ext4: handle the rest of ext4_mb_load_buddy() ENOMEM errors
authorKonstantin Khlebnikov <khlebnikov@yandex-team.ru>
Mon, 22 May 2017 02:35:23 +0000 (22:35 -0400)
committerTheodore Ts'o <tytso@mit.edu>
Mon, 22 May 2017 02:35:23 +0000 (22:35 -0400)
commit9651e6b2e20648d04d5e1fe6479a3056047e8781
treea7699442ffe923af87988f8631945f244d5ec3bd
parent3f1d5bad3fae983da07be01cff2fde13293bb7b9
ext4: handle the rest of ext4_mb_load_buddy() ENOMEM errors

I've got another report about breaking ext4 by ENOMEM error returned from
ext4_mb_load_buddy() caused by memory shortage in memory cgroup.
This time inside ext4_discard_preallocations().

This patch replaces ext4_error() with ext4_warning() where errors returned
from ext4_mb_load_buddy() are not fatal and handled by caller:
* ext4_mb_discard_group_preallocations() - called before generating ENOSPC,
  we'll try to discard other group or return ENOSPC into user-space.
* ext4_trim_all_free() - just stop trimming and return ENOMEM from ioctl.

Some callers cannot handle errors, thus __GFP_NOFAIL is used for them:
* ext4_discard_preallocations()
* ext4_mb_discard_lg_preallocations()

Fixes: adb7ef600cc9 ("ext4: use __GFP_NOFAIL in ext4_free_blocks()")
Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
fs/ext4/mballoc.c