[all-commits] [llvm/llvm-project] fad84c: [mlir][sparse] Support sparse2sparse collapse for ...

[Anlun Xu via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: fad84c3dbe85f3dd4a74ece6deab4306c813248e
      https://github.com/llvm/llvm-project/commit/fad84c3dbe85f3dd4a74ece6deab4306c813248e
  Author: Anlun Xu <anlunx at google.com>
  Date:   2022-09-27 (Tue, 27 Sep 2022)

  Changed paths:
    M mlir/lib/Dialect/SparseTensor/Transforms/SparseTensorConversion.cpp
    M mlir/test/Dialect/SparseTensor/sparse_reshape.mlir
    M mlir/test/Integration/Dialect/SparseTensor/CPU/sparse_reshape.mlir

  Log Message:
  -----------
  [mlir][sparse] Support sparse2sparse collapse for dynamic sizes

This patch implements sparse2sparse collapse for operands with dynamic shape.

Reviewed By: aartbik

Differential Revision: https://reviews.llvm.org/D131599