From: Denys Vlasenko <vda.linux@googlemail.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Kevin Shanahan <kmshanah@ucwb.org.au>, linux-kernel@vger.kernel.org
Subject: Re: OOM (HighMem) on linux 2.6.24.2
Date: Tue, 1 Apr 2008 10:17:35 +0200 [thread overview]
Message-ID: <200804011017.35352.vda.linux@googlemail.com> (raw)
In-Reply-To: <1207025847.8514.675.camel@twins>
On Tuesday 01 April 2008 06:57, Peter Zijlstra wrote:
> On Tue, 2008-04-01 at 04:31 +0200, Denys Vlasenko wrote:
> > On Thursday 20 March 2008 11:04, Kevin Shanahan wrote:
> > > Feb 28 06:32:20 hermes kernel: 5313 pages slab
> > > Feb 28 06:32:20 hermes kernel: 2632 pages pagetables
> > > Feb 28 06:32:20 hermes kernel: Out of memory: kill process 24013 (apache2) score 22070 or a child
> > > Feb 28 06:32:20 hermes kernel: Killed process 24013 (apache2)
> > >
> > > Looks like the HighMem low watermark was hit again (though I've no idea
> > > if that's actually significant).
> >
> > I kno that people routinely forget to do that, but
> > at least for purely paranoid reasons you are better
> > to run all daemons with setrlimit'ed memory size.
> > At least all daemons which you cannot reasonably
> > trust to be 100% leak-free and to have bounded
> > peak memory usage.
>
> Afaik we don't actually implement RLIMIT_RSS.
I don't know, but the following definitely works for me:
#include <stdlib.h>
int main() {
void *p;
unsigned size = 1 << 20;
unsigned long total = 0;
while (size) {
p = malloc(size);
if (!p) size >>= 1;
else {
memset(p, 0x77, size);
total += size;
printf("Allocated %9u bytes, %12lu total\n", size, total);
}
}
printf("Cannot malloc more. Exiting\n");
return 0;
}
# softlimit -a 12000000 ./oom
Allocated 1048576 bytes, 1048576 total
Allocated 1048576 bytes, 2097152 total
Allocated 1048576 bytes, 3145728 total
Allocated 1048576 bytes, 4194304 total
Allocated 1048576 bytes, 5242880 total
Allocated 1048576 bytes, 6291456 total
Allocated 1048576 bytes, 7340032 total
Allocated 1048576 bytes, 8388608 total
Allocated 1048576 bytes, 9437184 total
Allocated 1048576 bytes, 10485760 total
Allocated 262144 bytes, 10747904 total
Cannot malloc more. Exiting
--
vda
next prev parent reply other threads:[~2008-04-01 8:17 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-03-20 0:48 OOM (HighMem) on linux 2.6.24.2 Kevin Shanahan
2008-03-20 8:44 ` Jiri Slaby
2008-03-20 8:44 ` Jiri Slaby
2008-03-20 10:17 ` Kevin Shanahan
2008-03-20 10:17 ` Kevin Shanahan
2008-03-20 8:54 ` Peter Zijlstra
2008-03-20 10:04 ` Kevin Shanahan
2008-04-01 2:31 ` Denys Vlasenko
2008-04-01 4:57 ` Peter Zijlstra
2008-04-01 8:17 ` Denys Vlasenko [this message]
2008-04-01 8:38 ` KOSAKI Motohiro
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=200804011017.35352.vda.linux@googlemail.com \
--to=vda.linux@googlemail.com \
--cc=kmshanah@ucwb.org.au \
--cc=linux-kernel@vger.kernel.org \
--cc=peterz@infradead.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.