[all-commits] [llvm/llvm-project] dad9de: [mlir][vector] Improve lowering to LLVM for `minf`...

[Daniil Dudkin via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: dad9de0ae5360b18c890985d212bec266bf8c122
      https://github.com/llvm/llvm-project/commit/dad9de0ae5360b18c890985d212bec266bf8c122
  Author: Daniil Dudkin <unterumarmung at yandex.ru>
  Date:   2023-08-02 (Wed, 02 Aug 2023)

  Changed paths:
    M mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp
    M mlir/test/Conversion/VectorToLLVM/vector-to-llvm.mlir

  Log Message:
  -----------
  [mlir][vector] Improve lowering to LLVM for `minf`, `maxf` reductions

This patch improves the lowering by changing target LLVM intrinsics from
`reduce.fmax` and `reduce.fmin`,
which have different semantic for handling NaN,
to `reduce.fmaximum` and `reduce.fminimum` ones.

Fixes #63969

Depends on D155869

Reviewed By: dcaballe

Differential Revision: https://reviews.llvm.org/D155877