pytorch/cmake/Modules
linuxone f64906f470 ibm z14/15 SIMD support (#66407)
Summary:
https://github.com/pytorch/pytorch/issues/66406
implemented z arch 14/15 vector SIMD additions.
so far besides bfloat all other types have their SIMD implementation.

it has 99% coverage and currently passing the local test.
it is concise and the main SIMD file is only one header file
it's using template metaprogramming, mostly. but still, there are a few macrosses left with the intention not to modify PyTorch much
Sleef supports z15

Pull Request resolved: https://github.com/pytorch/pytorch/pull/66407

Reviewed By: mrshenli

Differential Revision: D33370163

Pulled By: malfet

fbshipit-source-id: 0e5a57f31b22a718cd2a9ac59753fb468cdda140
2022-01-04 09:40:18 -08:00
..
FindARM.cmake
FindAtlas.cmake Lint trailing newlines (#54737) 2021-03-30 13:09:52 -07:00
FindAVX.cmake Add AVX512 support in ATen & remove AVX support (#61903) 2021-07-22 08:51:49 -07:00
FindBenchmark.cmake
FindBLAS.cmake Modify "gemm" code to enable access to "sbgemm_" routine in OpenBLAS (#58831) 2021-11-03 08:53:27 -07:00
FindBLIS.cmake Adding a new include directory in BLIS search path (#58166) 2021-05-24 08:57:02 -07:00
FindCUB.cmake Update CMake and use native CUDA language support (#62445) 2021-10-11 09:05:48 -07:00
FindFFmpeg.cmake
FindFlexiBLAS.cmake Add FlexiBLAS build support per #64752 (#64815) 2021-10-28 11:28:00 -07:00
FindGloo.cmake
FindHiredis.cmake Lint trailing newlines (#54737) 2021-03-30 13:09:52 -07:00
FindLAPACK.cmake Add FlexiBLAS build support per #64752 (#64815) 2021-10-28 11:28:00 -07:00
FindLevelDB.cmake
FindLMDB.cmake
FindMAGMA.cmake
FindMatlabMex.cmake
FindMKL.cmake
FindMKLDNN.cmake Upgrade oneDNN to v2.3.3 and package oneDNN Graph API together (#63748) 2021-12-09 13:42:40 -08:00
FindNCCL.cmake
FindNuma.cmake Lint trailing newlines (#54737) 2021-03-30 13:09:52 -07:00
FindNumPy.cmake Lint trailing newlines (#54737) 2021-03-30 13:09:52 -07:00
FindOpenBLAS.cmake Lint trailing newlines (#54737) 2021-03-30 13:09:52 -07:00
FindOpenMP.cmake
Findpybind11.cmake
FindRocksDB.cmake Lint trailing newlines (#54737) 2021-03-30 13:09:52 -07:00
FindSnappy.cmake Lint trailing newlines (#54737) 2021-03-30 13:09:52 -07:00
FindvecLib.cmake
FindVSX.cmake Lint trailing newlines (#54737) 2021-03-30 13:09:52 -07:00
FindZMQ.cmake
FindZVECTOR.cmake ibm z14/15 SIMD support (#66407) 2022-01-04 09:40:18 -08:00
README.md

This folder contains various custom cmake modules for finding libraries and packages. Details about some of them are listed below.

FindOpenMP.cmake

This is modified from the file included in CMake 3.13 release, with the following changes:

  • Replace VERSION_GREATER_EQUAL with NOT ... VERSION_LESS as VERSION_GREATER_EQUAL is not supported in CMake 3.5 (our min supported version).

  • Update the separate_arguments commands to not use NATIVE_COMMAND which is not supported in CMake 3.5 (our min supported version).

  • Make it respect the QUIET flag so that, when it is set, try_compile failures are not reported.

  • For AppleClang compilers, use -Xpreprocessor instead of -Xclang as the later is not documented.

  • For AppleClang compilers, an extra flag option is tried, which is -Xpreprocessor -openmp -I${DIR_OF_omp_h}, where ${DIR_OF_omp_h} is a obtained using find_path on omp.h with brew's default include directory as a hint. Without this, the compiler will complain about missing headers as they are not natively included in Apple's LLVM.

  • For non-GNU compilers, whenever we try a candidate OpenMP flag, first try it with directly linking MKL's libomp if it has one. Otherwise, we may end up linking two libomps and end up with this nasty error:

    OMP: Error #15: Initializing libomp.dylib, but found libiomp5.dylib already
    initialized.
    
    OMP: Hint This means that multiple copies of the OpenMP runtime have been
    linked into the program. That is dangerous, since it can degrade performance
    or cause incorrect results. The best thing to do is to ensure that only a
    single OpenMP runtime is linked into the process, e.g. by avoiding static
    linking of the OpenMP runtime in any library. As an unsafe, unsupported,
    undocumented workaround you can set the environment variable
    KMP_DUPLICATE_LIB_OK=TRUE to allow the program to continue to execute, but
    that may cause crashes or silently produce incorrect results. For more
    information, please see http://openmp.llvm.org/
    

    See NOTE [ Linking both MKL and OpenMP ] for details.