block: return ELEVATOR_DISCARD_MERGE if possible
authorMing Lei <ming.lei@redhat.com>
Thu, 29 Jul 2021 03:42:26 +0000 (11:42 +0800)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 25 May 2022 07:14:38 +0000 (09:14 +0200)
commit9ea8e6a8323e063c42536f3f8c83d3768cff5f4d
treeb8d1e6596b2e41cabf45701a3e10ed1d79332995
parent36ac6caf742d9597a1ce3f2eeeb5de895fa75fd2
block: return ELEVATOR_DISCARD_MERGE if possible

commit 866663b7b52d2da267b28e12eed89ee781b8fed1 upstream.

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>
Signed-off-by: Gwendal Grignou <gwendal@chromium.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
block/bfq-iosched.c
block/blk-merge.c
block/elevator.c
block/mq-deadline.c
include/linux/blkdev.h