[Mlir-commits] [mlir] [mlir][linalg] Add mixed precision folding pattern in vectorize_children_and_apply_patterns TD Op (PR #148684)

[Adam Siemieniuk]

================
@@ -115,6 +194,267 @@ module attributes {transform.with_named_sequence} {
 
 // -----
 
+// Mixed precision vectorization tests.
+
+// CHECK-LABEL: @float_mixed_precision_matmul_as_contract
+// CHECK-COUNT-3: vector.transfer_read
+// CHECK-NOT:     arith.extf
+// CHECK:        vector.contract {{.*}} : vector<24x12xbf16>, vector<12x25xbf16> into vector<24x25xf32>
----------------
adam-smnk wrote:

nit: indentation

https://github.com/llvm/llvm-project/pull/148684