[mlir][linalg] Add named ops for depthwise 3d convolution

[bkramer]

Summary

Also complete the set by adding a variant of depthwise 1d convolution
with the multiplier != 1.

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

value: !ScalarExpression
                 kind: type
                 fn_name: cast_signed
                 type_var: U
                 operands:
                 - !ScalarExpression
                   scalar_arg: K
 --- !LinalgOpConfig
 metadata: !LinalgOpMetadata
+  name: depthwise_conv_1d_nwc_wcm
+  cpp_class_name: DepthwiseConv1DNwcWcmOp
+  doc: |-
+    Performs depth-wise 1-D convolution.
+
+    Numeric casting is performed on the operands to the inner multiply, promoting
+    them to the same data type as the accumulator/output.
+  implements:
+  - LinalgConvolutionOpInterface
+structured_op: !LinalgStructuredOpConfig
+  args:
+  - !LinalgOperandDefConfig
+    name: I
+    kind: input_tensor
+    type_var: T1
+    shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6] -> (s0, s1 * s2 + s3 * s4,
+      s5)>
+  - !LinalgOperandDefConfig
+    name: K
+    kind: input_tensor
+    type_var: T2
+    shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6] -> (s3, s5, s6)>
+  - !LinalgOperandDefConfig
+    name: O
+    kind: output_tensor
+    type_var: U
+    shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6] -> (s0, s1, s5, s6)>
+  - !LinalgOperandDefConfig
+    name: strides
+    kind: index_attr
+    index_attr_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6] -> (s2)>
+    default_indices:
+    - 1
+  - !LinalgOperandDefConfig
+    name: dilations
+    kind: index_attr
+    index_attr_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6] -> (s4)>
+    default_indices:
+    - 1
+  indexing_maps: !LinalgIndexingMapsConfig
+    static_indexing_maps:
+    - affine_map<(d0, d1, d2, d3, d4)[s0, s1, s2, s3, s4, s5, s6] -> (d0, d1 * s2
+      + d4 * s4, d2)>
+    - affine_map<(d0, d1, d2, d3, d4)[s0, s1, s2, s3, s4, s5, s6] -> (d4, d2, d3)>
+    - affine_map<(d0, d1, d2, d3, d4)[s0, s1, s2, s3, s4, s5, s6] -> (d0, d1, d2,
+      d3)>
+  iterator_types:
+  - parallel
+  - parallel
+  - parallel
+  - parallel
+  - reduction
+  assignments:
+  - !ScalarAssign
+    arg: O
+    value: !ScalarExpression
+      scalar_fn:
+        kind: binary
+        fn_name: add
+        operands:
+        - !ScalarExpression
+          scalar_arg: O
+        - !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: I
+            - !ScalarExpression
+              scalar_fn:
+                kind: type
+                fn_name: cast_signed
+                type_var: U
+                operands:
+                - !ScalarExpression
+                  scalar_arg: K
+--- !LinalgOpConfig
+metadata: !LinalgOpMetadata
   name: depthwise_conv_2d_nhwc_hwc
   cpp_class_name: DepthwiseConv2DNhwcHwcOp
   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. Multiplier is set to 1
     which is a special case for most depthwise convolutions.
value: !ScalarExpression
                     kind: type
                     fn_name: cast_signed
                     type_var: U
                     operands:
                     - !ScalarExpression
                       scalar_arg: KZp
 --- !LinalgOpConfig
 metadata: !LinalgOpMetadata
+  name: depthwise_conv_3d_ndhwc_dhwc
+  cpp_class_name: DepthwiseConv3DNdhwcDhwcOp
+  doc: |-
+    Performs depth-wise 3-D convolution.
+
+    Numeric casting is performed on the operands to the inner multiply, promoting
+    them to the same data type as the accumulator/output. Multiplier is set to 1
+    which is a special case for most depthwise convolutions.
+  implements:
+  - LinalgConvolutionOpInterface
+structured_op: !LinalgStructuredOpConfig
+  args:
+  - !LinalgOperandDefConfig
+    name: I
+    kind: input_tensor
+    type_var: T1
+    shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
+      s13] -> (s0, s1 * s2 + s3 * s4, s5 * s6 + s7 * s8, s9 * s10 + s11 * s12, s13)>
+  - !LinalgOperandDefConfig
+    name: K
+    kind: input_tensor
+    type_var: T2
+    shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
+      s13] -> (s3, s7, s11, s13)>
+  - !LinalgOperandDefConfig
+    name: O
+    kind: output_tensor
+    type_var: U
+    shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
+      s13] -> (s0, s1, s5, s9)>
+  - !LinalgOperandDefConfig
+    name: strides
+    kind: index_attr
+    index_attr_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
+      s12, s13] -> (s2, s6, s10)>
+    default_indices:
+    - 1
+    - 1
+    - 1
+  - !LinalgOperandDefConfig
+    name: dilations
+    kind: index_attr
+    index_attr_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
+      s12, s13] -> (s4, s8, s12)>
+    default_indices:
+    - 1
+    - 1
+    - 1
+  indexing_maps: !LinalgIndexingMapsConfig
+    static_indexing_maps:
+    - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7,
+      s8, s9, s10, s11, s12, s13] -> (d0, d1 * s2 + d4 * s4, d2 * s6 + d5 * s8, d3
+      * s10 + d6 * s12, d7)>
+    - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7,
+      s8, s9, s10, s11, s12, s13] -> (d4, d5, d6, d7)>
+    - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7,
+      s8, s9, s10, s11, s12, s13] -> (d0, d1, d2, d3, d7)>
+  iterator_types:
+  - parallel
+  - parallel
+  - parallel
+  - parallel
+  - reduction
+  - reduction
+  - reduction
+  - parallel
+  assignments:
+  - !ScalarAssign
+    arg: O
+    value: !ScalarExpression
+      scalar_fn:
+        kind: binary
+        fn_name: add
+        operands:
+        - !ScalarExpression
+          scalar_arg: O
+        - !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: I
+            - !ScalarExpression
+              scalar_fn:
+                kind: type
+                fn_name: cast_signed
+                type_var: U
+                operands:
+                - !ScalarExpression
+                  scalar_arg: K
+--- !LinalgOpConfig
+metadata: !LinalgOpMetadata
+  name: depthwise_conv_3d_ndhwc_dhwcm
+  cpp_class_name: DepthwiseConv3DNdhwcDhwcmOp
+  doc: |-
+    Performs depth-wise 3-D convolution.
+
+    Numeric casting is performed on the operands to the inner multiply, promoting
+    them to the same data type as the accumulator/output.
+  implements:
+  - LinalgConvolutionOpInterface
+structured_op: !LinalgStructuredOpConfig
+  args:
+  - !LinalgOperandDefConfig
+    name: I
+    kind: input_tensor
+    type_var: T1
+    shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
+      s13, s14] -> (s0, s1 * s2 + s3 * s4, s5 * s6 + s7 * s8, s9 * s10 + s11 * s12,
+      s13)>
+  - !LinalgOperandDefConfig
+    name: K
+    kind: input_tensor
+    type_var: T2
+    shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
+      s13, s14] -> (s3, s7, s11, s13, s14)>
+  - !LinalgOperandDefConfig
+    name: O
+    kind: output_tensor
+    type_var: U
+    shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
+      s13, s14] -> (s0, s1, s5, s9, s14)>
+  - !LinalgOperandDefConfig
+    name: strides
+    kind: index_attr
+    index_attr_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
+      s12, s13, s14] -> (s2, s6, s10)>
+    default_indices:
+    - 1
+    - 1
+    - 1
+  - !LinalgOperandDefConfig
+    name: dilations
+    kind: index_attr
+    index_attr_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
+      s12, s13, s14] -> (s4, s8, s12)>
+    default_indices:
+    - 1
+    - 1
+    - 1
+  indexing_maps: !LinalgIndexingMapsConfig
+    static_indexing_maps:
+    - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7, d8)[s0, s1, s2, s3, s4, s5, s6,
+      s7, s8, s9, s10, s11, s12, s13, s14] -> (d0, d1 * s2 + d5 * s4, d2 * s6 + d6
+      * s8, d3 * s10 + d7 * s12, d8)>
+    - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7, d8)[s0, s1, s2, s3, s4, s5, s6,
+      s7, s8, s9, s10, s11, s12, s13, s14] -> (d5, d6, d7, d8, d4)>
+    - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7, d8)[s0, s1, s2, s3, s4, s5, s6,
+      s7, s8, s9, s10, s11, s12, s13, s14] -> (d0, d1, d2, d3, d8, d4)>
+  iterator_types:
+  - parallel
+  - parallel
+  - parallel
+  - parallel
+  - parallel
+  - reduction
+  - reduction
+  - reduction
+  - parallel
+  assignments:
+  - !ScalarAssign
+    arg: O
+    value: !ScalarExpression
+      scalar_fn:
+        kind: binary
+        fn_name: add
+        operands:
+        - !ScalarExpression
+          scalar_arg: O
+        - !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: I
+            - !ScalarExpression
+              scalar_fn:
+                kind: type
+                fn_name: cast_signed
+                type_var: U
+                operands:
+                - !ScalarExpression
+                  scalar_arg: K
+--- !LinalgOpConfig
+metadata: !LinalgOpMetadata
   name: pooling_nhwc_sum
   cpp_class_name: PoolingNhwcSumOp
   doc: |-
     Performs sum pooling.
 
     Layout:
       * Input: NHWC.
       * Kernel: HW.

mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py

def depthwise_conv_1d_nwc_wc(I=TensorDef(T1, S.N, S.OW * S.SW + S.KW * S.DW,
   implements(ConvolutionOpInterface)
   domain(D.n, D.ow, D.ic, D.kw)
   O[D.n, D.ow, D.ic] += \
       TypeFn.cast_signed(U, I[D.n, D.ow * S.SW + D.kw * S.DW, D.ic]) * \
       TypeFn.cast_signed(U, K[D.kw, D.ic])
 
 
 @linalg_structured_op
+def depthwise_conv_1d_nwc_wcm(I=TensorDef(T1, S.N, S.OW * S.SW + S.KW * S.DW,
+                                          S.IC),
+                              K=TensorDef(T2, S.KW, S.IC, S.CM),
+                              O=TensorDef(U, S.N, S.OW, S.IC, S.CM,
+                                          output=True),
+                              strides=IndexAttrDef(S.SW, default=[1]),
+                              dilations=IndexAttrDef(S.DW, default=[1])):
+  """Performs depth-wise 1-D convolution.
+
+  Numeric casting is performed on the operands to the inner multiply, promoting
+  them to the same data type as the accumulator/output.
+  """
+  implements(ConvolutionOpInterface)
+  domain(D.n, D.ow, D.ic, D.cm, D.kw)
+  O[D.n, D.ow, D.ic, D.cm] += \
+      TypeFn.cast_signed(U, I[D.n, D.ow * S.SW + D.kw * S.DW, D.ic]) * \
+      TypeFn.cast_signed(U, K[D.kw, D.ic, D.cm])
+
+
+@linalg_structured_op
 def depthwise_conv_2d_nhwc_hwc(I=TensorDef(T1, S.N, S.OH * S.SH + S.KH * S.DH,
                                            S.OW * S.SW + S.KW * S.DW, S.IC),
                                K=TensorDef(T2, S.KH, S.KW, S.IC),
                                O=TensorDef(U,
                                            S.N,
                                            S.OH,
                                            S.OW,
                                            S.IC,
O[D.n, D.oh, D.ow, D.ic,
     D.cm] += ((TypeFn.cast_signed(
         U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, D.ic]) -
                TypeFn.cast_signed(U, IZp)) *
               (TypeFn.cast_signed(U, K[D.kh, D.kw, D.ic, D.cm]) -
                TypeFn.cast_signed(U, KZp)))
 
 
 @linalg_structured_op
+def depthwise_conv_3d_ndhwc_dhwc(I=TensorDef(T1, S.N, S.OD * S.SD + S.KD * S.DD,
+                                             S.OH * S.SH + S.KH * S.DH,
+                                             S.OW * S.SW + S.KW * S.DW, S.IC),
+                                 K=TensorDef(T2, S.KD, S.KH, S.KW, S.IC),
+                                 O=TensorDef(U, S.N, S.OD, S.OH, S.OW,
+                                             output=True),
+                                 strides=IndexAttrDef(S.SD,
+                                                      S.SH,
+                                                      S.SW,
+                                                      default=[1, 1, 1]),
+                                 dilations=IndexAttrDef(S.DD,
+                                                        S.DH,
+                                                        S.DW,
+                                                        default=[1, 1, 1])):
+  """Performs depth-wise 3-D convolution.
+
+  Numeric casting is performed on the operands to the inner multiply, promoting
+  them to the same data type as the accumulator/output. Multiplier is set to 1
+  which is a special case for most depthwise convolutions.
+  """
+  implements(ConvolutionOpInterface)
+  domain(D.n, D.od, D.oh, D.ow, D.kd, D.kh, D.kw, D.ic)
+  O[D.n, D.od, D.oh, D.ow, D.ic] += TypeFn.cast_signed(
+      U, I[D.n, D.od * S.SD + D.kd * S.DD, D.oh * S.SH + D.kh * S.DH,
+           D.ow * S.SW + D.kw * S.DW, D.ic]) * TypeFn.cast_signed(
+               U, K[D.kd, D.kh, D.kw, D.ic])
+
+
+@linalg_structured_op
+def depthwise_conv_3d_ndhwc_dhwcm(I=TensorDef(T1,
+                                              S.N, S.OD * S.SD + S.KD * S.DD,
+                                              S.OH * S.SH + S.KH * S.DH,
+                                              S.OW * S.SW + S.KW * S.DW, S.IC),
+                                  K=TensorDef(T2, S.KD, S.KH, S.KW, S.IC, S.CM),
+                                  O=TensorDef(U, S.N, S.OD, S.OH, S.OW, S.CM,
+                                              output=True),
+                                  strides=IndexAttrDef(S.SD,
+                                                       S.SH,
+                                                       S.SW,
+                                                       default=[1, 1, 1]),
+                                  dilations=IndexAttrDef(S.DD,
+                                                         S.DH,
+                                                         S.DW,
+                                                         default=[1, 1, 1])):
+  """Performs depth-wise 3-D convolution.
+
+  Numeric casting is performed on the operands to the inner multiply, promoting
+  them to the same data type as the accumulator/output.
+  """
+  implements(ConvolutionOpInterface)
+  domain(D.n, D.od, D.oh, D.ow, D.cm, D.kd, D.kh, D.kw, D.ic)
+  O[D.n, D.od, D.oh, D.ow, D.ic, D.cm] += TypeFn.cast_signed(
+      U, I[D.n, D.od * S.SD + D.kd * S.DD, D.oh * S.SH + D.kh * S.DH,
+           D.ow * S.SW + D.kw * S.DW, D.ic]) * TypeFn.cast_signed(
+               U, K[D.kd, D.kh, D.kw, D.ic, D.cm])
+
+
+@linalg_structured_op
 def pooling_nhwc_sum(I=TensorDef(T1, S.N, S.OH * S.SH + S.KH * S.DH,
                                  S.OW * S.SW + S.KW * S.DW, S.C),
                      K=TensorDef(T2, S.KH, S.KW, index_dims=[D.kh, D.kw]),
                      O=TensorDef(U, S.N, S.OH, S.OW, S.C, output=True),
                      strides=IndexAttrDef(S.SH, S.SW, default=[1, 1]),
                      dilations=IndexAttrDef(S.DH, S.DW, default=[1, 1])):
   """Performs sum pooling.
 

mlir/test/Dialect/Linalg/named-ops.mlir

 // RUN: mlir-opt -split-input-file -verify-diagnostics %s | FileCheck %s
 
+// CHECK-LABEL: func @depthwise_conv_1d_nwc_wcm
+func.func @depthwise_conv_1d_nwc_wcm(%input: tensor<1x12x8xf32>, %filter: tensor<3x8x8xf32>) -> tensor<1x10x8x8xf32> {
+  %zero = arith.constant 0.000000e+00 : f32
+  %init = linalg.init_tensor [1, 10, 8, 8] : tensor<1x10x8x8xf32>
+  %fill = linalg.fill ins(%zero : f32) outs(%init : tensor<1x10x8x8xf32>) -> tensor<1x10x8x8xf32>
+  // CHECK: depthwise_conv_1d_nwc_wcm
+  %0 = linalg.depthwise_conv_1d_nwc_wcm {dilations = dense<1> : tensor<1xi64>, strides = dense<1> : tensor<1xi64>}
+    ins(%input, %filter : tensor<1x12x8xf32>, tensor<3x8x8xf32>)
+    outs(%fill : tensor<1x10x8x8xf32>) -> tensor<1x10x8x8xf32>
+  return %0 : tensor<1x10x8x8xf32>
+}
+
+// -----
+
+// CHECK-LABEL: func @depthwise_conv_1d_nwc_wc
+func.func @depthwise_conv_1d_nwc_wc(%input: tensor<1x12x8xf32>, %filter: tensor<3x8xf32>) -> tensor<1x10x8xf32> {
+  %zero = arith.constant 0.000000e+00 : f32
+  %init = linalg.init_tensor [1, 10, 8] : tensor<1x10x8xf32>
+  %fill = linalg.fill ins(%zero : f32) outs(%init : tensor<1x10x8xf32>) -> tensor<1x10x8xf32>
+  // CHECK: depthwise_conv_1d_nwc_wc
+  %0 = linalg.depthwise_conv_1d_nwc_wc {dilations = dense<1> : tensor<1xi64>, strides = dense<1> : tensor<1xi64>}
+    ins(%input, %filter : tensor<1x12x8xf32>, tensor<3x8xf32>)
+    outs(%fill : tensor<1x10x8xf32>) -> tensor<1x10x8xf32>
+  return %0 : tensor<1x10x8xf32>
+}
+
+// -----
+
 // CHECK-LABEL: func @depthwise_conv_2d_nhwc_hwcm_tensor
 func.func @depthwise_conv_2d_nhwc_hwcm_tensor(%input: tensor<2x4x5x2xf32>, %filter: tensor<2x2x2x3xf32>) -> tensor<2x3x4x2x3xf32> {
   %zero = arith.constant 0.000000e+00 : f32
   %init = linalg.init_tensor [2, 3, 4, 2, 3] : tensor<2x3x4x2x3xf32>
   %fill = linalg.fill ins(%zero : f32) outs(%init : tensor<2x3x4x2x3xf32>) -> tensor<2x3x4x2x3xf32>
   // CHECK:      %{{.+}} = linalg.depthwise_conv_2d_nhwc_hwcm
   // CHECK-SAME:   {dilations = dense<1> : tensor<2xi64>, strides = dense<1> : tensor<2xi64>}
   // CHECK-SAME:   ins(%{{.+}}, %{{.+}} : tensor<2x4x5x2xf32>, tensor<2x2x2x3xf32>)
func.func @depthwise_conv_2d_input_nhwc_filter_wrong_stride_size(%input: memref<1x113x113x96xf32>, %filter: memref<3x3x96xf32>, %output: memref<1x56x56x96xf32>) {
   linalg.depthwise_conv_2d_nhwc_hwc {dilations = dense<1> : vector<2xi64>, strides = dense<2> : vector<3xi64> }
     ins(%input, %filter: memref<1x113x113x96xf32>, memref<3x3x96xf32>)
     outs(%output: memref<1x56x56x96xf32>)
   return
 }
 
 // -----
 
+// CHECK-LABEL: func @depthwise_conv_3d_ndhwc_dhwcm
+func.func @depthwise_conv_3d_ndhwc_dhwcm(%input: tensor<2x6x13x12x6xf32>, %filter: tensor<2x1x3x6x6xf32>) -> tensor<2x3x13x4x6x6xf32> {
+  %zero = arith.constant 0.000000e+00 : f32
+  %init = linalg.init_tensor [2, 3, 13, 4, 6, 6] : tensor<2x3x13x4x6x6xf32>
+  %fill = linalg.fill ins(%zero : f32) outs(%init : tensor<2x3x13x4x6x6xf32>) -> tensor<2x3x13x4x6x6xf32>
+  // CHECK: depthwise_conv_3d_ndhwc_dhwcm
+  %0 = linalg.depthwise_conv_3d_ndhwc_dhwcm {dilations = dense<1> : tensor<3xi64>, strides = dense<[2, 1, 3]> : tensor<3xi64>}
+    ins(%input, %filter : tensor<2x6x13x12x6xf32>, tensor<2x1x3x6x6xf32>)
+    outs(%fill : tensor<2x3x13x4x6x6xf32>) -> tensor<2x3x13x4x6x6xf32>
+  return %0 : tensor<2x3x13x4x6x6xf32>
+}
+
+// -----
+
+// CHECK-LABEL: func @depthwise_conv_3d_ndhwc_dhwc
+func.func @depthwise_conv_3d_ndhwc_dhwc(%input: tensor<2x6x13x12x6xf32>, %filter: tensor<2x1x3x6xf32>) -> tensor<2x3x13x4x6xf32> {
+  %zero = arith.constant 0.000000e+00 : f32
+  %init = linalg.init_tensor [2, 3, 13, 4, 6] : tensor<2x3x13x4x6xf32>
+  %fill = linalg.fill ins(%zero : f32) outs(%init : tensor<2x3x13x4x6xf32>) -> tensor<2x3x13x4x6xf32>
+  // CHECK: depthwise_conv_3d_ndhwc_dhwc
+  %0 = linalg.depthwise_conv_3d_ndhwc_dhwc {dilations = dense<1> : tensor<3xi64>, strides = dense<[2, 1, 3]> : tensor<3xi64>}
+    ins(%input, %filter : tensor<2x6x13x12x6xf32>, tensor<2x1x3x6xf32>)
+    outs(%fill : tensor<2x3x13x4x6xf32>) -> tensor<2x3x13x4x6xf32>
+  return %0 : tensor<2x3x13x4x6xf32>
+}
+
+// -----
+
 // CHECK-LABEL: func @conv_1d_nwc_wcf
 func.func @conv_1d_nwc_wcf(%input: tensor<?x?x?xf32>, %filter: tensor<?x?x?xf32>, %init: tensor<?x?x?xf32>) -> tensor<?x?x?xf32> {
   // CHECK:      %{{.+}} = linalg.conv_1d_nwc_wcf
   // CHECK-SAME:   dilations = dense<1> : tensor<1xi64>
   // CHECK-SAME:   strides = dense<1> : tensor<1xi64>
   // CHECK-SAME:   ins(%{{.+}}, %{{.+}} : tensor<?x?x?xf32>, tensor<?x?x?xf32>)
   // CHECK-SAME:   outs(%{{.+}} : tensor<?x?x?xf32>) -> tensor<?x?x?xf32>
   %0 = linalg.conv_1d_nwc_wcf {dilations = dense<1> : tensor<1xi64>,