[all-commits] [llvm/llvm-project] 1aaf8a: [mlir][Linalg] Conv1D, Conv2D and Conv3D added as ...

[Jakub Lichman via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 1aaf8aa53d694309087b322861038130490bdd5e
      https://github.com/llvm/llvm-project/commit/1aaf8aa53d694309087b322861038130490bdd5e
  Author: Jakub Lichman <limo at google.com>
  Date:   2020-07-29 (Wed, 29 Jul 2020)

  Changed paths:
    M mlir/include/mlir/Dialect/Linalg/IR/LinalgOps.h
    M mlir/include/mlir/Dialect/Linalg/IR/LinalgStructuredOps.td
    M mlir/lib/Conversion/LinalgToStandard/LinalgToStandard.cpp
    M mlir/lib/Dialect/Linalg/IR/LinalgOps.cpp
    M mlir/lib/Dialect/Linalg/Transforms/Loops.cpp
    M mlir/test/Dialect/Linalg/invalid.mlir
    M mlir/test/Dialect/Linalg/loops.mlir

  Log Message:
  -----------
  [mlir][Linalg] Conv1D, Conv2D and Conv3D added as named ops

This commit is part of a greater project which aims to add
full end-to-end support for convolutions inside mlir. The
reason behind having conv ops for each rank rather than
having one generic ConvOp is to enable better optimizations
for every N-D case which reflects memory layout of input/kernel
buffers better and simplifies code as well. We expect plain linalg.conv
to be progressively retired.

Reviewed By: ftynse

Differential Revision: https://reviews.llvm.org/D83879