[llvm] [AMDGPU] Add hazard workarounds to insertIndirectBranch (PR #109127)

Carl Ritson via llvm-commits llvm-commits at lists.llvm.org
Wed Sep 18 06:03:25 PDT 2024


perlfu wrote:

> we probably could run branch relaxation before the post RA scheduler

Presumably this would cause problems if new hazard related instructions pushed a branch over the representable size?

https://github.com/llvm/llvm-project/pull/109127


More information about the llvm-commits mailing list