On Nov 20, 2023, Jason Merrill <ja...@redhat.com> wrote: > So it only passed on that platform before because of the bug?
I'm not sure it passed (we've had patches for that testcase before), but I didn't look closely enough into their history to tell. I suspected the warning suppression machinery changed, or details on cloning did, or something. It's been fragile historically. But yeah, recently, the test for the warning was only passing because of the bug. But we were also getting excess warnings, so it wasn't fully passing. Thanks for the reviews! -- Alexandre Oliva, happy hacker https://FSFLA.org/blogs/lxo/ Free Software Activist GNU Toolchain Engineer More tolerance and less prejudice are key for inclusion and diversity Excluding neuro-others for not behaving ""normal"" is *not* inclusive
