dm: fix BLK_STS_DM_REQUEUE handling when dm_io represents split bio

Commit 7dd76d1fee ("dm: improve bio splitting and associated IO
accounting") removed using cloned bio when dm io splitting is needed.
Using bio_trim()+bio_inc_remaining() rather than bio_split()+bio_chain()
causes multiple dm_io instances to share the same original bio, and it
works fine if IOs are completed successfully.

But a regression was caused for the case when BLK_STS_DM_REQUEUE is
returned from any one of DM's cloned bios (whose dm_io share the same
orig_bio). In this BLK_STS_DM_REQUEUE case only the mapped subset of
the original bio for the current exact dm_io needs to be re-submitted.
However, since the original bio is shared among all dm_io instances,
the ->orig_bio actually only represents the last dm_io instance, so
requeue can't work as expected. Also when more than one dm_io is
requeued, the same original bio is requeued from all dm_io's
completion handler, then race is caused.

Fix this issue by still allocating one clone bio for completing io
only, then io accounting can rely on ->orig_bio being unmodified. This
is needed because the dm_io's sector_offset and sectors members are
recorded relative to an unmodified ->orig_bio.

In the future, we can go back to using bio_trim()+bio_inc_remaining()
for dm's io splitting but then delay needing a bio clone only when
handling BLK_STS_DM_REQUEUE, but that approach is a bit complicated
(so it needs a development cycle):
1) bio clone needs to be done in task context
2) a block interface for unwinding bio is required

Fixes: 7dd76d1fee ("dm: improve bio splitting and associated IO accounting")
Reported-by: Benjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@kernel.org>
This commit is contained in:
Ming Lei 2022-06-23 21:20:05 +08:00 committed by Mike Snitzer
parent 78ccef9123
commit 61b6e2e532
2 changed files with 8 additions and 4 deletions

View File

@ -272,6 +272,7 @@ struct dm_io {
atomic_t io_count; atomic_t io_count;
struct mapped_device *md; struct mapped_device *md;
struct bio *split_bio;
/* The three fields represent mapped part of original bio */ /* The three fields represent mapped part of original bio */
struct bio *orig_bio; struct bio *orig_bio;
unsigned int sector_offset; /* offset to end of orig_bio */ unsigned int sector_offset; /* offset to end of orig_bio */

View File

@ -594,6 +594,7 @@ static struct dm_io *alloc_io(struct mapped_device *md, struct bio *bio)
atomic_set(&io->io_count, 2); atomic_set(&io->io_count, 2);
this_cpu_inc(*md->pending_io); this_cpu_inc(*md->pending_io);
io->orig_bio = bio; io->orig_bio = bio;
io->split_bio = NULL;
io->md = md; io->md = md;
spin_lock_init(&io->lock); spin_lock_init(&io->lock);
io->start_time = jiffies; io->start_time = jiffies;
@ -887,7 +888,7 @@ static void dm_io_complete(struct dm_io *io)
{ {
blk_status_t io_error; blk_status_t io_error;
struct mapped_device *md = io->md; struct mapped_device *md = io->md;
struct bio *bio = io->orig_bio; struct bio *bio = io->split_bio ? io->split_bio : io->orig_bio;
if (io->status == BLK_STS_DM_REQUEUE) { if (io->status == BLK_STS_DM_REQUEUE) {
unsigned long flags; unsigned long flags;
@ -1693,9 +1694,11 @@ static void dm_split_and_process_bio(struct mapped_device *md,
* Remainder must be passed to submit_bio_noacct() so it gets handled * Remainder must be passed to submit_bio_noacct() so it gets handled
* *after* bios already submitted have been completely processed. * *after* bios already submitted have been completely processed.
*/ */
bio_trim(bio, io->sectors, ci.sector_count); WARN_ON_ONCE(!dm_io_flagged(io, DM_IO_WAS_SPLIT));
trace_block_split(bio, bio->bi_iter.bi_sector); io->split_bio = bio_split(bio, io->sectors, GFP_NOIO,
bio_inc_remaining(bio); &md->queue->bio_split);
bio_chain(io->split_bio, bio);
trace_block_split(io->split_bio, bio->bi_iter.bi_sector);
submit_bio_noacct(bio); submit_bio_noacct(bio);
out: out:
/* /*