[llvm-branch-commits] [clang] clang/AMDGPU: Remove artificial restriction on --gpu-max-threads-per-block (PR #204864)

[Shilei Tian via llvm-branch-commits]

shiltian wrote:

> It turns out there is a warning that it's hip only in CompilerInvocation

If we enable this regardless, we definitely need to remove the warning. I'd expect the code generation is to attach target dependent attributes.

https://github.com/llvm/llvm-project/pull/204864