* [PATCH] tools/perf/sched: Update process names of processes in zombie state for both -s and -S options
@ 2026-04-26 9:39 Athira Rajeev
2026-04-27 5:56 ` Namhyung Kim
2026-04-27 6:43 ` Venkat
0 siblings, 2 replies; 8+ messages in thread
From: Athira Rajeev @ 2026-04-26 9:39 UTC (permalink / raw)
To: acme, jolsa, adrian.hunter, mpetlan, tmricht, maddy, irogers,
namhyung
Cc: linux-perf-users, linuxppc-dev, atrajeev, hbathini, Tejas.Manhas1,
Tanushree.Shah, Shivani.Nittor
In redhat perftool testsuite, observed fail for this test:
-- [ FAIL ] -- perf_sched :: test_timehist :: --with-summary (output regexp parsing)
This led to analysis of "perf sched timehist" summary options.
# perf sched record -a -o ./perf.data -- sleep 0.1
This will record using perf sched record
perf sched timeliest has two options "-s" and "-S"
# perf sched -i ./perf.data timehist -S
-S : Captures summary also at the end
# perf sched -i ./perf.data timehist -s
-s : Captures only summary
The test saves -s result which has only summary and compares with
summary which comes at the end from -S . Since there is a difference
in these two, test fails.
Checking the behaviour change in -S and -s results, difference is:
rcu_sched[16] 2 4 0.013 0.001 0.003 0.006 33.23 0
migration/11[73] 2 1 0.006 0.006 0.006 0.006 0.00 0
migration/3[33] 2 1 0.006 0.006 0.006 0.006 0.00 0
- :216753[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
+ sleep[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
migration/8[58] 2 1 0.005 0.005 0.005 0.005 0.00 0
NetworkManager[811] 1 2 0.089 0.028 0.044 0.060 36.06 0
migration/13[83] 2 1 0.005 0.005 0.005 0.005 0.00 0
Here 216753 is pid for sleep which is a zombie process. This is
happening in latest kernel due to an update in "-S" result.
In -S, the process name appears in the results "sleep[216753]",
where as in the -s, only pid is present in the summary result
":216753[216753]".
After commit 39f473f6d0b2 ("perf sched timehist: decode process names
of processes in zombie state")
for -S option, if process name is using pid, it uses different way to
set it. So that we get the process name and not just Pid.
This change went in only for timehist_print_sample() function.
Add this improvement in generic place so that even -s option (which
captures summary) also will have meaningful information.
Signed-off-by: Athira Rajeev <atrajeev@linux.ibm.com>
---
tools/perf/builtin-sched.c | 14 +++++++++-----
1 file changed, 9 insertions(+), 5 deletions(-)
diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
index 555247568e7a..ee1b89a6af50 100644
--- a/tools/perf/builtin-sched.c
+++ b/tools/perf/builtin-sched.c
@@ -2197,11 +2197,6 @@ static void timehist_print_sample(struct perf_sched *sched,
printf(" ");
}
- if (!thread__comm_set(thread)) {
- const char *prev_comm = evsel__strval(evsel, sample, "prev_comm");
- thread__set_comm(thread, prev_comm, sample->time);
- }
-
printf(" %-*s ", comm_width, timehist_get_commstr(thread));
if (sched->show_prio)
@@ -2890,6 +2885,15 @@ static int timehist_sched_change_event(const struct perf_tool *tool,
itr->last_thread = NULL;
}
+ /*
+ * If the process name is not set for the thread, use "prev_comm"
+ * to set it. Otherwise the sched summary will have just pid information
+ */
+ if (!thread__comm_set(thread)) {
+ const char *prev_comm = evsel__strval(evsel, sample, "prev_comm");
+ thread__set_comm(thread, prev_comm, sample->time);
+ }
+
if (!sched->summary_only)
timehist_print_sample(sched, evsel, sample, &al, thread, t, state);
}
--
2.47.3
^ permalink raw reply related [flat|nested] 8+ messages in thread
* Re: [PATCH] tools/perf/sched: Update process names of processes in zombie state for both -s and -S options
2026-04-26 9:39 [PATCH] tools/perf/sched: Update process names of processes in zombie state for both -s and -S options Athira Rajeev
@ 2026-04-27 5:56 ` Namhyung Kim
2026-05-21 8:32 ` Athira Rajeev
2026-04-27 6:43 ` Venkat
1 sibling, 1 reply; 8+ messages in thread
From: Namhyung Kim @ 2026-04-27 5:56 UTC (permalink / raw)
To: Athira Rajeev
Cc: acme, jolsa, adrian.hunter, mpetlan, tmricht, maddy, irogers,
linux-perf-users, linuxppc-dev, hbathini, Tejas.Manhas1,
Tanushree.Shah, Shivani.Nittor
On Sun, Apr 26, 2026 at 03:09:30PM +0530, Athira Rajeev wrote:
> In redhat perftool testsuite, observed fail for this test:
> -- [ FAIL ] -- perf_sched :: test_timehist :: --with-summary (output regexp parsing)
>
> This led to analysis of "perf sched timehist" summary options.
>
> # perf sched record -a -o ./perf.data -- sleep 0.1
> This will record using perf sched record
>
> perf sched timeliest has two options "-s" and "-S"
> # perf sched -i ./perf.data timehist -S
> -S : Captures summary also at the end
>
> # perf sched -i ./perf.data timehist -s
> -s : Captures only summary
>
> The test saves -s result which has only summary and compares with
> summary which comes at the end from -S . Since there is a difference
> in these two, test fails.
>
> Checking the behaviour change in -S and -s results, difference is:
>
> rcu_sched[16] 2 4 0.013 0.001 0.003 0.006 33.23 0
> migration/11[73] 2 1 0.006 0.006 0.006 0.006 0.00 0
> migration/3[33] 2 1 0.006 0.006 0.006 0.006 0.00 0
> - :216753[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
> + sleep[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
> migration/8[58] 2 1 0.005 0.005 0.005 0.005 0.00 0
> NetworkManager[811] 1 2 0.089 0.028 0.044 0.060 36.06 0
> migration/13[83] 2 1 0.005 0.005 0.005 0.005 0.00 0
>
> Here 216753 is pid for sleep which is a zombie process. This is
> happening in latest kernel due to an update in "-S" result.
> In -S, the process name appears in the results "sleep[216753]",
> where as in the -s, only pid is present in the summary result
> ":216753[216753]".
>
> After commit 39f473f6d0b2 ("perf sched timehist: decode process names
> of processes in zombie state")
> for -S option, if process name is using pid, it uses different way to
> set it. So that we get the process name and not just Pid.
>
> This change went in only for timehist_print_sample() function.
> Add this improvement in generic place so that even -s option (which
> captures summary) also will have meaningful information.
>
> Signed-off-by: Athira Rajeev <atrajeev@linux.ibm.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Thanks,
Namhyung
> ---
> tools/perf/builtin-sched.c | 14 +++++++++-----
> 1 file changed, 9 insertions(+), 5 deletions(-)
>
> diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
> index 555247568e7a..ee1b89a6af50 100644
> --- a/tools/perf/builtin-sched.c
> +++ b/tools/perf/builtin-sched.c
> @@ -2197,11 +2197,6 @@ static void timehist_print_sample(struct perf_sched *sched,
> printf(" ");
> }
>
> - if (!thread__comm_set(thread)) {
> - const char *prev_comm = evsel__strval(evsel, sample, "prev_comm");
> - thread__set_comm(thread, prev_comm, sample->time);
> - }
> -
> printf(" %-*s ", comm_width, timehist_get_commstr(thread));
>
> if (sched->show_prio)
> @@ -2890,6 +2885,15 @@ static int timehist_sched_change_event(const struct perf_tool *tool,
> itr->last_thread = NULL;
> }
>
> + /*
> + * If the process name is not set for the thread, use "prev_comm"
> + * to set it. Otherwise the sched summary will have just pid information
> + */
> + if (!thread__comm_set(thread)) {
> + const char *prev_comm = evsel__strval(evsel, sample, "prev_comm");
> + thread__set_comm(thread, prev_comm, sample->time);
> + }
> +
> if (!sched->summary_only)
> timehist_print_sample(sched, evsel, sample, &al, thread, t, state);
> }
> --
> 2.47.3
>
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH] tools/perf/sched: Update process names of processes in zombie state for both -s and -S options
2026-04-26 9:39 [PATCH] tools/perf/sched: Update process names of processes in zombie state for both -s and -S options Athira Rajeev
2026-04-27 5:56 ` Namhyung Kim
@ 2026-04-27 6:43 ` Venkat
1 sibling, 0 replies; 8+ messages in thread
From: Venkat @ 2026-04-27 6:43 UTC (permalink / raw)
To: Athira Rajeev
Cc: acme, jolsa, adrian.hunter, mpetlan, tmricht, maddy, irogers,
namhyung, linux-perf-users, linuxppc-dev, hbathini, Tejas.Manhas1,
Tanushree.Shah, Shivani.Nittor
> On 26 Apr 2026, at 3:09 PM, Athira Rajeev <atrajeev@linux.ibm.com> wrote:
>
> In redhat perftool testsuite, observed fail for this test:
> -- [ FAIL ] -- perf_sched :: test_timehist :: --with-summary (output regexp parsing)
>
> This led to analysis of "perf sched timehist" summary options.
>
> # perf sched record -a -o ./perf.data -- sleep 0.1
> This will record using perf sched record
>
> perf sched timeliest has two options "-s" and "-S"
> # perf sched -i ./perf.data timehist -S
> -S : Captures summary also at the end
>
> # perf sched -i ./perf.data timehist -s
> -s : Captures only summary
>
> The test saves -s result which has only summary and compares with
> summary which comes at the end from -S . Since there is a difference
> in these two, test fails.
>
> Checking the behaviour change in -S and -s results, difference is:
>
> rcu_sched[16] 2 4 0.013 0.001 0.003 0.006 33.23 0
> migration/11[73] 2 1 0.006 0.006 0.006 0.006 0.00 0
> migration/3[33] 2 1 0.006 0.006 0.006 0.006 0.00 0
> - :216753[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
> + sleep[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
> migration/8[58] 2 1 0.005 0.005 0.005 0.005 0.00 0
> NetworkManager[811] 1 2 0.089 0.028 0.044 0.060 36.06 0
> migration/13[83] 2 1 0.005 0.005 0.005 0.005 0.00 0
>
> Here 216753 is pid for sleep which is a zombie process. This is
> happening in latest kernel due to an update in "-S" result.
> In -S, the process name appears in the results "sleep[216753]",
> where as in the -s, only pid is present in the summary result
> ":216753[216753]".
>
> After commit 39f473f6d0b2 ("perf sched timehist: decode process names
> of processes in zombie state")
> for -S option, if process name is using pid, it uses different way to
> set it. So that we get the process name and not just Pid.
>
> This change went in only for timehist_print_sample() function.
> Add this improvement in generic place so that even -s option (which
> captures summary) also will have meaningful information.
>
> Signed-off-by: Athira Rajeev <atrajeev@linux.ibm.com>
> ---
Tested this patch, by applying on top of mainline kernel.
WithOut Fix:
# ./perf sched -i perf.data timehist -s
Samples of sched_switch event do not have callchains.
Runtime summary
comm parent sched-in run-time min-run avg-run max-run stddev migrations
(count) (msec) (msec) (msec) (msec) %
---------------------------------------------------------------------------------------------------------------------
migration/14[88] 2 1 0.006 0.006 0.006 0.006 0.00 0
migration/6[48] 2 1 0.008 0.008 0.008 0.008 0.00 0
rcu_preempt[16] 2 7 0.042 0.003 0.006 0.014 23.80 0
migration/11[73] 2 1 0.010 0.010 0.010 0.010 0.00 0
migration/3[33] 2 1 0.009 0.009 0.009 0.009 0.00 0
sshd-session[384033] 384025 4 0.104 0.023 0.026 0.032 8.14 0
kworker/u68:3-e[316378] 2 1 0.003 0.003 0.003 0.003 0.00 0
kcompactd3[114] 2 1 0.003 0.003 0.003 0.003 0.00 0
migration/8[58] 2 1 0.012 0.012 0.012 0.012 0.00 0
migration/13[83] 2 1 0.008 0.008 0.008 0.008 0.00 0
migration/5[43] 2 1 0.007 0.007 0.007 0.007 0.00 0
migration/0[19] 2 1 0.010 0.010 0.010 0.010 0.00 0
migration/10[68] 2 1 0.007 0.007 0.007 0.007 0.00 0
ksoftirqd/5[44] 2 1 0.184 0.184 0.184 0.184 0.00 0
migration/2[28] 2 1 0.008 0.008 0.008 0.008 0.00 0
kworker/4:0-mm_[383676] 2 1 0.013 0.013 0.013 0.013 0.00 0
perf[390116] 384034 17 0.065 0.000 0.003 0.065 100.00 0
xfsaild/dm-2[861] 2 1 0.008 0.008 0.008 0.008 0.00 0
migration/15[93] 2 1 0.009 0.009 0.009 0.009 0.00 0
:390117[390117] -1 1 0.124 0.124 0.124 0.124 0.00 0
kworker/9:4-mm_[390101] 2 1 0.069 0.069 0.069 0.069 0.00 0
migration/7[53] 2 1 0.008 0.008 0.008 0.008 0.00 0
kworker/5:0-xfs[390077] 2 1 0.015 0.015 0.015 0.015 0.00 0
migration/12[78] 2 1 0.007 0.007 0.007 0.007 0.00 0
migration/4[38] 2 1 0.005 0.005 0.005 0.005 0.00 0
NetworkManager[903] 1 4 0.275 0.032 0.068 0.105 29.66 0
migration/1[23] 2 1 0.010 0.010 0.010 0.010 0.00 0
migration/9[63] 2 1 0.011 0.011 0.011 0.011 0.00 0
With FIX:
# ./perf sched -i perf.data timehist -s
Samples of sched_switch event do not have callchains.
Runtime summary
comm parent sched-in run-time min-run avg-run max-run stddev migrations
(count) (msec) (msec) (msec) (msec) %
---------------------------------------------------------------------------------------------------------------------
migration/14[88] 2 1 0.006 0.006 0.006 0.006 0.00 0
migration/6[48] 2 1 0.008 0.008 0.008 0.008 0.00 0
rcu_preempt[16] 2 7 0.042 0.003 0.006 0.014 23.80 0
migration/11[73] 2 1 0.010 0.010 0.010 0.010 0.00 0
migration/3[33] 2 1 0.009 0.009 0.009 0.009 0.00 0
sshd-session[384033] 384025 4 0.104 0.023 0.026 0.032 8.14 0
kworker/u68:3-e[316378] 2 1 0.003 0.003 0.003 0.003 0.00 0
kcompactd3[114] 2 1 0.003 0.003 0.003 0.003 0.00 0
migration/8[58] 2 1 0.012 0.012 0.012 0.012 0.00 0
migration/13[83] 2 1 0.008 0.008 0.008 0.008 0.00 0
migration/5[43] 2 1 0.007 0.007 0.007 0.007 0.00 0
migration/0[19] 2 1 0.010 0.010 0.010 0.010 0.00 0
migration/10[68] 2 1 0.007 0.007 0.007 0.007 0.00 0
ksoftirqd/5[44] 2 1 0.184 0.184 0.184 0.184 0.00 0
migration/2[28] 2 1 0.008 0.008 0.008 0.008 0.00 0
kworker/4:0-mm_[383676] 2 1 0.013 0.013 0.013 0.013 0.00 0
perf[390116] 384034 17 0.065 0.000 0.003 0.065 100.00 0
xfsaild/dm-2[861] 2 1 0.008 0.008 0.008 0.008 0.00 0
migration/15[93] 2 1 0.009 0.009 0.009 0.009 0.00 0
sleep[390117] -1 1 0.124 0.124 0.124 0.124 0.00 0
kworker/9:4-mm_[390101] 2 1 0.069 0.069 0.069 0.069 0.00 0
migration/7[53] 2 1 0.008 0.008 0.008 0.008 0.00 0
kworker/5:0-xfs[390077] 2 1 0.015 0.015 0.015 0.015 0.00 0
migration/12[78] 2 1 0.007 0.007 0.007 0.007 0.00 0
migration/4[38] 2 1 0.005 0.005 0.005 0.005 0.00 0
NetworkManager[903] 1 4 0.275 0.032 0.068 0.105 29.66 0
migration/1[23] 2 1 0.010 0.010 0.010 0.010 0.00 0
migration/9[63] 2 1 0.011 0.011 0.011 0.011 0.00 0
Please add below tag.
Tested-by: Venkat Rao Bagalkote <venkat88@linux.ibm.com>
Regards,
Venkat.
> tools/perf/builtin-sched.c | 14 +++++++++-----
> 1 file changed, 9 insertions(+), 5 deletions(-)
>
> diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
> index 555247568e7a..ee1b89a6af50 100644
> --- a/tools/perf/builtin-sched.c
> +++ b/tools/perf/builtin-sched.c
> @@ -2197,11 +2197,6 @@ static void timehist_print_sample(struct perf_sched *sched,
> printf(" ");
> }
>
> - if (!thread__comm_set(thread)) {
> - const char *prev_comm = evsel__strval(evsel, sample, "prev_comm");
> - thread__set_comm(thread, prev_comm, sample->time);
> - }
> -
> printf(" %-*s ", comm_width, timehist_get_commstr(thread));
>
> if (sched->show_prio)
> @@ -2890,6 +2885,15 @@ static int timehist_sched_change_event(const struct perf_tool *tool,
> itr->last_thread = NULL;
> }
>
> + /*
> + * If the process name is not set for the thread, use "prev_comm"
> + * to set it. Otherwise the sched summary will have just pid information
> + */
> + if (!thread__comm_set(thread)) {
> + const char *prev_comm = evsel__strval(evsel, sample, "prev_comm");
> + thread__set_comm(thread, prev_comm, sample->time);
> + }
> +
> if (!sched->summary_only)
> timehist_print_sample(sched, evsel, sample, &al, thread, t, state);
> }
> --
> 2.47.3
>
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH] tools/perf/sched: Update process names of processes in zombie state for both -s and -S options
2026-04-27 5:56 ` Namhyung Kim
@ 2026-05-21 8:32 ` Athira Rajeev
2026-05-21 14:17 ` Arnaldo Carvalho de Melo
0 siblings, 1 reply; 8+ messages in thread
From: Athira Rajeev @ 2026-05-21 8:32 UTC (permalink / raw)
To: Namhyung Kim, Ian Rogers, Arnaldo Carvalho de Melo
Cc: jolsa, adrian.hunter, mpetlan, tmricht, maddy, linux-perf-users,
linuxppc-dev, hbathini, Tejas.Manhas1, Tanushree.Shah,
Shivani.Nittor
> On 27 Apr 2026, at 11:26 AM, Namhyung Kim <namhyung@kernel.org> wrote:
>
> On Sun, Apr 26, 2026 at 03:09:30PM +0530, Athira Rajeev wrote:
>> In redhat perftool testsuite, observed fail for this test:
>> -- [ FAIL ] -- perf_sched :: test_timehist :: --with-summary (output regexp parsing)
>>
>> This led to analysis of "perf sched timehist" summary options.
>>
>> # perf sched record -a -o ./perf.data -- sleep 0.1
>> This will record using perf sched record
>>
>> perf sched timeliest has two options "-s" and "-S"
>> # perf sched -i ./perf.data timehist -S
>> -S : Captures summary also at the end
>>
>> # perf sched -i ./perf.data timehist -s
>> -s : Captures only summary
>>
>> The test saves -s result which has only summary and compares with
>> summary which comes at the end from -S . Since there is a difference
>> in these two, test fails.
>>
>> Checking the behaviour change in -S and -s results, difference is:
>>
>> rcu_sched[16] 2 4 0.013 0.001 0.003 0.006 33.23 0
>> migration/11[73] 2 1 0.006 0.006 0.006 0.006 0.00 0
>> migration/3[33] 2 1 0.006 0.006 0.006 0.006 0.00 0
>> - :216753[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
>> + sleep[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
>> migration/8[58] 2 1 0.005 0.005 0.005 0.005 0.00 0
>> NetworkManager[811] 1 2 0.089 0.028 0.044 0.060 36.06 0
>> migration/13[83] 2 1 0.005 0.005 0.005 0.005 0.00 0
>>
>> Here 216753 is pid for sleep which is a zombie process. This is
>> happening in latest kernel due to an update in "-S" result.
>> In -S, the process name appears in the results "sleep[216753]",
>> where as in the -s, only pid is present in the summary result
>> ":216753[216753]".
>>
>> After commit 39f473f6d0b2 ("perf sched timehist: decode process names
>> of processes in zombie state")
>> for -S option, if process name is using pid, it uses different way to
>> set it. So that we get the process name and not just Pid.
>>
>> This change went in only for timehist_print_sample() function.
>> Add this improvement in generic place so that even -s option (which
>> captures summary) also will have meaningful information.
>>
>> Signed-off-by: Athira Rajeev <atrajeev@linux.ibm.com>
>
> Acked-by: Namhyung Kim <namhyung@kernel.org>
>
> Thanks,
> Namhyung
Hi,
Can we please have this pulled in, if the patch looks fine ?
Thanks
Athira
>
>> ---
>> tools/perf/builtin-sched.c | 14 +++++++++-----
>> 1 file changed, 9 insertions(+), 5 deletions(-)
>>
>> diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
>> index 555247568e7a..ee1b89a6af50 100644
>> --- a/tools/perf/builtin-sched.c
>> +++ b/tools/perf/builtin-sched.c
>> @@ -2197,11 +2197,6 @@ static void timehist_print_sample(struct perf_sched *sched,
>> printf(" ");
>> }
>>
>> - if (!thread__comm_set(thread)) {
>> - const char *prev_comm = evsel__strval(evsel, sample, "prev_comm");
>> - thread__set_comm(thread, prev_comm, sample->time);
>> - }
>> -
>> printf(" %-*s ", comm_width, timehist_get_commstr(thread));
>>
>> if (sched->show_prio)
>> @@ -2890,6 +2885,15 @@ static int timehist_sched_change_event(const struct perf_tool *tool,
>> itr->last_thread = NULL;
>> }
>>
>> + /*
>> + * If the process name is not set for the thread, use "prev_comm"
>> + * to set it. Otherwise the sched summary will have just pid information
>> + */
>> + if (!thread__comm_set(thread)) {
>> + const char *prev_comm = evsel__strval(evsel, sample, "prev_comm");
>> + thread__set_comm(thread, prev_comm, sample->time);
>> + }
>> +
>> if (!sched->summary_only)
>> timehist_print_sample(sched, evsel, sample, &al, thread, t, state);
>> }
>> --
>> 2.47.3
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH] tools/perf/sched: Update process names of processes in zombie state for both -s and -S options
2026-05-21 8:32 ` Athira Rajeev
@ 2026-05-21 14:17 ` Arnaldo Carvalho de Melo
2026-06-04 14:17 ` Arnaldo Carvalho de Melo
0 siblings, 1 reply; 8+ messages in thread
From: Arnaldo Carvalho de Melo @ 2026-05-21 14:17 UTC (permalink / raw)
To: Athira Rajeev
Cc: Namhyung Kim, Ian Rogers, jolsa, adrian.hunter, mpetlan, tmricht,
maddy, linux-perf-users, linuxppc-dev, hbathini, Tejas.Manhas1,
Tanushree.Shah, Shivani.Nittor
On Thu, May 21, 2026 at 02:02:53PM +0530, Athira Rajeev wrote:
> > On 27 Apr 2026, at 11:26 AM, Namhyung Kim <namhyung@kernel.org> wrote:
> > On Sun, Apr 26, 2026 at 03:09:30PM +0530, Athira Rajeev wrote:
> >> In redhat perftool testsuite, observed fail for this test:
> >> -- [ FAIL ] -- perf_sched :: test_timehist :: --with-summary (output regexp parsing)
> >>
> >> This led to analysis of "perf sched timehist" summary options.
> >>
> >> # perf sched record -a -o ./perf.data -- sleep 0.1
> >> This will record using perf sched record
> >>
> >> perf sched timeliest has two options "-s" and "-S"
> >> # perf sched -i ./perf.data timehist -S
> >> -S : Captures summary also at the end
> >>
> >> # perf sched -i ./perf.data timehist -s
> >> -s : Captures only summary
> >>
> >> The test saves -s result which has only summary and compares with
> >> summary which comes at the end from -S . Since there is a difference
> >> in these two, test fails.
> >>
> >> Checking the behaviour change in -S and -s results, difference is:
> >>
> >> rcu_sched[16] 2 4 0.013 0.001 0.003 0.006 33.23 0
> >> migration/11[73] 2 1 0.006 0.006 0.006 0.006 0.00 0
> >> migration/3[33] 2 1 0.006 0.006 0.006 0.006 0.00 0
> >> - :216753[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
> >> + sleep[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
> >> migration/8[58] 2 1 0.005 0.005 0.005 0.005 0.00 0
> >> NetworkManager[811] 1 2 0.089 0.028 0.044 0.060 36.06 0
> >> migration/13[83] 2 1 0.005 0.005 0.005 0.005 0.00 0
> >>
> >> Here 216753 is pid for sleep which is a zombie process. This is
> >> happening in latest kernel due to an update in "-S" result.
> >> In -S, the process name appears in the results "sleep[216753]",
> >> where as in the -s, only pid is present in the summary result
> >> ":216753[216753]".
> >>
> >> After commit 39f473f6d0b2 ("perf sched timehist: decode process names
> >> of processes in zombie state")
> >> for -S option, if process name is using pid, it uses different way to
> >> set it. So that we get the process name and not just Pid.
> >>
> >> This change went in only for timehist_print_sample() function.
> >> Add this improvement in generic place so that even -s option (which
> >> captures summary) also will have meaningful information.
> >>
> >> Signed-off-by: Athira Rajeev <atrajeev@linux.ibm.com>
> >
> > Acked-by: Namhyung Kim <namhyung@kernel.org>
> >
> > Thanks,
> > Namhyung
> Hi,
>
> Can we please have this pulled in, if the patch looks fine ?
Can you please check applying it on top of current perf-tools-next?
Thanks,
- Arnaldo
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH] tools/perf/sched: Update process names of processes in zombie state for both -s and -S options
2026-05-21 14:17 ` Arnaldo Carvalho de Melo
@ 2026-06-04 14:17 ` Arnaldo Carvalho de Melo
2026-06-04 15:08 ` Athira Rajeev
0 siblings, 1 reply; 8+ messages in thread
From: Arnaldo Carvalho de Melo @ 2026-06-04 14:17 UTC (permalink / raw)
To: Athira Rajeev, Anubhav Shelat
Cc: Namhyung Kim, Ian Rogers, jolsa, adrian.hunter, mpetlan, tmricht,
maddy, linux-perf-users, linuxppc-dev, hbathini, Tejas.Manhas1,
Tanushree.Shah, Shivani.Nittor
On Thu, May 21, 2026 at 11:17:58AM -0300, Arnaldo Carvalho de Melo wrote:
> On Thu, May 21, 2026 at 02:02:53PM +0530, Athira Rajeev wrote:
> > > On 27 Apr 2026, at 11:26 AM, Namhyung Kim <namhyung@kernel.org> wrote:
> > > On Sun, Apr 26, 2026 at 03:09:30PM +0530, Athira Rajeev wrote:
> > >> In redhat perftool testsuite, observed fail for this test:
> > >> -- [ FAIL ] -- perf_sched :: test_timehist :: --with-summary (output regexp parsing)
> > >>
> > >> This led to analysis of "perf sched timehist" summary options.
> > >>
> > >> # perf sched record -a -o ./perf.data -- sleep 0.1
> > >> This will record using perf sched record
> > >>
> > >> perf sched timeliest has two options "-s" and "-S"
> > >> # perf sched -i ./perf.data timehist -S
> > >> -S : Captures summary also at the end
> > >>
> > >> # perf sched -i ./perf.data timehist -s
> > >> -s : Captures only summary
> > >>
> > >> The test saves -s result which has only summary and compares with
> > >> summary which comes at the end from -S . Since there is a difference
> > >> in these two, test fails.
> > >>
> > >> Checking the behaviour change in -S and -s results, difference is:
> > >>
> > >> rcu_sched[16] 2 4 0.013 0.001 0.003 0.006 33.23 0
> > >> migration/11[73] 2 1 0.006 0.006 0.006 0.006 0.00 0
> > >> migration/3[33] 2 1 0.006 0.006 0.006 0.006 0.00 0
> > >> - :216753[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
> > >> + sleep[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
> > >> migration/8[58] 2 1 0.005 0.005 0.005 0.005 0.00 0
> > >> NetworkManager[811] 1 2 0.089 0.028 0.044 0.060 36.06 0
> > >> migration/13[83] 2 1 0.005 0.005 0.005 0.005 0.00 0
> > >>
> > >> Here 216753 is pid for sleep which is a zombie process. This is
> > >> happening in latest kernel due to an update in "-S" result.
> > >> In -S, the process name appears in the results "sleep[216753]",
> > >> where as in the -s, only pid is present in the summary result
> > >> ":216753[216753]".
> > >>
> > >> After commit 39f473f6d0b2 ("perf sched timehist: decode process names
> > >> of processes in zombie state")
> > >> for -S option, if process name is using pid, it uses different way to
> > >> set it. So that we get the process name and not just Pid.
> > >>
> > >> This change went in only for timehist_print_sample() function.
> > >> Add this improvement in generic place so that even -s option (which
> > >> captures summary) also will have meaningful information.
> > >>
> > >> Signed-off-by: Athira Rajeev <atrajeev@linux.ibm.com>
> > >
> > > Acked-by: Namhyung Kim <namhyung@kernel.org>
> > >
> > > Thanks,
> > > Namhyung
> > Hi,
> >
> > Can we please have this pulled in, if the patch looks fine ?
>
> Can you please check applying it on top of current perf-tools-next?
So, this seems to be also addressed by:
commit 39f473f6d0b24cf375893f2110b1cc9d8a079a42
Author: Anubhav Shelat <ashelat@redhat.com>
Date: Wed Jul 16 16:39:15 2025 -0400
perf sched timehist: decode process names of processes in zombie state
Previously when running perf trace timehist --state, when recording
processes in the zombie state the process name would not be decoded
properly and appears with just the PID:
1140057.412177 [0006] Mutter Input Th[3139/3104] 0.956 0.019 0.041 S
1140057.412222 [0012] :1248612[1248612] 0.000 0.000 0.332 Z
1140057.412275 [0004] <idle> 0.052 0.052 0.953 I
1140057.412284 [0008] <idle> 0.070 0.070 0.932 I
1140057.412333 [0004] KMS thread[3126/3104] 0.953 0.112 0.058 S
Now some extra processing has been added to decode the process name:
1140057.412177 [0006] Mutter Input Th[3139/3104] 0.956 0.019 0.041 S
1140057.412222 [0012] sleep[1248612] 0.000 0.000 0.332 Z
1140057.412275 [0004] <idle> 0.052 0.052 0.953 I
1140057.412284 [0008] <idle> 0.070 0.070 0.932 I
1140057.412333 [0004] KMS thread[3126/3104] 0.953 0.112 0.058 S
Signed-off-by: Anubhav Shelat <ashelat@redhat.com>
Link: https://lore.kernel.org/r/20250716203914.45772-2-ashelat@redhat.com
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
No? It is not applying to perf-tools-next, a quick look found the patch
above.
- Arnaldo
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH] tools/perf/sched: Update process names of processes in zombie state for both -s and -S options
2026-06-04 14:17 ` Arnaldo Carvalho de Melo
@ 2026-06-04 15:08 ` Athira Rajeev
2026-06-04 15:26 ` Arnaldo Carvalho de Melo
0 siblings, 1 reply; 8+ messages in thread
From: Athira Rajeev @ 2026-06-04 15:08 UTC (permalink / raw)
To: Arnaldo Carvalho de Melo
Cc: Anubhav Shelat, Namhyung Kim, Ian Rogers, jolsa, adrian.hunter,
mpetlan, tmricht, maddy, linux-perf-users, linuxppc-dev, hbathini,
Tejas.Manhas1, Tanushree.Shah, Shivani.Nittor
> On 4 Jun 2026, at 7:47 PM, Arnaldo Carvalho de Melo <acme@kernel.org> wrote:
>
> On Thu, May 21, 2026 at 11:17:58AM -0300, Arnaldo Carvalho de Melo wrote:
>> On Thu, May 21, 2026 at 02:02:53PM +0530, Athira Rajeev wrote:
>>>> On 27 Apr 2026, at 11:26 AM, Namhyung Kim <namhyung@kernel.org> wrote:
>>>> On Sun, Apr 26, 2026 at 03:09:30PM +0530, Athira Rajeev wrote:
>>>>> In redhat perftool testsuite, observed fail for this test:
>>>>> -- [ FAIL ] -- perf_sched :: test_timehist :: --with-summary (output regexp parsing)
>>>>>
>>>>> This led to analysis of "perf sched timehist" summary options.
>>>>>
>>>>> # perf sched record -a -o ./perf.data -- sleep 0.1
>>>>> This will record using perf sched record
>>>>>
>>>>> perf sched timeliest has two options "-s" and "-S"
>>>>> # perf sched -i ./perf.data timehist -S
>>>>> -S : Captures summary also at the end
>>>>>
>>>>> # perf sched -i ./perf.data timehist -s
>>>>> -s : Captures only summary
>>>>>
>>>>> The test saves -s result which has only summary and compares with
>>>>> summary which comes at the end from -S . Since there is a difference
>>>>> in these two, test fails.
>>>>>
>>>>> Checking the behaviour change in -S and -s results, difference is:
>>>>>
>>>>> rcu_sched[16] 2 4 0.013 0.001 0.003 0.006 33.23 0
>>>>> migration/11[73] 2 1 0.006 0.006 0.006 0.006 0.00 0
>>>>> migration/3[33] 2 1 0.006 0.006 0.006 0.006 0.00 0
>>>>> - :216753[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
>>>>> + sleep[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
>>>>> migration/8[58] 2 1 0.005 0.005 0.005 0.005 0.00 0
>>>>> NetworkManager[811] 1 2 0.089 0.028 0.044 0.060 36.06 0
>>>>> migration/13[83] 2 1 0.005 0.005 0.005 0.005 0.00 0
>>>>>
>>>>> Here 216753 is pid for sleep which is a zombie process. This is
>>>>> happening in latest kernel due to an update in "-S" result.
>>>>> In -S, the process name appears in the results "sleep[216753]",
>>>>> where as in the -s, only pid is present in the summary result
>>>>> ":216753[216753]".
>>>>>
>>>>> After commit 39f473f6d0b2 ("perf sched timehist: decode process names
>>>>> of processes in zombie state")
>>>>> for -S option, if process name is using pid, it uses different way to
>>>>> set it. So that we get the process name and not just Pid.
>>>>>
>>>>> This change went in only for timehist_print_sample() function.
>>>>> Add this improvement in generic place so that even -s option (which
>>>>> captures summary) also will have meaningful information.
>>>>>
>>>>> Signed-off-by: Athira Rajeev <atrajeev@linux.ibm.com>
>>>>
>>>> Acked-by: Namhyung Kim <namhyung@kernel.org>
>>>>
>>>> Thanks,
>>>> Namhyung
>>> Hi,
>>>
>>> Can we please have this pulled in, if the patch looks fine ?
>>
>> Can you please check applying it on top of current perf-tools-next?
>
> So, this seems to be also addressed by:
>
> commit 39f473f6d0b24cf375893f2110b1cc9d8a079a42
> Author: Anubhav Shelat <ashelat@redhat.com>
> Date: Wed Jul 16 16:39:15 2025 -0400
>
> perf sched timehist: decode process names of processes in zombie state
>
> Previously when running perf trace timehist --state, when recording
> processes in the zombie state the process name would not be decoded
> properly and appears with just the PID:
>
> 1140057.412177 [0006] Mutter Input Th[3139/3104] 0.956 0.019 0.041 S
> 1140057.412222 [0012] :1248612[1248612] 0.000 0.000 0.332 Z
> 1140057.412275 [0004] <idle> 0.052 0.052 0.953 I
> 1140057.412284 [0008] <idle> 0.070 0.070 0.932 I
> 1140057.412333 [0004] KMS thread[3126/3104] 0.953 0.112 0.058 S
>
> Now some extra processing has been added to decode the process name:
>
> 1140057.412177 [0006] Mutter Input Th[3139/3104] 0.956 0.019 0.041 S
> 1140057.412222 [0012] sleep[1248612] 0.000 0.000 0.332 Z
> 1140057.412275 [0004] <idle> 0.052 0.052 0.953 I
> 1140057.412284 [0008] <idle> 0.070 0.070 0.932 I
> 1140057.412333 [0004] KMS thread[3126/3104] 0.953 0.112 0.058 S
>
> Signed-off-by: Anubhav Shelat <ashelat@redhat.com>
> Link: https://lore.kernel.org/r/20250716203914.45772-2-ashelat@redhat.com
> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
>
>
> No? It is not applying to perf-tools-next, a quick look found the patch
> above.
Hi Arnaldo
commit 39f473f6d0b2 ("perf sched timehist: decode process names
of processes in zombie state”)
added change for -S option. The patch I submitted is to add change in process name for “-s” option as well
I will check applying this on top of current perf-tools-next
Thanks
Athira
>
> - Arnaldo
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH] tools/perf/sched: Update process names of processes in zombie state for both -s and -S options
2026-06-04 15:08 ` Athira Rajeev
@ 2026-06-04 15:26 ` Arnaldo Carvalho de Melo
0 siblings, 0 replies; 8+ messages in thread
From: Arnaldo Carvalho de Melo @ 2026-06-04 15:26 UTC (permalink / raw)
To: Athira Rajeev
Cc: Anubhav Shelat, Namhyung Kim, Ian Rogers, jolsa, adrian.hunter,
mpetlan, tmricht, maddy, linux-perf-users, linuxppc-dev, hbathini,
Tejas.Manhas1, Tanushree.Shah, Shivani.Nittor
On Thu, Jun 04, 2026 at 08:38:46PM +0530, Athira Rajeev wrote:
> > On 4 Jun 2026, at 7:47 PM, Arnaldo Carvalho de Melo <acme@kernel.org> wrote:
> >
> > On Thu, May 21, 2026 at 11:17:58AM -0300, Arnaldo Carvalho de Melo wrote:
> >> On Thu, May 21, 2026 at 02:02:53PM +0530, Athira Rajeev wrote:
> >>>> On 27 Apr 2026, at 11:26 AM, Namhyung Kim <namhyung@kernel.org> wrote:
> >>>> On Sun, Apr 26, 2026 at 03:09:30PM +0530, Athira Rajeev wrote:
> >>>>> In redhat perftool testsuite, observed fail for this test:
> >>>>> -- [ FAIL ] -- perf_sched :: test_timehist :: --with-summary (output regexp parsing)
> >>>>>
> >>>>> This led to analysis of "perf sched timehist" summary options.
> >>>>>
> >>>>> # perf sched record -a -o ./perf.data -- sleep 0.1
> >>>>> This will record using perf sched record
> >>>>>
> >>>>> perf sched timeliest has two options "-s" and "-S"
> >>>>> # perf sched -i ./perf.data timehist -S
> >>>>> -S : Captures summary also at the end
> >>>>>
> >>>>> # perf sched -i ./perf.data timehist -s
> >>>>> -s : Captures only summary
> >>>>>
> >>>>> The test saves -s result which has only summary and compares with
> >>>>> summary which comes at the end from -S . Since there is a difference
> >>>>> in these two, test fails.
> >>>>>
> >>>>> Checking the behaviour change in -S and -s results, difference is:
> >>>>>
> >>>>> rcu_sched[16] 2 4 0.013 0.001 0.003 0.006 33.23 0
> >>>>> migration/11[73] 2 1 0.006 0.006 0.006 0.006 0.00 0
> >>>>> migration/3[33] 2 1 0.006 0.006 0.006 0.006 0.00 0
> >>>>> - :216753[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
> >>>>> + sleep[216753] -1 1 0.041 0.041 0.041 0.041 0.00 0
> >>>>> migration/8[58] 2 1 0.005 0.005 0.005 0.005 0.00 0
> >>>>> NetworkManager[811] 1 2 0.089 0.028 0.044 0.060 36.06 0
> >>>>> migration/13[83] 2 1 0.005 0.005 0.005 0.005 0.00 0
> >>>>>
> >>>>> Here 216753 is pid for sleep which is a zombie process. This is
> >>>>> happening in latest kernel due to an update in "-S" result.
> >>>>> In -S, the process name appears in the results "sleep[216753]",
> >>>>> where as in the -s, only pid is present in the summary result
> >>>>> ":216753[216753]".
> >>>>>
> >>>>> After commit 39f473f6d0b2 ("perf sched timehist: decode process names
> >>>>> of processes in zombie state")
> >>>>> for -S option, if process name is using pid, it uses different way to
> >>>>> set it. So that we get the process name and not just Pid.
> >>>>>
> >>>>> This change went in only for timehist_print_sample() function.
> >>>>> Add this improvement in generic place so that even -s option (which
> >>>>> captures summary) also will have meaningful information.
> >>>>>
> >>>>> Signed-off-by: Athira Rajeev <atrajeev@linux.ibm.com>
> >>>>
> >>>> Acked-by: Namhyung Kim <namhyung@kernel.org>
> >>>>
> >>>> Thanks,
> >>>> Namhyung
> >>> Hi,
> >>>
> >>> Can we please have this pulled in, if the patch looks fine ?
> >>
> >> Can you please check applying it on top of current perf-tools-next?
> >
> > So, this seems to be also addressed by:
> >
> > commit 39f473f6d0b24cf375893f2110b1cc9d8a079a42
> > Author: Anubhav Shelat <ashelat@redhat.com>
> > Date: Wed Jul 16 16:39:15 2025 -0400
> >
> > perf sched timehist: decode process names of processes in zombie state
> >
> > Previously when running perf trace timehist --state, when recording
> > processes in the zombie state the process name would not be decoded
> > properly and appears with just the PID:
> >
> > 1140057.412177 [0006] Mutter Input Th[3139/3104] 0.956 0.019 0.041 S
> > 1140057.412222 [0012] :1248612[1248612] 0.000 0.000 0.332 Z
> > 1140057.412275 [0004] <idle> 0.052 0.052 0.953 I
> > 1140057.412284 [0008] <idle> 0.070 0.070 0.932 I
> > 1140057.412333 [0004] KMS thread[3126/3104] 0.953 0.112 0.058 S
> >
> > Now some extra processing has been added to decode the process name:
> >
> > 1140057.412177 [0006] Mutter Input Th[3139/3104] 0.956 0.019 0.041 S
> > 1140057.412222 [0012] sleep[1248612] 0.000 0.000 0.332 Z
> > 1140057.412275 [0004] <idle> 0.052 0.052 0.953 I
> > 1140057.412284 [0008] <idle> 0.070 0.070 0.932 I
> > 1140057.412333 [0004] KMS thread[3126/3104] 0.953 0.112 0.058 S
> >
> > Signed-off-by: Anubhav Shelat <ashelat@redhat.com>
> > Link: https://lore.kernel.org/r/20250716203914.45772-2-ashelat@redhat.com
> > Signed-off-by: Namhyung Kim <namhyung@kernel.org>
> >
> >
> > No? It is not applying to perf-tools-next, a quick look found the patch
> > above.
>
> Hi Arnaldo
>
> commit 39f473f6d0b2 ("perf sched timehist: decode process names
> of processes in zombie state”)
> added change for -S option. The patch I submitted is to add change in process name for “-s” option as well
>
> I will check applying this on top of current perf-tools-next
Thanks for looking into this!
- Arnaldo
^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2026-06-04 15:26 UTC | newest]
Thread overview: 8+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-04-26 9:39 [PATCH] tools/perf/sched: Update process names of processes in zombie state for both -s and -S options Athira Rajeev
2026-04-27 5:56 ` Namhyung Kim
2026-05-21 8:32 ` Athira Rajeev
2026-05-21 14:17 ` Arnaldo Carvalho de Melo
2026-06-04 14:17 ` Arnaldo Carvalho de Melo
2026-06-04 15:08 ` Athira Rajeev
2026-06-04 15:26 ` Arnaldo Carvalho de Melo
2026-04-27 6:43 ` Venkat
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox