lkml.org 
[lkml]   [2012]   [Jan]   [9]   [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 Mon, 2012-01-09 at 19:14 -0500, Youquan Song wrote:
> > Yes it is.. also that knob should die! Like i've been saying for way too
> > long. I'm >< close to committing a patch removing all the power_saving
> > magic from the scheduler.
> Sorry. I do not notice it.

I hadn't posted it before, but just about every time people post
something related to smt/mc power balancing I say this needs to get
cleaned up. Since telling people doesn't seem to have any effect what so
ever, stronger measures are needed.

> But currently in many real tests, the knob prove to save power in semi-idle system.
> They are useful in many user scenarios currently.

I'm not saying the stuff is without merit, I'm just saying that a) the
interface is utter crap and b) nobody seems to spend enough time on it
to 1) understand the load-balancer so that 2) he can integrate the power
saving stuff properly.

Instead I get one sporadic band-aid after another.

Also, fundamentally, the split between {smt/mc/numa/book}_power_savings
is completely and fundamentally broken, nobody cares about the actual
topology.


\
 
 \ /
  Last update: 2012-01-09 12:15    [W:0.308 / U:0.184 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site