lkml.org 
[lkml]   [2005]   [Aug]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: cache regresions with 2.6.1x ?
jerome lacoste <jerome.lacoste@gmail.com> wrote:
>
> On 8/23/05, Andrew Morton <akpm@osdl.org> wrote:
> > jerome lacoste <jerome.lacoste@gmail.com> wrote:
> > >
> > > I am on a Dell Inspiron 8100 laptop with 512 M and 1G disk cache. I
> > > usually have at least 4 big applications running simultaneously: a
> > > Java IDE, firefox, firefox and X. All that under the Gnome desktop.
> > >
> > > I've sometimes seen periods where my laptop goes kind of nuts. While
> > > the cpu is still at 0%, the workload goes to 100% (as shown in the
> > > gnome process monitor) (I haven't checked in other means, e.g. top or
> > > /proc info as my machine is unusable).
> > >
> > > But with my latest upgrade to 2.6.12 from 2.6.10, the hanging happens
> > > much more often. It lasts for over 30 seconds.
> > >
> > > Could this hanging be related to swapping?
> > > Are there any VM regression lately that would make a kernel less
> > > appropriate for desktop use?
> > > How can I investigate that further?
> >
> > 10-20 lines of `vmstat 1' output while it's happening would help.
>
> Here it goes. Maybe just some bad swapping?

Maybe. There's certainly a ton of swapping happening.

> jerome@expresso> vmstat 1
> procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
> r b swpd free buff cache si so bi bo in cs us sy id wa
> 1 7 588164 7424 18612 106908 13 7 34 44 10 12 8 2 85 5
> 2 4 587996 6152 18624 108092 404 664 540 2892 1201 2631 70 9 0 21
> 0 12 588276 5160 18620 109188 664 1244 860 1244 1195 615 46 5 0 50
> 0 13 588140 4912 18628 109188 216 0 216 8 1156 245 0 0 0 100
> 0 17 588536 4892 18628 109972 132 576 132 576 1172 353 32 4 0 64
> 0 16 589096 5016 18628 110192 0 608 4 628 1169 247 7 2 0 91
> 0 16 589780 5636 18632 110136 0 716 0 808 1181 261 1 0 0 99

But maybe a memory leak. Can you take a copy of /proc/meminfo and
/proc/slabinfo when this is happening?

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-08-26 03:32    [W:4.346 / U:0.016 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site