RE: [PATCH] Fix race between oom kill and task exit

From: Tu, Xiaobing
Date: Thu Nov 28 2013 - 21:08:53 EST


We will do more stress test in more machine at the same time

-----Original Message-----
From: Ma, Xindong
Sent: Friday, November 29, 2013 10:06 AM
To: Oleg Nesterov; Michal Hocko
Cc: William Dauchy; Johannes Weiner; akpm@xxxxxxxxxxxxxxxxxxxx; rientjes@xxxxxxxxxx; rusty@xxxxxxxxxxxxxxx; linux-mm@xxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; Peter Zijlstra; gregkh@xxxxxxxxxxxxxxxxxxx; Tu, Xiaobing; azurIt; Sameer Nanda
Subject: RE: [PATCH] Fix race between oom kill and task exit

> From: Oleg Nesterov [mailto:oleg@xxxxxxxxxx]
> Sent: Friday, November 29, 2013 2:39 AM
> To: Michal Hocko
> Cc: William Dauchy; Johannes Weiner; Ma, Xindong;
> akpm@xxxxxxxxxxxxxxxxxxxx; rientjes@xxxxxxxxxx; rusty@xxxxxxxxxxxxxxx;
> linux-mm@xxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; Peter Zijlstra;
> gregkh@xxxxxxxxxxxxxxxxxxx; Tu, Xiaobing; azurIt; Sameer Nanda
> Subject: Re: [PATCH] Fix race between oom kill and task exit
>
> On 11/28, Michal Hocko wrote:
> >
> > They are both trying to solve the same issue. Neither of them is
> > optimal unfortunately.
>
> yes, but this one doesn't look right.
>
> > Oleg said he would look into this and I have seen some patches but
> > didn't geto check them.
>
> Only preparations so far.
>
> Oleg.

I was not aware there's a long story for this issue. I hit this issue a lot of times during stress test and root caused it. After applying my patch, I did extensive test on 5 machines for a long time, it does not reproduced anymore so I submitted the patch.

I will do more research on this issue.
--
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/