[Mlir-commits] [mlir] [mlir][x86vector] AVX Convert/Broadcast BF16 to F32 instructions (PR #135143)

[Adam Siemieniuk]

================
@@ -408,4 +408,110 @@ def DotOp : AVX_LowOp<"dot", [Pure,
   }];
 }
 
+
+//----------------------------------------------------------------------------//
+// AVX: Convert packed BF16 even-indexed/odd-indexed elements into packed F32
+//----------------------------------------------------------------------------//
+
+def CvtPackedEvenIndexedBF16ToF32Op : AVX_Op<"cvt.packed.even.indexed.bf16_to_f32", [Pure,
----------------
adam-smnk wrote:

These ops shouldn't be `Pure` but model memory reads.

https://github.com/llvm/llvm-project/pull/135143