Re: [linus:master] [perf parse] a4c7d7c502: perf-test.perf_pmu_event_count_group.count_group_cpu/branch-misses/_cpu/topdown-heavy-ops/_cpu/branch-misses/_R0.fail

[Ian Rogers]

On Mon, May 29, 2023 at 12:30 AM kernel test robot
<oliver.sang@xxxxxxxxx> wrote:
>
>
>
> Hello,
>
> kernel test robot noticed "perf-test.perf_pmu_event_count_group.count_group_cpu/branch-misses/_cpu/topdown-heavy-ops/_cpu/branch-misses/_R0.fail" on:
>
> commit: a4c7d7c502b935f3a8324d954de78aecf6940897 ("perf parse-events: Warn when events are regrouped")
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master

I'm having a hard time understanding the output from the test:

1) the output is filled with sanitizer output that makes it look like
a non-asan version of libc is being used, either that or we're missing
asan interceptors for functions called by perf. This means the noise
is much greater than the signal in the output.

2) the referenced test doesn't appear to be part of the "perf test"
set of tests. Are these additional tests that Intel have created?
Could we get them moved to "perf test" which is used as a pre-submit
and can help avoid regressions.

Why a warning would cause this test to start failing I'm not sure.
Given the amount of difficulty the harness has with asan I think
there's a good chance this is a false positive. It could also be that
the test expectations have changed and could have been fixed, which
would break external tests that make assumptions on things like
events. Moving the test to "perf test" would be the best way to
resolve this.

Thanks,
Ian

> [test failed on linux-next/master 6a3d37b4d885129561e1cef361216f00472f7d2e]
>
> in testcase: perf-test
> version: perf-test-x86_64-git-1_20220520
> with following parameters:
>
>         type: lkp
>         group: group-00
>
> test-description: The internal Perf Test suite.
>
>
> compiler: gcc-11
> test machine: 224 threads 2 sockets Intel(R) Xeon(R) Platinum 8480+ (Sapphire Rapids) with 256G memory
>
> (please refer to attached dmesg/kmsg for entire log/backtrace)
>
>
>
>
> If you fix the issue, kindly add following tag
> | Reported-by: kernel test robot <oliver.sang@xxxxxxxxx>
> | Closes: https://lore.kernel.org/oe-lkp/202305291525.df3ce28d-oliver.sang@xxxxxxxxx
>
>
>
> perf kernel cpu PMU event count group test
> Sun May 28 06:51:04 UTC 2023
> 1 test cases pass for perf_pmu_event_count_group test. 1 test cases fail for perf_pmu_event_count_group test.
> Test Case count_group_cpu/topdown-fetch-lat/_cpu/mem-loads-aux/_cpu/branch-instructions/_R1 PASS!
> Test Case count_group_cpu/branch-misses/_cpu/topdown-heavy-ops/_cpu/branch-misses/_R0 FAILED!
>
>
>
> To reproduce:
>
>         git clone https://github.com/intel/lkp-tests.git
>         cd lkp-tests
>         sudo bin/lkp install job.yaml           # job file is attached in this email
>         bin/lkp split-job --compatible job.yaml # generate the yaml file for lkp run
>         sudo bin/lkp run generated-yaml-file
>
>         # if come across any failure that blocks the test,
>         # please remove ~/.lkp and /lkp dir to run from a clean state.
>
>
>
> --
> 0-DAY CI Kernel Test Service
> https://github.com/intel/lkp-tests/wiki
>
>