md/raid5: fix 'out of memory' during raid cache recovery
authorAlexei Naberezhnov <anaberezhnov@fb.com>
Tue, 27 Mar 2018 23:54:16 +0000 (16:54 -0700)
committerSong Liu <songliubraving@fb.com>
Mon, 28 Jan 2019 19:44:40 +0000 (11:44 -0800)
commit483cbbeddd5fe2c80fd4141ff0748fa06c4ff146
tree3ebb98292523a656d752359fff20b39978d7e31d
parentf17b5f06cb92ef2250513a1e154c47b78df07d40
md/raid5: fix 'out of memory' during raid cache recovery

This fixes the case when md array assembly fails because of raid cache recovery
unable to allocate a stripe, despite attempts to replay stripes and increase
cache size. This happens because stripes released by r5c_recovery_replay_stripes
and raid5_set_cache_size don't become available for allocation immediately.
Released stripes first are placed on conf->released_stripes list and require
md thread to merge them on conf->inactive_list before they can be allocated.

Patch allows final allocation attempt during cache recovery to wait for
new stripes to become availabe for allocation.

Cc: linux-raid@vger.kernel.org
Cc: Shaohua Li <shli@kernel.org>
Cc: linux-stable <stable@vger.kernel.org> # 4.10+
Fixes: b4c625c67362 ("md/r5cache: r5cache recovery: part 1")
Signed-off-by: Alexei Naberezhnov <anaberezhnov@fb.com>
Signed-off-by: Song Liu <songliubraving@fb.com>
drivers/md/raid5-cache.c
drivers/md/raid5.c