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

Groups > linux.kernel > #1426474 > unrolled thread

RE: [PATCH 1/1] perf/x86/intel: Add extended event constraints for Knights Landing

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: [PATCH 1/1] perf/x86/intel: Add extended event constraints for  Knights Landing "Odzioba, Lukasz" <lukasz.odzioba@intel.com> - 2016-06-20 12:40 +0200
    Re: [PATCH 1/1] perf/x86/intel: Add extended event constraints for  Knights Landing Peter Zijlstra <peterz@infradead.org> - 2016-06-21 12:40 +0200

#1426474 — RE: [PATCH 1/1] perf/x86/intel: Add extended event constraints for Knights Landing

On 08.06.2016 Peter Zijlstra wrote:
> How does this work in the light of intel_alt_er() ?

Hi Peter,

If the constrained bit is valid on only one of the OCR MSRs (like in case of KNL), 
then OCR valid mask will forbid altering it by the other MSR in intel_alt_er.

If constrained bit is valid on both OCR MSR, so it can be safely altered by 
intel_alt_er, but in this case it should be defined twice - for each MSR i.e:

INTEL_EEVENT_CONSTRAINT(0x01b7, 0, 0x1); // bit0 can be used only on PMC0
INTEL_EEVENT_CONSTRAINT(0x02b7, 0, 0x1); // bit0 can be used only on PMC0

I hope that answers your question. 

Thanks,
Lukas

[toc] | [next] | [standalone]

#1427590

On Mon, Jun 20, 2016 at 10:26:30AM +0000, Odzioba, Lukasz wrote:
> On 08.06.2016 Peter Zijlstra wrote:
> > How does this work in the light of intel_alt_er() ?
> 
> Hi Peter,
> 
> If the constrained bit is valid on only one of the OCR MSRs (like in case of KNL), 
> then OCR valid mask will forbid altering it by the other MSR in intel_alt_er.

Yes, that is the intent, but how is this achieved? I'm not sure I see
how the patch ensure this.

Also, intel_get_event_constraints() has a path where it copies the
constraint, should it not also copy the new field?

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web