[MLIR][TORCH] Add support for 1-d group convolution (#3770)

This commit adds the support for the 1-d depthwise convolution as a
special case of 1-d group convolution.

Signed-Off By: Vivek Khandelwal <vivekkhandelwal1424@gmail.com>

lib/Conversion/TorchToLinalg/Linear.cpp

@@ -1184,10 +1184,6 @@ public:
       return success();
     }
 
-    if (numSpatialDims != 2)
-      return rewriter.notifyMatchFailure(
-          op, "unimplemented: only 2D grouped convolution supported");
-
     // Special depthwise case: Cin = Cout = groups.
     // Note: pytorch considers Cin == groups (Cout possibly a non-zero multiple
     // of groups) to be depthwise in their documentation, but the linalg ops
@@ -1199,21 +1195,45 @@ public:
     if (inShape[1] == numGroups && weightShape[0] == numGroups &&
         weightShape[1] == 1) {
       // Collapse weight shape (C/G == 1)
-      SmallVector<ReassociationIndices, 4> collapsedDims = {{0, 1}, {2}, {3}};
+      SmallVector<ReassociationIndices> collapsedDims = {{0, 1}};
-      SmallVector<int64_t> collapsedShape{weightShape[0] * weightShape[1],
+      SmallVector<int64_t> collapsedShape{weightShape[0] * weightShape[1]};
-                                          weightShape[2], weightShape[3]};
+      for (unsigned i = 0; i < numSpatialDims; i++) {
+        collapsedDims.push_back({i + 2});
+        collapsedShape.push_back(weightShape[i + 2]);
+      }
       Type collapsedType = RankedTensorType::get(
           makeShapeLLVMCompatible(collapsedShape), weightDTy);
       Value collapsedWeight = rewriter.create<tensor::CollapseShapeOp>(
           loc, collapsedType, weight, collapsedDims);
       if (!inputZp) {
-        conv = rewriter
+        switch (numSpatialDims) {
-                   .create<linalg::DepthwiseConv2DNchwChwOp>(
+        case 1:
-                       loc, outputTensor.getType(),
+          conv = rewriter
-                       ValueRange{paddedInput, collapsedWeight}, outputTensor,
+                     .create<linalg::DepthwiseConv1DNcwCwOp>(
-                       stridesAttr, dilationAttr)
+                         loc, outputTensor.getType(),
-                   .getResult(0);
+                         ValueRange{paddedInput, collapsedWeight}, outputTensor,
+                         stridesAttr, dilationAttr)
+                     .getResult(0);
+          break;
+        case 2:
+          conv = rewriter
+                     .create<linalg::DepthwiseConv2DNchwChwOp>(
+                         loc, outputTensor.getType(),
+                         ValueRange{paddedInput, collapsedWeight}, outputTensor,
+                         stridesAttr, dilationAttr)
+                     .getResult(0);
+          break;
+        default:
+          return rewriter.notifyMatchFailure(
+              op, "unimplemented: only 1D and 2D depthwise convolution "
+                  "supported for special case of group convolution");
+        };
       } else {
+        if (numSpatialDims != 2)
+          return rewriter.notifyMatchFailure(
+              op, "unimplemented: only 2D depthwise quantized convolution "
+                  "supported for special case of group convolution");
+
         // currently, the only named depthwise qconv op is nhwc_hwc
         // input: nchw -> nhwc; weight (collapsed): chw -> hwc
         // linalg conv result nhwc -> nchw
@@ -1260,6 +1280,10 @@ public:
       return success();
     }
 
+    if (numSpatialDims != 2)
+      return rewriter.notifyMatchFailure(
+          op, "unimplemented: only 2D grouped convolution supported");
+
     // Grouped case, use the grouped conv linalg op
     auto expandGroups = [&](Value tensor, size_t dim) {
       auto inType = cast<RankedTensorType>(tensor.getType());

projects/pt1/e2e_testing/xfail_sets.py

@@ -1048,6 +1048,7 @@ STABLEHLO_PASS_SET = {
     "ContainsIntList_False",
     "ContainsIntList_True",
     "ContiguousModule_basic",
+    "Conv1dDepthwiseWithPaddingDilationStrideStaticModule_basic",
     "Conv2dWithPaddingDilationStrideStaticModule_basic",
     "Conv2dWithPaddingDilationStrideStaticModule_depthwise",
     "Conv2dWithPaddingDilationStrideStaticModule_depthwise_multiplier",
@@ -3395,6 +3396,7 @@ FX_IMPORTER_TOSA_XFAIL_SET = {
     "ContainsIntList_False",
     "ContainsIntList_True",
     "Conv1dModule_basic",
+    "Conv1dDepthwiseWithPaddingDilationStrideStaticModule_basic",
     "Conv2dQInt8Module_basic",
     "Conv2dQInt8Module_depthwise",
     "Conv2dQInt8Module_grouped",
@@ -4087,6 +4089,7 @@ ONNX_TOSA_XFAIL_SET = {
     "ContainsIntList_False",
     "ContainsIntList_True",
     "Conv1dModule_basic",
+    "Conv1dDepthwiseWithPaddingDilationStrideStaticModule_basic",
     "Conv2dBiasNoPaddingModule_basic",
     "Conv2dModule_basic",
     "Conv2dNoPaddingModule_basic",

projects/pt1/python/torch_mlir_e2e_test/test_suite/conv.py

@@ -1067,6 +1067,33 @@ def Conv1dModule_basic(module, tu: TestUtils):
     module.forward(inputVec, weight, bias)
 
 
+class Conv1dDepthwiseWithPaddingDilationStrideStaticModule(torch.nn.Module):
+    def __init__(self):
+        super().__init__()
+
+    @export
+    @annotate_args(
+        [
+            None,
+            ([2, 4, 6], torch.float32, True),
+            ([4, 1, 3], torch.float32, True),
+        ]
+    )
+    def forward(self, inputVec, weight):
+        return torch.ops.aten.conv1d(
+            inputVec, weight, bias=None, stride=[1], padding=[4], dilation=[1], groups=4
+        )
+
+
+@register_test_case(
+    module_factory=lambda: Conv1dDepthwiseWithPaddingDilationStrideStaticModule()
+)
+def Conv1dDepthwiseWithPaddingDilationStrideStaticModule_basic(module, tu: TestUtils):
+    inputVec = tu.rand(2, 4, 6)
+    weight = torch.randn(4, 1, 3)
+    module.forward(inputVec, weight)
+
+
 class Conv2dModule(torch.nn.Module):
     def __init__(self):
         super().__init__()