue is going away, we can get stuck
-		 * forever on waiting for the queue reference to drop. But
-		 * that will never happen, as we're already holding a
-		 * reference to it.
-		 */
-		bio_set_flag(*bio, BIO_QUEUE_ENTERED);
-
 		bio_chain(split, *bio);
 		trace_block_split(q, split, (*bio)->bi_iter.bi_sector);
 		generic_make_request(*bio);
-- 
2.19.1

[PATCH AUTOSEL 4.19 65/65] Revert "block: cover another queue enter recursion via BIO_QUEUE_ENTERED"

[Sasha Levin]

From: Jens Axboe <axboe@kernel.dk>

[ Upstream commit 947b7ac135b16aa60f9141ff72bd494eda0edb5e ]

We can't touch a bio after ->make_request_fn(), for all we know it could
already have been completed by the time this function returns.

This reverts commit 698cef173983b086977e633e46476e0f925ca01e.

Reported-by: syzbot+4df6ca820108fd248943@syzkaller.appspotmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 block/blk-core.c  | 11 -----------
 block/blk-merge.c | 10 ++++++++++
 2 files changed, 10 insertions(+), 11 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index a9fde80606d7..eb8b52241453 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -2454,18 +2454,7 @@ blk_qc_t generic_make_request(struct bio *bio)
 			/* Create a fresh bio_list for all subordinate requests */
 			bio_list_on_stack[1] = bio_list_on_stack[0];
 			bio_list_init(&bio_list_on_stack[0]);
-
-			/*
-			 * Since we're recursing into make_request here, ensure
-			 * that we mark this bio as already having entered the queue.
-			 * If not, and the queue is going away, we can get stuck
-			 * forever on waiting for the queue reference to drop. But
-			 * that will never happen, as we're already holding a
-			 * reference to it.
-			 */
-			bio_set_flag(bio, BIO_QUEUE_ENTERED);
 			ret = q->make_request_fn(q, bio);
-			bio_clear_flag(bio, BIO_QUEUE_ENTERED);
 
 			/* sort new bios into those for a lower level
 			 * and those for the same level
diff --git a/block/blk-merge.c b/block/blk-merge.c
index ba7f732c519e..2e042190a4f1 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -211,6 +211,16 @@ void blk_queue_split(struct request_queue *q, struct bio **bio)
 		/* there isn't chance to merge the splitted bio */
 		split->bi_opf |= REQ_NOMERGE;
 
+		/*
+		 * Since we're recursing into make_request here, ensure
+		 * that we mark this bio as already having entered the queue.
+		 * If not, and the queue is going away, we can get stuck
+		 * forever on waiting for the queue reference to drop. But
+		 * that will never happen, as we're already holding a
+		 * reference to it.
+		 */
+		bio_set_flag(*bio, BIO_QUEUE_ENTERED);
+
 		bio_chain(split, *bio);
 		trace_block_split(q, split, (*bio)->bi_iter.bi_sector);
 		generic_make_request(*bio);
-- 
2.19.1