Re: [RFC PATCH 1/1] perf/core: Wake up parent event if inherited event has no ring buffer

[Ruben Ayrapetyan]

Hi everyone,

I've tried running a benchmark with and without this patch applied, with the following options:
 perf record -e arm_spe_0/event_filter=0,jitter=0,ts_enable=1,pa_enable=0,min_latency=0/ ...
repeating it 60 times with and without the patch.

Without the patch, the output perf.data sizes were: 1.0G 1.0G 1.0G 1.0G 1.0G 1.0G 1.0G 1.0G 47M 48M 48M 48M 48M 48M 48M 48M 48M 48M 49M 49M 49M 49M 49M 49M 49M 49M 49M 49M 49M 49M 49M 49M 49M 49M 49M 60M 60M 60M 60M 61M 61M 61M 61M 61M 61M 61M 62M 62M 62M 62M 62M 62M 62M 62M 62M 63M 63M 63M 65M 990M i.e. 47MB to 1.0GB

With the patch, the output sizes were: 1.4G 1.4G 1.4G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G 1.5G i.e. 1.4GB to 1.5GB.

Tested-by: Ruben Ayrapetyan <Ruben.Ayrapetyan@xxxxxxx>
IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.