From: Michal Hocko <mhocko@kernel.org>
To: linux-mm@kvack.org
Cc: Andrew Morton <akpm@linux-foundation.org>,
Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>,
Oleg Nesterov <oleg@redhat.com>,
David Rientjes <rientjes@google.com>,
Vladimir Davydov <vdavydov@parallels.com>,
"Michael S. Tsirkin" <mst@redhat.com>,
Michal Hocko <mhocko@suse.com>
Subject: [RFC PATCH 0/6] fortify oom killer even more
Date: Fri, 1 Jul 2016 11:26:24 +0200 [thread overview]
Message-ID: <1467365190-24640-1-git-send-email-mhocko@kernel.org> (raw)
Hi,
I am sending this pile as an RFC and I hope it will make a good
foundation to hopefully plug the remaining holes which could lead to oom
lockups for CONFIG_MMU.
There are two main parts patches 1-4 and the 5-6. The first pile focuses
on moving decisions about oom victims more to mm_struct. Especially the
part when there is an oom victim noticed and we decide whether to select
new victim. Patch 1 remembers the mm at the time oom victim is selected
and it is stable for the rest of the process group life time. This
allows some simplifications.
The later part is about kthread vs. oom_reaper interaction. It seems that
the only use_mm() user which needs fixing is vhost and that is fixable.
Then we can remove the kthread restriction and so basically the every
oom victim will be reapable now (well except the weird cases where the
mm is shared with init but I consider that uninteresting).
I haven't tested this properly yet. I will be mostly offline next week
but definitely plan to test it later on. Right now I would appreciated
feedback/review. If this looks OK then I would like to target it for 4.9.
The series is based on top of the current mmotm (2016-06-24-15-53) +
http://lkml.kernel.org/r/1467201562-6709-1-git-send-email-mhocko@kernel.org
Thanks!
Michal Hocko (6):
oom: keep mm of the killed task available
oom, suspend: fix oom_killer_disable vs. pm suspend properly
exit, oom: postpone exit_oom_victim to later
oom, oom_reaper: consider mmget_not_zero as a failure
vhost, mm: make sure that oom_reaper doesn't reap memory read by vhost
oom, oom_reaper: allow to reap mm shared by the kthreads
drivers/vhost/scsi.c | 2 +-
drivers/vhost/vhost.c | 18 +++----
include/linux/oom.h | 2 +-
include/linux/sched.h | 3 ++
include/linux/uaccess.h | 22 +++++++++
include/linux/uio.h | 10 ++++
kernel/exit.c | 5 +-
kernel/fork.c | 2 +
kernel/power/process.c | 17 ++-----
mm/mmu_context.c | 6 +++
mm/oom_kill.c | 127 ++++++++++++++++++++++++------------------------
11 files changed, 124 insertions(+), 90 deletions(-)
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next reply other threads:[~2016-07-01 9:26 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-07-01 9:26 Michal Hocko [this message]
2016-07-01 9:26 ` [RFC PATCH 1/6] oom: keep mm of the killed task available Michal Hocko
2016-07-03 2:45 ` Tetsuo Handa
2016-07-07 8:24 ` Michal Hocko
2016-07-07 11:48 ` Tetsuo Handa
2016-07-07 13:32 ` Michal Hocko
2016-07-01 9:26 ` [RFC PATCH 2/6] oom, suspend: fix oom_killer_disable vs. pm suspend properly Michal Hocko
2016-07-01 9:26 ` [RFC PATCH 3/6] exit, oom: postpone exit_oom_victim to later Michal Hocko
2016-07-01 9:26 ` [RFC PATCH 4/6] oom, oom_reaper: consider mmget_not_zero as a failure Michal Hocko
2016-07-01 9:26 ` [RFC PATCH 5/6] vhost, mm: make sure that oom_reaper doesn't reap memory read by vhost Michal Hocko
2016-07-03 13:47 ` Oleg Nesterov
2016-07-03 14:09 ` Michael S. Tsirkin
2016-07-03 15:18 ` Oleg Nesterov
2016-07-03 15:30 ` Michael S. Tsirkin
2016-07-03 16:47 ` Oleg Nesterov
2016-07-03 21:17 ` Michael S. Tsirkin
2016-07-07 8:28 ` Michal Hocko
2016-07-07 15:38 ` Michael S. Tsirkin
2016-07-08 12:29 ` Oleg Nesterov
2016-07-11 14:14 ` Michal Hocko
2016-07-12 14:33 ` Oleg Nesterov
2016-07-07 8:42 ` Michal Hocko
2016-07-07 16:46 ` Oleg Nesterov
2016-07-07 8:39 ` Michal Hocko
2016-07-22 11:09 ` Michal Hocko
2016-07-01 9:26 ` [RFC PATCH 6/6] oom, oom_reaper: allow to reap mm shared by the kthreads Michal Hocko
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1467365190-24640-1-git-send-email-mhocko@kernel.org \
--to=mhocko@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.com \
--cc=mst@redhat.com \
--cc=oleg@redhat.com \
--cc=penguin-kernel@I-love.SAKURA.ne.jp \
--cc=rientjes@google.com \
--cc=vdavydov@parallels.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).