Re: [PATCH sched_ext for-7.4-fixes v2] selftests/sched_ext: Fix flaky rt_stall measurement

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential
issue to consider:
- [Low] selftests/sched_ext: leak of BPF skeleton in setup() on affinity
  error
--

commit 24932bf17cf4d833030f3c4b3de4fdbf28e3d142
Author: Jonathan Bastari Thamrin <thamrinjonathan@gmail.com>

selftests/sched_ext: Fix flaky rt_stall measurement

This commit fixes flaky runtime measurements in the rt_stall selftest by
measuring per-task CPU clock deltas around the sleep window and pinning the
parent process off the test CPU.

Link: https://github.com/kernel-patches/vmtest/issues/458

> diff --git a/tools/testing/selftests/sched_ext/rt_stall.c b/tools/testing/selftests/sched_ext/rt_stall.c
> index a1552b38a6e8c..c5027fd6b6820 100644
> --- a/tools/testing/selftests/sched_ext/rt_stall.c
> +++ b/tools/testing/selftests/sched_ext/rt_stall.c

[ ... ]

> @@ -132,6 +149,9 @@ static enum scx_test_status setup(void **ctx)
>  
>  	*ctx = skel;
>  
> +	err = pin_parent_off_core();
> +	SCX_FAIL_IF(err == -1, "Failed to pin parent CPU");
> +
>  	return SCX_TEST_PASS;
>  }

[Severity: Low]
Does this leak skel if pin_parent_off_core() fails?  run_test() returns
immediately when test->setup() does not return SCX_TEST_PASS without
calling test->cleanup(), leaving the opened and loaded BPF skeleton
undestroyed across subsequent tests in the runner process.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261115-scx-rt-stall-v2-1-926678b214c6@gmail.com?part=1