From: Adrian Hunter <adrian.hunter@intel.com>
To: Namhyung Kim <namhyung@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>,
Jiri Olsa <jolsa@redhat.com>, Ian Rogers <irogers@google.com>,
Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>,
Leo Yan <leo.yan@linaro.org>,
linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH V2 22/23] perf tools: Allow system-wide events to keep their own CPUs
Date: Thu, 12 May 2022 13:34:47 +0300 [thread overview]
Message-ID: <52f201d4-4fab-c2f4-9c4d-af887900732f@intel.com> (raw)
In-Reply-To: <CAM9d7cg8JRK2oVUvmWit=F5zhLhpqP=gD6iYBAa8_O-+c=EjPQ@mail.gmail.com>
On 12/05/22 08:27, Namhyung Kim wrote:
> On Fri, May 6, 2022 at 5:27 AM Adrian Hunter <adrian.hunter@intel.com> wrote:
>>
>> Currently, user_requested_cpus supplants system-wide CPUs when the evlist
>> has_user_cpus. Change that so that system-wide events retain their own
>> CPUs and they are added to all_cpus.
>>
>> Acked-by: Ian Rogers <irogers@google.com>
>> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
>> ---
>> tools/lib/perf/evlist.c | 11 +++++------
>> 1 file changed, 5 insertions(+), 6 deletions(-)
>>
>> diff --git a/tools/lib/perf/evlist.c b/tools/lib/perf/evlist.c
>> index 1c801f8da44f..9a6801b53274 100644
>> --- a/tools/lib/perf/evlist.c
>> +++ b/tools/lib/perf/evlist.c
>> @@ -40,12 +40,11 @@ static void __perf_evlist__propagate_maps(struct perf_evlist *evlist,
>> * We already have cpus for evsel (via PMU sysfs) so
>> * keep it, if there's no target cpu list defined.
>> */
>> - if (!evsel->own_cpus || evlist->has_user_cpus) {
>> - perf_cpu_map__put(evsel->cpus);
>> - evsel->cpus = perf_cpu_map__get(evlist->user_requested_cpus);
>> - } else if (!evsel->system_wide &&
>> - !evsel->requires_cpu &&
>> - perf_cpu_map__empty(evlist->user_requested_cpus)) {
>> + if (!evsel->own_cpus ||
>> + (!evsel->system_wide && evlist->has_user_cpus) ||
>> + (!evsel->system_wide &&
>> + !evsel->requires_cpu &&
>> + perf_cpu_map__empty(evlist->user_requested_cpus))) {
>
> This is getting hard to understand. IIUC this propagation basically
> sets user requested cpus to evsel unless it has its own cpus, right?
I put the conditional logic altogether because that is kernel style but
it does make it practically unreadable.
If we start with the original logic:
if (!evsel->own_cpus || evlist->has_user_cpus) {
perf_cpu_map__put(evsel->cpus);
evsel->cpus = perf_cpu_map__get(evlist->user_requested_cpus);
} else if (!evsel->system_wide && perf_cpu_map__empty(evlist->user_requested_cpus)) {
perf_cpu_map__put(evsel->cpus);
evsel->cpus = perf_cpu_map__get(evlist->user_requested_cpus);
} else if (evsel->cpus != evsel->own_cpus) {
perf_cpu_map__put(evsel->cpus);
evsel->cpus = perf_cpu_map__get(evsel->own_cpus);
}
Then make it more readable, i.e. same functionality
struct perf_cpu_map *cpus;
if (!evsel->own_cpus || evlist->has_user_cpus)
cpus = evlist->user_requested_cpus;
else if (!evsel->system_wide && perf_cpu_map__empty(evlist->user_requested_cpus))
cpus = evlist->user_requested_cpus;
else
cpus = evsel->own_cpus;
if (evsel->cpus != cpus) {
perf_cpu_map__put(evsel->cpus);
evsel->cpus = perf_cpu_map__get(cpus);
}
Then separate out the conditions, i.e. still same functionality
if (!evsel->own_cpus)
cpus = evlist->user_requested_cpus;
else if (evlist->has_user_cpus)
cpus = evlist->user_requested_cpus;
else if (evsel->system_wide)
cpus = evsel->own_cpus;
else if (perf_cpu_map__empty(evlist->user_requested_cpus)) /* per-thread */
cpus = evlist->user_requested_cpus;
else
cpus = evsel->own_cpus;
Then add the new requires_cpu flag:
if (!evsel->own_cpus)
cpus = evlist->user_requested_cpus;
else if (evlist->has_user_cpus)
cpus = evlist->user_requested_cpus;
else if (evsel->system_wide)
cpus = evsel->own_cpus;
- else if (perf_cpu_map__empty(evlist->user_requested_cpus)) /* per-thread */
+ else if (!evsel->requres_cpu && perf_cpu_map__empty(evlist->user_requested_cpus)) /* per-thread */
cpus = evlist->user_requested_cpus;
else
cpus = evsel->own_cpus;
Then make system_wide keep own_cpus even if has_user_cpus:
if (!evsel->own_cpus)
cpus = evlist->user_requested_cpus;
+ else if (evsel->system_wide)
+ cpus = evsel->own_cpus;
else if (evlist->has_user_cpus)
cpus = evlist->user_requested_cpus;
- else if (evsel->system_wide)
- cpus = evsel->own_cpus;
else if (!evsel->requres_cpu && perf_cpu_map__empty(evlist->user_requested_cpus)) /* per-thread */
cpus = evlist->user_requested_cpus;
else
cpus = evsel->own_cpus;
Which leaves:
if (!evsel->own_cpus)
cpus = evlist->user_requested_cpus;
else if (evsel->system_wide)
cpus = evsel->own_cpus;
else if (evlist->has_user_cpus)
cpus = evlist->user_requested_cpus;
else if (!evsel->requres_cpu && perf_cpu_map__empty(evlist->user_requested_cpus)) /* per-thread */
cpus = evlist->user_requested_cpus;
else
cpus = evsel->own_cpus;
And putting it back together:
if (!evsel->own_cpus ||
(!evsel->system_wide && evlist->has_user_cpus) ||
(!evsel->system_wide &&
!evsel->requires_cpu &&
perf_cpu_map__empty(evlist->user_requested_cpus))) {
cpus = evlist->user_requested_cpus;
else
cpus = evsel->own_cpus;
Perhaps I shouldn't put it together?
>
> But the hybrid pmus make this complex. Maybe we can move the
> logic in evlist__fix_hybrid_cpus() here and simplify it like below
>
> if (evsel->own_cpus) {
> if (evsel->pmu->is_hybrid)
> evsel->cpus = fixup_hybrid_cpus(evsel>own_cpus,
> evlist->user_requested_cpus); //?
> else
> evsel->cpus = evlist->own_cpus; // put + get
> } else {
> evsel->cpus = evlist->user_requested_cpus; // put + get
> }
>
> Then we need to make sure evsel->pmu is set properly.
>
> What do you think?
Hybrid handling looks complicated. I would have to spend time
better understanding it.
So, in the context of this patch set, I don't want to look at
issues with hybrid CPUs, except that there should be no change
to how they are handled.
>
> Thanks,
> Namhyung
>
>
>> perf_cpu_map__put(evsel->cpus);
>> evsel->cpus = perf_cpu_map__get(evlist->user_requested_cpus);
>> } else if (evsel->cpus != evsel->own_cpus) {
>> --
>> 2.25.1
>>
next prev parent reply other threads:[~2022-05-12 10:35 UTC|newest]
Thread overview: 70+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-05-06 12:25 [PATCH V2 00/23] perf intel-pt: Better support for perf record --cpu Adrian Hunter
2022-05-06 12:25 ` [PATCH V2 01/23] perf intel-pt: Add a test for system-wide side band Adrian Hunter
2022-05-10 17:18 ` Arnaldo Carvalho de Melo
2022-05-10 17:21 ` Adrian Hunter
2022-05-06 12:25 ` [PATCH V2 02/23] libperf evsel: Add perf_evsel__enable_thread() Adrian Hunter
2022-05-06 17:06 ` Ian Rogers
2022-05-10 17:19 ` Arnaldo Carvalho de Melo
2022-05-06 12:25 ` [PATCH V2 03/23] perf evlist: Use libperf functions in evlist__enable_event_idx() Adrian Hunter
2022-05-10 17:23 ` Arnaldo Carvalho de Melo
2022-05-06 12:25 ` [PATCH V2 04/23] perf auxtrace: Move evlist__enable_event_idx() to auxtrace.c Adrian Hunter
2022-05-10 17:24 ` Arnaldo Carvalho de Melo
2022-05-06 12:25 ` [PATCH V2 05/23] perf auxtrace: Do not mix up mmap idx Adrian Hunter
2022-05-10 17:25 ` Arnaldo Carvalho de Melo
2022-05-06 12:25 ` [PATCH V2 06/23] libperf evlist: Remove ->idx() per_cpu parameter Adrian Hunter
2022-05-10 17:26 ` Arnaldo Carvalho de Melo
2022-05-06 12:25 ` [PATCH V2 07/23] libperf evlist: Move ->idx() into mmap_per_evsel() Adrian Hunter
2022-05-10 17:26 ` Arnaldo Carvalho de Melo
2022-05-06 12:25 ` [PATCH V2 08/23] libperf evlist: Add evsel as a parameter to ->idx() Adrian Hunter
2022-05-10 17:26 ` Arnaldo Carvalho de Melo
2022-05-06 12:25 ` [PATCH V2 09/23] perf auxtrace: Record whether an auxtrace mmap is needed Adrian Hunter
2022-05-10 17:27 ` Arnaldo Carvalho de Melo
2022-05-06 12:25 ` [PATCH V2 10/23] perf auxtrace: Add mmap_needed to auxtrace_mmap_params Adrian Hunter
2022-05-06 20:16 ` Ian Rogers
2022-05-11 7:02 ` Adrian Hunter
2022-05-11 7:01 ` [PATCH V3 " Adrian Hunter
2022-05-06 12:25 ` [PATCH V2 11/23] perf auxtrace: Remove auxtrace_mmap_params__set_idx() per_cpu parameter Adrian Hunter
2022-05-06 20:19 ` Ian Rogers
2022-05-06 12:25 ` [PATCH V2 12/23] perf evlist: Factor out evlist__dummy_event() Adrian Hunter
2022-05-06 12:25 ` [PATCH V2 13/23] perf evlist: Add evlist__add_dummy_on_all_cpus() Adrian Hunter
2022-05-06 13:47 ` Ian Rogers
2022-05-06 15:07 ` Adrian Hunter
2022-05-06 15:35 ` Ian Rogers
2022-05-10 14:55 ` Adrian Hunter
2022-05-10 16:19 ` Ian Rogers
2022-05-10 16:24 ` Arnaldo Carvalho de Melo
2022-05-10 17:32 ` Adrian Hunter
2022-05-11 7:02 ` [PATCH V3 " Adrian Hunter
2022-05-11 22:50 ` Namhyung Kim
2022-05-12 4:33 ` Adrian Hunter
2022-05-12 5:01 ` Namhyung Kim
2022-05-06 12:25 ` [PATCH V2 14/23] perf record: Use evlist__add_dummy_on_all_cpus() in record__config_text_poke() Adrian Hunter
2022-05-06 12:25 ` [PATCH V2 15/23] perf intel-pt: Use evlist__add_dummy_on_all_cpus() for switch tracking Adrian Hunter
2022-05-06 12:25 ` [PATCH V2 16/23] perf intel-pt: Track sideband system-wide when needed Adrian Hunter
2022-05-06 12:25 ` [PATCH V2 17/23] perf tools: Allow all_cpus to be a superset of user_requested_cpus Adrian Hunter
2022-05-06 12:25 ` [PATCH V2 18/23] libperf evlist: Allow mixing per-thread and per-cpu mmaps Adrian Hunter
2022-05-06 12:25 ` [PATCH V2 19/23] libperf evlist: Check nr_mmaps is correct Adrian Hunter
2022-05-06 20:20 ` Ian Rogers
2022-05-06 12:25 ` [PATCH V2 20/23] perf stat: Add requires_cpu flag for uncore Adrian Hunter
2022-05-06 12:25 ` [PATCH V2 21/23] libperf evsel: Add comments for booleans Adrian Hunter
2022-05-06 20:51 ` Ian Rogers
2022-05-11 7:03 ` Adrian Hunter
2022-05-12 5:34 ` Ian Rogers
2022-05-12 11:40 ` Adrian Hunter
2022-05-06 12:26 ` [PATCH V2 22/23] perf tools: Allow system-wide events to keep their own CPUs Adrian Hunter
2022-05-12 5:27 ` Namhyung Kim
2022-05-12 10:34 ` Adrian Hunter [this message]
2022-05-12 18:53 ` Namhyung Kim
2022-05-13 4:48 ` Adrian Hunter
2022-05-13 14:12 ` Liang, Kan
2022-05-13 15:21 ` Adrian Hunter
2022-05-13 15:46 ` Liang, Kan
2022-05-13 16:11 ` Adrian Hunter
2022-05-13 16:42 ` Namhyung Kim
2022-05-13 17:32 ` Liang, Kan
2022-05-14 13:35 ` Arnaldo Carvalho de Melo
2022-05-17 23:31 ` Namhyung Kim
2022-05-06 12:26 ` [PATCH V2 23/23] perf tools: Allow system-wide events to keep their own threads Adrian Hunter
2022-05-08 15:08 ` [PATCH V2 00/23] perf intel-pt: Better support for perf record --cpu Leo Yan
2022-05-09 5:44 ` Adrian Hunter
2022-05-09 8:46 ` Leo Yan
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=52f201d4-4fab-c2f4-9c4d-af887900732f@intel.com \
--to=adrian.hunter@intel.com \
--cc=acme@kernel.org \
--cc=alexey.v.bayduraev@linux.intel.com \
--cc=irogers@google.com \
--cc=jolsa@redhat.com \
--cc=leo.yan@linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=namhyung@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®