[all-commits] [llvm/llvm-project] 5c4e39: [mlir][sparse] add parallelization strategies to s...

[Aart Bik via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 5c4e397e6ce5c89d63f590857e5cb0e80237de62
      https://github.com/llvm/llvm-project/commit/5c4e397e6ce5c89d63f590857e5cb0e80237de62
  Author: Aart Bik <ajcbik at google.com>
  Date:   2020-11-24 (Tue, 24 Nov 2020)

  Changed paths:
    M mlir/include/mlir/Dialect/Linalg/Transforms/Transforms.h
    M mlir/lib/Dialect/Linalg/Transforms/Sparsification.cpp
    A mlir/test/Dialect/Linalg/sparse_parallel.mlir
    M mlir/test/lib/Transforms/TestSparsification.cpp

  Log Message:
  -----------
  [mlir][sparse] add parallelization strategies to sparse compiler

This CL adds the ability to request different parallelization strategies
for the generate code. Every "parallel" loop is a candidate, and converted
to a parallel op if it is an actual for-loop (not a while) and the strategy
allows dense/sparse outer/inner parallelization.

This will connect directly with the work of @ezhulenev on parallel loops.

Still TBD: vectorization strategy

Reviewed By: penpornk

Differential Revision: https://reviews.llvm.org/D91978