[all-commits] [llvm/llvm-project] e2e6e7: [mlir][sparse] start using size_hint provided in a...

[Aart Bik via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: e2e6e7a6a3a575153e2c3a4782c89c2ca1e09c02
      https://github.com/llvm/llvm-project/commit/e2e6e7a6a3a575153e2c3a4782c89c2ca1e09c02
  Author: Aart Bik <ajcbik at google.com>
  Date:   2023-02-03 (Fri, 03 Feb 2023)

  Changed paths:
    M mlir/lib/Dialect/SparseTensor/Transforms/SparseTensorCodegen.cpp
    M mlir/test/Dialect/SparseTensor/codegen.mlir

  Log Message:
  -----------
  [mlir][sparse] start using size_hint provided in allocation op

Even though we introduced the size_hint, we never used it.
This is a very first step, using the hint during the codegen path.
Note that we can refine the heuristics. Also, we need to start
adding the hint on all allocation generated for reading tensors,
converting tensors, etc.

Reviewed By: Peiming, bixia

Differential Revision: https://reviews.llvm.org/D143292