block: ignore RWF_HIPRI hint for sync dio
authorMing Lei <ming.lei@redhat.com>
Wed, 20 Apr 2022 14:31:10 +0000 (22:31 +0800)
committerJens Axboe <axboe@kernel.dk>
Mon, 2 May 2022 16:07:42 +0000 (10:07 -0600)
commit9650b453a3d4b1b8ed4ea8bcb9b40109608d1faf
treec43dfc5033aec249e22536898a6a043390df7782
parente233fe1aa02815f38588a5a965a197bbcabfb125
block: ignore RWF_HIPRI hint for sync dio

So far bio is marked as REQ_POLLED if RWF_HIPRI/IOCB_HIPRI is passed
from userspace sync io interface, then block layer tries to poll until
the bio is completed. But the current implementation calls
blk_io_schedule() if bio_poll() returns 0, and this way causes io hang or
timeout easily.

But looks no one reports this kind of issue, which should have been
triggered in normal io poll sanity test or blktests block/007 as
observed by Changhui, that means it is very likely that no one uses it
or no one cares it.

Also after io_uring is invented, io poll for sync dio becomes legacy
interface.

So ignore RWF_HIPRI hint for sync dio.

CC: linux-mm@kvack.org
Cc: linux-xfs@vger.kernel.org
Reported-by: Changhui Zhong <czhong@redhat.com>
Suggested-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Tested-by: Changhui Zhong <czhong@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20220420143110.2679002-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
block/fops.c
fs/iomap/direct-io.c
mm/page_io.c