[llvm] [AMDGPU] Register allocation anti-hints to reduce MFMA hazard NOPs (PR #156943)

[Matt Arsenault via llvm-commits]

================
@@ -0,0 +1,177 @@
+# RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx942 -debug -run-pass=greedy,machineverifier,virtregrewriter %s -o - | FileCheck -check-prefix=CHECK %s
+--- |
+  ; ModuleID = '/work/mdssefat/FullTimeWork/MLSCHED/composable_kernel/noopexample/llvm.amdgcn.mfma.hint.haard.barrier.gfx942_short.mir'
+  source_filename = "/work/mdssefat/FullTimeWork/MLSCHED/composable_kernel/noopexample/llvm.amdgcn.mfma.hint.haard.barrier.gfx942_short.mir"
+  target datalayout = "e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-p7:160:256:256:32-p8:128:128:128:48-p9:192:256:256:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-S32-A5-G1-ni:7:8:9"
+  target triple = "amdgcn-amd-amdhsa"
+
+  ; Function Attrs: nounwind
----------------
arsenm wrote:

```suggestion
```

https://github.com/llvm/llvm-project/pull/156943