On Feb 3, 2026, Jeffrey Law <[email protected]> wrote: > Because with Zbs in the arch string actually tests for a real bug we > had in the backend.
That's clear enough. I expected you'd run tests for various arches and boards, including some with that option and some without, so v2 would give us all higher coverage and less redundancy. You seem to prefer the test to have a narrow focus on the regression aspect for that specific architecture. That means you prefer v3. I'd be happy with either one. I don't really have a horse on this race, I just want this buggy test that causes a noisy failure to get out of my radar ;-) > Maybe part of the problem here is a misunderstanding of how codegen No misunderstanding, it was just an expectation that others would test with a variety of TOOL_OPTIONS, target_list or whatever, like we do. -- Alexandre Oliva, happy hacker https://blog.lx.oliva.nom.br/ Free Software Activist FSFLA co-founder GNU Toolchain Engineer More tolerance and less prejudice are key for inclusion and diversity. Excluding neuro-others for not behaving ""normal"" is *not* inclusive!
