[llvm-branch-commits] [llvm] [AMDGPU] Set MOThreadPrivate on memory accesses for spills (PR #179414)

[Pierre van Houtryve via llvm-branch-commits]

Pierre-vh wrote:

> The patch description and even the comments say "non volatile", but the code actually sets `MOThreadPrivate`. In general, I would prefer to move away from the term "volatile" and its equally unwieldy negation. In this patch, is it correct to just say "thread private" everywhere?

Yup, I just forgot to update the commit description. I did it now

https://github.com/llvm/llvm-project/pull/179414