[llvm] [AMDGPU] Inplace FI elimination during PEI for scalar copy instruction (PR #99556)

[Matt Arsenault via llvm-commits]

================
@@ -211,3 +211,327 @@ body:             |
     renamable $vgpr0 = V_ADD_CO_U32_e32 %stack.0, killed $vgpr0, implicit-def dead $vcc, implicit $exec
 ...
 
+---
----------------
arsenm wrote:

In a pre-commit, can you generate GFX8 and GFX9 checks in this test 

https://github.com/llvm/llvm-project/pull/99556