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

[Matt Arsenault via llvm-branch-commits]

arsenm wrote:

> Yeah, this limit is an implementation detail of the hardware. Though, I don't see it changing from 1024 anytime soon.

The default is the hardware limit, the point of the flag is to decrease the limit to what you want. The only interaction to sort out is whatever explicit openmp annotations there are 

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