[libcxx-commits] [libcxx] [libcxx] Adjust inline assembly constraints for the AMDGPU target (PR #101747)

[Louis Dionne via libcxx-commits]

https://github.com/ldionne approved this pull request.

We should really have a builtin like `__builtin_escape(variable)` that implements this logic instead. But I don't want to block this PR on that.

https://github.com/llvm/llvm-project/pull/101747