On Sun, Apr 20, 2025 at 05:48:15PM -0400, Waiman Long <llong@redhat.com> wrote:

I was referring to the suggestion that the setting of memory_recursiveprot
mount option has a material impact of the child 2 test result. Roman
probably didn't have memory_recursiveprot set when developing this selftest.

The patch in its v7 form is effectively a revert of
	1d09069f5313f ("selftests: memcg: expect no low events in unprotected sibling")

i.e. this would be going in circles (that commit is also a revert) hence
I suggested to exempt looking at memory.events:low entirely with
memory_recursiveprot (and check for 0 when !memory_recursiveprot) --
which is something new (and hopefully universally better :-)

Ah, you had done a lot of work for the test_memcontrol selftest. Am I correct to assume that the purpose of 1d09069f5313f ("selftests: memcg: expect no low events in unprotected sibling") is to force a failure in the test_memcg_low test to force a change in the current behavior? Or was it the case that it didn't fail when you submit your patch?

Cheers, Longman