2024-06-05 16:10:31

by Liang, Kan

[permalink] [raw]
Subject: [PATCH] perf stat: Fix the hard-coded metrics calculation on the hybrid

From: Kan Liang <[email protected]>

The hard-coded metrics is wrongly calculated on the hybrid machine.

$ perf stat -e cycles,instructions -a sleep 1

Performance counter stats for 'system wide':

18,205,487 cpu_atom/cycles/
9,733,603 cpu_core/cycles/
9,423,111 cpu_atom/instructions/ # 0.52 insn per cycle
4,268,965 cpu_core/instructions/ # 0.23 insn per cycle

The insn per cycle for cpu_core should be 4,268,965 / 9,733,603 = 0.44.

When finding the metric events, the find_stat() doesn't take the PMU
type into account. The cpu_atom/cycles/ is wrongly used to calculate
the IPC of the cpu_core.

Fixes: 0a57b910807a ("perf stat: Use counts rather than saved_value")
Reported-by: "Khalil, Amiri" <[email protected]>
Signed-off-by: Kan Liang <[email protected]>
Cc: [email protected]
---
tools/perf/util/stat-shadow.c | 4 ++++
1 file changed, 4 insertions(+)

diff --git a/tools/perf/util/stat-shadow.c b/tools/perf/util/stat-shadow.c
index 3466aa952442..4d0edc061f1a 100644
--- a/tools/perf/util/stat-shadow.c
+++ b/tools/perf/util/stat-shadow.c
@@ -176,6 +176,10 @@ static double find_stat(const struct evsel *evsel, int aggr_idx, enum stat_type
if (type != evsel__stat_type(cur))
continue;

+ /* Ignore if not the PMU we're looking for. */
+ if (evsel->pmu != cur->pmu)
+ continue;
+
aggr = &cur->stats->aggr[aggr_idx];
if (type == STAT_NSECS)
return aggr->counts.val;
--
2.35.1



2024-06-05 17:21:37

by Ian Rogers

[permalink] [raw]
Subject: Re: [PATCH] perf stat: Fix the hard-coded metrics calculation on the hybrid

On Wed, Jun 5, 2024 at 9:10 AM <[email protected]> wrote:
>
> From: Kan Liang <[email protected]>
>
> The hard-coded metrics is wrongly calculated on the hybrid machine.
>
> $ perf stat -e cycles,instructions -a sleep 1
>
> Performance counter stats for 'system wide':
>
> 18,205,487 cpu_atom/cycles/
> 9,733,603 cpu_core/cycles/
> 9,423,111 cpu_atom/instructions/ # 0.52 insn per cycle
> 4,268,965 cpu_core/instructions/ # 0.23 insn per cycle
>
> The insn per cycle for cpu_core should be 4,268,965 / 9,733,603 = 0.44.
>
> When finding the metric events, the find_stat() doesn't take the PMU
> type into account. The cpu_atom/cycles/ is wrongly used to calculate
> the IPC of the cpu_core.
>
> Fixes: 0a57b910807a ("perf stat: Use counts rather than saved_value")
> Reported-by: "Khalil, Amiri" <[email protected]>
> Signed-off-by: Kan Liang <[email protected]>

Reviewed-by: Ian Rogers <[email protected]>

Thanks,
Ian

> Cc: [email protected]
> ---
> tools/perf/util/stat-shadow.c | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/tools/perf/util/stat-shadow.c b/tools/perf/util/stat-shadow.c
> index 3466aa952442..4d0edc061f1a 100644
> --- a/tools/perf/util/stat-shadow.c
> +++ b/tools/perf/util/stat-shadow.c
> @@ -176,6 +176,10 @@ static double find_stat(const struct evsel *evsel, int aggr_idx, enum stat_type
> if (type != evsel__stat_type(cur))
> continue;
>
> + /* Ignore if not the PMU we're looking for. */
> + if (evsel->pmu != cur->pmu)
> + continue;
> +
> aggr = &cur->stats->aggr[aggr_idx];
> if (type == STAT_NSECS)
> return aggr->counts.val;
> --
> 2.35.1
>

2024-06-06 07:34:40

by Namhyung Kim

[permalink] [raw]
Subject: Re: [PATCH] perf stat: Fix the hard-coded metrics calculation on the hybrid

On Wed, Jun 5, 2024 at 10:21 AM Ian Rogers <[email protected]> wrote:
>
> On Wed, Jun 5, 2024 at 9:10 AM <[email protected]> wrote:
> >
> > From: Kan Liang <[email protected]>
> >
> > The hard-coded metrics is wrongly calculated on the hybrid machine.
> >
> > $ perf stat -e cycles,instructions -a sleep 1
> >
> > Performance counter stats for 'system wide':
> >
> > 18,205,487 cpu_atom/cycles/
> > 9,733,603 cpu_core/cycles/
> > 9,423,111 cpu_atom/instructions/ # 0.52 insn per cycle
> > 4,268,965 cpu_core/instructions/ # 0.23 insn per cycle
> >
> > The insn per cycle for cpu_core should be 4,268,965 / 9,733,603 = 0.44.
> >
> > When finding the metric events, the find_stat() doesn't take the PMU
> > type into account. The cpu_atom/cycles/ is wrongly used to calculate
> > the IPC of the cpu_core.
> >
> > Fixes: 0a57b910807a ("perf stat: Use counts rather than saved_value")
> > Reported-by: "Khalil, Amiri" <[email protected]>
> > Signed-off-by: Kan Liang <[email protected]>
>
> Reviewed-by: Ian Rogers <[email protected]>
>
> Thanks,
> Ian
>
> > Cc: [email protected]
> > ---
> > tools/perf/util/stat-shadow.c | 4 ++++
> > 1 file changed, 4 insertions(+)
> >
> > diff --git a/tools/perf/util/stat-shadow.c b/tools/perf/util/stat-shadow.c
> > index 3466aa952442..4d0edc061f1a 100644
> > --- a/tools/perf/util/stat-shadow.c
> > +++ b/tools/perf/util/stat-shadow.c
> > @@ -176,6 +176,10 @@ static double find_stat(const struct evsel *evsel, int aggr_idx, enum stat_type
> > if (type != evsel__stat_type(cur))
> > continue;
> >
> > + /* Ignore if not the PMU we're looking for. */
> > + if (evsel->pmu != cur->pmu)
> > + continue;

Hmm.. Don't some metrics need events from different PMU?
Like cycles per sec or branch instructions per sec..

Thanks,
Namhyung


> > +
> > aggr = &cur->stats->aggr[aggr_idx];
> > if (type == STAT_NSECS)
> > return aggr->counts.val;
> > --
> > 2.35.1
> >

2024-06-06 13:44:51

by Liang, Kan

[permalink] [raw]
Subject: Re: [PATCH] perf stat: Fix the hard-coded metrics calculation on the hybrid



On 2024-06-06 3:34 a.m., Namhyung Kim wrote:
> On Wed, Jun 5, 2024 at 10:21 AM Ian Rogers <[email protected]> wrote:
>>
>> On Wed, Jun 5, 2024 at 9:10 AM <[email protected]> wrote:
>>>
>>> From: Kan Liang <[email protected]>
>>>
>>> The hard-coded metrics is wrongly calculated on the hybrid machine.
>>>
>>> $ perf stat -e cycles,instructions -a sleep 1
>>>
>>> Performance counter stats for 'system wide':
>>>
>>> 18,205,487 cpu_atom/cycles/
>>> 9,733,603 cpu_core/cycles/
>>> 9,423,111 cpu_atom/instructions/ # 0.52 insn per cycle
>>> 4,268,965 cpu_core/instructions/ # 0.23 insn per cycle
>>>
>>> The insn per cycle for cpu_core should be 4,268,965 / 9,733,603 = 0.44.
>>>
>>> When finding the metric events, the find_stat() doesn't take the PMU
>>> type into account. The cpu_atom/cycles/ is wrongly used to calculate
>>> the IPC of the cpu_core.
>>>
>>> Fixes: 0a57b910807a ("perf stat: Use counts rather than saved_value")
>>> Reported-by: "Khalil, Amiri" <[email protected]>
>>> Signed-off-by: Kan Liang <[email protected]>
>>
>> Reviewed-by: Ian Rogers <[email protected]>
>>
>> Thanks,
>> Ian
>>
>>> Cc: [email protected]
>>> ---
>>> tools/perf/util/stat-shadow.c | 4 ++++
>>> 1 file changed, 4 insertions(+)
>>>
>>> diff --git a/tools/perf/util/stat-shadow.c b/tools/perf/util/stat-shadow.c
>>> index 3466aa952442..4d0edc061f1a 100644
>>> --- a/tools/perf/util/stat-shadow.c
>>> +++ b/tools/perf/util/stat-shadow.c
>>> @@ -176,6 +176,10 @@ static double find_stat(const struct evsel *evsel, int aggr_idx, enum stat_type
>>> if (type != evsel__stat_type(cur))
>>> continue;
>>>
>>> + /* Ignore if not the PMU we're looking for. */
>>> + if (evsel->pmu != cur->pmu)
>>> + continue;
>
> Hmm.. Don't some metrics need events from different PMU?
> Like cycles per sec or branch instructions per sec..
>

Right.

In the hard-coded metrics, the events from a different PMU are
SW_CPU_CLOCK and SW_TASK_CLOCK. They both have the stat type,
STAT_NSECS. Perf should ignore the PMU checking for the type as below.
I will send a V2 to fix it.

diff --git a/tools/perf/util/stat-shadow.c b/tools/perf/util/stat-shadow.c
index 3466aa952442..d01335f18808 100644
--- a/tools/perf/util/stat-shadow.c
+++ b/tools/perf/util/stat-shadow.c
@@ -176,6 +176,9 @@ static double find_stat(const struct evsel *evsel,
int aggr_idx, enum stat_type
if (type != evsel__stat_type(cur))
continue;

+ if ((type != STAT_NSECS) && (evsel->pmu != cur->pmu))
+ continue;
+
aggr = &cur->stats->aggr[aggr_idx];
if (type == STAT_NSECS)
return aggr->counts.val;


Thanks,
Kan

> Thanks,
> Namhyung
>
>
>>> +
>>> aggr = &cur->stats->aggr[aggr_idx];
>>> if (type == STAT_NSECS)
>>> return aggr->counts.val;
>>> --
>>> 2.35.1
>>>
>