boomanaiden154 wrote: > Somewhat off-topic for this PR, but why can't REQUIRES: shell force use of > shell even if internal shell is the default? Then we don't lose test coverage > and don't block this on migration of the long-tail of tests.
That's another option. The long tail isn't actually that long though. There are only a handful of tests left in the monorepo that don't pass with the internal shell (~10 for clang, similar numbers for compiler-rt as far as I'm aware, PRs are already up for all the LLVM tests). Most of the tests don't require any complicated adjustments to work with the internal shell and end up cleaner (or at least I would argue) after being ported. https://github.com/llvm/llvm-project/pull/157232 _______________________________________________ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
