[mlir][tosa] Remove optional for pad_const and remove input_zp attr for PadOp #129336

Jerry-Ge · 2025-02-28T23:16:26Z

Always generated pad_const and remove input_zp attr for PadOp.

Co-authored-by: Udaya Ranga [email protected]
Co-authored-by: Tai Ly [email protected]

llvmbot · 2025-02-28T23:17:01Z

@llvm/pr-subscribers-mlir-tosa

Author: Jerry-Ge (Jerry-Ge)

Changes

Always generated pad_const and remove input_zp attr for PadOp

Patch is 36.88 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/129336.diff

14 Files Affected:

(modified) mlir/include/mlir/Dialect/Tosa/IR/TosaOpBase.td (-8)
(modified) mlir/include/mlir/Dialect/Tosa/IR/TosaOps.h (+4)
(modified) mlir/include/mlir/Dialect/Tosa/IR/TosaOps.td (+2-5)
(modified) mlir/lib/Conversion/TosaToTensor/TosaToTensor.cpp (+2-19)
(modified) mlir/lib/Dialect/Tosa/IR/TosaCanonicalizations.cpp (-47)
(modified) mlir/lib/Dialect/Tosa/IR/TosaOps.cpp (+23-22)
(modified) mlir/lib/Dialect/Tosa/Transforms/TosaDecomposeTransposeConv.cpp (+14-20)
(modified) mlir/test/Conversion/TosaToTensor/tosa-to-tensor.mlir (+15-10)
(modified) mlir/test/Dialect/Tosa/availability.mlir (+2-1)
(modified) mlir/test/Dialect/Tosa/canonicalize.mlir (+14-8)
(modified) mlir/test/Dialect/Tosa/invalid.mlir (+10-7)
(modified) mlir/test/Dialect/Tosa/ops.mlir (-8)
(modified) mlir/test/Dialect/Tosa/tosa-decompose-transpose-conv.mlir (+25-24)
(modified) mlir/test/Dialect/Tosa/tosa-infer-shapes.mlir (+6-4)

diff --git a/mlir/include/mlir/Dialect/Tosa/IR/TosaOpBase.td b/mlir/include/mlir/Dialect/Tosa/IR/TosaOpBase.td
index ce17ad9362227..15def695f6a54 100644
--- a/mlir/include/mlir/Dialect/Tosa/IR/TosaOpBase.td
+++ b/mlir/include/mlir/Dialect/Tosa/IR/TosaOpBase.td
@@ -197,14 +197,6 @@ def Tosa_PadOpQuantInfoBuilder : OpBuilder<
                             input, paddings);
   }]>;
 
-def Tosa_ExplicitValuePadOpQuantInfoBuilder : OpBuilder<
-  (ins "Type":$outputType, "Value":$input, "Value":$paddings,
-       "Value":$pad_value),
-  [{
-    buildExplicitValuePadOpWithQuantInfo($_builder, $_state, outputType,
-                                         input, paddings, pad_value);
-  }]>;
-
 // Wrapper over base I32EnumAttr to set common fields.
 class Tosa_I32Enum<string name, string description, list<I32EnumAttrCase> cases>
      : I32EnumAttr<name, description, cases> {
diff --git a/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.h b/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.h
index 344a54f0bb1c9..f0797f97fd842 100644
--- a/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.h
+++ b/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.h
@@ -168,6 +168,10 @@ namespace tosa {
 std::optional<Value> createZeroPointTensor(OpBuilder &builder, Location loc,
                                            Type srcElemType, int64_t zp = 0);
 
+// Create a pad-const const tensor with value of `val` of required data-type
+std::optional<Value> createPadConstTensor(OpBuilder &builder, Location loc,
+  Value src, int32_t val = 0);
+
 } // namespace tosa
 } // namespace mlir
 
diff --git a/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.td b/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.td
index abdd8347cb2b5..aedea883396f8 100644
--- a/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.td
+++ b/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.td
@@ -1882,8 +1882,7 @@ def Tosa_PadOp : Tosa_InferShapedTypeOp<"pad"> {
   let arguments = (ins
     Tosa_RankedTensor:$input1,
     Tosa_Shape:$padding,
-    Optional<Tosa_ScalarTensor>:$pad_const,
-    OptionalAttr<I32Attr>:$input_zp
+    Tosa_ScalarTensor:$pad_const
   );
 
   let results = (outs
@@ -1895,10 +1894,8 @@ def Tosa_PadOp : Tosa_InferShapedTypeOp<"pad"> {
     Extension<[Tosa_EXT_FP8E4M3, Tosa_EXT_FP8E5M2, Tosa_EXT_BF16]>,
   ];
 
-  let builders = [Tosa_PadOpQuantInfoBuilder,
-                  Tosa_ExplicitValuePadOpQuantInfoBuilder];
+  let builders = [Tosa_PadOpQuantInfoBuilder];
 
-  let hasCanonicalizer = 1;
   let hasFolder = 1;
   let hasVerifier = 1;
 }
diff --git a/mlir/lib/Conversion/TosaToTensor/TosaToTensor.cpp b/mlir/lib/Conversion/TosaToTensor/TosaToTensor.cpp
index 7f029d56e2582..6a65904272991 100644
--- a/mlir/lib/Conversion/TosaToTensor/TosaToTensor.cpp
+++ b/mlir/lib/Conversion/TosaToTensor/TosaToTensor.cpp
@@ -350,29 +350,12 @@ class PadConverter : public OpConversionPattern<tosa::PadOp> {
     }
 
     ShapedType inputTy = cast<ShapedType>(input.getType());
-    Type elementTy = inputTy.getElementType();
     int64_t rank = inputTy.getRank();
 
     // Setup the default constantAttr.
 
-    Value padConstant;
-
-    if (padOp.getPadConst()) {
-      padConstant = rewriter.createOrFold<tensor::ExtractOp>(
-          loc, padOp.getPadConst(), ValueRange({}));
-    } else {
-      TypedAttr constantAttr;
-      if (isa<FloatType>(elementTy)) {
-        constantAttr = rewriter.getFloatAttr(elementTy, 0.0);
-      } else if (isa<IntegerType>(elementTy) && !padOp.getInputZpAttr()) {
-        constantAttr = rewriter.getIntegerAttr(elementTy, 0);
-      } else if (isa<IntegerType>(elementTy) && padOp.getInputZpAttr()) {
-        int64_t value = padOp.getInputZpAttr().getInt();
-        constantAttr = rewriter.getIntegerAttr(elementTy, value);
-      }
-      if (constantAttr)
-        padConstant = rewriter.create<arith::ConstantOp>(loc, constantAttr);
-    }
+    Value padConstant = rewriter.createOrFold<tensor::ExtractOp>(
+      loc, padOp.getPadConst(), ValueRange({}));
 
     if (!padConstant) {
       return rewriter.notifyMatchFailure(
diff --git a/mlir/lib/Dialect/Tosa/IR/TosaCanonicalizations.cpp b/mlir/lib/Dialect/Tosa/IR/TosaCanonicalizations.cpp
index 363b5958bc0fd..2c0376134b599 100644
--- a/mlir/lib/Dialect/Tosa/IR/TosaCanonicalizations.cpp
+++ b/mlir/lib/Dialect/Tosa/IR/TosaCanonicalizations.cpp
@@ -175,53 +175,6 @@ void TransposeOp::getCanonicalizationPatterns(RewritePatternSet &results,
   results.add<ConsolidateTransposeOptimization, TransposeIsReshape>(context);
 }
 
-struct MaterializePadValue : public OpRewritePattern<tosa::PadOp> {
-  using OpRewritePattern::OpRewritePattern;
-
-  LogicalResult matchAndRewrite(tosa::PadOp op,
-                                PatternRewriter &rewriter) const override {
-    if (op.getPadConst())
-      return failure();
-
-    auto input = op.getInput1();
-    auto padding = op.getPadding();
-
-    ShapedType inputTy = llvm::cast<ShapedType>(input.getType());
-    Type elementTy = inputTy.getElementType();
-
-    Attribute constantAttr;
-    if (llvm::isa<FloatType>(elementTy)) {
-      constantAttr = rewriter.getFloatAttr(elementTy, 0.0);
-    } else if (llvm::isa<IntegerType>(elementTy) && !op.getInputZpAttr()) {
-      constantAttr = rewriter.getIntegerAttr(elementTy, 0);
-    } else if (llvm::isa<IntegerType>(elementTy) && op.getInputZpAttr()) {
-      int64_t value = op.getInputZpAttr().getInt();
-      constantAttr = rewriter.getIntegerAttr(elementTy, value);
-    }
-
-    if (!constantAttr) {
-      return rewriter.notifyMatchFailure(
-          op,
-          "tosa.pad to linalg lowering encountered an unknown element type");
-    }
-
-    auto denseAttr = DenseElementsAttr::get(
-        RankedTensorType::get({1}, elementTy), constantAttr);
-    auto constantVal = rewriter.create<tosa::ConstOp>(
-        op.getLoc(), denseAttr.getType(), denseAttr);
-
-    rewriter.replaceOpWithNewOp<tosa::PadOp>(
-        op, op.getType(), ValueRange{input, padding, constantVal},
-        op->getAttrs());
-    return success();
-  }
-};
-
-void PadOp::getCanonicalizationPatterns(RewritePatternSet &results,
-                                        MLIRContext *context) {
-  results.add<MaterializePadValue>(context);
-}
-
 struct MaxPool2dIsNoOp : public OpRewritePattern<tosa::MaxPool2dOp> {
   using OpRewritePattern::OpRewritePattern;
 
diff --git a/mlir/lib/Dialect/Tosa/IR/TosaOps.cpp b/mlir/lib/Dialect/Tosa/IR/TosaOps.cpp
index 54f9fa917f2e0..a76a687c3f1eb 100644
--- a/mlir/lib/Dialect/Tosa/IR/TosaOps.cpp
+++ b/mlir/lib/Dialect/Tosa/IR/TosaOps.cpp
@@ -214,6 +214,23 @@ void mlir::tosa::printTypeOrAttr(OpAsmPrinter &p, Operation *op, TypeAttr type,
   }
 }
 
+// Create a pad-const const tensor with value of `val` of required data-type
+std::optional<Value> mlir::tosa::createPadConstTensor(OpBuilder &builder,
+                                                      Location loc, Value src,
+                                                      int32_t val) {
+  auto const srcType = getElementTypeOrSelf(src);
+  auto const srcElemType = getElementTypeOrSelf(src);
+  auto const padConstType = mlir::RankedTensorType::get({1}, srcType);
+  auto const padConstEType = mlir::RankedTensorType::get({1}, srcElemType);
+  auto const pad_const_attr{
+      llvm::isa<FloatType>(srcElemType)
+          ? DenseElementsAttr::get(padConstEType,
+                                   builder.getFloatAttr(srcElemType, val))
+          : DenseElementsAttr::get(padConstEType,
+                                   builder.getIntegerAttr(srcElemType, val))};
+  return builder.create<tosa::ConstOp>(loc, padConstType, pad_const_attr);
+}
+
 //===----------------------------------------------------------------------===//
 // Tosa utilities.
 //===----------------------------------------------------------------------===//
@@ -679,30 +696,14 @@ static void buildUnaryOpWithQuantInfo(OpBuilder &builder,
 static void buildPadOpWithQuantInfo(OpBuilder &builder, OperationState &result,
                                     Type outputType, Value input,
                                     Value paddings) {
-  result.addOperands({input, paddings});
-  auto quantAttr = buildPadOpQuantizationAttr(builder, input);
+  const Location loc{result.location};
+  int32_t zp{0};
+  auto const quantAttr = buildPadOpQuantizationAttr(builder, input);
   if (quantAttr) {
-    result.addAttribute("input_zp",
-                        builder.getI32IntegerAttr(
-                            static_cast<int32_t>(quantAttr.getInputZp())));
-  }
-  result.types.push_back(outputType);
-}
-
-/// This builder is called on TOSA pad operator when an explicit pad_const
-/// value is passed in. It also optionally constructs quantization_attr.
-static void buildExplicitValuePadOpWithQuantInfo(OpBuilder &builder,
-                                                 OperationState &result,
-                                                 Type outputType, Value input,
-                                                 Value paddings,
-                                                 Value padConst) {
-  result.addOperands({input, paddings, padConst});
-  auto quantAttr = buildPadOpQuantizationAttr(builder, input);
-  if (quantAttr) {
-    result.addAttribute("input_zp",
-                        builder.getI32IntegerAttr(
-                            static_cast<int32_t>(quantAttr.getInputZp())));
+    zp = static_cast<int32_t>(quantAttr.getInputZp());
   }
+  auto const pad_const_op{createPadConstTensor(builder, loc, input, zp)};
+  result.addOperands({input, paddings, pad_const_op.value()});
   result.types.push_back(outputType);
 }
 
diff --git a/mlir/lib/Dialect/Tosa/Transforms/TosaDecomposeTransposeConv.cpp b/mlir/lib/Dialect/Tosa/Transforms/TosaDecomposeTransposeConv.cpp
index 83bdbce5d1857..b629c3e7df510 100644
--- a/mlir/lib/Dialect/Tosa/Transforms/TosaDecomposeTransposeConv.cpp
+++ b/mlir/lib/Dialect/Tosa/Transforms/TosaDecomposeTransposeConv.cpp
@@ -148,16 +148,16 @@ class TransposeConvStridedConverter
       return rewriter.notifyMatchFailure(
           op, "zero point must be zero for non-int8 integer types");
 
-    if (weightZpVal != 0) {
-      weight = CreateOpAndInferShape<tosa::PadOp>(
-          rewriter, loc, UnrankedTensorType::get(weightETy), weight,
-          weightPaddingVal, nullptr, rewriter.getI32IntegerAttr(weightZpVal));
-
-    } else {
-      weight = CreateOpAndInferShape<tosa::PadOp>(
-          rewriter, loc, UnrankedTensorType::get(weightETy), weight,
-          weightPaddingVal);
-    }
+    // construct pad_const values from zp values
+    ImplicitLocOpBuilder builder(op->getLoc(), rewriter);
+    Value const inputPadConst =
+        createPadConstTensor(builder, op->getLoc(), input, inputZpVal).value();
+    Value const weightPadConst =
+        createPadConstTensor(builder, op->getLoc(), input, weightZpVal).value();
+    
+    weight = CreateOpAndInferShape<tosa::PadOp>(
+        rewriter, loc, UnrankedTensorType::get(weightETy), weight,
+        weightPaddingVal, weightPadConst);
 
     weightTy = cast<ShapedType>(weight.getType());
     weightHeight = weightTy.getDimSize(1);
@@ -169,7 +169,7 @@ class TransposeConvStridedConverter
         stride[0],      weightWidth / stride[1],
         stride[1],      inputChannels};
 
-    ImplicitLocOpBuilder builder(op->getLoc(), rewriter);
+
     weight = CreateOpAndInferShape<tosa::ReshapeOp>(
         builder, UnrankedTensorType::get(weightETy), weight,
         getTosaConstShape(rewriter, loc, weightReshapeDims0));
@@ -206,15 +206,9 @@ class TransposeConvStridedConverter
     Value inputPaddingVal =
         getTosaConstShape(rewriter, op->getLoc(), inputPadding);
 
-    if (inputZpVal != 0) {
-      input = CreateOpAndInferShape<tosa::PadOp>(
-          rewriter, loc, UnrankedTensorType::get(inputETy), input,
-          inputPaddingVal, nullptr, rewriter.getI32IntegerAttr(inputZpVal));
-    } else {
-      input = CreateOpAndInferShape<tosa::PadOp>(
-          rewriter, loc, UnrankedTensorType::get(inputETy), input,
-          inputPaddingVal);
-    }
+    input = CreateOpAndInferShape<tosa::PadOp>(
+        rewriter, loc, UnrankedTensorType::get(inputETy), input,
+        inputPaddingVal, inputPadConst);
 
     // We use a zero bias as we need to broadcast the bias.
     auto zeroBias = rewriter.create<tosa::ConstOp>(
diff --git a/mlir/test/Conversion/TosaToTensor/tosa-to-tensor.mlir b/mlir/test/Conversion/TosaToTensor/tosa-to-tensor.mlir
index 6b7f622d3303f..c7a689f5a9ae9 100644
--- a/mlir/test/Conversion/TosaToTensor/tosa-to-tensor.mlir
+++ b/mlir/test/Conversion/TosaToTensor/tosa-to-tensor.mlir
@@ -498,35 +498,38 @@ func.func @slice_dyn(%arg0: tensor<?xf32>) -> (tensor<?xf32>) {
 // CHECK-SAME: (%[[ARG0:[0-9a-zA-Z_]*]]:
 func.func @pad_float(%arg0 : tensor<1x2xf32>) -> (tensor<4x9xf32>) {
   %0 = tosa.const_shape {value = dense<[1, 2, 3, 4]> : tensor<4xindex>} : () -> !tosa.shape<4>
+  %pad_const = "tosa.const"() {value = dense<3.14> : tensor<1xf32>} : () -> tensor<1xf32>
   // CHECK-DAG: [[INDEX1:%.+]] = arith.constant 1 : index
   // CHECK-DAG: [[INDEX2:%.+]] = arith.constant 2 : index
   // CHECK-DAG: [[INDEX3:%.+]] = arith.constant 3 : index
   // CHECK-DAG: [[INDEX4:%.+]] = arith.constant 4 : index
-  // CHECK-DAG: [[CST:%.+]] = arith.constant 0.000000e+00 : f32
+  // CHECK-DAG: [[CST:%.+]] = arith.constant 3.140000e+00 : f32
   // CHECK: tensor.pad %[[ARG0]] low{{\[}}[[INDEX1]], [[INDEX3]]] high{{\[}}[[INDEX2]], [[INDEX4]]]  {
   // CHECK:   tensor.yield [[CST]]
   // CHECK: } : tensor<1x2xf32> to tensor<4x9xf32>
-  %1 = "tosa.pad"(%arg0, %0)  : (tensor<1x2xf32>, !tosa.shape<4>)  -> (tensor<4x9xf32>)
+  %1 = "tosa.pad"(%arg0, %0, %pad_const)  : (tensor<1x2xf32>, !tosa.shape<4>, tensor<1xf32>)  -> (tensor<4x9xf32>)
   return %1 : tensor<4x9xf32>
 }
 // -----
 
 func.func @pad_int(%arg0 : tensor<1x2xi32>) -> (tensor<4x9xi32>) {
   %0 = tosa.const_shape {value = dense<[1, 2, 3, 4]> : tensor<4xindex>} : () -> !tosa.shape<4>
-  // CHECK: [[CST:%.+]] = arith.constant 0 : i32
+  %pad_const = "tosa.const"() {value = dense<3> : tensor<1xi32>} : () -> tensor<1xi32>
+  // CHECK: [[CST:%.+]] = arith.constant 3 : i32
   // CHECK: tensor.pad
   // CHECK:   tensor.yield [[CST]]
-  %1 = "tosa.pad"(%arg0, %0)  : (tensor<1x2xi32>, !tosa.shape<4>)  -> (tensor<4x9xi32>)
+  %1 = "tosa.pad"(%arg0, %0, %pad_const)  : (tensor<1x2xi32>, !tosa.shape<4>, tensor<1xi32>)  -> (tensor<4x9xi32>)
   return %1 : tensor<4x9xi32>
 }
 // -----
 
 func.func @pad_quant(%arg0 : tensor<1x2xi32>) -> (tensor<4x9xi32>) {
   %0 = tosa.const_shape {value = dense<[1, 2, 3, 4]> : tensor<4xindex>} : () -> !tosa.shape<4>
-  // CHECK: [[CST:%.+]] = arith.constant 42 : i32
+  %pad_const = "tosa.const"() {value = dense<0> : tensor<1xi32>} : () -> tensor<1xi32>
+  // CHECK: [[CST:%.+]] = arith.constant 0 : i32
   // CHECK: tensor.pad
   // CHECK:   tensor.yield [[CST]]
-  %1 = "tosa.pad"(%arg0, %0) {input_zp = 42 : i32} : (tensor<1x2xi32>, !tosa.shape<4>)  -> (tensor<4x9xi32>)
+  %1 = "tosa.pad"(%arg0, %0, %pad_const) {input_zp = 42 : i32} : (tensor<1x2xi32>, !tosa.shape<4>, tensor<1xi32>)  -> (tensor<4x9xi32>)
   return %1 : tensor<4x9xi32>
 }
 
@@ -551,30 +554,32 @@ func.func @pad_float_explicit(%arg0 : tensor<1x2xf32>) -> (tensor<4x9xf32>) {
 
 func.func @pad_dyn_input(%arg0 : tensor<?x2xf32>) -> (tensor<?x9xf32>) {
   %0 = tosa.const_shape {value = dense<[1, 2, 3, 4]> : tensor<4xindex>} : () -> !tosa.shape<4>
+  %pad_const = "tosa.const"() {value = dense<3.14> : tensor<1xf32>} : () -> tensor<1xf32>
   // CHECK-DAG: [[INDEX1:%.+]] = arith.constant 1 : index
   // CHECK-DAG: [[INDEX2:%.+]] = arith.constant 2 : index
   // CHECK-DAG: [[INDEX3:%.+]] = arith.constant 3 : index
   // CHECK-DAG: [[INDEX4:%.+]] = arith.constant 4 : index
-  // CHECK-DAG: [[CST:%.+]] = arith.constant 0.000000e+00 : f32
+  // CHECK-DAG: [[CST:%.+]] = arith.constant 3.140000e+00 : f32
   // CHECK: tensor.pad %[[ARG0]] low{{\[}}[[INDEX1]], [[INDEX3]]] high{{\[}}[[INDEX2]], [[INDEX4]]]  {
   // CHECK:   tensor.yield [[CST]]
   // CHECK: } : tensor<?x2xf32> to tensor<?x9xf32>
-  %1 = "tosa.pad"(%arg0, %0)  : (tensor<?x2xf32>, !tosa.shape<4>)  -> (tensor<?x9xf32>)
+  %1 = "tosa.pad"(%arg0, %0, %pad_const)  : (tensor<?x2xf32>, !tosa.shape<4>, tensor<1xf32>)  -> (tensor<?x9xf32>)
   return %1 : tensor<?x9xf32>
 }
 // -----
 
 func.func @pad_dyn_padding(%arg0 : tensor<1x2xf32>) -> (tensor<?x9xf32>) {
   %0 = tosa.const_shape {value = dense<[-1, 2, 3, 4]> : tensor<4xindex>} : () -> !tosa.shape<4>
+  %pad_const = "tosa.const"() {value = dense<3.14> : tensor<1xf32>} : () -> tensor<1xf32>
   // CHECK-DAG: [[INDEX1:%.+]] = arith.constant -1 : index
   // CHECK-DAG: [[INDEX2:%.+]] = arith.constant 2 : index
   // CHECK-DAG: [[INDEX3:%.+]] = arith.constant 3 : index
   // CHECK-DAG: [[INDEX4:%.+]] = arith.constant 4 : index
-  // CHECK-DAG: [[CST:%.+]] = arith.constant 0.000000e+00 : f32
+  // CHECK-DAG: [[CST:%.+]] = arith.constant 3.140000e+00 : f32
   // CHECK: tensor.pad %[[ARG0]] low{{\[}}[[INDEX1]], [[INDEX3]]] high{{\[}}[[INDEX2]], [[INDEX4]]]  {
   // CHECK:   tensor.yield [[CST]]
   // CHECK: } : tensor<1x2xf32> to tensor<?x9xf32>
-  %1 = "tosa.pad"(%arg0, %0)  : (tensor<1x2xf32>, !tosa.shape<4>)  -> (tensor<?x9xf32>)
+  %1 = "tosa.pad"(%arg0, %0, %pad_const)  : (tensor<1x2xf32>, !tosa.shape<4>, tensor<1xf32>)  -> (tensor<?x9xf32>)
   return %1 : tensor<?x9xf32>
 }
 
diff --git a/mlir/test/Dialect/Tosa/availability.mlir b/mlir/test/Dialect/Tosa/availability.mlir
index 7324b0ea52e89..4203132e9f702 100644
--- a/mlir/test/Dialect/Tosa/availability.mlir
+++ b/mlir/test/Dialect/Tosa/availability.mlir
@@ -512,9 +512,10 @@ func.func @test_concat(%arg0: tensor<13x21x3xf32>, %arg1: tensor<13x21x3xf32>) -
 // CHECK-LABEL: pad
 func.func @test_pad(%arg0: tensor<13x21x3xf32>) -> tensor<13x21x3xf32> {
   %padding = tosa.const_shape {value = dense<0> : tensor<6xindex>} : () -> !tosa.shape<6>
+  %pad_const = "tosa.const"() {value = dense<3.14> : tensor<1xf32>} : () -> tensor<1xf32>
   // CHECK: profiles: [ [pro_int, pro_fp] ]
   // CHECK: extensions: [ [fp8e4m3, fp8e5m2, bf16] ]
-  %0 = tosa.pad %arg0, %padding : (tensor<13x21x3xf32>, !tosa.shape<6>) -> tensor<13x21x3xf32>
+  %0 = tosa.pad %arg0, %padding, %pad_const : (tensor<13x21x3xf32>, !tosa.shape<6>, tensor<1xf32>) -> tensor<13x21x3xf32>
   return %0 : tensor<13x21x3xf32>
 }
 
diff --git a/mlir/test/Dialect/Tosa/canonicalize.mlir b/mlir/test/Dialect/Tosa/canonicalize.mlir
index 175145f332f8e..f7874aaebee21 100644
--- a/mlir/test/Dialect/Tosa/canonicalize.mlir
+++ b/mlir/test/Dialect/Tosa/canonicalize.mlir
@@ -258,7 +258,8 @@ func.func @max_pool2d_is_noop(%arg0: tensor<10x1x1x3xf32>) -> tensor<10x1x1x3xf3
 func.func @pad_noop(%arg0: tensor<?x?xf32>) -> tensor<?x?xf32> {
   // CHECK: return %arg0
   %0 = tosa.const_shape { value = dense<0> : tensor<4xindex>} : () -> !tosa.shape<4>
-  %1 = tosa.pad %arg0, %0 : (tensor<?x?xf32>, !tosa.shape<4>) -> tensor<?x?xf32>
+  %pad_const = "tosa.const"() {value = dense<3.14> : tensor<1xf32>} : () -> tensor<1xf32>
+  %1 = tosa.pad %arg0, %0, %pad_const : (tensor<?x?xf32>, !tosa.shape<4>, tensor<1xf32>) -> tensor<?x?xf32>
   return %1 : tensor<?x?xf32>
 }
 
@@ -269,7 +270,8 @@ func.func @pad_noop_padding_mismatch_nofold(%arg0: tensor<?x?xf32>) -> tensor<?x
   // CHECK: %[[PAD:.+]] = tosa.pad
   // CHECK: return %[[PAD]]
   %shape = tosa.const_shape { value = dense<[1, 0, 0, 1]> : tensor<4xindex>} : () -> !tosa.shape<4>
-  %1 = tosa.pad %arg0, %shape : (tensor<?x?xf32>, !tosa.shape<4>) -> tensor<?x?xf32>
+  %pad_const = "tosa.const"() {value = dense<3.14> : tensor<1xf32>} : () -> tensor<1xf32>
+  %1 = tosa.pad %arg0, %shape, %pad_const : (tensor<?x?xf32>, !tosa.shape<4>, tensor<1xf32>) -> tensor<?x?xf32>
   return %1 : tensor<?x?xf32>
 }
 
@@ -280,7 +282,8 @@ func.func @pad_noop_type_mismatch_nofold(%arg0: tensor<10xf32>) -> tensor<?xf32>
   // CHECK: %[[PAD:.+]] = tosa.pad
   // CHECK: return %[[PAD]]
   %shape = tosa.const_shape { value = dense<[1, 2]> : tensor<2xindex>} : () -> !tosa.shape<2>
-  %0 = tosa.pad %arg0, %shape : (tensor<10xf32>, !tosa.shape<2>) -> tensor<?xf32>
+  %pad_const = "tosa.const"() {value = dense<3.14> : tensor<1xf32>} : () -> tensor<1xf32>
+  %0 = tosa.pad %arg0, %shape, %pad_const : (tensor<10xf32>, !tosa.shape<2>, tensor<1xf32>) -> tensor<?xf32>
   return %0 : tensor<?xf32>
 }
 
@@ -291,8 +294,9 @@ func.func @pad_determine_val_i32(%arg0: tensor<?x?xi32>, %arg1 : tensor<2x2xi32>
   // CHECK-DAG: %[[ZERO:.+]...
[truncated]

llvmbot · 2025-02-28T23:17:02Z

@llvm/pr-subscribers-mlir

Author: Jerry-Ge (Jerry-Ge)

Changes

Always generated pad_const and remove input_zp attr for PadOp

Patch is 36.88 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/129336.diff

14 Files Affected:

(modified) mlir/include/mlir/Dialect/Tosa/IR/TosaOpBase.td (-8)
(modified) mlir/include/mlir/Dialect/Tosa/IR/TosaOps.h (+4)
(modified) mlir/include/mlir/Dialect/Tosa/IR/TosaOps.td (+2-5)
(modified) mlir/lib/Conversion/TosaToTensor/TosaToTensor.cpp (+2-19)
(modified) mlir/lib/Dialect/Tosa/IR/TosaCanonicalizations.cpp (-47)
(modified) mlir/lib/Dialect/Tosa/IR/TosaOps.cpp (+23-22)
(modified) mlir/lib/Dialect/Tosa/Transforms/TosaDecomposeTransposeConv.cpp (+14-20)
(modified) mlir/test/Conversion/TosaToTensor/tosa-to-tensor.mlir (+15-10)
(modified) mlir/test/Dialect/Tosa/availability.mlir (+2-1)
(modified) mlir/test/Dialect/Tosa/canonicalize.mlir (+14-8)
(modified) mlir/test/Dialect/Tosa/invalid.mlir (+10-7)
(modified) mlir/test/Dialect/Tosa/ops.mlir (-8)
(modified) mlir/test/Dialect/Tosa/tosa-decompose-transpose-conv.mlir (+25-24)
(modified) mlir/test/Dialect/Tosa/tosa-infer-shapes.mlir (+6-4)

diff --git a/mlir/include/mlir/Dialect/Tosa/IR/TosaOpBase.td b/mlir/include/mlir/Dialect/Tosa/IR/TosaOpBase.td
index ce17ad9362227..15def695f6a54 100644
--- a/mlir/include/mlir/Dialect/Tosa/IR/TosaOpBase.td
+++ b/mlir/include/mlir/Dialect/Tosa/IR/TosaOpBase.td
@@ -197,14 +197,6 @@ def Tosa_PadOpQuantInfoBuilder : OpBuilder<
                             input, paddings);
   }]>;
 
-def Tosa_ExplicitValuePadOpQuantInfoBuilder : OpBuilder<
-  (ins "Type":$outputType, "Value":$input, "Value":$paddings,
-       "Value":$pad_value),
-  [{
-    buildExplicitValuePadOpWithQuantInfo($_builder, $_state, outputType,
-                                         input, paddings, pad_value);
-  }]>;
-
 // Wrapper over base I32EnumAttr to set common fields.
 class Tosa_I32Enum<string name, string description, list<I32EnumAttrCase> cases>
      : I32EnumAttr<name, description, cases> {
diff --git a/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.h b/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.h
index 344a54f0bb1c9..f0797f97fd842 100644
--- a/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.h
+++ b/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.h
@@ -168,6 +168,10 @@ namespace tosa {
 std::optional<Value> createZeroPointTensor(OpBuilder &builder, Location loc,
                                            Type srcElemType, int64_t zp = 0);
 
+// Create a pad-const const tensor with value of `val` of required data-type
+std::optional<Value> createPadConstTensor(OpBuilder &builder, Location loc,
+  Value src, int32_t val = 0);
+
 } // namespace tosa
 } // namespace mlir
 
diff --git a/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.td b/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.td
index abdd8347cb2b5..aedea883396f8 100644
--- a/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.td
+++ b/mlir/include/mlir/Dialect/Tosa/IR/TosaOps.td
@@ -1882,8 +1882,7 @@ def Tosa_PadOp : Tosa_InferShapedTypeOp<"pad"> {
   let arguments = (ins
     Tosa_RankedTensor:$input1,
     Tosa_Shape:$padding,
-    Optional<Tosa_ScalarTensor>:$pad_const,
-    OptionalAttr<I32Attr>:$input_zp
+    Tosa_ScalarTensor:$pad_const
   );
 
   let results = (outs
@@ -1895,10 +1894,8 @@ def Tosa_PadOp : Tosa_InferShapedTypeOp<"pad"> {
     Extension<[Tosa_EXT_FP8E4M3, Tosa_EXT_FP8E5M2, Tosa_EXT_BF16]>,
   ];
 
-  let builders = [Tosa_PadOpQuantInfoBuilder,
-                  Tosa_ExplicitValuePadOpQuantInfoBuilder];
+  let builders = [Tosa_PadOpQuantInfoBuilder];
 
-  let hasCanonicalizer = 1;
   let hasFolder = 1;
   let hasVerifier = 1;
 }
diff --git a/mlir/lib/Conversion/TosaToTensor/TosaToTensor.cpp b/mlir/lib/Conversion/TosaToTensor/TosaToTensor.cpp
index 7f029d56e2582..6a65904272991 100644
--- a/mlir/lib/Conversion/TosaToTensor/TosaToTensor.cpp
+++ b/mlir/lib/Conversion/TosaToTensor/TosaToTensor.cpp
@@ -350,29 +350,12 @@ class PadConverter : public OpConversionPattern<tosa::PadOp> {
     }
 
     ShapedType inputTy = cast<ShapedType>(input.getType());
-    Type elementTy = inputTy.getElementType();
     int64_t rank = inputTy.getRank();
 
     // Setup the default constantAttr.
 
-    Value padConstant;
-
-    if (padOp.getPadConst()) {
-      padConstant = rewriter.createOrFold<tensor::ExtractOp>(
-          loc, padOp.getPadConst(), ValueRange({}));
-    } else {
-      TypedAttr constantAttr;
-      if (isa<FloatType>(elementTy)) {
-        constantAttr = rewriter.getFloatAttr(elementTy, 0.0);
-      } else if (isa<IntegerType>(elementTy) && !padOp.getInputZpAttr()) {
-        constantAttr = rewriter.getIntegerAttr(elementTy, 0);
-      } else if (isa<IntegerType>(elementTy) && padOp.getInputZpAttr()) {
-        int64_t value = padOp.getInputZpAttr().getInt();
-        constantAttr = rewriter.getIntegerAttr(elementTy, value);
-      }
-      if (constantAttr)
-        padConstant = rewriter.create<arith::ConstantOp>(loc, constantAttr);
-    }
+    Value padConstant = rewriter.createOrFold<tensor::ExtractOp>(
+      loc, padOp.getPadConst(), ValueRange({}));
 
     if (!padConstant) {
       return rewriter.notifyMatchFailure(
diff --git a/mlir/lib/Dialect/Tosa/IR/TosaCanonicalizations.cpp b/mlir/lib/Dialect/Tosa/IR/TosaCanonicalizations.cpp
index 363b5958bc0fd..2c0376134b599 100644
--- a/mlir/lib/Dialect/Tosa/IR/TosaCanonicalizations.cpp
+++ b/mlir/lib/Dialect/Tosa/IR/TosaCanonicalizations.cpp
@@ -175,53 +175,6 @@ void TransposeOp::getCanonicalizationPatterns(RewritePatternSet &results,
   results.add<ConsolidateTransposeOptimization, TransposeIsReshape>(context);
 }
 
-struct MaterializePadValue : public OpRewritePattern<tosa::PadOp> {
-  using OpRewritePattern::OpRewritePattern;
-
-  LogicalResult matchAndRewrite(tosa::PadOp op,
-                                PatternRewriter &rewriter) const override {
-    if (op.getPadConst())
-      return failure();
-
-    auto input = op.getInput1();
-    auto padding = op.getPadding();
-
-    ShapedType inputTy = llvm::cast<ShapedType>(input.getType());
-    Type elementTy = inputTy.getElementType();
-
-    Attribute constantAttr;
-    if (llvm::isa<FloatType>(elementTy)) {
-      constantAttr = rewriter.getFloatAttr(elementTy, 0.0);
-    } else if (llvm::isa<IntegerType>(elementTy) && !op.getInputZpAttr()) {
-      constantAttr = rewriter.getIntegerAttr(elementTy, 0);
-    } else if (llvm::isa<IntegerType>(elementTy) && op.getInputZpAttr()) {
-      int64_t value = op.getInputZpAttr().getInt();
-      constantAttr = rewriter.getIntegerAttr(elementTy, value);
-    }
-
-    if (!constantAttr) {
-      return rewriter.notifyMatchFailure(
-          op,
-          "tosa.pad to linalg lowering encountered an unknown element type");
-    }
-
-    auto denseAttr = DenseElementsAttr::get(
-        RankedTensorType::get({1}, elementTy), constantAttr);
-    auto constantVal = rewriter.create<tosa::ConstOp>(
-        op.getLoc(), denseAttr.getType(), denseAttr);
-
-    rewriter.replaceOpWithNewOp<tosa::PadOp>(
-        op, op.getType(), ValueRange{input, padding, constantVal},
-        op->getAttrs());
-    return success();
-  }
-};
-
-void PadOp::getCanonicalizationPatterns(RewritePatternSet &results,
-                                        MLIRContext *context) {
-  results.add<MaterializePadValue>(context);
-}
-
 struct MaxPool2dIsNoOp : public OpRewritePattern<tosa::MaxPool2dOp> {
   using OpRewritePattern::OpRewritePattern;
 
diff --git a/mlir/lib/Dialect/Tosa/IR/TosaOps.cpp b/mlir/lib/Dialect/Tosa/IR/TosaOps.cpp
index 54f9fa917f2e0..a76a687c3f1eb 100644
--- a/mlir/lib/Dialect/Tosa/IR/TosaOps.cpp
+++ b/mlir/lib/Dialect/Tosa/IR/TosaOps.cpp
@@ -214,6 +214,23 @@ void mlir::tosa::printTypeOrAttr(OpAsmPrinter &p, Operation *op, TypeAttr type,
   }
 }
 
+// Create a pad-const const tensor with value of `val` of required data-type
+std::optional<Value> mlir::tosa::createPadConstTensor(OpBuilder &builder,
+                                                      Location loc, Value src,
+                                                      int32_t val) {
+  auto const srcType = getElementTypeOrSelf(src);
+  auto const srcElemType = getElementTypeOrSelf(src);
+  auto const padConstType = mlir::RankedTensorType::get({1}, srcType);
+  auto const padConstEType = mlir::RankedTensorType::get({1}, srcElemType);
+  auto const pad_const_attr{
+      llvm::isa<FloatType>(srcElemType)
+          ? DenseElementsAttr::get(padConstEType,
+                                   builder.getFloatAttr(srcElemType, val))
+          : DenseElementsAttr::get(padConstEType,
+                                   builder.getIntegerAttr(srcElemType, val))};
+  return builder.create<tosa::ConstOp>(loc, padConstType, pad_const_attr);
+}
+
 //===----------------------------------------------------------------------===//
 // Tosa utilities.
 //===----------------------------------------------------------------------===//
@@ -679,30 +696,14 @@ static void buildUnaryOpWithQuantInfo(OpBuilder &builder,
 static void buildPadOpWithQuantInfo(OpBuilder &builder, OperationState &result,
                                     Type outputType, Value input,
                                     Value paddings) {
-  result.addOperands({input, paddings});
-  auto quantAttr = buildPadOpQuantizationAttr(builder, input);
+  const Location loc{result.location};
+  int32_t zp{0};
+  auto const quantAttr = buildPadOpQuantizationAttr(builder, input);
   if (quantAttr) {
-    result.addAttribute("input_zp",
-                        builder.getI32IntegerAttr(
-                            static_cast<int32_t>(quantAttr.getInputZp())));
-  }
-  result.types.push_back(outputType);
-}
-
-/// This builder is called on TOSA pad operator when an explicit pad_const
-/// value is passed in. It also optionally constructs quantization_attr.
-static void buildExplicitValuePadOpWithQuantInfo(OpBuilder &builder,
-                                                 OperationState &result,
-                                                 Type outputType, Value input,
-                                                 Value paddings,
-                                                 Value padConst) {
-  result.addOperands({input, paddings, padConst});
-  auto quantAttr = buildPadOpQuantizationAttr(builder, input);
-  if (quantAttr) {
-    result.addAttribute("input_zp",
-                        builder.getI32IntegerAttr(
-                            static_cast<int32_t>(quantAttr.getInputZp())));
+    zp = static_cast<int32_t>(quantAttr.getInputZp());
   }
+  auto const pad_const_op{createPadConstTensor(builder, loc, input, zp)};
+  result.addOperands({input, paddings, pad_const_op.value()});
   result.types.push_back(outputType);
 }
 
diff --git a/mlir/lib/Dialect/Tosa/Transforms/TosaDecomposeTransposeConv.cpp b/mlir/lib/Dialect/Tosa/Transforms/TosaDecomposeTransposeConv.cpp
index 83bdbce5d1857..b629c3e7df510 100644
--- a/mlir/lib/Dialect/Tosa/Transforms/TosaDecomposeTransposeConv.cpp
+++ b/mlir/lib/Dialect/Tosa/Transforms/TosaDecomposeTransposeConv.cpp
@@ -148,16 +148,16 @@ class TransposeConvStridedConverter
       return rewriter.notifyMatchFailure(
           op, "zero point must be zero for non-int8 integer types");
 
-    if (weightZpVal != 0) {
-      weight = CreateOpAndInferShape<tosa::PadOp>(
-          rewriter, loc, UnrankedTensorType::get(weightETy), weight,
-          weightPaddingVal, nullptr, rewriter.getI32IntegerAttr(weightZpVal));
-
-    } else {
-      weight = CreateOpAndInferShape<tosa::PadOp>(
-          rewriter, loc, UnrankedTensorType::get(weightETy), weight,
-          weightPaddingVal);
-    }
+    // construct pad_const values from zp values
+    ImplicitLocOpBuilder builder(op->getLoc(), rewriter);
+    Value const inputPadConst =
+        createPadConstTensor(builder, op->getLoc(), input, inputZpVal).value();
+    Value const weightPadConst =
+        createPadConstTensor(builder, op->getLoc(), input, weightZpVal).value();
+    
+    weight = CreateOpAndInferShape<tosa::PadOp>(
+        rewriter, loc, UnrankedTensorType::get(weightETy), weight,
+        weightPaddingVal, weightPadConst);
 
     weightTy = cast<ShapedType>(weight.getType());
     weightHeight = weightTy.getDimSize(1);
@@ -169,7 +169,7 @@ class TransposeConvStridedConverter
         stride[0],      weightWidth / stride[1],
         stride[1],      inputChannels};
 
-    ImplicitLocOpBuilder builder(op->getLoc(), rewriter);
+
     weight = CreateOpAndInferShape<tosa::ReshapeOp>(
         builder, UnrankedTensorType::get(weightETy), weight,
         getTosaConstShape(rewriter, loc, weightReshapeDims0));
@@ -206,15 +206,9 @@ class TransposeConvStridedConverter
     Value inputPaddingVal =
         getTosaConstShape(rewriter, op->getLoc(), inputPadding);
 
-    if (inputZpVal != 0) {
-      input = CreateOpAndInferShape<tosa::PadOp>(
-          rewriter, loc, UnrankedTensorType::get(inputETy), input,
-          inputPaddingVal, nullptr, rewriter.getI32IntegerAttr(inputZpVal));
-    } else {
-      input = CreateOpAndInferShape<tosa::PadOp>(
-          rewriter, loc, UnrankedTensorType::get(inputETy), input,
-          inputPaddingVal);
-    }
+    input = CreateOpAndInferShape<tosa::PadOp>(
+        rewriter, loc, UnrankedTensorType::get(inputETy), input,
+        inputPaddingVal, inputPadConst);
 
     // We use a zero bias as we need to broadcast the bias.
     auto zeroBias = rewriter.create<tosa::ConstOp>(
diff --git a/mlir/test/Conversion/TosaToTensor/tosa-to-tensor.mlir b/mlir/test/Conversion/TosaToTensor/tosa-to-tensor.mlir
index 6b7f622d3303f..c7a689f5a9ae9 100644
--- a/mlir/test/Conversion/TosaToTensor/tosa-to-tensor.mlir
+++ b/mlir/test/Conversion/TosaToTensor/tosa-to-tensor.mlir
@@ -498,35 +498,38 @@ func.func @slice_dyn(%arg0: tensor<?xf32>) -> (tensor<?xf32>) {
 // CHECK-SAME: (%[[ARG0:[0-9a-zA-Z_]*]]:
 func.func @pad_float(%arg0 : tensor<1x2xf32>) -> (tensor<4x9xf32>) {
   %0 = tosa.const_shape {value = dense<[1, 2, 3, 4]> : tensor<4xindex>} : () -> !tosa.shape<4>
+  %pad_const = "tosa.const"() {value = dense<3.14> : tensor<1xf32>} : () -> tensor<1xf32>
   // CHECK-DAG: [[INDEX1:%.+]] = arith.constant 1 : index
   // CHECK-DAG: [[INDEX2:%.+]] = arith.constant 2 : index
   // CHECK-DAG: [[INDEX3:%.+]] = arith.constant 3 : index
   // CHECK-DAG: [[INDEX4:%.+]] = arith.constant 4 : index
-  // CHECK-DAG: [[CST:%.+]] = arith.constant 0.000000e+00 : f32
+  // CHECK-DAG: [[CST:%.+]] = arith.constant 3.140000e+00 : f32
   // CHECK: tensor.pad %[[ARG0]] low{{\[}}[[INDEX1]], [[INDEX3]]] high{{\[}}[[INDEX2]], [[INDEX4]]]  {
   // CHECK:   tensor.yield [[CST]]
   // CHECK: } : tensor<1x2xf32> to tensor<4x9xf32>
-  %1 = "tosa.pad"(%arg0, %0)  : (tensor<1x2xf32>, !tosa.shape<4>)  -> (tensor<4x9xf32>)
+  %1 = "tosa.pad"(%arg0, %0, %pad_const)  : (tensor<1x2xf32>, !tosa.shape<4>, tensor<1xf32>)  -> (tensor<4x9xf32>)
   return %1 : tensor<4x9xf32>
 }
 // -----
 
 func.func @pad_int(%arg0 : tensor<1x2xi32>) -> (tensor<4x9xi32>) {
   %0 = tosa.const_shape {value = dense<[1, 2, 3, 4]> : tensor<4xindex>} : () -> !tosa.shape<4>
-  // CHECK: [[CST:%.+]] = arith.constant 0 : i32
+  %pad_const = "tosa.const"() {value = dense<3> : tensor<1xi32>} : () -> tensor<1xi32>
+  // CHECK: [[CST:%.+]] = arith.constant 3 : i32
   // CHECK: tensor.pad
   // CHECK:   tensor.yield [[CST]]
-  %1 = "tosa.pad"(%arg0, %0)  : (tensor<1x2xi32>, !tosa.shape<4>)  -> (tensor<4x9xi32>)
+  %1 = "tosa.pad"(%arg0, %0, %pad_const)  : (tensor<1x2xi32>, !tosa.shape<4>, tensor<1xi32>)  -> (tensor<4x9xi32>)
   return %1 : tensor<4x9xi32>
 }
 // -----
 
 func.func @pad_quant(%arg0 : tensor<1x2xi32>) -> (tensor<4x9xi32>) {
   %0 = tosa.const_shape {value = dense<[1, 2, 3, 4]> : tensor<4xindex>} : () -> !tosa.shape<4>
-  // CHECK: [[CST:%.+]] = arith.constant 42 : i32
+  %pad_const = "tosa.const"() {value = dense<0> : tensor<1xi32>} : () -> tensor<1xi32>
+  // CHECK: [[CST:%.+]] = arith.constant 0 : i32
   // CHECK: tensor.pad
   // CHECK:   tensor.yield [[CST]]
-  %1 = "tosa.pad"(%arg0, %0) {input_zp = 42 : i32} : (tensor<1x2xi32>, !tosa.shape<4>)  -> (tensor<4x9xi32>)
+  %1 = "tosa.pad"(%arg0, %0, %pad_const) {input_zp = 42 : i32} : (tensor<1x2xi32>, !tosa.shape<4>, tensor<1xi32>)  -> (tensor<4x9xi32>)
   return %1 : tensor<4x9xi32>
 }
 
@@ -551,30 +554,32 @@ func.func @pad_float_explicit(%arg0 : tensor<1x2xf32>) -> (tensor<4x9xf32>) {
 
 func.func @pad_dyn_input(%arg0 : tensor<?x2xf32>) -> (tensor<?x9xf32>) {
   %0 = tosa.const_shape {value = dense<[1, 2, 3, 4]> : tensor<4xindex>} : () -> !tosa.shape<4>
+  %pad_const = "tosa.const"() {value = dense<3.14> : tensor<1xf32>} : () -> tensor<1xf32>
   // CHECK-DAG: [[INDEX1:%.+]] = arith.constant 1 : index
   // CHECK-DAG: [[INDEX2:%.+]] = arith.constant 2 : index
   // CHECK-DAG: [[INDEX3:%.+]] = arith.constant 3 : index
   // CHECK-DAG: [[INDEX4:%.+]] = arith.constant 4 : index
-  // CHECK-DAG: [[CST:%.+]] = arith.constant 0.000000e+00 : f32
+  // CHECK-DAG: [[CST:%.+]] = arith.constant 3.140000e+00 : f32
   // CHECK: tensor.pad %[[ARG0]] low{{\[}}[[INDEX1]], [[INDEX3]]] high{{\[}}[[INDEX2]], [[INDEX4]]]  {
   // CHECK:   tensor.yield [[CST]]
   // CHECK: } : tensor<?x2xf32> to tensor<?x9xf32>
-  %1 = "tosa.pad"(%arg0, %0)  : (tensor<?x2xf32>, !tosa.shape<4>)  -> (tensor<?x9xf32>)
+  %1 = "tosa.pad"(%arg0, %0, %pad_const)  : (tensor<?x2xf32>, !tosa.shape<4>, tensor<1xf32>)  -> (tensor<?x9xf32>)
   return %1 : tensor<?x9xf32>
 }
 // -----
 
 func.func @pad_dyn_padding(%arg0 : tensor<1x2xf32>) -> (tensor<?x9xf32>) {
   %0 = tosa.const_shape {value = dense<[-1, 2, 3, 4]> : tensor<4xindex>} : () -> !tosa.shape<4>
+  %pad_const = "tosa.const"() {value = dense<3.14> : tensor<1xf32>} : () -> tensor<1xf32>
   // CHECK-DAG: [[INDEX1:%.+]] = arith.constant -1 : index
   // CHECK-DAG: [[INDEX2:%.+]] = arith.constant 2 : index
   // CHECK-DAG: [[INDEX3:%.+]] = arith.constant 3 : index
   // CHECK-DAG: [[INDEX4:%.+]] = arith.constant 4 : index
-  // CHECK-DAG: [[CST:%.+]] = arith.constant 0.000000e+00 : f32
+  // CHECK-DAG: [[CST:%.+]] = arith.constant 3.140000e+00 : f32
   // CHECK: tensor.pad %[[ARG0]] low{{\[}}[[INDEX1]], [[INDEX3]]] high{{\[}}[[INDEX2]], [[INDEX4]]]  {
   // CHECK:   tensor.yield [[CST]]
   // CHECK: } : tensor<1x2xf32> to tensor<?x9xf32>
-  %1 = "tosa.pad"(%arg0, %0)  : (tensor<1x2xf32>, !tosa.shape<4>)  -> (tensor<?x9xf32>)
+  %1 = "tosa.pad"(%arg0, %0, %pad_const)  : (tensor<1x2xf32>, !tosa.shape<4>, tensor<1xf32>)  -> (tensor<?x9xf32>)
   return %1 : tensor<?x9xf32>
 }
 
diff --git a/mlir/test/Dialect/Tosa/availability.mlir b/mlir/test/Dialect/Tosa/availability.mlir
index 7324b0ea52e89..4203132e9f702 100644
--- a/mlir/test/Dialect/Tosa/availability.mlir
+++ b/mlir/test/Dialect/Tosa/availability.mlir
@@ -512,9 +512,10 @@ func.func @test_concat(%arg0: tensor<13x21x3xf32>, %arg1: tensor<13x21x3xf32>) -
 // CHECK-LABEL: pad
 func.func @test_pad(%arg0: tensor<13x21x3xf32>) -> tensor<13x21x3xf32> {
   %padding = tosa.const_shape {value = dense<0> : tensor<6xindex>} : () -> !tosa.shape<6>
+  %pad_const = "tosa.const"() {value = dense<3.14> : tensor<1xf32>} : () -> tensor<1xf32>
   // CHECK: profiles: [ [pro_int, pro_fp] ]
   // CHECK: extensions: [ [fp8e4m3, fp8e5m2, bf16] ]
-  %0 = tosa.pad %arg0, %padding : (tensor<13x21x3xf32>, !tosa.shape<6>) -> tensor<13x21x3xf32>
+  %0 = tosa.pad %arg0, %padding, %pad_const : (tensor<13x21x3xf32>, !tosa.shape<6>, tensor<1xf32>) -> tensor<13x21x3xf32>
   return %0 : tensor<13x21x3xf32>
 }
 
diff --git a/mlir/test/Dialect/Tosa/canonicalize.mlir b/mlir/test/Dialect/Tosa/canonicalize.mlir
index 175145f332f8e..f7874aaebee21 100644
--- a/mlir/test/Dialect/Tosa/canonicalize.mlir
+++ b/mlir/test/Dialect/Tosa/canonicalize.mlir
@@ -258,7 +258,8 @@ func.func @max_pool2d_is_noop(%arg0: tensor<10x1x1x3xf32>) -> tensor<10x1x1x3xf3
 func.func @pad_noop(%arg0: tensor<?x?xf32>) -> tensor<?x?xf32> {
   // CHECK: return %arg0
   %0 = tosa.const_shape { value = dense<0> : tensor<4xindex>} : () -> !tosa.shape<4>
-  %1 = tosa.pad %arg0, %0 : (tensor<?x?xf32>, !tosa.shape<4>) -> tensor<?x?xf32>
+  %pad_const = "tosa.const"() {value = dense<3.14> : tensor<1xf32>} : () -> tensor<1xf32>
+  %1 = tosa.pad %arg0, %0, %pad_const : (tensor<?x?xf32>, !tosa.shape<4>, tensor<1xf32>) -> tensor<?x?xf32>
   return %1 : tensor<?x?xf32>
 }
 
@@ -269,7 +270,8 @@ func.func @pad_noop_padding_mismatch_nofold(%arg0: tensor<?x?xf32>) -> tensor<?x
   // CHECK: %[[PAD:.+]] = tosa.pad
   // CHECK: return %[[PAD]]
   %shape = tosa.const_shape { value = dense<[1, 0, 0, 1]> : tensor<4xindex>} : () -> !tosa.shape<4>
-  %1 = tosa.pad %arg0, %shape : (tensor<?x?xf32>, !tosa.shape<4>) -> tensor<?x?xf32>
+  %pad_const = "tosa.const"() {value = dense<3.14> : tensor<1xf32>} : () -> tensor<1xf32>
+  %1 = tosa.pad %arg0, %shape, %pad_const : (tensor<?x?xf32>, !tosa.shape<4>, tensor<1xf32>) -> tensor<?x?xf32>
   return %1 : tensor<?x?xf32>
 }
 
@@ -280,7 +282,8 @@ func.func @pad_noop_type_mismatch_nofold(%arg0: tensor<10xf32>) -> tensor<?xf32>
   // CHECK: %[[PAD:.+]] = tosa.pad
   // CHECK: return %[[PAD]]
   %shape = tosa.const_shape { value = dense<[1, 2]> : tensor<2xindex>} : () -> !tosa.shape<2>
-  %0 = tosa.pad %arg0, %shape : (tensor<10xf32>, !tosa.shape<2>) -> tensor<?xf32>
+  %pad_const = "tosa.const"() {value = dense<3.14> : tensor<1xf32>} : () -> tensor<1xf32>
+  %0 = tosa.pad %arg0, %shape, %pad_const : (tensor<10xf32>, !tosa.shape<2>, tensor<1xf32>) -> tensor<?xf32>
   return %0 : tensor<?xf32>
 }
 
@@ -291,8 +294,9 @@ func.func @pad_determine_val_i32(%arg0: tensor<?x?xi32>, %arg1 : tensor<2x2xi32>
   // CHECK-DAG: %[[ZERO:.+]...
[truncated]

github-actions · 2025-02-28T23:21:27Z

✅ With the latest revision this PR passed the C/C++ code formatter.

lhutton1

Overall looks great, just had a few nitpicks

mlir/lib/Dialect/Tosa/IR/TosaOps.cpp

mlir/test/Dialect/Tosa/ops.mlir

mlir/lib/Dialect/Tosa/IR/TosaOps.cpp

FranklandJack

This commit message is worded in a slightly confusing way, what was the intended meaning behind: Always generated pad_const?

mlir/lib/Conversion/TosaToTensor/TosaToTensor.cpp

mlir/lib/Dialect/Tosa/IR/TosaOps.cpp

FranklandJack · 2025-03-04T08:01:53Z

mlir/include/mlir/Dialect/Tosa/IR/TosaOps.td

@@ -1882,8 +1882,7 @@ def Tosa_PadOp : Tosa_InferShapedTypeOp<"pad"> {
  let arguments = (ins
    Tosa_RankedTensor:$input1,
    Tosa_Shape:$padding,
-    Optional<Tosa_ScalarTensor>:$pad_const,
-    OptionalAttr<I32Attr>:$input_zp


Should this change be split into a separate commit, it seems orthognal to making the padding constant non optional.

The upstreaming plan is to combine patches related to each operator into a single patch. Those two are the last for PadOp to do some removal.

Jerry-Ge · 2025-03-04T16:39:18Z

This commit message is worded in a slightly confusing way, what was the intended meaning behind: Always generated pad_const?

I agree. Will update the commit message.

…or PadOp Co-authored-by: Udaya Ranga <[email protected]> Co-authored-by: Tai Ly <[email protected]> Signed-off-by: Jerry Ge <[email protected]> Change-Id: I2b7a0169b7ec1158d28779713ad125c061e04592

lhutton1

LGTM!

* pad_const is now required for tosa.pad op after PR[@129336](#129336) * updated the examples in the dialect description section to reflect this change. Signed-off-by: Jerry Ge <[email protected]>

…ion (#131919) * pad_const is now required for tosa.pad op after PR[@129336](llvm/llvm-project#129336) * updated the examples in the dialect description section to reflect this change. Signed-off-by: Jerry Ge <[email protected]>

…or PadOp (llvm#129336) Always generated pad_const and remove input_zp attr for PadOp. - Co-authored-by: Udaya Ranga <[email protected]> - Co-authored-by: Tai Ly <[email protected]> Signed-off-by: Jerry Ge <[email protected]>

…attr for PadOp Upstream PR: llvm#129336 Co-authored-by: Udaya Ranga <[email protected]> Co-authored-by: Tai Ly <[email protected]> Signed-off-by: Jerry Ge <[email protected]> Change-Id: I2b7a0169b7ec1158d28779713ad125c061e04592

llvmbot added mlir mlir:tosa labels Feb 28, 2025

Jerry-Ge force-pushed the pad branch from 4f405b9 to 91f7b7b Compare February 28, 2025 23:17

Jerry-Ge force-pushed the pad branch from 91f7b7b to 4e2d374 Compare February 28, 2025 23:25

Jerry-Ge requested review from GeorgeARM and lhutton1 February 28, 2025 23:39

lhutton1 reviewed Mar 3, 2025

View reviewed changes

mlir/lib/Dialect/Tosa/IR/TosaOps.cpp Outdated Show resolved Hide resolved

mlir/test/Dialect/Tosa/ops.mlir Show resolved Hide resolved

mlir/lib/Dialect/Tosa/IR/TosaOps.cpp Outdated Show resolved Hide resolved

Jerry-Ge force-pushed the pad branch 7 times, most recently from 5099f1f to 5e94f5d Compare March 3, 2025 19:18

FranklandJack requested changes Mar 4, 2025

View reviewed changes

[mlir][tosa] Remove optional for pad_const and remove input_zp attr f…

e6102d2

…or PadOp Co-authored-by: Udaya Ranga <[email protected]> Co-authored-by: Tai Ly <[email protected]> Signed-off-by: Jerry Ge <[email protected]> Change-Id: I2b7a0169b7ec1158d28779713ad125c061e04592

Jerry-Ge force-pushed the pad branch from 5e94f5d to e6102d2 Compare March 4, 2025 16:50

Jerry-Ge changed the title ~~[mlir][tosa] Always generated pad_const and remove input_zp attr for PadOp~~ [mlir][tosa] Remove optional for pad_const and remove input_zp attr for PadOp Mar 4, 2025

lhutton1 approved these changes Mar 5, 2025

View reviewed changes

Jerry-Ge merged commit 7e10ecd into llvm:main Mar 5, 2025
11 checks passed

wonjeon mentioned this pull request Mar 10, 2025

[mlir][tosa] Update control flow variable names and Fix lit tests for max_pool2d and pad op tensorflow/tensorflow#88987

Merged

Jerry-Ge mentioned this pull request Mar 18, 2025

[mlir][tosa] Update PadOp examples in the dialect description #131919

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[mlir][tosa] Remove optional for pad_const and remove input_zp attr for PadOp #129336

[mlir][tosa] Remove optional for pad_const and remove input_zp attr for PadOp #129336

Uh oh!

Jerry-Ge commented Feb 28, 2025 •

edited

Loading

Uh oh!

llvmbot commented Feb 28, 2025

Uh oh!

llvmbot commented Feb 28, 2025

Uh oh!

github-actions bot commented Feb 28, 2025 •

edited

Loading

Uh oh!

lhutton1 left a comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

FranklandJack left a comment

Uh oh!

Uh oh!

Uh oh!

FranklandJack Mar 4, 2025

Uh oh!

Jerry-Ge Mar 4, 2025 •

edited

Loading

Uh oh!

Jerry-Ge commented Mar 4, 2025

Uh oh!

lhutton1 left a comment

Uh oh!

Uh oh!

Uh oh!

[mlir][tosa] Remove optional for pad_const and remove input_zp attr for PadOp #129336

[mlir][tosa] Remove optional for pad_const and remove input_zp attr for PadOp #129336

Uh oh!

Conversation

Jerry-Ge commented Feb 28, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

llvmbot commented Feb 28, 2025

Uh oh!

llvmbot commented Feb 28, 2025

Uh oh!

github-actions bot commented Feb 28, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

lhutton1 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

FranklandJack left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

FranklandJack Mar 4, 2025

Choose a reason for hiding this comment

Uh oh!

Jerry-Ge Mar 4, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Jerry-Ge commented Mar 4, 2025

Uh oh!

lhutton1 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Jerry-Ge commented Feb 28, 2025 •

edited

Loading

github-actions bot commented Feb 28, 2025 •

edited

Loading

Jerry-Ge Mar 4, 2025 •

edited

Loading