[all-commits] [llvm/llvm-project] 2fda62: [mlir][sparse][vectorization] implement "index" ve...

[Aart Bik via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 2fda6207118d1d1c19e3b66f615f332ffc2792d0
      https://github.com/llvm/llvm-project/commit/2fda6207118d1d1c19e3b66f615f332ffc2792d0
  Author: Aart Bik <ajcbik at google.com>
  Date:   2022-11-30 (Wed, 30 Nov 2022)

  Changed paths:
    M mlir/lib/Dialect/SparseTensor/Transforms/SparseVectorization.cpp
    A mlir/test/Dialect/SparseTensor/sparse_vector_index.mlir
    A mlir/test/Dialect/SparseTensor/sparse_vector_ops.mlir

  Log Message:
  -----------
  [mlir][sparse][vectorization] implement "index" vectorization

This adds the capability to vectorize computations like a[i] = i.
This also generalizes the supported unary and binary ops and
adds a test for each to ensure actual SIMD code can result.

Reviewed By: bixia

Differential Revision: https://reviews.llvm.org/D138956