[llvm] [AMDGPU] (x or y) xor -1 -> x nor y (PR #130264)

[Petar Avramovic via llvm-commits]

================
@@ -0,0 +1,115 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 3
+; RUN: llc -mtriple=amdgcn-amd-amdpal -mcpu=gfx1200 -mattr="+wavefrontsize32,-wavefrontsize64" -o - < %s | FileCheck -check-prefix=GFX12W32 %s
+; RUN: llc -mtriple=amdgcn-amd-amdpal -mcpu=gfx1100 -mattr="+wavefrontsize32,-wavefrontsize64" -o - < %s | FileCheck -check-prefixes=GFX11W32 %s
+
+define amdgpu_ps i32 @test_w32(i32 %x, i32 %y) {
----------------
petar-avramovic wrote:

%x and %y should be inreg, inverse.ballot.i32 arguments are uniform i32/i64

https://github.com/llvm/llvm-project/pull/130264