Re: [PATCH for testing] Re: Decreasing stime running confuses top

[Frans Pop <elendil@planet.nl>]

On Friday 05 October 2007, Chuck Ebbert wrote:
> procfs: Don't read runtime twice when computing task's stime
>
> Current code reads p->se.sum_exec_runtime twice and goes through
> multiple type conversions to calculate stime. Read it once and
> skip some of the conversions.
>
> Signed-off-by: Chuck Ebbert <cebbert@redhat.com>

This second patch is a major improvement. But both for kontact and amarok I 
still see stime decreasing occasionally. Sometimes even still quite 
frequently like in this series:

Oct 05 17:00:42 698 178
Oct 05 17:00:43 700 177
Oct 05 17:00:44 700 177
Oct 05 17:00:45 700 177
Oct 05 17:00:46 700 178
Oct 05 17:00:47 700 178
Oct 05 17:00:48 700 177
Oct 05 17:00:49 700 177
Oct 05 17:00:50 700 178
Oct 05 17:00:51 700 178
Oct 05 17:00:52 700 179
Oct 05 17:00:53 698 180
Oct 05 17:00:54 700 179
Oct 05 17:00:55 700 179
Oct 05 17:00:56 700 179
Oct 05 17:00:57 700 179
Oct 05 17:00:58 700 180
Oct 05 17:00:59 700 179

This was after 6 minutes of steady increases.
Should we try a debug patch that shows the raw data behind the calculations?

I'll give the first patch a try next.
-
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/