[llvm] [AMDGPU][GFX12] VOP encoding and codegen - add support for v_cvt fp8/… (PR #78414)

[Mariusz Sikora via llvm-commits]

================
@@ -1495,14 +1495,16 @@ def FeatureISAVersion12 : FeatureSet<
    FeatureFlatAtomicFaddF32Inst,
    FeatureImageInsts,
    FeatureExtendedImageInsts,
+   FeatureFP8Insts,
----------------
mariusz-sikora-at-amd wrote:

This feature is not limited only to conversion instructions. I see that it is used also for __builtin_amdgcn_smfmac_f32_32x32x32_bf8_bf8

https://github.com/llvm/llvm-project/pull/78414