lkml.org 
[lkml]   [2014]   [Apr]   [14]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [tip:x86/urgent] x86 idle: Repair large-server 50-watt idle-power regression
From
Date
On Tue, 2014-04-08 at 21:43 +0000, Brown, Len wrote:
> Davidlohr,
>
> Thanks for the note.
>
> Ideally (on Linux in general, and on servers, in particular) we strive
> for the performance impact of power saving features to be small enough
> to be considered in "measurement noise".
>
> Your report for 160 core Westmere AIM numbers being hit at 10-25%
> shows 15% measurement noise? But even if true, this looks bad.
>
> Any chance you can re-run, with the following two tweaks,
> one at a time?
>
> I'd be curious if you can wrap the invocation in turbostat -v
> and capture that output to how what states we are seeing
> during the benchmark run.

Hi Len, apologies for the delay, I've been having to use my machine for
other things and haven't gotten time to get around to this yet. I'll get
you the requested information sometime this week.



\
 
 \ /
  Last update: 2014-04-15 05:41    [W:0.261 / U:1.320 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site