[all-commits] [llvm/llvm-project] 0d82ab: [mlir] [VectorOps] Improve vector.constant_mask lo...

[Aart Bik via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 0d82ab788534bfc05a4455292e27ed035caac639
      https://github.com/llvm/llvm-project/commit/0d82ab788534bfc05a4455292e27ed035caac639
  Author: aartbik <ajcbik at google.com>
  Date:   2020-06-19 (Fri, 19 Jun 2020)

  Changed paths:
    M mlir/include/mlir/IR/Builders.h
    M mlir/lib/Dialect/Vector/VectorTransforms.cpp
    M mlir/lib/IR/Builders.cpp
    M mlir/test/Conversion/VectorToLLVM/vector-to-llvm.mlir
    M mlir/test/Dialect/Vector/vector-contract-transforms.mlir

  Log Message:
  -----------
  [mlir] [VectorOps] Improve vector.constant_mask lowering

Use direct vector constants for the 1-D case. This approach
scales much better than generating elaborate insertion operations
that are eventually folded into a constant. We could of course
generalize the 1-D case to higher ranks, but this simplification
already helps in scaling some microbenchmarks that would formerly
crash on the intermediate IR length.

Reviewed By: reidtatge

Differential Revision: https://reviews.llvm.org/D82144