[llvm] [AMDGPU] Set MaxAtomicSizeInBitsSupported. (PR #75185)

Matt Arsenault via llvm-commits llvm-commits at lists.llvm.org
Mon Dec 18 07:11:24 PST 2023


arsenm wrote:

What's the point of doing this? There aren't any lib calls, and the lib call handling is unaware of the address space anyway. This is just changing a hard error into producing nonworking code 

https://github.com/llvm/llvm-project/pull/75185


More information about the llvm-commits mailing list