[all-commits] [llvm/llvm-project] a48adc: [mlir][math] Promote (b)f16 to f32 when lowering t...

Benjamin Kramer via All-commits all-commits at lists.llvm.org
Mon May 9 03:03:40 PDT 2022


  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: a48adc565864e0ce10becf301de5455308bd7d6c
      https://github.com/llvm/llvm-project/commit/a48adc565864e0ce10becf301de5455308bd7d6c
  Author: Benjamin Kramer <benny.kra at googlemail.com>
  Date:   2022-05-09 (Mon, 09 May 2022)

  Changed paths:
    M mlir/lib/Conversion/MathToLibm/MathToLibm.cpp
    M mlir/test/Conversion/MathToLibm/convert-to-libm.mlir

  Log Message:
  -----------
  [mlir][math] Promote (b)f16 to f32 when lowering to libm calls

libm doesn't have overloads for the small types, so promote them to a
bigger type and use the f32 function.

Differential Revision: https://reviews.llvm.org/D125093




More information about the All-commits mailing list