[all-commits] [llvm/llvm-project] 95934c: [libc] Add hardware implementations of fma and fma...

[Siva Chandra via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 95934c3a37cb00e4fff6042c52837b42f6219772
      https://github.com/llvm/llvm-project/commit/95934c3a37cb00e4fff6042c52837b42f6219772
  Author: Siva Chandra <sivachandra at google.com>
  Date:   2021-04-21 (Wed, 21 Apr 2021)

  Changed paths:
    M libc/config/linux/aarch64/entrypoints.txt
    M libc/config/linux/x86_64/entrypoints.txt
    M libc/src/math/CMakeLists.txt
    A libc/src/math/fma.cpp
    A libc/src/math/fma.h
    A libc/src/math/fmaf.cpp
    M libc/src/math/generic/CMakeLists.txt
    R libc/src/math/generic/fmaf.cpp
    M libc/test/src/math/CMakeLists.txt
    A libc/test/src/math/fma_test.cpp
    A libc/utils/FPUtil/FMA.h
    A libc/utils/FPUtil/aarch64/FMA.h
    A libc/utils/FPUtil/generic/FMA.h
    A libc/utils/FPUtil/generic/README.md
    A libc/utils/FPUtil/x86_64/FMA.h

  Log Message:
  -----------
  [libc] Add hardware implementations of fma and fmaf for x86_64 and aarch64.

The current generic implementation of the fmaf function has been moved
to the FPUtil directory. This allows one use the fma operation from
implementations of other math functions like the trignometric functions
without depending on/requiring the fma/fmaf/fmal function targets. If
this pattern ends being convenient, we will switch all generic math
implementations to this pattern.

Reviewed By: lntue

Differential Revision: https://reviews.llvm.org/D100811