> On Fri, Aug 28, 2015 at 8:00 AM, Liang, Kan wrote:
> >
> >
> >> On Tue, Aug 25, 2015 at 1:15 PM, Liang, Kan
> wrote:
> >> >
> >> >> >> >
> >> >> >> I understand that these metrics are useful and needed however
> >> >> >> if I look at the broader picture I see many PMUs doing similar
> >> >> >
On Fri, Aug 28, 2015 at 8:00 AM, Liang, Kan wrote:
>
>
>> On Tue, Aug 25, 2015 at 1:15 PM, Liang, Kan wrote:
>> >
>> >> >> >
>> >> >> I understand that these metrics are useful and needed however if I
>> >> >> look at the broader picture I see many PMUs doing similar things
>> >> >> or appearing
> On Tue, Aug 25, 2015 at 1:15 PM, Liang, Kan wrote:
> >
> >> >> >
> >> >> I understand that these metrics are useful and needed however if I
> >> >> look at the broader picture I see many PMUs doing similar things
> >> >> or appearing different when they are actually very close. It would
> >> >
On Tue, Aug 25, 2015 at 1:15 PM, Liang, Kan wrote:
>
>> >> >
>> >> I understand that these metrics are useful and needed however if I
>> >> look at the broader picture I see many PMUs doing similar things or
>> >> appearing different when they are actually very close. It would be
>> >> nice to hav
> >> >
> >> I understand that these metrics are useful and needed however if I
> >> look at the broader picture I see many PMUs doing similar things or
> >> appearing different when they are actually very close. It would be
> >> nice to have a more unified approach. You have RAPL (client, server)
> >> On Thu, Aug 6, 2015 at 1:25 PM, Liang, Kan wrote:
> >> >
> >> >> >> >> >> +static cpumask_t power_cstate_core_cpu_mask;
> >> >> >> >> >
> >> >> >> >> > That one typically does not need a cpumask.
> >> >> >> >> >
> >> >> >> >> You need to pick one CPU out of the multi-core. But it is
> >> >>
Melo;
>> a...@linux.intel.com; LKML
>> Subject: Re: [PATCH V2 1/1] perf/x86: Add Intel power cstate PMUs
>> support
>>
>> On Thu, Aug 6, 2015 at 1:25 PM, Liang, Kan wrote:
>> >
>> >> >> >> >> +static cpu
> -Original Message-
> From: Stephane Eranian [mailto:eran...@google.com]
> Sent: Thursday, August 06, 2015 4:38 PM
> To: Liang, Kan
> Cc: Peter Zijlstra; mi...@redhat.com; Arnaldo Carvalho de Melo;
> a...@linux.intel.com; LKML
> Subject: Re: [PATCH V2 1/1] perf
On Thu, Aug 06, 2015 at 11:14:17AM -0700, Stephane Eranian wrote:
> On Thu, Aug 6, 2015 at 8:44 AM, Peter Zijlstra wrote:
> >> +static cpumask_t power_cstate_core_cpu_mask;
> >
> > That one typically does not need a cpumask.
> >
> You need to pick one CPU out of the multi-core. But it is for clien
On Thu, Aug 6, 2015 at 1:25 PM, Liang, Kan wrote:
>
>> >> >> >> +static cpumask_t power_cstate_core_cpu_mask;
>> >> >> >
>> >> >> > That one typically does not need a cpumask.
>> >> >> >
>> >> >> You need to pick one CPU out of the multi-core. But it is for
>> >> >> client parts thus there is only
> >> >> >> +static cpumask_t power_cstate_core_cpu_mask;
> >> >> >
> >> >> > That one typically does not need a cpumask.
> >> >> >
> >> >> You need to pick one CPU out of the multi-core. But it is for
> >> >> client parts thus there is only one socket. At least this is my
> understanding.
> >> >>
On Thu, Aug 6, 2015 at 12:25 PM, Liang, Kan wrote:
>
>>
>> On Thu, Aug 6, 2015 at 11:52 AM, Liang, Kan wrote:
>> >
>> >
>> >> >> +static cpumask_t power_cstate_core_cpu_mask;
>> >> >
>> >> > That one typically does not need a cpumask.
>> >> >
>> >> You need to pick one CPU out of the multi-core.
>
> On Thu, Aug 6, 2015 at 11:52 AM, Liang, Kan wrote:
> >
> >
> >> >> +static cpumask_t power_cstate_core_cpu_mask;
> >> >
> >> > That one typically does not need a cpumask.
> >> >
> >> You need to pick one CPU out of the multi-core. But it is for client
> >> parts thus there is only one socket
On Thu, Aug 6, 2015 at 11:52 AM, Liang, Kan wrote:
>
>
>> >> +static cpumask_t power_cstate_core_cpu_mask;
>> >
>> > That one typically does not need a cpumask.
>> >
>> You need to pick one CPU out of the multi-core. But it is for client parts
>> thus there is only one socket. At least this is my
> >> +static cpumask_t power_cstate_core_cpu_mask;
> >
> > That one typically does not need a cpumask.
> >
> You need to pick one CPU out of the multi-core. But it is for client parts
> thus there is only one socket. At least this is my understanding.
>
CORE_C*_RESIDENCY are available for physi
On Thu, Aug 6, 2015 at 8:44 AM, Peter Zijlstra wrote:
> On Mon, Jul 27, 2015 at 04:46:16AM -0400, kan.li...@intel.com wrote:
>
> As a general comment; this thing is unreadable. Far too much macro foo
> to instantiate the different PMUs.
>
>> +struct perf_power_cstate_event_msr {
>> + int i
On Mon, Jul 27, 2015 at 04:46:16AM -0400, kan.li...@intel.com wrote:
As a general comment; this thing is unreadable. Far too much macro foo
to instantiate the different PMUs.
> +struct perf_power_cstate_event_msr {
> + int id;
> + u64 msr;
> +};
> +
> +enum perf_power_cstate_id {
From: Kan Liang
This patch adds new PMUs to support power cstate related free running
(read-only) counters. These counters may be used simultaneously by other
tools, such as turbostat. However, it still make sense to implement them
in perf. Because we can conveniently collect them together with o
18 matches
Mail list logo
