Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1446789

Re: [PATCH 10/10] mm, oom: hide mm which is shared with kthread or global init

From Andrew Morton <akpm@linux-foundation.org>
Newsgroups linux.kernel
Subject Re: [PATCH 10/10] mm, oom: hide mm which is shared with kthread or global init
Date 2016-07-20 01:30 +0200
Message-ID <rWMCC-1Wj-17@gated-at.bofh.it> (permalink)
References <rM6Om-8j0-11@gated-at.bofh.it> <rM6Om-8j0-39@gated-at.bofh.it> <rWC0x-3PW-1@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


On Tue, 19 Jul 2016 14:05:39 +0200 Michal Hocko <mhocko@kernel.org> wrote:

> > After this patch we should guarantee a forward progress for the OOM
> > killer even when the selected victim is sharing memory with a kernel
> > thread or global init.
> 
> Could you replace the last two paragraphs with the following. Tetsuo
> didn't like the guarantee mentioned there because that is a too strong
> statement as find_lock_task_mm might not find any mm and so we still
> could end up looping on the oom victim if it gets stuck somewhere in
> __mmput. This particular patch didn't aim at closing that case. Plugging
> that hole is planned later after the next upcoming merge window closes.
> 
> "
> In order to help a forward progress for the OOM killer, make sure
> that this really rare cases will not get into the way and hide
> the mm from the oom killer by setting MMF_OOM_REAPED flag for it.
> oom_scan_process_thread will ignore any TIF_MEMDIE task if it has
> MMF_OOM_REAPED flag set to catch these oom victims.
> 		        
> After this patch we should guarantee a forward progress for the OOM
> killer even when the selected victim is sharing memory with a kernel
> thread or global init as long as the victims mm is still alive.
> "

I tweaked it a bit:

: In order to help forward progress for the OOM killer, make sure that
: this really rare case will not get in the way - we do this by hiding
: the mm from the oom killer by setting MMF_OOM_REAPED flag for it. 
: oom_scan_process_thread will ignore any TIF_MEMDIE task if it has
: MMF_OOM_REAPED flag set to catch these oom victims.
: 
: After this patch we should guarantee forward progress for the OOM
: killer even when the selected victim is sharing memory with a kernel
: thread or global init as long as the victims mm is still alive.

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

Re: [PATCH 10/10] mm, oom: hide mm which is shared with kthread or  global init Michal Hocko <mhocko@kernel.org> - 2016-07-19 14:10 +0200
  Re: [PATCH 10/10] mm, oom: hide mm which is shared with kthread or  global init Andrew Morton <akpm@linux-foundation.org> - 2016-07-20 01:30 +0200
    Re: [PATCH 10/10] mm, oom: hide mm which is shared with kthread or  global init Michal Hocko <mhocko@kernel.org> - 2016-07-20 08:30 +0200

csiph-web