Re: [PATCH] perf cs-etm: Delay decode of non-timeless data until cs_etm__flush_events()

From: Arnaldo Carvalho de Melo
Date: Thu Jul 01 2021 - 13:15:32 EST


Em Mon, Jun 14, 2021 at 09:54:42AM -0600, Mathieu Poirier escreveu:
> On Wed, Jun 09, 2021 at 04:04:20PM +0300, James Clark wrote:
> > Currently, timeless mode starts the decode on PERF_RECORD_EXIT, and
> > non-timeless mode starts decoding on the fist PERF_RECORD_AUX record.
> >
> > This can cause the "data has no samples!" error if the first
> > PERF_RECORD_AUX record comes before the first (or any relevant)
> > PERF_RECORD_MMAP2 record because the mmaps are required by the decoder
> > to access the binary data.
> >
> > This change pushes the start of non-timeless decoding to the very end of
> > parsing the file. The PERF_RECORD_EXIT event can't be used because it
> > might not exist in system-wide or snapshot modes.

<SNIP>

> This looks good to me but I'd like to have other people testing it.

So Leo reviewed and tested this, I'm taking your "looks good to me" as
an Acked-by, as per Documentation/process/submitting-patches.rst.

- Arnaldo

> Thanks,
> Mathieu