On Wed, Nov 25, 2020 at 05:17:07PM +0800, JeffleXu wrote:

On 11/25/20 4:29 PM, Ming Lei wrote:

quoted

On Wed, Nov 25, 2020 at 02:41:47PM +0800, Jeffle Xu wrote:

quoted

iopoll is initially for small size, latency sensitive IO. It doesn't
work well for big IO, especially when it needs to be split to multiple
bios. In this case, the returned cookie of __submit_bio_noacct_mq() is
indeed the cookie of the last split bio. The completion of *this* last
split bio done by iopoll doesn't mean the whole original bio has
completed. Callers of iopoll still need to wait for completion of other
split bios.

Besides bio splitting may cause more trouble for iopoll which isn't
supposed to be used in case of big IO.

iopoll for split bio may cause potential race if CPU migration happens
during bio submission. Since the returned cookie is that of the last
split bio, polling on the corresponding hardware queue doesn't help
complete other split bios, if these split bios are enqueued into
different hardware queues. Since interrupts are disabled for polling
queues, the completion of these other split bios depends on timeout
mechanism, thus causing a potential hang.

iopoll for split bio may also cause hang for sync polling. Currently
both the blkdev and iomap-based fs (ext4/xfs, etc) support sync polling
in direct IO routine. These routines will submit bio without REQ_NOWAIT
flag set, and then start sync polling in current process context. The
process may hang in blk_mq_get_tag() if the submitted bio has to be
split into multiple bios and can rapidly exhaust the queue depth. The
process are waiting for the completion of the previously allocated
requests, which should be reaped by the following polling, and thus
causing a deadlock.

To avoid these subtle trouble described above, just disable iopoll for
split bio.

Suggested-by: Ming Lei <redacted>
Signed-off-by: Jeffle Xu <jefflexu@linux.alibaba.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
---
 block/bio.c               |  2 ++
 block/blk-merge.c         | 12 ++++++++++++
 block/blk-mq.c            |  3 +++
 include/linux/blk_types.h |  1 +
 4 files changed, 18 insertions(+)

diff --git a/block/bio.c b/block/bio.c
index fa01bef35bb1..7f7ddc22a30d 100644
--- a/block/bio.c
+++ b/block/bio.c

@@ -684,6 +684,8 @@ void __bio_clone_fast(struct bio *bio, struct bio *bio_src)
 	bio_set_flag(bio, BIO_CLONED);
 	if (bio_flagged(bio_src, BIO_THROTTLED))
 		bio_set_flag(bio, BIO_THROTTLED);
+	if (bio_flagged(bio_src, BIO_SPLIT))
+		bio_set_flag(bio, BIO_SPLIT);
 	bio->bi_opf = bio_src->bi_opf;
 	bio->bi_ioprio = bio_src->bi_ioprio;
 	bio->bi_write_hint = bio_src->bi_write_hint;

diff --git a/block/blk-merge.c b/block/blk-merge.c
index bcf5e4580603..a2890cebf99f 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c

@@ -279,6 +279,18 @@ static struct bio *blk_bio_segment_split(struct request_queue *q,
 	return NULL;
 split:
 	*segs = nsegs;
+
+	/*
+	 * Bio splitting may cause subtle trouble such as hang when doing sync
+	 * iopoll in direct IO routine. Given performance gain of iopoll for
+	 * big IO can be trival, disable iopoll when split needed. We need
+	 * BIO_SPLIT to identify bios need this workaround. Since currently
+	 * only normal IO under mq routine may suffer this issue, BIO_SPLIT is
+	 * only marked here.
+	 */
+	bio->bi_opf &= ~REQ_HIPRI;
+	bio_set_flag(bio, BIO_SPLIT);

You may need to put the above into one helper, and call the helper for
other splitted cases(discard, write zero and write same) too.

It could be, though currently only normal IO could be marked with REQ_HIPRI.

You are right, so far RWF_HIPRI is only applied on RW IO, then no need to
do that for other non-RW split.

Thanks,
Ming