Re: [PATCH 2/5] Revert "oom: give the dying task a higher priority"
From: KOSAKI Motohiro
Date: Wed Mar 23 2011 - 20:07:16 EST
> On Wed, Mar 23, 2011 at 04:42:29PM +0900, KAMEZAWA Hiroyuki wrote:
> | On Tue, 22 Mar 2011 20:06:48 +0900 (JST)
> | KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx> wrote:
> | > This reverts commit 93b43fa55088fe977503a156d1097cc2055449a2.
> | >
> | > The commit dramatically improve oom killer logic when fork-bomb
> | > occur. But, I've found it has nasty corner case. Now cpu cgroup
> | > has strange default RT runtime. It's 0! That said, if a process
> | > under cpu cgroup promote RT scheduling class, the process never
> | > run at all.
> | >
> | > Eventually, kernel may hang up when oom kill occur.
> | >
> | > The author need to resubmit it as adding knob and disabled
> | > by default if he really need this feature.
> | >
> | > Cc: Luis Claudio R. Goncalves <lclaudio@xxxxxxxx>
> | > Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>
> | Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
> The original patch was written to fix an issue observed in 188.8.131.52-rt.
> As the logic sounded useful, I ported it to upstream. Anyway,I am trying
> a few ideas to rework that patch. In the meantime, I'm pretty fine with
> reverting the commit.
> Acked-by: Luis Claudio R. Gonçalves <lgoncalv@xxxxxxxx>
Ok, and then, I'll drop [patch 3/5] too. I hope to focus to discuss your idea.
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/