On Friday 02 May 2008, Peter Zijlstra wrote:
quoted text
> On Fri, 2008-05-02 at 13:37 +0200, Frans Pop wrote:
> > On Friday 02 May 2008, Peter Zijlstra wrote:
> > > > Ok, that's _WAY_ too big, something went seriously wrong. Thanks!
> > >
> > > In fact, that is exactly 1024 time too large.
> > >
> > > How does this work for you:
> >
> > Changes nothing. Values of bonus_max are unchanged and still skips.
>
> Humm, I saw some bonus_max spikes on my laptop which were fixed by that
> patchlet, ah well, the hunt goes on.

Looks like you've got it after all! With the patch in the correct place 
there are no more skips and I now get:

/proc/5182/task/5182/sched:se.bonus_max  :        39.960966
/proc/5182/task/5263/sched:se.bonus_max  :        39.960966
/proc/5182/task/5267/sched:se.bonus_max  :        39.960966
/proc/5182/task/5268/sched:se.bonus_max  :        39.960966
/proc/5182/task/5269/sched:se.bonus_max  :        39.960966
/proc/5182/task/5270/sched:se.bonus_max  :        19.990234
/proc/5182/task/5661/sched:se.bonus_max  :        39.960966

Note (mostly for Parag) that I had GROUP_SCHED disabled for these tests, but 
for me that did make not any difference with my 2.6.25-rc tests for this 
issue, nor with my tests using sched-devel earlier in the .26 merge window.

Feel free to add my
Tested-by: Frans Pop <elendil@planet.nl>

Cheers,
FJP
--
unsubscribe notice
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/