Is SANITIZE=leak make test unreliable for anyone else?

"Eric W. Biederman" <ebiederm@xxxxxxxxxxxx> · Tue, 03 Oct 2023 20:33:26 -0500

Jeff,

Please accept my apologies for slightly hijacking your posting, but I
see you have been fixing some leaks, and so presumably you are familiar
with building git with "SANITIZE=leak".

I have fixed some leaks in my SHA1+SHA256 patchset recently and while
tracking them down I found that simply enabling SANITIZE=leak caused
"make test" on git v2.42 without patches to give different failures
from test run to test run.

Well actually I wound up with the following command line:
GIT_TEST_PASSING_SANITIZE_LEAK=true GIT_TEST_SANITIZE_LEAK_LOG=true SANITIZE=leak DEVELOPER=1 make test

I had removed "-j32" to make things more reproducible.

I observed this unreliability with SANITIZE=leak when building git on
an fully updated version of debian 12.

My big question is:

    Do other people see random test failures when SANITIZE=leak is enabled?

Is it just me?

Thanks,
Eric