[MLIR][Linalg] Introduce broadcast/transpose semantic to batch_matmul (#122275)

Goals:
1. To add syntax and semantic to 'batch_matmul' without changing any of
the existing syntax expectations for current usage. batch_matmul is
still just batch_matmul.

2. Move the definition of batch_matmul from linalg OpDsl to tablegen ODS
infra.

Scope of this patch:
To expose broadcast and transpose semantics on the 'batch_matmul'.

The broadcast and transpose semantic are as follows:

By default, 'linalg.batch_matmul' behavior will remain as is. Broadcast
and Transpose semantics can be applied by specifying the explicit
attribute 'indexing_maps' as shown below. This is a list attribute, so
the list must include all the maps if specified.

    Example Transpose:
    ```
    linalg.batch_matmul indexing_maps = [
affine_map< (d0, d1, d2, d3) -> (d0, d3, d1)>, //transpose
                   affine_map< (d0, d1, d2, d3) -> (d0, d3, d2)>,
                   affine_map< (d0, d1, d2, d3) -> (d0, d1, d2)>
                   ]
ins (%arg0, %arg1: memref<2x5x3xf32>,memref<2x5x7xf32>)
                   outs (%arg2: memref<2x3x7xf32>)
    ```

    Example Broadcast:
    ```
    linalg.batch_matmul indexing_maps = [
affine_map< (d0, d1, d2, d3) -> (d3)>, //broadcast
                       affine_map< (d0, d1, d2, d3) -> (d0, d3, d2)>,
                       affine_map< (d0, d1, d2, d3) -> (d0, d1, d2)>
                     ]
                     ins (%arg0, %arg1: memref<5xf32>,memref<2x5x7xf32>)
                     outs (%arg2: memref<2x3x7xf32>)
    ```

    Example Broadcast and transpose:
    ```
    linalg.batch_matmul indexing_maps = [
affine_map< (d0, d1, d2, d3) -> (d1, d3)>, //broadcast
affine_map< (d0, d1, d2, d3) -> (d0, d2, d3)>, //transpose
                       affine_map< (d0, d1, d2, d3) -> (d0, d1, d2)>
                     ]
ins (%arg0, %arg1: memref<3x5xf32>, memref<2x7x5xf32>)
                     outs (%arg2: memref<2x3x7xf32>)
    ```

RFCs and related PR:

https://discourse.llvm.org/t/rfc-linalg-opdsl-constant-list-attribute-definition/80149
https://discourse.llvm.org/t/rfc-op-explosion-in-linalg/82863
https://discourse.llvm.org/t/rfc-mlir-linalg-operation-tree/83586
#115319

Author: shahidact

mlir/include/mlir/Dialect/Linalg/IR/LinalgInterfaces.td

@@ -710,7 +710,10 @@ def LinalgStructuredInterface
     >,
     InterfaceMethod<
       /*desc=*/[{
-        Return true if the user has supplied an explicit indexing maps for this op.
+        Returns true if the user has supplied explicit indexing maps that are
+        different from default indexing maps for this op. Returns `false` otherwise.
+        Note, if the user define maps that are identical to the default maps,
+        this method returns `false`.
       }],
       /*retTy=*/"bool",
       /*methodName=*/"hasUserDefinedMaps",

mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml

@@ -1472,75 +1472,6 @@ structured_op: !LinalgStructuredOpConfig
                 - !ScalarExpression
                   scalar_arg: rhs
 --- !LinalgOpConfig
-metadata: !LinalgOpMetadata
-  name: batch_matmul
-  cpp_class_name: BatchMatmulOp
-  doc: |-
-    Performs a batched matrix multiplication of two 3D inputs.
-
-    Numeric casting is performed on the operands to the inner multiply, promoting
-    them to the same data type as the accumulator/output.
-  implements:
-  - LinalgContractionOpInterface
-structured_op: !LinalgStructuredOpConfig
-  args:
-  - !LinalgOperandDefConfig
-    name: A
-    kind: input_tensor
-    type_var: T1
-    shape_map: affine_map<()[s0, s1, s2, s3] -> (s0, s1, s2)>
-  - !LinalgOperandDefConfig
-    name: B
-    kind: input_tensor
-    type_var: T2
-    shape_map: affine_map<()[s0, s1, s2, s3] -> (s0, s2, s3)>
-  - !LinalgOperandDefConfig
-    name: C
-    kind: output_tensor
-    type_var: U
-    shape_map: affine_map<()[s0, s1, s2, s3] -> (s0, s1, s3)>
-  indexing_maps: !LinalgIndexingMapsConfig
-    static_indexing_maps:
-    - affine_map<(d0, d1, d2, d3)[s0, s1, s2, s3] -> (d0, d1, d3)>
-    - affine_map<(d0, d1, d2, d3)[s0, s1, s2, s3] -> (d0, d3, d2)>
-    - affine_map<(d0, d1, d2, d3)[s0, s1, s2, s3] -> (d0, d1, d2)>
-  iterator_types:
-  - parallel
-  - parallel
-  - parallel
-  - reduction
-  assignments:
-  - !ScalarAssign
-    arg: C
-    value: !ScalarExpression
-      scalar_fn:
-        kind: binary
-        fn_name: add
-        operands:
-        - !ScalarExpression
-          scalar_arg: C
-        - !ScalarExpression
-          scalar_fn:
-            kind: binary
-            fn_name: mul
-            operands:
-            - !ScalarExpression
-              scalar_fn:
-                kind: type
-                fn_name: cast_signed
-                type_var: U
-                operands:
-                - !ScalarExpression
-                  scalar_arg: A
-            - !ScalarExpression
-              scalar_fn:
-                kind: type
-                fn_name: cast_signed
-                type_var: U
-                operands:
-                - !ScalarExpression
-                  scalar_arg: B
---- !LinalgOpConfig
 metadata: !LinalgOpMetadata
   name: batch_matmul_transpose_a
   cpp_class_name: BatchMatmulTransposeAOp

mlir/include/mlir/Dialect/Linalg/IR/LinalgStructuredOps.td

@@ -674,8 +674,7 @@ def MatmulOp : LinalgStructuredBase_Op<"matmul", [
       static unsigned getNumRegionArgs();
       std::string getLibraryCallName();
       bool hasDynamicIndexingMaps();
-      /// Check if the op has broadcast and/or transpose semantic. Returns true if the
-      /// user defined indexing maps are not equal to default map.
+      /// Returns true if the user defined indexing maps are not equal to default maps.
       bool hasUserDefinedMaps();
     }];
 }
@@ -816,6 +815,129 @@ def ContractOp : LinalgStructuredBase_Op<"contract", [
   }];
 }
 
+//===----------------------------------------------------------------------===//
+// Op definition for BatchMatmulOp
+//===----------------------------------------------------------------------===//
+
+def BatchMatmulOp : LinalgStructuredBase_Op<"batch_matmul", !listconcat([AttrSizedOperandSegments],
+  /*extraInterfaces=*/[LinalgContractionOpInterface])> {
+    
+  let summary = [{Performs a batched matrix multiplication of two 3D inputs.}];
+  let description = [{Numeric casting is performed on the operands to the inner multiply, promoting
+    them to the same data type as the accumulator/output.
+
+    Broadcast and Transpose semantics can be appiled by specifying the explicit attribute
+    'indexing_maps' as shown below. This is a list attribute, so must include maps for all
+    arguments if specified.
+
+    Example Transpose:
+    ```
+    linalg.batch_matmul indexing_maps = [
+                   affine_map<(d0, d1, d2, d3) -> (d0, d3, d1)>, // transpose
+                   affine_map<(d0, d1, d2, d3) -> (d0, d3, d2)>,
+                   affine_map<(d0, d1, d2, d3) -> (d0, d1, d2)>
+                   ]
+                   ins(%arg0, %arg1 : memref<2x5x3xf32>,memref<2x5x7xf32>)
+                   outs(%arg2: memref<2x3x7xf32>)
+    ```
+
+    Example Broadcast:
+    ```
+    linalg.batch_matmul indexing_maps = [
+                       affine_map<(d0, d1, d2, d3) -> (d3)>,         // broadcast
+                       affine_map<(d0, d1, d2, d3) -> (d0, d3, d2)>,
+                       affine_map<(d0, d1, d2, d3) -> (d0, d1, d2)>
+                     ]
+                     ins(%arg0, %arg1 : memref<5xf32>, memref<2x5x7xf32>)
+                     outs(%arg2: memref<2x3x7xf32>)
+    ```
+
+    Example Broadcast and Transpose:
+    ```
+    linalg.batch_matmul indexing_maps = [
+                       affine_map<(d0, d1, d2, d3) -> (d1, d3)>,     // broadcast
+                       affine_map<(d0, d1, d2, d3) -> (d0, d2, d3)>, // transpose
+                       affine_map<(d0, d1, d2, d3) -> (d0, d1, d2)>
+                     ]
+                     ins(%arg0, %arg1 : memref<3x5xf32>, memref<2x7x5xf32>)
+                     outs(%arg2: memref<2x3x7xf32>)
+    ```
+}];
+
+    let arguments = (ins
+      Variadic<AnyType>:$inputs,
+      Variadic<AnyShaped>:$outputs,
+      DefaultValuedOptionalAttr<AffineMapArrayAttr, "{}">:$indexing_maps
+    );
+    let results = (outs Variadic<AnyRankedTensor>:$result_tensors);
+    let regions = (region AnyRegion:$region);
+
+    let skipDefaultBuilders = 1;
+    let builders = [
+      OpBuilder<
+      (ins "ValueRange":$inputs, "ValueRange":$outputs,
+            CArg<"ArrayRef<NamedAttribute>", "{}">:$attributes),
+      [{
+        buildBatchMatmulOp($_builder, $_state, std::nullopt, inputs, outputs,
+          attributes, BatchMatmulOp::getRegionBuilder(),
+          BatchMatmulOp::getDefaultIndexingMaps($_builder.getContext()));
+      }]>,
+      OpBuilder<
+      (ins "TypeRange":$resultTensorTypes, "ValueRange":$inputs,
+            "ValueRange":$outputs,
+            CArg<"ArrayRef<NamedAttribute>", "{}">:$attributes),
+      [{
+        buildBatchMatmulOp($_builder, $_state, resultTensorTypes,
+          inputs, outputs, attributes, BatchMatmulOp::getRegionBuilder(),
+          BatchMatmulOp::getDefaultIndexingMaps($_builder.getContext()));
+      }]>,
+      OpBuilder<
+      (ins "TypeRange":$resultTensorTypes, "ValueRange":$operands,
+            CArg<"ArrayRef<NamedAttribute>", "{}">:$attributes),
+      [{
+        $_state.addOperands(operands);
+        $_state.addAttributes(attributes);
+        $_state.addTypes(resultTensorTypes);
+        (void)$_state.addRegion(),
+        BatchMatmulOp::getDefaultIndexingMaps($_builder.getContext());
+      }]>
+      
+    ];
+    let hasCustomAssemblyFormat = 1;
+    let hasFolder = 1;
+    let hasVerifier = 1;
+
+    let extraClassDeclaration = structuredOpsBaseDecls # [{
+
+      SmallVector<utils::IteratorType> getIteratorTypesArray();
+      static void regionBuilder(ImplicitLocOpBuilder &b,
+                                Block &block, ArrayRef<NamedAttribute> attrs);
+      static std::function<void(ImplicitLocOpBuilder &,
+                                Block &, ArrayRef<NamedAttribute>)>
+      getRegionBuilder() {
+        return regionBuilder;
+      }
+
+      /// Returns a list with default AffineMap(s), i.e. without broadcasts and transpositions.
+      static SmallVector<AffineMap> getDefaultIndexingMaps(MLIRContext *context);
+
+      /// Returns true if the given broadcast map \p bcastMap is valid for this op.
+      bool isValidLhsRhsBroadcastMap(AffineMap bcastMap, bool isLHS = true);
+
+      ::mlir::MutableOperandRange getDpsInitsMutable() {
+        return getOutputsMutable();
+      }
+
+      // Generic methods.
+      static unsigned getNumRegionArgs();
+      bool hasDynamicIndexingMaps() { return true; }
+      std::string getLibraryCallName();
+      /// Returns true if the user defined indexing maps are not equal to default maps.
+      bool hasUserDefinedMaps();
+    }];
+}
+
+
 //===----------------------------------------------------------------------===//
 // Named Linalg ops, implemented as a declarative configurations of generic ops.
 //===----------------------------------------------------------------------===//