OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-06 12:20:52 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity

add missing space to first cmake append (#125294)

Browse Source 
the first append not having a space incorrectly merges it to any previous arguments, like `-allow-unsupported-compiler` in my case which results in a silly error: `unrecognized command-line option '-allow-unsupported-compiler-DLIBCUDACXX_ENABLE_SIMPLIFIED_COMPLEX_OPERATIONS'`

full log:
```
python setup.py develop
Building wheel torch-2.4.0a0+git75fa54a
-- Building version 2.4.0a0+git75fa54a
cmake3 -GNinja -DBUILD_PYTHON=True -DBUILD_TEST=True -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX=/code/pytorch/torch -DCMAKE_PREFIX_PATH=/code/pytorch/.venv/lib/python3.12/site-packages;/code/spack/opt/spack/linux-fedora40-zen2/gcc-14.0.1/gcc-13.2.0-noa2f4oqalxzqvsebhuntndewgt4gq4h:/code/spack/opt/spack/linux-fedora40-zen2/gcc-14.0.1/zstd-1.5.6-z3guwm4l5rmmsv4g4wvkej3ri3bppeja:/code/spack/opt/spack/linux-fedora40-zen2/gcc-14.0.1/zlib-ng-2.1.6-kwi4ljobodjgv5eetnga4bow6crdlacl:/code/spack/opt/spack/linux-fedora40-zen2/gcc-14.0.1/mpc-1.3.1-nuwa2snyzm265lsupa2dkmxxyhiqcv7e:/code/spack/opt/spack/linux-fedora40-zen2/gcc-14.0.1/mpfr-4.2.1-wepuwobwttxbtz3nguimxa2mlljjozsi:/code/spack/opt/spack/linux-fedora40-zen2/gcc-14.0.1/gmp-6.2.1-ashy6kiitonxv2f365f4q3beggzf3646:/code/spack/opt/spack/linux-fedora40-zen2/gcc-14.0.1/gcc-runtime-14.0.1-wmogkqrzn7t57dogaake2hmhjbod27gs -DNUMPY_INCLUDE_DIR=/code/pytorch/.venv/lib64/python3.12/site-packages/numpy/core/include -DPYTHON_EXECUTABLE=/code/pytorch/.venv/bin/python -DPYTHON_INCLUDE_DIR=/usr/include/python3.12 -DPYTHON_LIBRARY=/usr/lib64/libpython3.12.so.1.0 -DTORCH_BUILD_VERSION=2.4.0a0+git75fa54a -DUSE_NUMPY=True /code/pytorch
-- /usr/lib64/ccache/c++ /code/pytorch/torch/abi-check.cpp -o /code/pytorch/build/abi-check
-- Determined _GLIBCXX_USE_CXX11_ABI=1
-- Current compiler supports avx2 extension. Will build perfkernels.
-- Current compiler supports avx512f extension. Will build fbgemm.
-- The CUDA compiler identification is NVIDIA 12.4.131
-- Detecting CUDA compiler ABI info
-- Detecting CUDA compiler ABI info - failed
-- Check for working CUDA compiler: /usr/local/cuda-12/bin/nvcc
-- Check for working CUDA compiler: /usr/local/cuda-12/bin/nvcc - broken
CMake Error at /usr/share/cmake/Modules/CMakeTestCUDACompiler.cmake:59 (message):
  The CUDA compiler

    "/usr/local/cuda-12/bin/nvcc"

  is not able to compile a simple test program.
  It fails with the following output:
    Change Dir: '/code/pytorch/build/CMakeFiles/CMakeScratch/TryCompile-mSGoFl'

    Run Build Command(s): /code/pytorch/.venv/bin/ninja -v cmTC_ee207
    [1/2] /usr/local/cuda-12/bin/nvcc -forward-unknown-to-host-compiler   -allow-unsupported-compiler-DLIBCUDACXX_ENABLE_SIMPLIFIED_COMPLEX_OPERATIONS -D_GLIBCXX_USE_CXX11_ABI=1 -Xfatbin -compress-all  "--generate-code=arch=compute_52,code=[compute_52,sm_52]" -MD -MT CMakeFiles/cmTC_ee207.dir/main.cu.o -MF CMakeFiles/cmTC_ee207.dir/main.cu.o.d -x cu -c /code/pytorch/build/CMakeFiles/CMakeScratch/TryCompile-mSGoFl/main.cu -o CMakeFiles/cmTC_ee207.dir/main.cu.o
    FAILED: CMakeFiles/cmTC_ee207.dir/main.cu.o
    /usr/local/cuda-12/bin/nvcc -forward-unknown-to-host-compiler   -allow-unsupported-compiler-DLIBCUDACXX_ENABLE_SIMPLIFIED_COMPLEX_OPERATIONS -D_GLIBCXX_USE_CXX11_ABI=1 -Xfatbin -compress-all  "--generate-code=arch=compute_52,code=[compute_52,sm_52]" -MD -MT CMakeFiles/cmTC_ee207.dir/main.cu.o -MF CMakeFiles/cmTC_ee207.dir/main.cu.o.d -x cu -c /code/pytorch/build/CMakeFiles/CMakeScratch/TryCompile-mSGoFl/main.cu -o CMakeFiles/cmTC_ee207.dir/main.cu.o
    gcc: error: unrecognized command-line option '-allow-unsupported-compiler-DLIBCUDACXX_ENABLE_SIMPLIFIED_COMPLEX_OPERATIONS'
    ninja: build stopped: subcommand failed.

  CMake will not be able to correctly generate this project.
Call Stack (most recent call first):
  cmake/public/cuda.cmake:47 (enable_language)
  cmake/Dependencies.cmake:44 (include)
  CMakeLists.txt:758 (include)

-- Configuring incomplete, errors occurred!
```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/125294
Approved by: https://github.com/albanD
This commit is contained in:
Mihai 2024-05-01 18:35:49 +00:00 committed by PyTorch MergeBot
parent bf6acf9add
commit 07422fd0b9
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
CMakeLists.txt
View File

@ -56,7 +56,7 @@ endif()
# This define is needed to preserve behavior given anticpated changes to cccl/thrust
# https://nvidia.github.io/libcudacxx/standard_api/numerics_library/complex.html
string(APPEND CMAKE_CUDA_FLAGS "-DLIBCUDACXX_ENABLE_SIMPLIFIED_COMPLEX_OPERATIONS")
string(APPEND CMAKE_CUDA_FLAGS " -DLIBCUDACXX_ENABLE_SIMPLIFIED_COMPLEX_OPERATIONS")
if(LINUX)
include(cmake/CheckAbi.cmake)