Uh, I have a little problem here, too.
On Thu, Aug 24, 2023, 10:41 PM Yang Jihong <yangjihong1@xxxxxxxxxx <mailto:yangjihong1@xxxxxxxxxx>> wrote:
Hello,
On 2023/8/25 12:51, Ian Rogers wrote:
> On Sun, Aug 20, 2023 at 6:30 PM Yang Jihong
<yangjihong1@xxxxxxxxxx <mailto:yangjihong1@xxxxxxxxxx>> wrote:
>>
>> For dummy events that keep tracking, we may need to modify its
cpu_maps.
>> For example, change the cpu_maps to record sideband events for
all CPUS.
>> Add perf_evlist__go_system_wide() helper to support this scenario.
>>
>> Signed-off-by: Yang Jihong <yangjihong1@xxxxxxxxxx
<mailto:yangjihong1@xxxxxxxxxx>>
>> Acked-by: Adrian Hunter <adrian.hunter@xxxxxxxxx
<mailto:adrian.hunter@xxxxxxxxx>>
>> ---
>> tools/lib/perf/evlist.c | 9 +++++++++
>> tools/lib/perf/include/internal/evlist.h | 2 ++
>> 2 files changed, 11 insertions(+)
>>
>> diff --git a/tools/lib/perf/evlist.c b/tools/lib/perf/evlist.c
>> index b8b066d0dc5e..3acbbccc1901 100644
>> --- a/tools/lib/perf/evlist.c
>> +++ b/tools/lib/perf/evlist.c
>> @@ -738,3 +738,12 @@ int perf_evlist__nr_groups(struct
perf_evlist *evlist)
>> }
>> return nr_groups;
>> }
>> +
>> +void perf_evlist__go_system_wide(struct perf_evlist *evlist,
struct perf_evsel *evsel)
>> +{
>> + if (!evsel->system_wide) {
>> + evsel->system_wide = true;
>> + if (evlist->needs_map_propagation)
>> + __perf_evlist__propagate_maps(evlist,
evsel);
>> + }
>> +}
>
> I think this should be:
>
> void evsel__set_system_wide(struct evsel *evsel)
> {
> if (evsel->system_wide)
> return;
> evsel->system_wide = true;
> if (evsel->evlist->core.needs_map_propagation)
> ...
>
> The API being on evlist makes it look like all evsels are affected.
>
This part of the code is implemented according to Adrian's suggestion.
Refer to:
https://lore.kernel.org/linux-perf-users/206972a3-d44d-1c75-3fbc-426427614543@xxxxxxxxx/
<https://lore.kernel.org/linux-perf-users/206972a3-d44d-1c75-3fbc-426427614543@xxxxxxxxx/>
The logic of both is the same, and either is OK for me.
If really want to change it, please let me know.
Yes, I think the naming isn't correct and the function being on evlist is misleading.