On Tue, May 18, 2021 at 10:00:40PM +0800, Leo Yan wrote:
[...]
We could probably utilize the ETM strobing feature and reduce frequency of data collection but I see a problem when I'm using both. Within a minute of profiling the ETM generates a reasonable profile size (with strobing autofdo,preset=9 with period 0x1000 it is up to 20MB). But then the size grows unproportionally. With a 4 minute run I got a 6.3GB profile.
I don't see such a problem with the ETR polling patch.
Leo, could you please take a look at this problem?
Or in the other way, could you share the perf data files?
Base on the perf data files, I think at least we can confirm what's the AUX trace data size for every period, and we might also can quickly confirm if the strobing configurations are lost or not in the middle of tracing (To be honest, I have no knowledge for strobing, so need Mike's help for confirmation).
Thanks, Leo