Bugfix for grouped convolution op

[gpetters94]

Summary

Corrects the output order for Conv2dNgchwFgchw to the order expected by Torch-MLIR.

[silvas]

What is the bug, and how does this fix it?

[gpetters94]

In D128880#3633757, @silvas wrote:

What is the bug, and how does this fix it?

I was wrong about the order that Torch-MLIR expects for the output of this op because the test model I was using masked it. This corrects the output order to work with any grouped case, not just that specific model.

[silvas]

Okay, can you update the commit message to reflect that? Right now it's a bit hard to understand.

[gpetters94]

Can do. Is this better?

[silvas]

sg

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

value: !ScalarExpression
                 operands:
                 - !ScalarExpression
                   scalar_arg: K
 --- !LinalgOpConfig
 metadata: !LinalgOpMetadata
   name: conv_2d_ngchw_fgchw
   cpp_class_name: Conv2DNgchwFgchwOp
   doc: |-
-    Performs 2-D convolution.
+    Performs 2-D grouped convolution.
 
     Layout:
       * Input: NGCHW.
       * Kernel: FGCHW.
 
     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] -> (s0,
-      s1, s2 * s3 + s4 * s5, s6 * s7 + s8 * s9)>
+    shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] ->
+      (s0, s1, s2, s3 * s4 + s5 * s6, s7 * s8 + s9 * s10)>
   - !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] -> (s10,
-      s1, s11, s4, s8)>
+    shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] ->
+      (s11, s1, s2, s5, s9)>
   - !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] -> (s0,
-      s1, s10, s2, s6)>
+    shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] ->
+      (s0, s11, s1, s3, s7)>
   - !LinalgOperandDefConfig
     name: strides
     kind: index_attr
-    index_attr_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] ->
-      (s3, s7)>
+    index_attr_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11]
+      -> (s4, s8)>
     default_indices:
     - 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] ->
-      (s5, s9)>
+    index_attr_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11]
+      -> (s6, s10)>
     default_indices:
     - 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] -> (d0, d1, d5, d3 * s3 + d6 * s5, d4 * s7 + d7 * s9)>
-    - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
-      s9, s10, s11] -> (d2, d1, 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] -> (d0, d1, d2, d3, d4)>
+    - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7,
+      s8, s9, s10, s11] -> (d0, d1, d5, d3 * s4 + d6 * s6, d4 * s8 + d7 * s10)>
+    - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7,
+      s8, s9, s10, s11] -> (d1, d2, 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] -> (d0, d1, d2, d3, d4)>
   iterator_types:
   - parallel
   - parallel
   - parallel
   - parallel
   - parallel
   - reduction
   - reduction

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

def conv_2d_nchw_fchw(I=TensorDef(T1, S.N, S.C, S.OH * S.SH + S.KH * S.DH,
   O[D.n, D.f, D.oh, D.ow] += TypeFn.cast_signed(
       U, I[D.n, D.c, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW +
            D.kw * S.DW]) * TypeFn.cast_signed(U, K[D.f, D.c, D.kh, D.kw])
 
 @linalg_structured_op
 def conv_2d_ngchw_fgchw(I=TensorDef(T1, S.N, S.G, S.C, S.OH * S.SH + S.KH * S.DH,
                                   S.OW * S.SW + S.KW * S.DW),
                       K=TensorDef(T2, S.FG, S.G, S.C, S.KH, S.KW),
-                      O=TensorDef(U, S.N, S.G, S.FG, S.OH, S.OW, output=True),
+                      O=TensorDef(U, S.N, S.FG, S.G, S.OH, S.OW, output=True),
                       strides=IndexAttrDef(S.SH, S.SW, default=[1, 1]),
                       dilations=IndexAttrDef(S.DH, S.DW, default=[1, 1])):
   """Performs 2-D grouped convolution.
 
   Layout:
     * Input: NGCHW.
     * Kernel: FGCHW.
 
   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.g, D.fg, D.oh, D.ow, D.c, D.kh, D.kw)
   O[D.n, D.g, D.fg, D.oh, D.ow] += TypeFn.cast_signed(
       U, I[D.n, D.g, D.c, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW +
-          D.kw * S.DW]) * TypeFn.cast_signed(U, K[D.fg, D.g, D.c, D.kh, D.kw])
+          D.kw * S.DW]) * TypeFn.cast_signed(U, K[D.g, D.fg, D.c, D.kh, D.kw])
 
 @linalg_structured_op
 def conv_3d_ndhwc_dhwcf(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.C),
                         K=TensorDef(T2, S.KD, S.KH, S.KW, S.C, S.F),
                         O=TensorDef(U, S.N, S.OD, S.OH, S.OW, S.F, output=True),
                         strides=IndexAttrDef(S.SD,