[MLIR][Shape] Concretize broadcast result type if possible

As a canonicalization, infer the resulting shape rank if possible. Differential Revision: https://reviews.llvm.org/D101377
llvm · Apr 28, 2021 · dca5361 · dca5361
1 parent 2d37f21
commit dca5361
Show file tree

Hide file tree

Showing 3 changed files with 52 additions and 5 deletions.
diff --git a/mlir/include/mlir/Dialect/Shape/IR/Shape.h b/mlir/include/mlir/Dialect/Shape/IR/Shape.h
@@ -29,7 +29,8 @@ class PatternRewriter;
 namespace shape {
 
 /// Alias type for extent tensors.
-RankedTensorType getExtentTensorType(MLIRContext *ctx);
+RankedTensorType getExtentTensorType(MLIRContext *ctx,
+                                     int64_t rank = ShapedType::kDynamicSize);
 
 // Check if a type is an extent tensor, e.g., tensor<?xindex>.
 bool isExtentTensorType(Type);

diff --git a/mlir/lib/Dialect/Shape/IR/Shape.cpp b/mlir/lib/Dialect/Shape/IR/Shape.cpp
@@ -27,8 +27,8 @@ namespace {
 #include "ShapeCanonicalization.inc"
 }
 
-RankedTensorType shape::getExtentTensorType(MLIRContext *ctx) {
-  return RankedTensorType::get({ShapedType::kDynamicSize}, IndexType::get(ctx));
+RankedTensorType shape::getExtentTensorType(MLIRContext *ctx, int64_t rank) {
+  return RankedTensorType::get({rank}, IndexType::get(ctx));
 }
 
 bool shape::isExtentTensorType(Type type) {
@@ -660,11 +660,42 @@ struct CanonicalizeCastExtentTensorOperandsPattern
     return success();
   }
 };
+
+struct BroadcastConcretizeResultTypePattern
+    : public OpRewritePattern<BroadcastOp> {
+  using OpRewritePattern<BroadcastOp>::OpRewritePattern;
+
+  LogicalResult matchAndRewrite(BroadcastOp op,
+                                PatternRewriter &rewriter) const override {
+    // Only concretize dynamic extent tensor result types.
+    auto resultTy = op.getType().dyn_cast<RankedTensorType>();
+    if (!resultTy || !resultTy.isDynamicDim(0))
+      return failure();
+
+    // Infer resulting shape rank if possible.
+    int64_t maxRank = 0;
+    for (Value shape : op.shapes()) {
+      if (auto extentTensorTy = shape.getType().dyn_cast<RankedTensorType>()) {
+        // Cannot infer resulting shape rank if any operand is dynamically
+        // ranked.
+        if (extentTensorTy.isDynamicDim(0))
+          return failure();
+        maxRank = std::max(maxRank, extentTensorTy.getDimSize(0));
+      }
+    }
+
+    auto newOp = rewriter.create<BroadcastOp>(
+        op.getLoc(), getExtentTensorType(getContext(), maxRank), op.shapes());
+    rewriter.replaceOpWithNewOp<tensor::CastOp>(op, op.getType(), newOp);
+    return success();
+  }
+};
 } // namespace
 
 void BroadcastOp::getCanonicalizationPatterns(RewritePatternSet &patterns,
                                               MLIRContext *context) {
-  patterns.add<BroadcastFoldConstantOperandsPattern,
+  patterns.add<BroadcastConcretizeResultTypePattern,
+               BroadcastFoldConstantOperandsPattern,
                BroadcastForwardSingleOperandPattern,
                CanonicalizeCastExtentTensorOperandsPattern<BroadcastOp>,
                RemoveDuplicateOperandsPattern<BroadcastOp>,

diff --git a/mlir/test/Dialect/Shape/canonicalize.mlir b/mlir/test/Dialect/Shape/canonicalize.mlir
@@ -1344,7 +1344,8 @@ func @cast_extent_tensor_operands(%arg0 : tensor<?xindex>,
     %arg1 : tensor<3xindex>) -> (!shape.witness, tensor<?xindex>) {
   // CHECK: %[[CAST_ARG0:.*]] = tensor.cast %[[ARG0]] : tensor<?xindex> to tensor<3xindex>
   // CHECK: %[[WIT:.*]] = shape.cstr_broadcastable %[[CAST_ARG0]], %[[ARG1]] : tensor<3xindex>, tensor<3xindex>
-  // CHECK: %[[RES:.*]] = shape.broadcast %[[CAST_ARG0]], %[[ARG1]] : tensor<3xindex>, tensor<3xindex>
+  // CHECK: %[[UNCAST_RES:.*]] = shape.broadcast %[[CAST_ARG0]], %[[ARG1]] : tensor<3xindex>, tensor<3xindex> -> tensor<3xindex>
+  // CHECK: %[[RES:.*]] = tensor.cast %[[UNCAST_RES]] : tensor<3xindex> to tensor<?xindex>
   // CHECK: return %[[WIT]], %[[RES]]
   %0 = tensor.cast %arg0 : tensor<?xindex> to tensor<3xindex>
   %1 = tensor.cast %arg1 : tensor<3xindex> to tensor<?xindex>
@@ -1353,3 +1354,17 @@ func @cast_extent_tensor_operands(%arg0 : tensor<?xindex>,
       -> tensor<?xindex>
   return %2, %3 : !shape.witness, tensor<?xindex>
 }
+
+// -----
+
+// CHECK-LABEL: @concretize_broadcast_result_type
+// CHECK-SAME:  (%[[ARG0:.*]]: tensor<2xindex>, %[[ARG1:.*]]: tensor<3xindex>)
+func @concretize_broadcast_result_type(%arg0 : tensor<2xindex>,
+    %arg1 : tensor<3xindex>) -> tensor<?xindex> {
+  // CHECK: %[[CONCR:.*]] = shape.broadcast %[[ARG0]], %[[ARG1]] : tensor<2xindex>, tensor<3xindex> -> tensor<3xindex>
+  // CHECK: %[[RES:.*]] = tensor.cast %[[CONCR]] : tensor<3xindex> to tensor<?xindex>
+  // CHECK: return %[[RES]]
+  %0 = shape.broadcast %arg0, %arg1 : tensor<2xindex>, tensor<3xindex>
+      -> tensor<?xindex>
+  return %0 : tensor<?xindex>
+}