[all-commits] [llvm/llvm-project] 4bc88a: Enable support for floating-point division reductions

[Yichao Yu via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 4bc88a0e9a2ee29959a9053e867ae6f051348554
      https://github.com/llvm/llvm-project/commit/4bc88a0e9a2ee29959a9053e867ae6f051348554
  Author: Yichao Yu <yyc1992 at gmail.com>
  Date:   2020-11-23 (Mon, 23 Nov 2020)

  Changed paths:
    M llvm/lib/Analysis/IVDescriptors.cpp
    M llvm/test/Transforms/LoopVectorize/float-reduction.ll

  Log Message:
  -----------
  Enable support for floating-point division reductions

Similar to fsub, fdiv can also be vectorized using fmul.

Also http://llvm.org/viewvc/llvm-project?view=revision&revision=215200

Differential Revision: https://reviews.llvm.org/D34078

Co-authored-by: Jameson Nash <jameson at juliacomputing.com>