* [PATCH v2] perf stat: Improve runtime stat for interval mode
@ 2020-04-20 14:54 Jin Yao
2020-04-21 7:10 ` kajoljain
` (3 more replies)
0 siblings, 4 replies; 8+ messages in thread
From: Jin Yao @ 2020-04-20 14:54 UTC (permalink / raw)
To: acme, jolsa, peterz, mingo, alexander.shishkin
Cc: Linux-kernel, ak, kan.liang, yao.jin, Jin Yao
For interval mode, the metric is printed after # if it exists. But
it's not calculated by the counts generated in this interval. See
following examples,
root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
# time counts unit events
1.000422803 764,809 inst_retired.any # 2.9 CPI
1.000422803 2,234,932 cycles
2.001464585 1,960,061 inst_retired.any # 1.6 CPI
2.001464585 4,022,591 cycles
The second CPI should not be 1.6 (4,022,591/1,960,061 is 2.1)
root@kbl-ppc:~# perf stat -e cycles,instructions -I1000 --interval-count 2
# time counts unit events
1.000429493 2,869,311 cycles
1.000429493 816,875 instructions # 0.28 insn per cycle
2.001516426 9,260,973 cycles
2.001516426 5,250,634 instructions # 0.87 insn per cycle
The second 'insn per cycle' should not be 0.87 (5,250,634/9,260,973 is 0.57).
The current code uses a global variable rt_stat for tracking and
updating the std dev of runtime stat. Unlike the counts, rt_stat is
not reset for interval. While the counts are reset for interval.
perf_stat_process_counter()
{
if (config->interval)
init_stats(ps->res_stats);
}
So for interval, the rt_stat should be reset either.
This patch resets rt_stat before read_counters, so the runtime
stat is only calculated by the counts generated in this interval.
With this patch,
root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
# time counts unit events
1.000420924 2,408,818 inst_retired.any # 2.1 CPI
1.000420924 5,010,111 cycles
2.001448579 2,798,407 inst_retired.any # 1.6 CPI
2.001448579 4,599,861 cycles
root@kbl-ppc:~# perf stat -e cycles,instructions -I1000 --interval-count 2
# time counts unit events
1.000428555 2,769,714 cycles
1.000428555 774,462 instructions # 0.28 insn per cycle
2.001471562 3,595,904 cycles
2.001471562 1,243,703 instructions # 0.35 insn per cycle
Now the second 'insn per cycle' and CPI are calculated by the counts
generated in this interval.
v2:
---
Use just existing perf_stat__reset_shadow_per_stat(&rt_stat).
We don't need to define new function perf_stat__reset_rt_stat.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
---
tools/perf/Documentation/perf-stat.txt | 2 ++
tools/perf/builtin-stat.c | 1 +
2 files changed, 3 insertions(+)
diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
index 4d56586b2fb9..3fb5028aef08 100644
--- a/tools/perf/Documentation/perf-stat.txt
+++ b/tools/perf/Documentation/perf-stat.txt
@@ -176,6 +176,8 @@ Print count deltas every N milliseconds (minimum: 1ms)
The overhead percentage could be high in some cases, for instance with small, sub 100ms intervals. Use with caution.
example: 'perf stat -I 1000 -e cycles -a sleep 5'
+If the metric exists, it is calculated by the counts generated in this interval and the metric is printed after #.
+
--interval-count times::
Print count deltas for fixed number of times.
This option should be used together with "-I" option.
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 9207b6c45475..3f050d85c277 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -359,6 +359,7 @@ static void process_interval(void)
clock_gettime(CLOCK_MONOTONIC, &ts);
diff_timespec(&rs, &ts, &ref_time);
+ perf_stat__reset_shadow_per_stat(&rt_stat);
read_counters(&rs);
if (STAT_RECORD) {
--
2.17.1
^ permalink raw reply related [flat|nested] 8+ messages in thread
* Re: [PATCH v2] perf stat: Improve runtime stat for interval mode
2020-04-20 14:54 [PATCH v2] perf stat: Improve runtime stat for interval mode Jin Yao
@ 2020-04-21 7:10 ` kajoljain
2020-04-21 13:53 ` Arnaldo Carvalho de Melo
` (2 subsequent siblings)
3 siblings, 0 replies; 8+ messages in thread
From: kajoljain @ 2020-04-21 7:10 UTC (permalink / raw)
To: Jin Yao, acme, jolsa, peterz, mingo, alexander.shishkin
Cc: Linux-kernel, ak, kan.liang, yao.jin
On 4/20/20 8:24 PM, Jin Yao wrote:
> For interval mode, the metric is printed after # if it exists. But
> it's not calculated by the counts generated in this interval. See
> following examples,
>
> root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
> # time counts unit events
> 1.000422803 764,809 inst_retired.any # 2.9 CPI
> 1.000422803 2,234,932 cycles
> 2.001464585 1,960,061 inst_retired.any # 1.6 CPI
> 2.001464585 4,022,591 cycles
>
> The second CPI should not be 1.6 (4,022,591/1,960,061 is 2.1)
>
> root@kbl-ppc:~# perf stat -e cycles,instructions -I1000 --interval-count 2
> # time counts unit events
> 1.000429493 2,869,311 cycles
> 1.000429493 816,875 instructions # 0.28 insn per cycle
> 2.001516426 9,260,973 cycles
> 2.001516426 5,250,634 instructions # 0.87 insn per cycle
>
> The second 'insn per cycle' should not be 0.87 (5,250,634/9,260,973 is 0.57).
>
> The current code uses a global variable rt_stat for tracking and
> updating the std dev of runtime stat. Unlike the counts, rt_stat is
> not reset for interval. While the counts are reset for interval.
>
> perf_stat_process_counter()
> {
> if (config->interval)
> init_stats(ps->res_stats);
> }
>
> So for interval, the rt_stat should be reset either.
>
> This patch resets rt_stat before read_counters, so the runtime
> stat is only calculated by the counts generated in this interval.
>
> With this patch,
>
> root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
> # time counts unit events
> 1.000420924 2,408,818 inst_retired.any # 2.1 CPI
> 1.000420924 5,010,111 cycles
> 2.001448579 2,798,407 inst_retired.any # 1.6 CPI
> 2.001448579 4,599,861 cycles
>
> root@kbl-ppc:~# perf stat -e cycles,instructions -I1000 --interval-count 2
> # time counts unit events
> 1.000428555 2,769,714 cycles
> 1.000428555 774,462 instructions # 0.28 insn per cycle
> 2.001471562 3,595,904 cycles
> 2.001471562 1,243,703 instructions # 0.35 insn per cycle
>
> Now the second 'insn per cycle' and CPI are calculated by the counts
> generated in this interval.
>
> v2:
> ---
> Use just existing perf_stat__reset_shadow_per_stat(&rt_stat).
> We don't need to define new function perf_stat__reset_rt_stat.
>
> Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
> ---
> tools/perf/Documentation/perf-stat.txt | 2 ++
> tools/perf/builtin-stat.c | 1 +
> 2 files changed, 3 insertions(+)
>
> diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
> index 4d56586b2fb9..3fb5028aef08 100644
> --- a/tools/perf/Documentation/perf-stat.txt
> +++ b/tools/perf/Documentation/perf-stat.txt
> @@ -176,6 +176,8 @@ Print count deltas every N milliseconds (minimum: 1ms)
> The overhead percentage could be high in some cases, for instance with small, sub 100ms intervals. Use with caution.
> example: 'perf stat -I 1000 -e cycles -a sleep 5'
>
> +If the metric exists, it is calculated by the counts generated in this interval and the metric is printed after #.
> +
> --interval-count times::
> Print count deltas for fixed number of times.
> This option should be used together with "-I" option.
> diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
> index 9207b6c45475..3f050d85c277 100644
> --- a/tools/perf/builtin-stat.c
> +++ b/tools/perf/builtin-stat.c
> @@ -359,6 +359,7 @@ static void process_interval(void)
> clock_gettime(CLOCK_MONOTONIC, &ts);
> diff_timespec(&rs, &ts, &ref_time);
>
> + perf_stat__reset_shadow_per_stat(&rt_stat);
Tested-By: Kajol Jain <kjain@linux.ibm.com>
Thanks,
Kajol
> read_counters(&rs);
>
> if (STAT_RECORD) {
>
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH v2] perf stat: Improve runtime stat for interval mode
2020-04-20 14:54 [PATCH v2] perf stat: Improve runtime stat for interval mode Jin Yao
2020-04-21 7:10 ` kajoljain
@ 2020-04-21 13:53 ` Arnaldo Carvalho de Melo
2020-04-22 0:53 ` Jin, Yao
2020-04-22 7:22 ` Jiri Olsa
2020-05-08 13:05 ` [tip: perf/core] " tip-bot2 for Jin Yao
3 siblings, 1 reply; 8+ messages in thread
From: Arnaldo Carvalho de Melo @ 2020-04-21 13:53 UTC (permalink / raw)
To: Jin Yao
Cc: jolsa, peterz, mingo, alexander.shishkin, Linux-kernel, ak,
kan.liang, yao.jin
Em Mon, Apr 20, 2020 at 10:54:17PM +0800, Jin Yao escreveu:
> For interval mode, the metric is printed after # if it exists. But
> it's not calculated by the counts generated in this interval. See
> following examples,
>
> root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
> # time counts unit events
> 1.000422803 764,809 inst_retired.any # 2.9 CPI
> 1.000422803 2,234,932 cycles
> 2.001464585 1,960,061 inst_retired.any # 1.6 CPI
> 2.001464585 4,022,591 cycles
>
> The second CPI should not be 1.6 (4,022,591/1,960,061 is 2.1)
>
> root@kbl-ppc:~# perf stat -e cycles,instructions -I1000 --interval-count 2
> # time counts unit events
> 1.000429493 2,869,311 cycles
> 1.000429493 816,875 instructions # 0.28 insn per cycle
> 2.001516426 9,260,973 cycles
> 2.001516426 5,250,634 instructions # 0.87 insn per cycle
>
> The second 'insn per cycle' should not be 0.87 (5,250,634/9,260,973 is 0.57).
>
> The current code uses a global variable rt_stat for tracking and
> updating the std dev of runtime stat. Unlike the counts, rt_stat is
> not reset for interval. While the counts are reset for interval.
>
> perf_stat_process_counter()
> {
> if (config->interval)
> init_stats(ps->res_stats);
> }
>
> So for interval, the rt_stat should be reset either.
s/either/too/g right?
And please try and find what was the cset that introduced the problem,
so that we can have a Fixes: line and the stable series can pick it, ok?
- Arnaldo
> This patch resets rt_stat before read_counters, so the runtime
> stat is only calculated by the counts generated in this interval.
>
> With this patch,
>
> root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
> # time counts unit events
> 1.000420924 2,408,818 inst_retired.any # 2.1 CPI
> 1.000420924 5,010,111 cycles
> 2.001448579 2,798,407 inst_retired.any # 1.6 CPI
> 2.001448579 4,599,861 cycles
>
> root@kbl-ppc:~# perf stat -e cycles,instructions -I1000 --interval-count 2
> # time counts unit events
> 1.000428555 2,769,714 cycles
> 1.000428555 774,462 instructions # 0.28 insn per cycle
> 2.001471562 3,595,904 cycles
> 2.001471562 1,243,703 instructions # 0.35 insn per cycle
>
> Now the second 'insn per cycle' and CPI are calculated by the counts
> generated in this interval.
>
> v2:
> ---
> Use just existing perf_stat__reset_shadow_per_stat(&rt_stat).
> We don't need to define new function perf_stat__reset_rt_stat.
>
> Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
> ---
> tools/perf/Documentation/perf-stat.txt | 2 ++
> tools/perf/builtin-stat.c | 1 +
> 2 files changed, 3 insertions(+)
>
> diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
> index 4d56586b2fb9..3fb5028aef08 100644
> --- a/tools/perf/Documentation/perf-stat.txt
> +++ b/tools/perf/Documentation/perf-stat.txt
> @@ -176,6 +176,8 @@ Print count deltas every N milliseconds (minimum: 1ms)
> The overhead percentage could be high in some cases, for instance with small, sub 100ms intervals. Use with caution.
> example: 'perf stat -I 1000 -e cycles -a sleep 5'
>
> +If the metric exists, it is calculated by the counts generated in this interval and the metric is printed after #.
> +
> --interval-count times::
> Print count deltas for fixed number of times.
> This option should be used together with "-I" option.
> diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
> index 9207b6c45475..3f050d85c277 100644
> --- a/tools/perf/builtin-stat.c
> +++ b/tools/perf/builtin-stat.c
> @@ -359,6 +359,7 @@ static void process_interval(void)
> clock_gettime(CLOCK_MONOTONIC, &ts);
> diff_timespec(&rs, &ts, &ref_time);
>
> + perf_stat__reset_shadow_per_stat(&rt_stat);
> read_counters(&rs);
>
> if (STAT_RECORD) {
> --
> 2.17.1
>
--
- Arnaldo
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH v2] perf stat: Improve runtime stat for interval mode
2020-04-21 13:53 ` Arnaldo Carvalho de Melo
@ 2020-04-22 0:53 ` Jin, Yao
2020-04-22 1:08 ` Arnaldo Melo
2020-04-23 13:44 ` Arnaldo Carvalho de Melo
0 siblings, 2 replies; 8+ messages in thread
From: Jin, Yao @ 2020-04-22 0:53 UTC (permalink / raw)
To: Arnaldo Carvalho de Melo
Cc: jolsa, peterz, mingo, alexander.shishkin, Linux-kernel, ak,
kan.liang, yao.jin
Hi Arnaldo,
On 4/21/2020 9:53 PM, Arnaldo Carvalho de Melo wrote:
> Em Mon, Apr 20, 2020 at 10:54:17PM +0800, Jin Yao escreveu:
>> For interval mode, the metric is printed after # if it exists. But
>> it's not calculated by the counts generated in this interval. See
>> following examples,
>>
>> root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
>> # time counts unit events
>> 1.000422803 764,809 inst_retired.any # 2.9 CPI
>> 1.000422803 2,234,932 cycles
>> 2.001464585 1,960,061 inst_retired.any # 1.6 CPI
>> 2.001464585 4,022,591 cycles
>>
>> The second CPI should not be 1.6 (4,022,591/1,960,061 is 2.1)
>>
>> root@kbl-ppc:~# perf stat -e cycles,instructions -I1000 --interval-count 2
>> # time counts unit events
>> 1.000429493 2,869,311 cycles
>> 1.000429493 816,875 instructions # 0.28 insn per cycle
>> 2.001516426 9,260,973 cycles
>> 2.001516426 5,250,634 instructions # 0.87 insn per cycle
>>
>> The second 'insn per cycle' should not be 0.87 (5,250,634/9,260,973 is 0.57).
>>
>> The current code uses a global variable rt_stat for tracking and
>> updating the std dev of runtime stat. Unlike the counts, rt_stat is
>> not reset for interval. While the counts are reset for interval.
>>
>> perf_stat_process_counter()
>> {
>> if (config->interval)
>> init_stats(ps->res_stats);
>> }
>>
>> So for interval, the rt_stat should be reset either.
>
> s/either/too/g right?
>
Yes, should use "too" here. :)
> And please try and find what was the cset that introduced the problem,
> so that we can have a Fixes: line and the stable series can pick it, ok?
>
> - Arnaldo
>
I have tried to find the patch which introduced this issue.
51fd2df1e882 ("perf stat: Fix interval output values").
This patch zeros stats for interval mode. I just think it should reset
rt_stat too.
But I really don't know if it's fair to this patch so I don't add it in
my patch description.
Thanks
Jin Yao
>> This patch resets rt_stat before read_counters, so the runtime
>> stat is only calculated by the counts generated in this interval.
>>
>> With this patch,
>>
>> root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
>> # time counts unit events
>> 1.000420924 2,408,818 inst_retired.any # 2.1 CPI
>> 1.000420924 5,010,111 cycles
>> 2.001448579 2,798,407 inst_retired.any # 1.6 CPI
>> 2.001448579 4,599,861 cycles
>>
>> root@kbl-ppc:~# perf stat -e cycles,instructions -I1000 --interval-count 2
>> # time counts unit events
>> 1.000428555 2,769,714 cycles
>> 1.000428555 774,462 instructions # 0.28 insn per cycle
>> 2.001471562 3,595,904 cycles
>> 2.001471562 1,243,703 instructions # 0.35 insn per cycle
>>
>> Now the second 'insn per cycle' and CPI are calculated by the counts
>> generated in this interval.
>>
>> v2:
>> ---
>> Use just existing perf_stat__reset_shadow_per_stat(&rt_stat).
>> We don't need to define new function perf_stat__reset_rt_stat.
>>
>> Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
>> ---
>> tools/perf/Documentation/perf-stat.txt | 2 ++
>> tools/perf/builtin-stat.c | 1 +
>> 2 files changed, 3 insertions(+)
>>
>> diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
>> index 4d56586b2fb9..3fb5028aef08 100644
>> --- a/tools/perf/Documentation/perf-stat.txt
>> +++ b/tools/perf/Documentation/perf-stat.txt
>> @@ -176,6 +176,8 @@ Print count deltas every N milliseconds (minimum: 1ms)
>> The overhead percentage could be high in some cases, for instance with small, sub 100ms intervals. Use with caution.
>> example: 'perf stat -I 1000 -e cycles -a sleep 5'
>>
>> +If the metric exists, it is calculated by the counts generated in this interval and the metric is printed after #.
>> +
>> --interval-count times::
>> Print count deltas for fixed number of times.
>> This option should be used together with "-I" option.
>> diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
>> index 9207b6c45475..3f050d85c277 100644
>> --- a/tools/perf/builtin-stat.c
>> +++ b/tools/perf/builtin-stat.c
>> @@ -359,6 +359,7 @@ static void process_interval(void)
>> clock_gettime(CLOCK_MONOTONIC, &ts);
>> diff_timespec(&rs, &ts, &ref_time);
>>
>> + perf_stat__reset_shadow_per_stat(&rt_stat);
>> read_counters(&rs);
>>
>> if (STAT_RECORD) {
>> --
>> 2.17.1
>>
>
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH v2] perf stat: Improve runtime stat for interval mode
2020-04-22 0:53 ` Jin, Yao
@ 2020-04-22 1:08 ` Arnaldo Melo
2020-04-23 13:44 ` Arnaldo Carvalho de Melo
1 sibling, 0 replies; 8+ messages in thread
From: Arnaldo Melo @ 2020-04-22 1:08 UTC (permalink / raw)
To: Jin, Yao
Cc: jolsa, peterz, mingo, alexander.shishkin, Linux-kernel, ak,
kan.liang, yao.jin
On April 21, 2020 9:53:41 PM GMT-03:00, "Jin, Yao" <yao.jin@linux.intel.com> wrote:
>Hi Arnaldo,
>
>On 4/21/2020 9:53 PM, Arnaldo Carvalho de Melo wrote:
>> Em Mon, Apr 20, 2020 at 10:54:17PM +0800, Jin Yao escreveu:
>>> For interval mode, the metric is printed after # if it exists. But
>>> it's not calculated by the counts generated in this interval. See
>>> following examples,
>>>
>>> root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
>>> # time counts unit events
>>> 1.000422803 764,809 inst_retired.any
># 2.9 CPI
>>> 1.000422803 2,234,932 cycles
>>> 2.001464585 1,960,061 inst_retired.any
># 1.6 CPI
>>> 2.001464585 4,022,591 cycles
>>>
>>> The second CPI should not be 1.6 (4,022,591/1,960,061 is 2.1)
>>>
>>> root@kbl-ppc:~# perf stat -e cycles,instructions -I1000
>--interval-count 2
>>> # time counts unit events
>>> 1.000429493 2,869,311 cycles
>>> 1.000429493 816,875 instructions
># 0.28 insn per cycle
>>> 2.001516426 9,260,973 cycles
>>> 2.001516426 5,250,634 instructions
># 0.87 insn per cycle
>>>
>>> The second 'insn per cycle' should not be 0.87 (5,250,634/9,260,973
>is 0.57).
>>>
>>> The current code uses a global variable rt_stat for tracking and
>>> updating the std dev of runtime stat. Unlike the counts, rt_stat is
>>> not reset for interval. While the counts are reset for interval.
>>>
>>> perf_stat_process_counter()
>>> {
>>> if (config->interval)
>>> init_stats(ps->res_stats);
>>> }
>>>
>>> So for interval, the rt_stat should be reset either.
>>
>> s/either/too/g right?
>>
>
>Yes, should use "too" here. :)
Ok
>
>> And please try and find what was the cset that introduced the
>problem,
>> so that we can have a Fixes: line and the stable series can pick it,
>ok?
> >
>> - Arnaldo
>>
>
>I have tried to find the patch which introduced this issue.
>
>51fd2df1e882 ("perf stat: Fix interval output values").
>
>This patch zeros stats for interval mode. I just think it should reset
>rt_stat too.
>
>But I really don't know if it's fair to this patch so I don't add it in
>
>my patch description.
That's ok, this just speeds up the process, I'll check it :-)
- Arnaldo
>
>Thanks
>Jin Yao
>
>>> This patch resets rt_stat before read_counters, so the runtime
>>> stat is only calculated by the counts generated in this interval.
>>>
>>> With this patch,
>>>
>>> root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
>>> # time counts unit events
>>> 1.000420924 2,408,818 inst_retired.any
># 2.1 CPI
>>> 1.000420924 5,010,111 cycles
>>> 2.001448579 2,798,407 inst_retired.any
># 1.6 CPI
>>> 2.001448579 4,599,861 cycles
>>>
>>> root@kbl-ppc:~# perf stat -e cycles,instructions -I1000
>--interval-count 2
>>> # time counts unit events
>>> 1.000428555 2,769,714 cycles
>>> 1.000428555 774,462 instructions
># 0.28 insn per cycle
>>> 2.001471562 3,595,904 cycles
>>> 2.001471562 1,243,703 instructions
># 0.35 insn per cycle
>>>
>>> Now the second 'insn per cycle' and CPI are calculated by the counts
>>> generated in this interval.
>>>
>>> v2:
>>> ---
>>> Use just existing perf_stat__reset_shadow_per_stat(&rt_stat).
>>> We don't need to define new function perf_stat__reset_rt_stat.
>>>
>>> Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
>>> ---
>>> tools/perf/Documentation/perf-stat.txt | 2 ++
>>> tools/perf/builtin-stat.c | 1 +
>>> 2 files changed, 3 insertions(+)
>>>
>>> diff --git a/tools/perf/Documentation/perf-stat.txt
>b/tools/perf/Documentation/perf-stat.txt
>>> index 4d56586b2fb9..3fb5028aef08 100644
>>> --- a/tools/perf/Documentation/perf-stat.txt
>>> +++ b/tools/perf/Documentation/perf-stat.txt
>>> @@ -176,6 +176,8 @@ Print count deltas every N milliseconds
>(minimum: 1ms)
>>> The overhead percentage could be high in some cases, for instance
>with small, sub 100ms intervals. Use with caution.
>>> example: 'perf stat -I 1000 -e cycles -a sleep 5'
>>>
>>> +If the metric exists, it is calculated by the counts generated in
>this interval and the metric is printed after #.
>>> +
>>> --interval-count times::
>>> Print count deltas for fixed number of times.
>>> This option should be used together with "-I" option.
>>> diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
>>> index 9207b6c45475..3f050d85c277 100644
>>> --- a/tools/perf/builtin-stat.c
>>> +++ b/tools/perf/builtin-stat.c
>>> @@ -359,6 +359,7 @@ static void process_interval(void)
>>> clock_gettime(CLOCK_MONOTONIC, &ts);
>>> diff_timespec(&rs, &ts, &ref_time);
>>>
>>> + perf_stat__reset_shadow_per_stat(&rt_stat);
>>> read_counters(&rs);
>>>
>>> if (STAT_RECORD) {
>>> --
>>> 2.17.1
>>>
>>
--
Sent from my Android device with K-9 Mail. Please excuse my brevity.
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH v2] perf stat: Improve runtime stat for interval mode
2020-04-20 14:54 [PATCH v2] perf stat: Improve runtime stat for interval mode Jin Yao
2020-04-21 7:10 ` kajoljain
2020-04-21 13:53 ` Arnaldo Carvalho de Melo
@ 2020-04-22 7:22 ` Jiri Olsa
2020-05-08 13:05 ` [tip: perf/core] " tip-bot2 for Jin Yao
3 siblings, 0 replies; 8+ messages in thread
From: Jiri Olsa @ 2020-04-22 7:22 UTC (permalink / raw)
To: Jin Yao
Cc: acme, jolsa, peterz, mingo, alexander.shishkin, Linux-kernel, ak,
kan.liang, yao.jin
On Mon, Apr 20, 2020 at 10:54:17PM +0800, Jin Yao wrote:
> For interval mode, the metric is printed after # if it exists. But
> it's not calculated by the counts generated in this interval. See
> following examples,
>
> root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
> # time counts unit events
> 1.000422803 764,809 inst_retired.any # 2.9 CPI
> 1.000422803 2,234,932 cycles
> 2.001464585 1,960,061 inst_retired.any # 1.6 CPI
> 2.001464585 4,022,591 cycles
>
> The second CPI should not be 1.6 (4,022,591/1,960,061 is 2.1)
>
> root@kbl-ppc:~# perf stat -e cycles,instructions -I1000 --interval-count 2
> # time counts unit events
> 1.000429493 2,869,311 cycles
> 1.000429493 816,875 instructions # 0.28 insn per cycle
> 2.001516426 9,260,973 cycles
> 2.001516426 5,250,634 instructions # 0.87 insn per cycle
>
> The second 'insn per cycle' should not be 0.87 (5,250,634/9,260,973 is 0.57).
>
> The current code uses a global variable rt_stat for tracking and
> updating the std dev of runtime stat. Unlike the counts, rt_stat is
> not reset for interval. While the counts are reset for interval.
>
> perf_stat_process_counter()
> {
> if (config->interval)
> init_stats(ps->res_stats);
> }
>
> So for interval, the rt_stat should be reset either.
>
> This patch resets rt_stat before read_counters, so the runtime
> stat is only calculated by the counts generated in this interval.
>
> With this patch,
>
> root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
> # time counts unit events
> 1.000420924 2,408,818 inst_retired.any # 2.1 CPI
> 1.000420924 5,010,111 cycles
> 2.001448579 2,798,407 inst_retired.any # 1.6 CPI
> 2.001448579 4,599,861 cycles
>
> root@kbl-ppc:~# perf stat -e cycles,instructions -I1000 --interval-count 2
> # time counts unit events
> 1.000428555 2,769,714 cycles
> 1.000428555 774,462 instructions # 0.28 insn per cycle
> 2.001471562 3,595,904 cycles
> 2.001471562 1,243,703 instructions # 0.35 insn per cycle
>
> Now the second 'insn per cycle' and CPI are calculated by the counts
> generated in this interval.
>
> v2:
> ---
> Use just existing perf_stat__reset_shadow_per_stat(&rt_stat).
> We don't need to define new function perf_stat__reset_rt_stat.
Acked-by: Jiri Olsa <jolsa@redhat.com>
thanks,
jirka
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH v2] perf stat: Improve runtime stat for interval mode
2020-04-22 0:53 ` Jin, Yao
2020-04-22 1:08 ` Arnaldo Melo
@ 2020-04-23 13:44 ` Arnaldo Carvalho de Melo
1 sibling, 0 replies; 8+ messages in thread
From: Arnaldo Carvalho de Melo @ 2020-04-23 13:44 UTC (permalink / raw)
To: Jin, Yao
Cc: Arnaldo Carvalho de Melo, jolsa, peterz, mingo,
alexander.shishkin, Linux-kernel, ak, kan.liang, yao.jin
Em Wed, Apr 22, 2020 at 08:53:41AM +0800, Jin, Yao escreveu:
> Hi Arnaldo,
>
> On 4/21/2020 9:53 PM, Arnaldo Carvalho de Melo wrote:
> > Em Mon, Apr 20, 2020 at 10:54:17PM +0800, Jin Yao escreveu:
> > > For interval mode, the metric is printed after # if it exists. But
> > > it's not calculated by the counts generated in this interval. See
> > > following examples,
> > >
> > > root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
> > > # time counts unit events
> > > 1.000422803 764,809 inst_retired.any # 2.9 CPI
> > > 1.000422803 2,234,932 cycles
> > > 2.001464585 1,960,061 inst_retired.any # 1.6 CPI
> > > 2.001464585 4,022,591 cycles
> > >
> > > The second CPI should not be 1.6 (4,022,591/1,960,061 is 2.1)
> > >
> > > root@kbl-ppc:~# perf stat -e cycles,instructions -I1000 --interval-count 2
> > > # time counts unit events
> > > 1.000429493 2,869,311 cycles
> > > 1.000429493 816,875 instructions # 0.28 insn per cycle
> > > 2.001516426 9,260,973 cycles
> > > 2.001516426 5,250,634 instructions # 0.87 insn per cycle
> > >
> > > The second 'insn per cycle' should not be 0.87 (5,250,634/9,260,973 is 0.57).
> > >
> > > The current code uses a global variable rt_stat for tracking and
> > > updating the std dev of runtime stat. Unlike the counts, rt_stat is
> > > not reset for interval. While the counts are reset for interval.
> > >
> > > perf_stat_process_counter()
> > > {
> > > if (config->interval)
> > > init_stats(ps->res_stats);
> > > }
> > >
> > > So for interval, the rt_stat should be reset either.
> >
> > s/either/too/g right?
> >
>
> Yes, should use "too" here. :)
>
> > And please try and find what was the cset that introduced the problem,
> > so that we can have a Fixes: line and the stable series can pick it, ok?
> >
> > - Arnaldo
> >
>
> I have tried to find the patch which introduced this issue.
>
> 51fd2df1e882 ("perf stat: Fix interval output values").
>
> This patch zeros stats for interval mode. I just think it should reset
> rt_stat too.
> But I really don't know if it's fair to this patch so I don't add it in my
> patch description.
I think you are right that 51fd2df1e882 patch missed the fix you're
providing now, so I think it is fair to say that your current patch
fixes 51fd2df1e882, so I added it in the Fixes tag for the patch
discussed in this message, thanks.
- Arnaldo
> Thanks
> Jin Yao
>
> > > This patch resets rt_stat before read_counters, so the runtime
> > > stat is only calculated by the counts generated in this interval.
> > >
> > > With this patch,
> > >
> > > root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
> > > # time counts unit events
> > > 1.000420924 2,408,818 inst_retired.any # 2.1 CPI
> > > 1.000420924 5,010,111 cycles
> > > 2.001448579 2,798,407 inst_retired.any # 1.6 CPI
> > > 2.001448579 4,599,861 cycles
> > >
> > > root@kbl-ppc:~# perf stat -e cycles,instructions -I1000 --interval-count 2
> > > # time counts unit events
> > > 1.000428555 2,769,714 cycles
> > > 1.000428555 774,462 instructions # 0.28 insn per cycle
> > > 2.001471562 3,595,904 cycles
> > > 2.001471562 1,243,703 instructions # 0.35 insn per cycle
> > >
> > > Now the second 'insn per cycle' and CPI are calculated by the counts
> > > generated in this interval.
> > >
> > > v2:
> > > ---
> > > Use just existing perf_stat__reset_shadow_per_stat(&rt_stat).
> > > We don't need to define new function perf_stat__reset_rt_stat.
> > >
> > > Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
> > > ---
> > > tools/perf/Documentation/perf-stat.txt | 2 ++
> > > tools/perf/builtin-stat.c | 1 +
> > > 2 files changed, 3 insertions(+)
> > >
> > > diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
> > > index 4d56586b2fb9..3fb5028aef08 100644
> > > --- a/tools/perf/Documentation/perf-stat.txt
> > > +++ b/tools/perf/Documentation/perf-stat.txt
> > > @@ -176,6 +176,8 @@ Print count deltas every N milliseconds (minimum: 1ms)
> > > The overhead percentage could be high in some cases, for instance with small, sub 100ms intervals. Use with caution.
> > > example: 'perf stat -I 1000 -e cycles -a sleep 5'
> > > +If the metric exists, it is calculated by the counts generated in this interval and the metric is printed after #.
> > > +
> > > --interval-count times::
> > > Print count deltas for fixed number of times.
> > > This option should be used together with "-I" option.
> > > diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
> > > index 9207b6c45475..3f050d85c277 100644
> > > --- a/tools/perf/builtin-stat.c
> > > +++ b/tools/perf/builtin-stat.c
> > > @@ -359,6 +359,7 @@ static void process_interval(void)
> > > clock_gettime(CLOCK_MONOTONIC, &ts);
> > > diff_timespec(&rs, &ts, &ref_time);
> > > + perf_stat__reset_shadow_per_stat(&rt_stat);
> > > read_counters(&rs);
> > > if (STAT_RECORD) {
> > > --
> > > 2.17.1
> > >
> >
--
- Arnaldo
^ permalink raw reply [flat|nested] 8+ messages in thread
* [tip: perf/core] perf stat: Improve runtime stat for interval mode
2020-04-20 14:54 [PATCH v2] perf stat: Improve runtime stat for interval mode Jin Yao
` (2 preceding siblings ...)
2020-04-22 7:22 ` Jiri Olsa
@ 2020-05-08 13:05 ` tip-bot2 for Jin Yao
3 siblings, 0 replies; 8+ messages in thread
From: tip-bot2 for Jin Yao @ 2020-05-08 13:05 UTC (permalink / raw)
To: linux-tip-commits
Cc: Jin Yao, Jiri Olsa, Alexander Shishkin, Andi Kleen, Jin Yao,
Kan Liang, Peter Zijlstra, Arnaldo Carvalho de Melo, x86, LKML
The following commit has been merged into the perf/core branch of tip:
Commit-ID: 197ba86fdc888dc0d3d6b89b402c9c6851d4c6fb
Gitweb: https://git.kernel.org/tip/197ba86fdc888dc0d3d6b89b402c9c6851d4c6fb
Author: Jin Yao <yao.jin@linux.intel.com>
AuthorDate: Mon, 20 Apr 2020 22:54:17 +08:00
Committer: Arnaldo Carvalho de Melo <acme@redhat.com>
CommitterDate: Thu, 23 Apr 2020 11:03:46 -03:00
perf stat: Improve runtime stat for interval mode
For interval mode, the metric is printed after the '#' character if it
exists. But it's not calculated by the counts generated in this
interval.
See the following examples:
root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
# time counts unit events
1.000422803 764,809 inst_retired.any # 2.9 CPI
1.000422803 2,234,932 cycles
2.001464585 1,960,061 inst_retired.any # 1.6 CPI
2.001464585 4,022,591 cycles
The second CPI should not be 1.6 (4,022,591/1,960,061 is 2.1)
root@kbl-ppc:~# perf stat -e cycles,instructions -I1000 --interval-count 2
# time counts unit events
1.000429493 2,869,311 cycles
1.000429493 816,875 instructions # 0.28 insn per cycle
2.001516426 9,260,973 cycles
2.001516426 5,250,634 instructions # 0.87 insn per cycle
The second 'insn per cycle' should not be 0.87 (5,250,634/9,260,973 is
0.57).
The current code uses a global variable 'rt_stat' for tracking and
updating the std dev of runtime stat. Unlike the counts, 'rt_stat' is not
reset for interval. While the counts are reset for interval.
perf_stat_process_counter()
{
if (config->interval)
init_stats(ps->res_stats);
}
So for interval mode, the 'rt_stat' variable should be reset too.
This patch resets 'rt_stat' before read_counters(), so the runtime stat
is only calculated by the counts generated in this interval.
With this patch:
root@kbl-ppc:~# perf stat -M CPI -I1000 --interval-count 2
# time counts unit events
1.000420924 2,408,818 inst_retired.any # 2.1 CPI
1.000420924 5,010,111 cycles
2.001448579 2,798,407 inst_retired.any # 1.6 CPI
2.001448579 4,599,861 cycles
root@kbl-ppc:~# perf stat -e cycles,instructions -I1000 --interval-count 2
# time counts unit events
1.000428555 2,769,714 cycles
1.000428555 774,462 instructions # 0.28 insn per cycle
2.001471562 3,595,904 cycles
2.001471562 1,243,703 instructions # 0.35 insn per cycle
Now the second 'insn per cycle' and CPI are calculated by the counts
generated in this interval.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Tested-By: Kajol Jain <kjain@linux.ibm.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20200420145417.6864-1-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/Documentation/perf-stat.txt | 2 ++
tools/perf/builtin-stat.c | 1 +
2 files changed, 3 insertions(+)
diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
index 4d56586..3fb5028 100644
--- a/tools/perf/Documentation/perf-stat.txt
+++ b/tools/perf/Documentation/perf-stat.txt
@@ -176,6 +176,8 @@ Print count deltas every N milliseconds (minimum: 1ms)
The overhead percentage could be high in some cases, for instance with small, sub 100ms intervals. Use with caution.
example: 'perf stat -I 1000 -e cycles -a sleep 5'
+If the metric exists, it is calculated by the counts generated in this interval and the metric is printed after #.
+
--interval-count times::
Print count deltas for fixed number of times.
This option should be used together with "-I" option.
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 9207b6c..3f050d8 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -359,6 +359,7 @@ static void process_interval(void)
clock_gettime(CLOCK_MONOTONIC, &ts);
diff_timespec(&rs, &ts, &ref_time);
+ perf_stat__reset_shadow_per_stat(&rt_stat);
read_counters(&rs);
if (STAT_RECORD) {
^ permalink raw reply related [flat|nested] 8+ messages in thread
end of thread, other threads:[~2020-05-08 13:08 UTC | newest]
Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-04-20 14:54 [PATCH v2] perf stat: Improve runtime stat for interval mode Jin Yao
2020-04-21 7:10 ` kajoljain
2020-04-21 13:53 ` Arnaldo Carvalho de Melo
2020-04-22 0:53 ` Jin, Yao
2020-04-22 1:08 ` Arnaldo Melo
2020-04-23 13:44 ` Arnaldo Carvalho de Melo
2020-04-22 7:22 ` Jiri Olsa
2020-05-08 13:05 ` [tip: perf/core] " tip-bot2 for Jin Yao
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).