[Mlir-commits] [mlir] [mlir][AMDGPU] Add scaled floating point conversion ops fp8 (PR #141554)

[Krzysztof Drewniak]

================
@@ -139,6 +171,36 @@ def AMDGPU_PackedTrunc2xFp8Op :
   let hasVerifier = 1;
 }
 
+def AMDGPU_PackedScaledTruncOp
+    : AMDGPU_Op<"packed_scaled_trunc", [Pure]>,
----------------
krzysz00 wrote:

We have `ext_packed` (the things being extended are packed) but `packed_trunc` (the result is packed)

https://github.com/llvm/llvm-project/pull/141554