Re: [PATCH v1 4/6] perf auxtrace: Iterate all AUX events when finish reading

[Leo Yan]

On 8/1/24 16:04, Arnaldo Carvalho de Melo wrote:
> On Wed, Jul 31, 2024 at 06:38:59PM -0700, Ian Rogers wrote:
> > Just a heads up. Arnaldo added this to tmp.perf-tools-next and it
> > caused the intel-pt tests to start failing:

Sorry for causing regression and thanks for reporting the issue.

> My plan right now is just to remove that cset that Ian bisected since it
> is not on perf-tools-next, just on the scratch branch
> tmp.perf-tools-next.

Please leave the patch 04 out. I have sent out v2 but Adrian pointed out a 
concern for per-thread mode, now I am working on this and after ready I will 
send new patches.

I have verified the latest perf-tools-next branch, it works well on Arm SPE 
(thanks for picking up patches 05 and 06).

> Trying to do that now as it will help us with bisection in the future.

I am not clear this. Could you elaborate a bit what I should follow up?

> > ```
> > $ perf test 118 -v
> > 118: Miscellaneous Intel PT testing:
> > --- start ---
> > test child forked, pid 148999
> > --- Test system-wide sideband ---
> > Checking for CPU-wide recording on CPU 0
> > OK
> > Checking for CPU-wide recording on CPU 1
> > OK
> > Linux
> > Failed to enable event (idx=0): -22
> > Failed to record MMAP events on CPU 1 when tracing CPU 0
> > ...
> > ```
> > It's likely Adrian's comments already address this but you may also
> > want to double check this test is passing with v2.

Sure. I will give a test for this test when I send new patches.

Thanks,
Leo