[Mlir-commits] [mlir] [MLIR][AMDGPU] Add OCP FP8 support for new hardware (PR #106160)

[Krzysztof Drewniak]

================
@@ -68,6 +72,15 @@ struct TruncfToFloat16RewritePattern final
 
 } // end namespace
 
+static LogicalResult isSupportedFp8(Type elementType, Chipset chipset) {
----------------
krzysz00 wrote:

I think we've been using Fp8 in this file? But my memory of having written this code ages ago could be hazy

https://github.com/llvm/llvm-project/pull/106160