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

[Benjamin Maxwell]

================
@@ -0,0 +1,127 @@
+//===- TransposeConv2D.cpp - Convoultion transposition  -------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "mlir/Dialect/Func/IR/FuncOps.h"
+#include "mlir/Dialect/Linalg/IR/Linalg.h"
+#include "mlir/Dialect/Linalg/Passes.h"
+#include "mlir/IR/BuiltinTypes.h"
+#include "mlir/IR/PatternMatch.h"
+#include "mlir/IR/ValueRange.h"
+#include "mlir/Pass/Pass.h"
+#include "mlir/Support/LogicalResult.h"
+#include "mlir/Transforms/DialectConversion.h"
+#include "mlir/Transforms/GreedyPatternRewriteDriver.h"
+#include "llvm/ADT/SmallVector.h"
+#include <memory>
+#include <numeric>
+
+namespace mlir {
+#define GEN_PASS_DEF_LINALGTRANSPOSECONV2D
+#include "mlir/Dialect/Linalg/Passes.h.inc"
+} // namespace mlir
+
+using namespace mlir;
+
+namespace {
+// Convolution converter that applies the following rewrite:
+//
+// clang-format off
+// Before:
+//
+//   %0 = linalg.conv_2d_nhwc_fhwc {dilations = dense<1> : tensor<2xi64>,
+//                                               strides = dense<2> : tensor<2xi64>}
+//      ins (%input, %filter: tensor<1x4x4x6xf32>, tensor<8x2x2x6xf32>)
+//     outs (%init: tensor<1x2x2x8xf32>) -> tensor<1x2x2x8xf32>
+//
+// After:
+//
+//    %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>)
+//                  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>)
+//         -> tensor<1x2x2x8xf32>
+//
+// clang-format on
+// with an analogous example for the quantized case.
----------------
MacDue wrote:

nit: Use `///` for doc comments and move the clang-format directives to outside the docs
e.g.
```c++
// clang-format off
/// Convolution converter that applies the following rewrite:
///
/// Before:
/// ...
// clang-format on
```

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