[all-commits] [llvm/llvm-project] 83cf0d: [mlir][sparse] implement direct IR alloc/empty/new...

[Aart Bik via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 83cf0dc98234bbd8cb0d0959baa570477a8daf92
      https://github.com/llvm/llvm-project/commit/83cf0dc98234bbd8cb0d0959baa570477a8daf92
  Author: Aart Bik <39774503+aartbik at users.noreply.github.com>
  Date:   2023-11-16 (Thu, 16 Nov 2023)

  Changed paths:
    M mlir/include/mlir/ExecutionEngine/SparseTensor/File.h
    M mlir/lib/Dialect/SparseTensor/Transforms/SparseTensorCodegen.cpp
    M mlir/test/Dialect/SparseTensor/codegen.mlir
    M mlir/test/Dialect/SparseTensor/codegen_buffer_initialization.mlir
    M mlir/test/Integration/Dialect/SparseTensor/CPU/block.mlir
    M mlir/test/Integration/Dialect/SparseTensor/GPU/CUDA/sparse-sddmm-lib.mlir

  Log Message:
  -----------
  [mlir][sparse] implement direct IR alloc/empty/new for non-permutations (#72585)

This change implements the correct *level* sizes set up for the direct
IR codegen fields in the sparse storage scheme. This brings libgen and
codegen together again.

This is step 3 out of 3 to make sparse_tensor.new work for BSR