[Mlir-commits] [mlir] a1f4656 - [mlir][tosa] Add quantized and unquantized versions for tosa.depthwise_conv2d lowering
Rob Suderman
llvmlistbot at llvm.org
Tue Aug 10 14:35:32 PDT 2021
Author: natashaknk
Date: 2021-08-10T14:29:26-07:00
New Revision: a1f46569a186cad80ebae87280292ad4cdf703eb
URL: https://github.com/llvm/llvm-project/commit/a1f46569a186cad80ebae87280292ad4cdf703eb
DIFF: https://github.com/llvm/llvm-project/commit/a1f46569a186cad80ebae87280292ad4cdf703eb.diff
LOG: [mlir][tosa] Add quantized and unquantized versions for tosa.depthwise_conv2d lowering
Reviewed By: rsuderman
Differential Revision: https://reviews.llvm.org/D107855
Added:
Modified:
mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml
mlir/lib/Conversion/TosaToLinalg/TosaToLinalg.cpp
mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py
mlir/test/Conversion/TosaToLinalg/tosa-to-linalg.mlir
Removed:
################################################################################
diff --git a/mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml b/mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml
index 6053b0707c962..53b54e1bff9fb 100644
--- a/mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml
+++ b/mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml
@@ -905,6 +905,202 @@ structured_op: !LinalgStructuredOpConfig
- !ScalarExpression
scalar_arg: K
--- !LinalgOpConfig
+metadata: !LinalgOpMetadata
+ name: depthwise_conv_2D_nchw
+ cpp_class_name: DepthwiseConv2DNchwOp
+ doc: |-
+ Performs depth-wise 2-D convolution.
+
+ Numeric casting is performed on the operands to the inner multiply, promoting
+ them to the same data type as the accumulator/output.
+structured_op: !LinalgStructuredOpConfig
+ args:
+ - !LinalgOperandDefConfig
+ name: I
+ usage: InputOperand
+ type_var: T1
+ shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12]
+ -> (s0, s1, s2, s3)>
+ - !LinalgOperandDefConfig
+ name: K
+ usage: InputOperand
+ type_var: T2
+ shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12]
+ -> (s4, s5, s3, s6)>
+ - !LinalgOperandDefConfig
+ name: O
+ usage: OutputOperand
+ type_var: U
+ shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12]
+ -> (s0, s7, s8, s3, s6)>
+ - !LinalgOperandDefConfig
+ name: strides
+ usage: IndexAttribute
+ type_var: I64
+ attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
+ s12] -> (s9, s10)>
+ - !LinalgOperandDefConfig
+ name: dilations
+ usage: IndexAttribute
+ type_var: I64
+ attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
+ s12] -> (s11, s12)>
+ indexing_maps: !LinalgIndexingMapsConfig
+ static_indexing_maps:
+ - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
+ s9, s10, s11, s12] -> (d0, d1 * s9 + d3 * s11, d2 * s10 + d4 * s12, d5)>
+ - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
+ s9, s10, s11, s12] -> (d3, d4, d5, d6)>
+ - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
+ s9, s10, s11, s12] -> (d0, d1, d2, d5, d6)>
+ iterator_types:
+ - parallel
+ - parallel
+ - parallel
+ - reduction
+ - reduction
+ - parallel
+ - parallel
+ assignments:
+ - !ScalarAssign
+ arg: O
+ value: !ScalarExpression
+ scalar_apply:
+ fn_name: add
+ operands:
+ - !ScalarExpression
+ scalar_arg: O
+ - !ScalarExpression
+ scalar_apply:
+ fn_name: mul
+ operands:
+ - !ScalarExpression
+ symbolic_cast:
+ type_var: U
+ operands:
+ - !ScalarExpression
+ scalar_arg: I
+ - !ScalarExpression
+ symbolic_cast:
+ type_var: U
+ operands:
+ - !ScalarExpression
+ scalar_arg: K
+--- !LinalgOpConfig
+metadata: !LinalgOpMetadata
+ name: depthwise_conv2D_nchw_q
+ cpp_class_name: DepthwiseConv2DNchwQOp
+ doc: |-
+ Performs depth-wise 2-D convolution.
+
+ Numeric casting is performed on the operands to the inner multiply, promoting
+ them to the same data type as the accumulator/output.
+structured_op: !LinalgStructuredOpConfig
+ args:
+ - !LinalgOperandDefConfig
+ name: I
+ usage: InputOperand
+ type_var: T1
+ shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12]
+ -> (s0, s1, s2, s3)>
+ - !LinalgOperandDefConfig
+ name: K
+ usage: InputOperand
+ type_var: T2
+ shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12]
+ -> (s4, s5, s3, s6)>
+ - !LinalgOperandDefConfig
+ name: IZp
+ usage: InputOperand
+ type_var: I32
+ - !LinalgOperandDefConfig
+ name: KZp
+ usage: InputOperand
+ type_var: I32
+ - !LinalgOperandDefConfig
+ name: O
+ usage: OutputOperand
+ type_var: U
+ shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12]
+ -> (s0, s7, s8, s3, s6)>
+ - !LinalgOperandDefConfig
+ name: strides
+ usage: IndexAttribute
+ type_var: I64
+ attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
+ s12] -> (s9, s10)>
+ - !LinalgOperandDefConfig
+ name: dilations
+ usage: IndexAttribute
+ type_var: I64
+ attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
+ s12] -> (s11, s12)>
+ indexing_maps: !LinalgIndexingMapsConfig
+ static_indexing_maps:
+ - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
+ s9, s10, s11, s12] -> (d0, d1 * s9 + d3 * s11, d2 * s10 + d4 * s12, d5)>
+ - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
+ s9, s10, s11, s12] -> (d3, d4, d5, d6)>
+ - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
+ s9, s10, s11, s12] -> ()>
+ - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
+ s9, s10, s11, s12] -> ()>
+ - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
+ s9, s10, s11, s12] -> (d0, d1, d2, d5, d6)>
+ iterator_types:
+ - parallel
+ - parallel
+ - parallel
+ - reduction
+ - reduction
+ - parallel
+ - parallel
+ assignments:
+ - !ScalarAssign
+ arg: O
+ value: !ScalarExpression
+ scalar_apply:
+ fn_name: add
+ operands:
+ - !ScalarExpression
+ scalar_arg: O
+ - !ScalarExpression
+ scalar_apply:
+ fn_name: mul
+ operands:
+ - !ScalarExpression
+ scalar_apply:
+ fn_name: sub
+ operands:
+ - !ScalarExpression
+ symbolic_cast:
+ type_var: U
+ operands:
+ - !ScalarExpression
+ scalar_arg: I
+ - !ScalarExpression
+ symbolic_cast:
+ type_var: U
+ operands:
+ - !ScalarExpression
+ scalar_arg: IZp
+ - !ScalarExpression
+ scalar_apply:
+ fn_name: sub
+ operands:
+ - !ScalarExpression
+ symbolic_cast:
+ type_var: U
+ operands:
+ - !ScalarExpression
+ scalar_arg: K
+ - !ScalarExpression
+ symbolic_cast:
+ type_var: U
+ operands:
+ - !ScalarExpression
+ scalar_arg: KZp
+--- !LinalgOpConfig
metadata: !LinalgOpMetadata
name: conv_2d_nchw
cpp_class_name: Conv2DNchwOp
@@ -1700,4 +1896,3 @@ structured_op: !LinalgStructuredOpConfig
operands:
- !ScalarExpression
scalar_arg: I
-
diff --git a/mlir/lib/Conversion/TosaToLinalg/TosaToLinalg.cpp b/mlir/lib/Conversion/TosaToLinalg/TosaToLinalg.cpp
index ea3fe0aec59fd..d0b6274ddc0a1 100644
--- a/mlir/lib/Conversion/TosaToLinalg/TosaToLinalg.cpp
+++ b/mlir/lib/Conversion/TosaToLinalg/TosaToLinalg.cpp
@@ -946,7 +946,7 @@ convolutionMatchAndRewriterHelper(Operation *op,
return success();
}
- if (isa<tosa::DepthwiseConv2DOp>(op) && !isQuantized) {
+ if (isa<tosa::DepthwiseConv2DOp>(op)) {
ShapedType linalgConvTy =
RankedTensorType::get({resultShape[0], resultShape[1], resultShape[2],
weightShape[2], weightShape[3]},
@@ -954,11 +954,23 @@ convolutionMatchAndRewriterHelper(Operation *op,
Value biasReshape =
rewriter.create<tosa::ReshapeOp>(loc, linalgConvTy, biasBroadcast);
- Value conv = rewriter
- .create<linalg::DepthwiseConvInputNHWCFilterHWCFOp>(
- loc, linalgConvTy, ValueRange{input, weight},
- ValueRange{biasReshape}, dilationAttr, strideAttr)
- .getResult(0);
+ Value conv;
+ if (!isQuantized) {
+ conv = rewriter
+ .create<linalg::DepthwiseConv2DNchwOp>(
+ loc, linalgConvTy, ValueRange{input, weight},
+ ValueRange{biasReshape}, dilationAttr, strideAttr)
+ .getResult(0);
+ } else {
+ auto iZpVal = rewriter.create<ConstantOp>(loc, iZp);
+ auto kZpVal = rewriter.create<ConstantOp>(loc, kZp);
+ conv =
+ rewriter
+ .create<linalg::DepthwiseConv2DNchwQOp>(
+ loc, linalgConvTy, ValueRange{input, weight, iZpVal, kZpVal},
+ ValueRange{biasReshape}, dilationAttr, strideAttr)
+ .getResult(0);
+ }
Value reshape = rewriter.create<tosa::ReshapeOp>(loc, resultTy, conv);
rewriter.replaceOp(op, reshape);
diff --git a/mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py b/mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py
index 11fcf9033b487..fc92c196a059b 100644
--- a/mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py
+++ b/mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py
@@ -223,6 +223,43 @@ def conv_2d_nchw(
]) * cast(U, K[D.f, D.c, D.kh, D.kw])
+def depthwise_conv2D_nchw( #TODO: Fix name
+ I=TensorDef(T1, S.N, S.IH, S.IW, S.IC),
+ K=TensorDef(T2, S.KH, S.KW, S.IC, S.CM),
+ O=TensorDef(U, S.N, S.OH, S.OW, S.IC, S.CM, output=True),
+ strides=AttributeDef(S.SH, S.SW),
+ dilations=AttributeDef(S.DH, S.DW)):
+ """Performs depth-wise 2-D convolution.
+
+ Numeric casting is performed on the operands to the inner multiply, promoting
+ them to the same data type as the accumulator/output.
+ """
+ domain(D.n, D.oh, D.ow, D.kh, D.kw, D.ic, D.cm)
+ O[D.n, D.oh, D.ow, D.ic, D.cm] += cast(
+ U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW,
+ D.ic]) * cast(U, K[D.kh, D.kw, D.ic, D.cm])
+
+
+def depthwise_conv2D_nchw_q( #TODO: Fix name
+ I=TensorDef(T1, S.N, S.IH, S.IW, S.IC),
+ K=TensorDef(T2, S.KH, S.KW, S.IC, S.CM),
+ IZp=ScalarDef(I32),
+ KZp=ScalarDef(I32),
+ O=TensorDef(U, S.N, S.OH, S.OW, S.IC, S.CM, output=True),
+ strides=AttributeDef(S.SH, S.SW),
+ dilations=AttributeDef(S.DH, S.DW)):
+ """Performs depth-wise 2-D convolution.
+
+ Numeric casting is performed on the operands to the inner multiply, promoting
+ them to the same data type as the accumulator/output.
+ """
+ domain(D.n, D.oh, D.ow, D.kh, D.kw, D.ic, D.cm)
+ O[D.n, D.oh, D.ow, D.ic, D.cm] += (
+ (cast(U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW,
+ D.ic]) - cast(U, IZp)) *
+ (cast(U, K[D.kh, D.kw, D.ic, D.cm]) - cast(U, KZp)))
+
+
@linalg_structured_op
def pooling_nhwc_sum(
I=TensorDef(T1, S.N, S.H, S.W, S.C),
diff --git a/mlir/test/Conversion/TosaToLinalg/tosa-to-linalg.mlir b/mlir/test/Conversion/TosaToLinalg/tosa-to-linalg.mlir
index 59454f2328a59..554fa2d4ff151 100644
--- a/mlir/test/Conversion/TosaToLinalg/tosa-to-linalg.mlir
+++ b/mlir/test/Conversion/TosaToLinalg/tosa-to-linalg.mlir
@@ -1219,7 +1219,7 @@ func @conv2d_quant(%arg0 : tensor<1x12x12x1xi8>, %arg1 : tensor<1024x3x3x1xi8>,
// CHECK: linalg.yield %arg3 : i32
// CHECK: %[[C128:.+]] = constant -128
// CHECK: %[[C42:.+]] = constant 42
- // CHECK: linalg.conv_2d_input_nhwc_filter_ohwi_poly_q {dilations = dense<1> : tensor<2xi64>, strides = dense<1> : tensor<2xi64>} ins(%arg0, %arg1, %[[C128]], %[[C42]] : tensor<1x12x12x1xi8>, tensor<1024x3x3x1xi8>, i32, i32) outs(%1 : tensor<1x10x10x1024xi32>)
+ // CHECK: linalg.conv_2d_input_nhwc_filter_ohwi_poly_q {dilations = dense<1> : tensor<2xi64>, strides = dense<1> : tensor<2xi64>} ins(%arg0, %arg1, %[[C128]], %[[C42]] : tensor<1x12x12x1xi8>, tensor<1024x3x3x1xi8>, i32, i32) outs(%1 : tensor<1x10x10x1024xi32>)
%0 = "tosa.conv2d"(%arg0, %arg1, %arg2) {dilation = [1, 1], pad = [0, 0, 0, 0], quantization_info = {input_zp = -128 : i32, weight_zp = 42 : i32}, stride = [1, 1]} : (tensor<1x12x12x1xi8>, tensor<1024x3x3x1xi8>, tensor<1024xi32>) -> tensor<1x10x10x1024xi32>
return
}
@@ -1237,7 +1237,7 @@ func @depthwise_conv(%arg0 : tensor<1x7x5x3xf32>, %arg1 : tensor<3x1x3x11xf32>,
// CHECK: linalg.yield %arg3 : f32
// CHECK: } -> tensor<1x5x5x33xf32>
// CHECK: [[DBIAS:%.+]] = linalg.tensor_expand_shape [[BIAS]] {{\[}}[0], [1], [2], [3, 4]]
- // CHECK: [[DEPTH:%.+]] = linalg.depthwise_conv_2d_input_nhwc_filter_hwcf {dilations = dense<1> : tensor<2xi64>, strides = dense<1> : tensor<2xi64>} ins(%arg0, %arg1 : tensor<1x7x5x3xf32>, tensor<3x1x3x11xf32>) outs([[DBIAS]] : tensor<1x5x5x3x11xf32>)
+ // CHECK: [[DEPTH:%.+]] = linalg.depthwise_conv_2D_nchw {dilations = dense<1> : tensor<2xi64>, strides = dense<1> : tensor<2xi64>} ins(%arg0, %arg1 : tensor<1x7x5x3xf32>, tensor<3x1x3x11xf32>) outs([[DBIAS]] : tensor<1x5x5x3x11xf32>)
// CHECK: linalg.tensor_collapse_shape %3 {{\[}}[0], [1], [2], [3, 4]]
%2 = "tosa.depthwise_conv2d"(%arg0, %arg1, %arg2) { pad = [0, 0, 0, 0], stride = [1, 1], dilation = [1, 1] } : (tensor<1x7x5x3xf32>, tensor<3x1x3x11xf32>, tensor<33xf32>) -> (tensor<1x5x5x33xf32>)
return
@@ -1245,6 +1245,27 @@ func @depthwise_conv(%arg0 : tensor<1x7x5x3xf32>, %arg1 : tensor<3x1x3x11xf32>,
// -----
+// CHECK: #[[$MAP0:.*]] = affine_map<(d0, d1, d2, d3) -> (d3)>
+// CHECK: #[[$MAP1:.*]] = affine_map<(d0, d1, d2, d3) -> (d0, d1, d2, d3)>
+
+// CHECK-LABEL: @depthwise_conv_quant
+func @depthwise_conv_quant(%arg0 : tensor<1x12x12x4xi8>, %arg1 : tensor<3x3x4x128xi8>, %arg2 : tensor<512xi32>) -> () {
+ // CHECK: [[INIT:%.+]] = linalg.init_tensor [1, 10, 10, 512]
+ // CHECK: [[BIAS:%.+]] = linalg.generic {indexing_maps = [#[[$MAP0]], #[[$MAP1]]], iterator_types = ["parallel", "parallel", "parallel", "parallel"]} ins(%arg2 : tensor<512xi32>) outs([[INIT]] : tensor<1x10x10x512xi32>) {
+ // CHECK: ^bb0(%arg3: i32, %arg4: i32): // no predecessors
+ // CHECK: linalg.yield %arg3 : i32
+ // CHECK: } -> tensor<1x10x10x512xi32>
+ // CHECK: [[DBIAS:%.+]] = linalg.tensor_expand_shape [[BIAS]] {{\[}}[0], [1], [2], [3, 4]]
+ // CHECK: %[[C128:.+]] = constant -128
+ // CHECK: %[[C42:.+]] = constant 42
+ // CHECK: [[DEPTH:%.+]] = linalg.depthwise_conv2D_nchw_q {dilations = dense<1> : tensor<2xi64>, strides = dense<1> : tensor<2xi64>} ins(%arg0, %arg1, %[[C128]], %[[C42]] : tensor<1x12x12x4xi8>, tensor<3x3x4x128xi8>, i32, i32) outs([[DBIAS]] : tensor<1x10x10x4x128xi32>)
+ // CHECK: linalg.tensor_collapse_shape %3 {{\[}}[0], [1], [2], [3, 4]]
+ %0 = "tosa.depthwise_conv2d"(%arg0, %arg1, %arg2) {pad = [0, 0, 0, 0], quantization_info = {input_zp = -128 : i32, weight_zp = 42 : i32}, stride = [1, 1], dilation = [1, 1] } : (tensor<1x12x12x4xi8>, tensor<3x3x4x128xi8>, tensor<512xi32>) -> tensor<1x10x10x512xi32>
+ return
+}
+
+// -----
+
// CHECK-LABEL: @transpose_conv
func @transpose_conv(%arg0 : tensor<1x12x12x2xf32>, %arg1 : tensor<4x3x3x2xf32>, %arg2 : tensor<4xf32>) -> () {
// CHECK: [[PAD:%.+]] = linalg.pad_tensor %arg0 low[0, 2, 2, 0] high[0, 2, 2, 0]
More information about the Mlir-commits
mailing list