[all-commits] [llvm/llvm-project] f36e90: [mlir][VectorOps] Use SCF for vector.print and all...

Fri Aug 11 02:30:41 PDT 2023

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: f36e909da03767c317eaa5ded9abbf3214299ff5
      https://github.com/llvm/llvm-project/commit/f36e909da03767c317eaa5ded9abbf3214299ff5
  Author: Benjamin Maxwell <benjamin.maxwell at arm.com>
  Date:   2023-08-11 (Fri, 11 Aug 2023)

  Changed paths:
    M mlir/include/mlir/Dialect/Vector/IR/VectorOps.td
    M mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp
    M mlir/lib/Conversion/VectorToSCF/VectorToSCF.cpp
    M mlir/test/Conversion/VectorToLLVM/vector-to-llvm.mlir
    M mlir/test/Conversion/VectorToSCF/vector-to-scf.mlir
    M mlir/test/Integration/Dialect/Arith/CPU/test-wide-int-emulation-compare-results-i16.mlir
    M mlir/test/Integration/Dialect/Arith/CPU/test-wide-int-emulation-constants-i16.mlir
    M mlir/test/Integration/Dialect/LLVMIR/CPU/X86/test-inline-asm-vector.mlir
    M mlir/test/Integration/Dialect/LLVMIR/CPU/test-vp-intrinsic.mlir
    M mlir/test/Integration/Dialect/SparseTensor/GPU/CUDA/sm80-lt/sparse-matmul-2-4-lib.mlir
    M mlir/test/Integration/Dialect/SparseTensor/GPU/CUDA/sparse-mma-2-4-f16.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/ArmSVE/test-sve.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/X86Vector/test-dot.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/X86Vector/test-mask-compress.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/X86Vector/test-rsqrt.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/X86Vector/test-vp2intersect-i32.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-0-d-vectors.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-broadcast.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-compress.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-constant-mask.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-contraction.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-create-mask-v4i1.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-create-mask.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-expand.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-extract-strided-slice.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-flat-transpose-col.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-flat-transpose-row.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-fma.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-gather.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-index-vectors.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-insert-strided-slice.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-maskedload.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-maskedstore.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-matrix-multiply-col.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-matrix-multiply-row.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-outerproduct-f32.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-outerproduct-i64.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-print-fp.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-print-int.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-realloc.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-reductions-f32-reassoc.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-reductions-f32.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-reductions-f64-reassoc.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-reductions-f64.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-reductions-i32.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-reductions-i4.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-reductions-i64.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-reductions-si4.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-reductions-ui4.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-scan.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-scatter.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-shape-cast.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-shuffle.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-shuffle16x16.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-sparse-dot-matvec.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-sparse-saxpy-jagged-matvec.mlir
    M mlir/test/Integration/Dialect/Vector/CPU/test-transpose.mlir
    M mlir/test/Integration/Dialect/Vector/GPU/CUDA/test-reduction-distribute.mlir
    M mlir/test/Integration/Dialect/Vector/GPU/CUDA/test-warp-distribute.mlir
    M mlir/test/Integration/GPU/CUDA/gpu-to-cubin.mlir
    M mlir/test/mlir-cpu-runner/X86Vector/math-polynomial-approx-avx2.mlir
    M mlir/test/mlir-cpu-runner/math-polynomial-approx.mlir
    M mlir/test/mlir-cpu-runner/test-expand-math-approx.mlir
    M mlir/test/python/dialects/vector.py

  Log Message:
  -----------
  [mlir][VectorOps] Use SCF for vector.print and allow scalable vectors

Reland of the original patch after updating the Python binding tests,
a few CUDA/GPU MLIR tests, and ensuring the assembly format is
round-trippable.

This patch splits the lowering of vector.print into first converting
an n-D print into a loop of scalar prints of the elements, then a second
pass that converts those scalar prints into the runtime calls. The
former is done in VectorToSCF and the latter in VectorToLLVM.

The main reason for this is to allow printing scalable vector types,
which are not possible to fully unroll at compile time, though this
also avoids fully unrolling very large vectors.

To allow VectorToSCF to add the necessary punctuation between vectors
and elements, a "punctuation" attribute has been added to vector.print.
This abstracts calling the runtime functions such as printNewline(),
without leaking the LLVM details into the higher abstraction levels.
For example:

  vector.print punctuation <comma>

lowers to

  llvm.call @printComma() : () -> ()

The output format and runtime functions remain the same, which avoids
the need to alter a large number of tests (aside from the pipelines).

Reviewed By: awarzynski, c-rhodes, aartbik

Differential Revision: https://reviews.llvm.org/D156519