Re: [PATCH] oom_kill: use rss value instead of vm size for badness

[Minchan Kim <minchan.kim@gmail.com>]

On Thu, Oct 29, 2009 at 10:00 AM, KAMEZAWA Hiroyuki
<kamezawa.hiroyu@jp.fujitsu.com> wrote:
>> I'll wait until the next week to post a new patch.
>> We don't need rapid way.
>>
> I wrote above...but for my mental health, this is bug-fixed version.
> Sorry for my carelessness. David, thank you for your review.
> Regards,
> -Kame
> ==
> From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
>
> It's reported that OOM-Killer kills Gnone/KDE at first...
> And yes, we can reproduce it easily.
>
> Now, oom-killer uses mm->total_vm as its base value. But in recent
> applications, there are a big gap between VM size and RSS size.
> Because
>  - Applications attaches much dynamic libraries. (Gnome, KDE, etc...)
>  - Applications may alloc big VM area but use small part of them.
>    (Java, and multi-threaded applications has this tendency because
>     of default-size of stack.)
>
> I think using mm->total_vm as score for oom-kill is not good.
> By the same reason, overcommit memory can't work as expected.
> (In other words, if we depends on total_vm, using overcommit more positive
>  is a good choice.)
>
> This patch uses mm->anon_rss/file_rss as base value for calculating badness.
>
> Following is changes to OOM score(badness) on an environment with 1.6G memory
> plus memory-eater(500M & 1G).
>
> Top 10 of badness score. (The highest one is the first candidate to be killed)
> Before
> badness program
> 91228   gnome-settings-
> 94210   clock-applet
> 103202  mixer_applet2
> 106563  tomboy
> 112947  gnome-terminal
> 128944  mmap              <----------- 500M malloc
> 129332  nautilus
> 215476  bash              <----------- parent of 2 mallocs.
> 256944  mmap              <----------- 1G malloc
> 423586  gnome-session
>
> After
> badness
> 1911    mixer_applet2
> 1955    clock-applet
> 1986    xinit
> 1989    gnome-session
> 2293    nautilus
> 2955    gnome-terminal
> 4113    tomboy
> 104163  mmap             <----------- 500M malloc.
> 168577  bash             <----------- parent of 2 mallocs
> 232375  mmap             <----------- 1G malloc
>
> seems good for me. Maybe we can tweak this patch more,
> but this one will be a good one as a start point.
>
> Changelog: 2009/10/29
>  - use get_mm_rss() instead of get_mm_counter()
>
> Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: Minchan Kim <minchan.kim@gmail.com>

Let's start from this.

-- 
Kind regards,
Minchan Kim

--
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>