memcg, fsnotify: no oom-kill for remote memcg charging
authorShakeel Butt <shakeelb@google.com>
Fri, 12 Jul 2019 03:55:52 +0000 (20:55 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Fri, 12 Jul 2019 18:05:43 +0000 (11:05 -0700)
commitec165450968b26298bd1c373de37b0ab6d826b33
treeb1ab4b67cdac726624b880cd17d521554e95a2cf
parent38d384932ed1b58f0ed8dd17c66e0cd89acb354a
memcg, fsnotify: no oom-kill for remote memcg charging

Commit d46eb14b735b ("fs: fsnotify: account fsnotify metadata to
kmemcg") added remote memcg charging for fanotify and inotify event
objects.  The aim was to charge the memory to the listener who is
interested in the events but without triggering the OOM killer.
Otherwise there would be security concerns for the listener.

At the time, oom-kill trigger was not in the charging path.  A parallel
work added the oom-kill back to charging path i.e.  commit 29ef680ae7c2
("memcg, oom: move out_of_memory back to the charge path").  So to not
trigger oom-killer in the remote memcg, explicitly add
__GFP_RETRY_MAYFAIL to the fanotigy and inotify event allocations.

Link: http://lkml.kernel.org/r/20190514212259.156585-2-shakeelb@google.com
Signed-off-by: Shakeel Butt <shakeelb@google.com>
Reviewed-by: Roman Gushchin <guro@fb.com>
Acked-by: Jan Kara <jack@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
fs/notify/fanotify/fanotify.c
fs/notify/inotify/inotify_fsnotify.c