[Mlir-commits] [mlir] [mlir][linalg] Add TransposeConv2D Pass (PR #68567)

[llvmlistbot at llvm.org]

github-actions[bot] wrote:


<!--LLVM CODE FORMAT COMMENT: {clang-format}-->

:warning: C/C++ code formatter, clang-format found issues in your code. :warning:

<details>
<summary>
You can test this locally with the following command:
</summary>

``````````bash
git-clang-format --diff 906d3ff054b0ca6439521e8871bac54843cc73a2 944c68b681753300d57b24a06de168c184c3f1c2 -- mlir/lib/Dialect/Linalg/Transforms/TransposeConv2D.cpp mlir/include/mlir/Dialect/Linalg/Passes.h
``````````

</details>

<details>
<summary>
View the diff from clang-format here.
</summary>

``````````diff
diff --git a/mlir/lib/Dialect/Linalg/Transforms/TransposeConv2D.cpp b/mlir/lib/Dialect/Linalg/Transforms/TransposeConv2D.cpp
index ec9e590750ff..22e592a11342 100644
--- a/mlir/lib/Dialect/Linalg/Transforms/TransposeConv2D.cpp
+++ b/mlir/lib/Dialect/Linalg/Transforms/TransposeConv2D.cpp
@@ -34,7 +34,8 @@ namespace {
 /// Before:
 ///
 ///   %0 = linalg.conv_2d_nhwc_fhwc {dilations = dense<1> : tensor<2xi64>,
-///                                               strides = dense<2> : tensor<2xi64>}
+///                                               strides = dense<2> :
+///                                               tensor<2xi64>}
 ///      ins (%input, %filter: tensor<1x4x4x6xf32>, tensor<8x2x2x6xf32>)
 ///     outs (%init: tensor<1x2x2x8xf32>) -> tensor<1x2x2x8xf32>
 ///
@@ -42,11 +43,14 @@ namespace {
 ///
 ///    %cst = arith.constant 0.000000e+00 : f32
 ///    %0 = tensor.empty() : tensor<2x2x6x8xf32>
-///    %1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<2x2x6x8xf32>) -> tensor<2x2x6x8xf32>
-///    %transposed = linalg.transpose ins(%arg1 : tensor<8x2x2x6xf32>) outs(%1 : tensor<2x2x6x8xf32>)
+///    %1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<2x2x6x8xf32>) ->
+///    tensor<2x2x6x8xf32> %transposed = linalg.transpose ins(%arg1 :
+///    tensor<8x2x2x6xf32>) outs(%1 : tensor<2x2x6x8xf32>)
 ///                  permutation = [1, 2, 3, 0]
-///    %2 = linalg.conv_2d_nhwc_hwcf {dilations = dense<1> : tensor<2xi64>, strides = dense<2> : tensor<2xi64>}
-///         ins(%arg0, %transposed : tensor<1x4x4x6xf32>, tensor<2x2x6x8xf32>) outs(%arg2 : tensor<1x2x2x8xf32>)
+///    %2 = linalg.conv_2d_nhwc_hwcf {dilations = dense<1> : tensor<2xi64>,
+///    strides = dense<2> : tensor<2xi64>}
+///         ins(%arg0, %transposed : tensor<1x4x4x6xf32>, tensor<2x2x6x8xf32>)
+///         outs(%arg2 : tensor<1x2x2x8xf32>)
 ///         -> tensor<1x2x2x8xf32>
 ///
 /// with an analogous example for the quantized case.

``````````

</details>


https://github.com/llvm/llvm-project/pull/68567