[Mlir-commits] [mlir] [MLIR][Linalg] Add pass to convert linalg.generic back to named ops (PR #95656)

[Javed Absar]

================
@@ -58,6 +68,176 @@ static bool areBinOpsSwapped(GenericOp genericOp) {
   return swapped;
 }
 
+//===----------------------------------------------------------------------===//
+// Specialize linalg generic to matmul variants.
+//===----------------------------------------------------------------------===//
+/// Identifies linalg.generic that is essentially named op of the form:
+//    ` linalg.{batch_}?matmul{_transpose_a | _transpose_b}? `
+//
+// It is possible that a linalg.generic may be implementing one of matmul
+// variants but not in a straight-forward way, or the linalg.generic's
+// affine map per operand capture more semantics than is possible with
+// named op (which has implicit map interpreted via name).
+//
+// But a named linalg matmul variant that was 'generalized' should be
+// convertible back to named op here.
+//
+namespace {
+enum class IndexMatchResult {
+  Match = 0,  // identity map.
+  Transposed, // transposed map.
+  Mismatch    // none of the above.
+};
+
+// Looks at the affine map of an operand and works out if generic accesses
+// the element as identity-map, transposed, or 'cant work out'.
+// This check skips the `offset` batch indices and focuses on the matmul part.
----------------
javedabsar1 wrote:

Changed the implementation and comments to make the implementation easier to understand.
Yes my first idea was to use some identity and set intersection to implement this. But that didnt work as 'identity' is not whats needed.

https://github.com/llvm/llvm-project/pull/95656