Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1450539 > unrolled thread

Re: Strange behavior of perf top with PEBS

Started byNikolay Borisov <kernel@kyup.com>
First post2016-07-26 13:40 +0200
Last post2016-08-08 15:20 +0200
Articles 6 — 4 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: Strange behavior of perf top with PEBS Nikolay Borisov <kernel@kyup.com> - 2016-07-26 13:40 +0200
    Re: Strange behavior of perf top with PEBS Jiri Olsa <jolsa@redhat.com> - 2016-08-04 17:40 +0200
      Re: Strange behavior of perf top with PEBS Nikolay Borisov <kernel@kyup.com> - 2016-08-05 11:40 +0200
        Re: Strange behavior of perf top with PEBS Jiri Olsa <jolsa@redhat.com> - 2016-08-05 12:30 +0200
          Re: Strange behavior of perf top with PEBS Peter Zijlstra <peterz@infradead.org> - 2016-08-08 15:10 +0200
            Re: Strange behavior of perf top with PEBS Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-08-08 15:20 +0200

#1450539 — Re: Strange behavior of perf top with PEBS

FromNikolay Borisov <kernel@kyup.com>
Date2016-07-26 13:40 +0200
SubjectRe: Strange behavior of perf top with PEBS
Message-ID<rZ8Sm-1mU-5@gated-at.bofh.it>

On 07/20/2016 05:38 PM, Jiri Olsa wrote:
> On Wed, Jul 20, 2016 at 04:34:17PM +0200, Jiri Olsa wrote:
>> On Wed, Jul 20, 2016 at 04:28:34PM +0300, Nikolay Borisov wrote:
>>> Hello,
>>>
>>> Running perf version 4.4.14.g0cb188d (no modification to the PMU/perf
>>> code) I observed that "perf top" counts no cycles and produces no
>>> output. After a bit of head scratching and testing I figured that
>>> running "perf top -e cycles" actually works whereas the default option
>>> is equivalent to running "perf top -e cycles:p". So the latter version
>>> seems to not work on my machine.
>>
>> hum, I think Core2 has PEBs valid only for instructions not cycles..
>>
>> I'll check why perf top forcing the precise for cycles
>> I thought we had that automated already
> 
> oops, too soon ;) we have:
> 
> perf/x86/intel: Fix Core2,Atom,NHM,WSM cycles:pp events
> commit 517e6341fa123ec3a2f9ea78ad547be910529881
> Author: Peter Zijlstra <peterz@infradead.org>
> Date:   Sat Apr 11 12:16:22 2015 +0200
> 
> 
> so i guess it should work.. checking ;-)

Any update on that?

> 
> jirka
>

[toc] | [next] | [standalone]

#1456496

FromJiri Olsa <jolsa@redhat.com>
Date2016-08-04 17:40 +0200
Message-ID<s2sUy-2Lw-31@gated-at.bofh.it>
In reply to#1450539
On Tue, Jul 26, 2016 at 02:30:46PM +0300, Nikolay Borisov wrote:
> 
> 
> On 07/20/2016 05:38 PM, Jiri Olsa wrote:
> > On Wed, Jul 20, 2016 at 04:34:17PM +0200, Jiri Olsa wrote:
> >> On Wed, Jul 20, 2016 at 04:28:34PM +0300, Nikolay Borisov wrote:
> >>> Hello,
> >>>
> >>> Running perf version 4.4.14.g0cb188d (no modification to the PMU/perf
> >>> code) I observed that "perf top" counts no cycles and produces no
> >>> output. After a bit of head scratching and testing I figured that
> >>> running "perf top -e cycles" actually works whereas the default option
> >>> is equivalent to running "perf top -e cycles:p". So the latter version
> >>> seems to not work on my machine.
> >>
> >> hum, I think Core2 has PEBs valid only for instructions not cycles..
> >>
> >> I'll check why perf top forcing the precise for cycles
> >> I thought we had that automated already
> > 
> > oops, too soon ;) we have:
> > 
> > perf/x86/intel: Fix Core2,Atom,NHM,WSM cycles:pp events
> > commit 517e6341fa123ec3a2f9ea78ad547be910529881
> > Author: Peter Zijlstra <peterz@infradead.org>
> > Date:   Sat Apr 11 12:16:22 2015 +0200
> > 
> > 
> > so i guess it should work.. checking ;-)
> 
> Any update on that?

sorry for late response..

I checked on f22 kernel and it's missing the core2 PEBs fix:
  1424a09a9e18 perf/x86: fix PEBS issues on Intel Atom/Core2

which was introduced in 4.5.. you should upgrade or include
this patch

jirka

[toc] | [prev] | [next] | [standalone]

#1456985

FromNikolay Borisov <kernel@kyup.com>
Date2016-08-05 11:40 +0200
Message-ID<s2JLH-5EJ-9@gated-at.bofh.it>
In reply to#1456496

On 08/04/2016 06:29 PM, Jiri Olsa wrote:
> On Tue, Jul 26, 2016 at 02:30:46PM +0300, Nikolay Borisov wrote:
[SNIP]
> 
> sorry for late response..
> 
> I checked on f22 kernel and it's missing the core2 PEBs fix:
>   1424a09a9e18 perf/x86: fix PEBS issues on Intel Atom/Core2
> 
> which was introduced in 4.5.. you should upgrade or include
> this patch
> 

Thanks for the reply. Isn't this supposed to be tagged as stable since
it essentially breaks the default invocation of perf top?


> jirka
>

[toc] | [prev] | [next] | [standalone]

#1457007

FromJiri Olsa <jolsa@redhat.com>
Date2016-08-05 12:30 +0200
Message-ID<s2Ky6-6dg-13@gated-at.bofh.it>
In reply to#1456985
On Fri, Aug 05, 2016 at 12:30:32PM +0300, Nikolay Borisov wrote:
> 
> 
> On 08/04/2016 06:29 PM, Jiri Olsa wrote:
> > On Tue, Jul 26, 2016 at 02:30:46PM +0300, Nikolay Borisov wrote:
> [SNIP]
> > 
> > sorry for late response..
> > 
> > I checked on f22 kernel and it's missing the core2 PEBs fix:
> >   1424a09a9e18 perf/x86: fix PEBS issues on Intel Atom/Core2
> > 
> > which was introduced in 4.5.. you should upgrade or include
> > this patch
> > 
> 
> Thanks for the reply. Isn't this supposed to be tagged as stable since
> it essentially breaks the default invocation of perf top?

yep, it seems like good one to have in stable, Peter?

thanks,
jirka

[toc] | [prev] | [next] | [standalone]

#1457769

FromPeter Zijlstra <peterz@infradead.org>
Date2016-08-08 15:10 +0200
Message-ID<s3StA-1Oa-19@gated-at.bofh.it>
In reply to#1457007
On Fri, Aug 05, 2016 at 12:23:20PM +0200, Jiri Olsa wrote:
> On Fri, Aug 05, 2016 at 12:30:32PM +0300, Nikolay Borisov wrote:
> > 
> > 
> > On 08/04/2016 06:29 PM, Jiri Olsa wrote:
> > > On Tue, Jul 26, 2016 at 02:30:46PM +0300, Nikolay Borisov wrote:
> > [SNIP]
> > > 
> > > sorry for late response..
> > > 
> > > I checked on f22 kernel and it's missing the core2 PEBs fix:
> > >   1424a09a9e18 perf/x86: fix PEBS issues on Intel Atom/Core2
> > > 
> > > which was introduced in 4.5.. you should upgrade or include
> > > this patch
> > > 
> > 
> > Thanks for the reply. Isn't this supposed to be tagged as stable since
> > it essentially breaks the default invocation of perf top?
> 
> yep, it seems like good one to have in stable, Peter?

Greg, could you pick:

  1424a09a9e18 ("perf/x86: fix PEBS issues on Intel Atom/Core2")

Into 4.4-stable ?

[toc] | [prev] | [next] | [standalone]

#1457782

FromGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Date2016-08-08 15:20 +0200
Message-ID<s3SDf-1Rl-15@gated-at.bofh.it>
In reply to#1457769
On Mon, Aug 08, 2016 at 03:03:43PM +0200, Peter Zijlstra wrote:
> On Fri, Aug 05, 2016 at 12:23:20PM +0200, Jiri Olsa wrote:
> > On Fri, Aug 05, 2016 at 12:30:32PM +0300, Nikolay Borisov wrote:
> > > 
> > > 
> > > On 08/04/2016 06:29 PM, Jiri Olsa wrote:
> > > > On Tue, Jul 26, 2016 at 02:30:46PM +0300, Nikolay Borisov wrote:
> > > [SNIP]
> > > > 
> > > > sorry for late response..
> > > > 
> > > > I checked on f22 kernel and it's missing the core2 PEBs fix:
> > > >   1424a09a9e18 perf/x86: fix PEBS issues on Intel Atom/Core2
> > > > 
> > > > which was introduced in 4.5.. you should upgrade or include
> > > > this patch
> > > > 
> > > 
> > > Thanks for the reply. Isn't this supposed to be tagged as stable since
> > > it essentially breaks the default invocation of perf top?
> > 
> > yep, it seems like good one to have in stable, Peter?
> 
> Greg, could you pick:
> 
>   1424a09a9e18 ("perf/x86: fix PEBS issues on Intel Atom/Core2")
> 
> Into 4.4-stable ?

Now queued up, thanks.

greg k-h

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web