[llvm-branch-commits] [llvm] [AMDGPU] Commute instructions to avoid VGPR MSB changes (PR #181918)

[Stanislav Mekhanoshin via llvm-branch-commits]

rampitec wrote:

I think it can be also further extended to ignore unused bits in comparison, i.e. for VOP2 commute and ignored src2. Although not in the same patch.

https://github.com/llvm/llvm-project/pull/181918