lkml.org 
[lkml]   [2013]   [Aug]   [1]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 6/8] perf: Account freq events per cpu
On Thu, Aug 01, 2013 at 04:06:15PM +0200, Peter Zijlstra wrote:
> On Thu, Aug 01, 2013 at 04:03:52PM +0200, Peter Zijlstra wrote:
> > On Thu, Aug 01, 2013 at 03:55:27PM +0200, Frederic Weisbecker wrote:
> > > On Thu, Aug 01, 2013 at 03:31:55PM +0200, Peter Zijlstra wrote:
> > > >
> > > > Where the freq thing is new and shiney, but we already had the other
> > > > two. Of those, cgroup events must be per cpu so that should be good,
> > > > the branch_stack thing tests ATTACH_TASK, which should also be good, but
> > > > leaves me wonder wth they do for those that are attached to tasks.
> > > >
> > > > But yes, the frequency thing is borken.
> > >
> > > Aie, so the freq thing, I can either account to all CPUs (inc to all
> > > and send an IPI to all), or when the event scheds in/out. Probably we
> > > should do the former to avoid sending an IPI at all context switches.
> >
> > Yeah, just go with a single global state for now..
>
> The perf default is to create inherited counter, which are per cpu
> anyway. So we'll not loose much.

So you mean that I keep the per cpu state when event->cpu != -1 and also have
a global counter for the others. Right?


\
 
 \ /
  Last update: 2013-08-01 17:21    [W:0.076 / U:0.156 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site