lkml.org 
[lkml]   [2012]   [Jan]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH] x86,sched: Fix sched_smt_power_savings totally broken
From
Date
On Tue, 2012-01-10 at 22:21 +0530, Vaidyanathan Srinivasan wrote:
> * Youquan Song <youquan.song@intel.com> [2012-01-10 11:54:26]:
>
> >
> > > Youquan, As far as I know both the
> > > sched_smt_power_savings/sched_mc_power_savings are broken for atleast an
> > > year.
> > >
> > > Have you checked the state of 'sched_mc_power_savings' to see if it is
> > > working or gets addressed with your proposed patches?
> > Thanks Suresh!
> >
> > We have verified that sched_mc_power_savings works well in mainline and
> > -tip tree, only sched_smt_power_savings broken.
>
> Hi Suresh,
>
> My testing also shows that sched_mc works find and only sched_smt
> suffers from the group capacity problem at the core level.
>

No. I just checked and it doesn't work as intended.

On a 2 socket 8-core SMT system, I ran 8 cpu hoggers and just toggled
'sched_mc_power_savings' between 0 and 1 multiple times. Irrespective of
the value, those 8 hoggers are distributed evenly between two sockets -
4 each and doesn't change their positions.

So I can bet it doesn't work.

thanks,
suresh



\
 
 \ /
  Last update: 2012-01-10 20:03    [W:0.070 / U:0.944 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site