[llvm-branch-commits] [llvm] [AMDGPU] Introduce a "new" target feature `xf32-insts` (PR #115214)

[Matt Arsenault via llvm-branch-commits]

================
@@ -1110,6 +1110,13 @@ def FeatureRequiresCOV6 : SubtargetFeature<"requires-cov6",
   "Target Requires Code Object V6"
 >;
 
+def FeatureXF32Insts : SubtargetFeature<"xf32-insts",
+   "HasXF32Insts",
+   "true",
+   "Has instructions that support xf32 format, such as "
+   "v_mfma_f32_16x16x8xf32 and v_mfma_f32_32x32x4xf32"
----------------
arsenm wrote:

We should fix the canonical names to have the underscore. I don't know what they were thinking when they published the instruction names without it 

https://github.com/llvm/llvm-project/pull/115214