[llvm] [AMDGPU] Try to reuse in v_cndmask register with constant from compare. (PR #131146)

[Daniil Fukalov via llvm-commits]

dfukalov wrote:

> Can we do this earlier, before selection? The same problem exists for the scalar case

I tried to detect these cases and revert the optimisation in amdgpu-codegenprepare, but realised that in some cases (e.g. when immediate can be encoded into instruction, depending on the target and the constant) it can increase registers usage (as well as instructions count).

https://github.com/llvm/llvm-project/pull/131146