Package Details: torch7-cutorch-git r1026.5e9d86c-6

Git Clone URL: https://aur.archlinux.org/torch7-cutorch-git.git (read-only, click to copy)
Package Base: torch7-cutorch-git
Description: A CUDA backend for Torch7
Upstream URL: https://github.com/torch/cutorch
Licenses: BSD
Conflicts: torch7-cutorch
Provides: torch7-cutorch
Submitter: petronny
Maintainer: None
Last Packager: Levitating
Votes: 8
Popularity: 0.000000
First Submitted: 2016-08-28 13:17 (UTC)
Last Updated: 2023-05-11 19:12 (UTC)

Latest Comments

« First ‹ Previous 1 2 3 4 Next › Last »

Talzahr commented on 2020-11-04 12:51 (UTC) (edited on 2020-11-04 13:19 (UTC) by Talzahr)

Appears that gcc8 package is required now but not in the makedeps.

**edit: error near end of compiling. The cutorch project hasn't been touched in several years now. Maybe I can find a patch

spendingmarrow commented on 2020-08-25 14:32 (UTC)

I get the following error:

 
:: (1/1) Parsing SRCINFO: torch7-cutorch-git
==> Making package: torch7-cutorch-git r1026.5e9d86c-5 (Tue 25 Aug 2020 10:23:18 AM)
==> Retrieving sources...
  -> Updating torch7-cutorch-git git repo...
Fetching origin
  -> Found THCAtomics.cuh.patch
==> Validating source files with sha512sums...
    torch7-cutorch-git ... Skipped
    THCAtomics.cuh.patch ... Passed
==> Making package: torch7-cutorch-git r1026.5e9d86c-5 (Tue 25 Aug 2020 10:23:20 AM)
==> Checking runtime dependencies...
==> Checking buildtime dependencies...
==> Retrieving sources...
  -> Updating torch7-cutorch-git git repo...
Fetching origin
  -> Found THCAtomics.cuh.patch
==> Validating source files with sha512sums...
    torch7-cutorch-git ... Skipped
    THCAtomics.cuh.patch ... Passed
==> Removing existing $srcdir/ directory...
==> Extracting sources...
  -> Creating working copy of torch7-cutorch-git git repo...
Cloning into 'torch7-cutorch-git'...
done.
==> Starting prepare()...
patching file lib/THC/THCAtomics.cuh
Hunk #2 succeeded at 118 with fuzz 2.
==> Starting pkgver()...
==> Sources are ready.
==> Making package: torch7-cutorch-git r1026.5e9d86c-5 (Tue 25 Aug 2020 10:23:23 AM)
==> Checking runtime dependencies...
==> Checking buildtime dependencies...
==> WARNING: Using existing $srcdir/ tree
==> Starting pkgver()...
==> Removing existing $pkgdir/ directory...
==> Starting build()...
CMake Warning (dev) in CMakeLists.txt:
  No project() command is present.  The top-level CMakeLists.txt file must
  contain a literal, direct call to the project() command.  Add a line of
  code such as

    project(ProjectName)

  near the top of the file, but after cmake_minimum_required().

  CMake is pretending there is a "project(Project)" command on the first
  line.
This warning is for project developers.  Use -Wno-dev to suppress it.

-- The C compiler identification is GNU 8.4.0
-- The CXX compiler identification is GNU 10.1.0
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Check for working C compiler: /usr/bin/gcc-8 - skipped
-- Detecting C compile features
-- Detecting C compile features - done
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Check for working CXX compiler: /usr/bin/c++ - skipped
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Found Torch7 in /usr
-- Looking for pthread.h
-- Looking for pthread.h - found
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD - Failed
-- Looking for pthread_create in pthreads
-- Looking for pthread_create in pthreads - not found
-- Looking for pthread_create in pthread
-- Looking for pthread_create in pthread - found
-- Found Threads: TRUE  
-- Found CUDA: /opt/cuda (found suitable version "11.0", minimum required is "6.5") 
-- Removing -DNDEBUG from compile flags
-- TH_LIBRARIES: TH
-- MAGMA not found. Compiling without MAGMA support
-- Autodetected CUDA architecture(s):  6.1
-- got cuda version 11.0
-- Found CUDA with FP16 support, compiling with torch.CudaHalfTensor
-- CUDA_NVCC_FLAGS: -D__CUDA_NO_HALF_OPERATORS__;-gencode;arch=compute_61,code=sm_61;-DCUDA_HAS_FP16=1
-- THC_SO_VERSION: 0
-- Performing Test HAS_LUAL_SETFUNCS
-- Performing Test HAS_LUAL_SETFUNCS - Failed
-- Configuring done
-- Generating done
-- Build files have been written to: /home/angad/.cache/yay/torch7-cutorch-git/src/torch7-cutorch-git
[  1%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCReduceApplyUtils.cu.o
[  2%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorCopy.cu.o
[  3%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCSleep.cu.o
[  4%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCStorage.cu.o
[  5%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCHalf.cu.o
[  6%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCBlas.cu.o
[  7%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCStorageCopy.cu.o
[  8%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensor.cu.o
[ 10%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorMath.cu.o
[ 11%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorMath2.cu.o
[ 12%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorMathBlas.cu.o
[ 13%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorMathMagma.cu.o
[ 14%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorMathPairwise.cu.o
[ 15%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorMathReduce.cu.o
[ 16%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorMathScan.cu.o
[ 17%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorIndex.cu.o
[ 18%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorConv.cu.o
[ 20%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorRandom.cu.o
[ 21%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorScatterGather.cu.o
[ 22%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorTopK.cu.o
[ 23%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorSort.cu.o
[ 24%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorTypeUtils.cu.o
[ 25%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCSortUtils.cu.o
[ 26%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/THC_generated_THCTensorMode.cu.o
[ 27%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorSortByte.cu.o
[ 28%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareTByte.cu.o
[ 30%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathPointwiseByte.cu.o
[ 31%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareByte.cu.o
[ 32%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathReduceByte.cu.o
[ 33%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMaskedByte.cu.o
[ 34%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorSortChar.cu.o
[ 35%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareTChar.cu.o
[ 36%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathPointwiseChar.cu.o
[ 37%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareChar.cu.o
[ 38%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathReduceChar.cu.o
[ 40%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMaskedChar.cu.o
[ 41%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorSortShort.cu.o
[ 42%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareTShort.cu.o
[ 43%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathPointwiseShort.cu.o
[ 44%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareShort.cu.o
[ 45%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathReduceShort.cu.o
[ 46%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMaskedShort.cu.o
[ 47%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorSortInt.cu.o
[ 48%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareTInt.cu.o
[ 50%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathPointwiseInt.cu.o
[ 51%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareInt.cu.o
[ 52%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathReduceInt.cu.o
[ 53%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMaskedInt.cu.o
[ 54%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorSortLong.cu.o
[ 55%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareTLong.cu.o
[ 56%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathPointwiseLong.cu.o
[ 57%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareLong.cu.o
[ 58%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathReduceLong.cu.o
[ 60%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMaskedLong.cu.o
[ 61%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorSortHalf.cu.o
[ 62%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareTHalf.cu.o
[ 63%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathPointwiseHalf.cu.o
[ 64%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareHalf.cu.o
[ 65%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathReduceHalf.cu.o
[ 66%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMaskedHalf.cu.o
[ 67%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorSortFloat.cu.o
[ 68%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareTFloat.cu.o
[ 70%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathPointwiseFloat.cu.o
[ 71%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareFloat.cu.o
[ 72%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathReduceFloat.cu.o
[ 73%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMaskedFloat.cu.o
[ 74%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorSortDouble.cu.o
[ 75%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareTDouble.cu.o
[ 76%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathPointwiseDouble.cu.o
[ 77%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathCompareDouble.cu.o
[ 78%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMathReduceDouble.cu.o
[ 80%] Building NVCC (Device) object lib/THC/CMakeFiles/THC.dir/generated/THC_generated_THCTensorMaskedDouble.cu.o
Scanning dependencies of target THC
[ 81%] Building CXX object lib/THC/CMakeFiles/THC.dir/THCCachingHostAllocator.cpp.o
[ 83%] Building CXX object lib/THC/CMakeFiles/THC.dir/THCCachingAllocator.cpp.o
[ 83%] Building C object lib/THC/CMakeFiles/THC.dir/THCStorageCopy.c.o
[ 85%] Building C object lib/THC/CMakeFiles/THC.dir/THCGeneral.c.o
[ 84%] Building CXX object lib/THC/CMakeFiles/THC.dir/THCStream.cpp.o
[ 87%] Building CXX object lib/THC/CMakeFiles/THC.dir/THCTensorRandom.cpp.o
[ 87%] Building C object lib/THC/CMakeFiles/THC.dir/THCTensor.c.o
[ 88%] Building C object lib/THC/CMakeFiles/THC.dir/THCTensorCopy.c.o
[ 90%] Building C object lib/THC/CMakeFiles/THC.dir/THCThreadLocal.c.o
[ 91%] Linking CXX shared library libTHC.so
[ 91%] Built target THC
[ 92%] Generating TensorMath.c
Scanning dependencies of target cutorch
[ 93%] Building C object CMakeFiles/cutorch.dir/Storage.c.o
[ 94%] Building C object CMakeFiles/cutorch.dir/Tensor.c.o
[ 96%] Building C object CMakeFiles/cutorch.dir/init.c.o
[ 96%] Building C object CMakeFiles/cutorch.dir/TensorOperator.c.o
[ 97%] Building C object CMakeFiles/cutorch.dir/torch/utils.c.o
[ 98%] Building C object CMakeFiles/cutorch.dir/TensorMath.c.o
/home/angad/.cache/yay/torch7-cutorch-git/src/torch7-cutorch-git/init.c: In function ‘cutorch_isManagedPtr’:
/home/angad/.cache/yay/torch7-cutorch-git/src/torch7-cutorch-git/init.c:938:34: error: ‘struct cudaPointerAttributes’ has no member named ‘isManaged’
     lua_pushboolean(L, attributes.isManaged);
                                  ^
make[2]: *** [CMakeFiles/cutorch.dir/build.make:99: CMakeFiles/cutorch.dir/init.c.o] Error 1
make[2]: *** Waiting for unfinished jobs....
make[1]: *** [CMakeFiles/Makefile2:134: CMakeFiles/cutorch.dir/all] Error 2
make: *** [Makefile:149: all] Error 2
==> ERROR: A failure occurred in build().
    Aborting...
error making: torch7-cutorch-git

spendingmarrow commented on 2020-08-24 21:14 (UTC) (edited on 2020-08-24 21:14 (UTC) by spendingmarrow)

Hi, I am getting the following error:

ptxas warning : Value of threads per SM for entry Z21kernelPointwiseApply1I20Tensor_log_Double_OpdjLi2EEv10TensorInfoIT0_T1_ES3_T is out of range. .minnctapersm will be ignored ptxas warning : Value of threads per SM for entry Z21kernelPointwiseApply1I20Tensor_log_Double_OpdjLi1EEv10TensorInfoIT0_T1_ES3_T is out of range. .minnctapersm will be ignored ptxas warning : Value of threads per SM for entry Z21kernelPointwiseApply1I20Tensor_log_Double_OpdjLin2EEv10TensorInfoIT0_T1_ES3_T is out of range. .minnctapersm will be ignored Scanning dependencies of target THC [ 81%] Building CXX object lib/THC/CMakeFiles/THC.dir/THCCachingAllocator.cpp.o [ 82%] Building CXX object lib/THC/CMakeFiles/THC.dir/THCStream.cpp.o [ 84%] Building C object lib/THC/CMakeFiles/THC.dir/THCTensor.c.o [ 84%] Building C object lib/THC/CMakeFiles/THC.dir/THCStorageCopy.c.o [ 85%] Building CXX object lib/THC/CMakeFiles/THC.dir/THCCachingHostAllocator.cpp.o [ 86%] Building C object lib/THC/CMakeFiles/THC.dir/THCGeneral.c.o [ 87%] Building CXX object lib/THC/CMakeFiles/THC.dir/THCTensorRandom.cpp.o [ 88%] Building C object lib/THC/CMakeFiles/THC.dir/THCTensorCopy.c.o [ 90%] Building C object lib/THC/CMakeFiles/THC.dir/THCThreadLocal.c.o [ 91%] Linking CXX shared library libTHC.so [ 91%] Built target THC [ 92%] Generating TensorMath.c Scanning dependencies of target cutorch [ 93%] Building C object CMakeFiles/cutorch.dir/Tensor.c.o [ 95%] Building C object CMakeFiles/cutorch.dir/TensorMath.c.o [ 95%] Building C object CMakeFiles/cutorch.dir/Storage.c.o [ 96%] Building C object CMakeFiles/cutorch.dir/TensorOperator.c.o [ 97%] Building C object CMakeFiles/cutorch.dir/init.c.o [ 98%] Building C object CMakeFiles/cutorch.dir/torch/utils.c.o /home/angad/.cache/yay/torch7-cutorch-git/src/torch7-cutorch-git/init.c: In function ‘cutorch_isManagedPtr’: /home/angad/.cache/yay/torch7-cutorch-git/src/torch7-cutorch-git/init.c:938:34: error: ‘struct cudaPointerAttributes’ has no member named ‘isManaged’ lua_pushboolean(L, attributes.isManaged); ^ make[2]: [CMakeFiles/cutorch.dir/build.make:99: CMakeFiles/cutorch.dir/init.c.o] Error 1 make[2]: Waiting for unfinished jobs.... make[1]: [CMakeFiles/Makefile2:134: CMakeFiles/cutorch.dir/all] Error 2 make: [Makefile:149: all] Error 2 ==> ERROR: A failure occurred in build(). Aborting...

mathiassteiger commented on 2019-08-11 22:40 (UTC)

error: identifier "__builtin_is_constant_evaluated" is undefined

Modify this line:

cmake . -DCMAKE_INSTALL_PREFIX=/usr -DCMAKE_BUILD_TYPE=Release -DCUDA_HOST_COMPILER='/usr/bin/gcc-8'

thgravo commented on 2019-08-09 11:49 (UTC) (edited on 2019-08-09 11:52 (UTC) by thgravo)

I get this error:

==> Continue building torch7-cutorch-git ? [Y/n]
==> --------------------------------------------
==> 
==> Building and installing package
==> Making package: torch7-cutorch-git r1026.5e9d86c-4 (2019-08-09T13:41:44 CEST)
==> Checking runtime dependencies...
==> Checking buildtime dependencies...
==> Retrieving sources...
  -> Cloning torch7-cutorch-git git repo...
Cloning into bare repository '<FOLDER>/aur-torch7-cutorch-git/torch7-cutorch-git'...
remote: Enumerating objects: 2, done.
remote: Counting objects: 100% (2/2), done.
remote: Compressing objects: 100% (2/2), done.
remote: Total 7722 (delta 1), reused 0 (delta 0), pack-reused 7720
Receiving objects: 100% (7722/7722), 2.73 MiB | 1.45 MiB/s, done.
Resolving deltas: 100% (4708/4708), done.
  -> Found THCAtomics.cuh.patch
==> Validating source files with sha512sums...
    torch7-cutorch-git ... Skipped
    THCAtomics.cuh.patch ... Passed
==> Extracting sources...
  -> Creating working copy of torch7-cutorch-git git repo...
Cloning into 'torch7-cutorch-git'...
done.
==> Starting prepare()...
patching file lib/THC/THCAtomics.cuh
Hunk #2 succeeded at 118 with fuzz 2.
==> Starting pkgver()...
==> Starting build()...
CMake Warning (dev) in CMakeLists.txt:
  No project() command is present.  The top-level CMakeLists.txt file must
  contain a literal, direct call to the project() command.  Add a line of
  code such as

    project(ProjectName)

  near the top of the file, but after cmake_minimum_required().

  CMake is pretending there is a "project(Project)" command on the first
  line.
This warning is for project developers.  Use -Wno-dev to suppress it.

-- The C compiler identification is GNU 9.1.0
-- The CXX compiler identification is GNU 9.1.0
-- Check for working C compiler: /usr/bin/cc
-- Check for working C compiler: /usr/bin/cc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Check for working CXX compiler: /usr/bin/c++
-- Check for working CXX compiler: /usr/bin/c++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Looking for pthread.h
-- Looking for pthread.h - found
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD - Failed
-- Looking for pthread_create in pthreads
-- Looking for pthread_create in pthreads - not found
-- Looking for pthread_create in pthread
-- Looking for pthread_create in pthread - found
-- Found Threads: TRUE  
-- Caffe2: Found gflags with new-style gflags target.
-- Caffe2: Cannot find glog automatically. Using legacy find.
-- Found glog: /usr/include  
-- Caffe2: Found glog (include: /usr/include, library: /usr/lib/libglog.so)
-- Found CUDA: /opt/cuda (found version "10.1") 
-- Caffe2: CUDA detected: 10.1
-- Caffe2: CUDA nvcc is: /opt/cuda/bin/nvcc
-- Caffe2: CUDA toolkit directory: /opt/cuda
-- Caffe2: Header version is: 10.1
-- Found CUDNN: /usr/include  
-- Found cuDNN: v7.6.1  (include: /usr/include, library: /usr/lib/libcudnn.so)
-- Autodetected CUDA architecture(s):  6.1
-- Added CUDA NVCC flags for: -gencode;arch=compute_61,code=sm_61
-- Found torch: /usr/lib/libtorch.so  
-- Found CUDA: /opt/cuda (found suitable version "10.1", minimum required is "6.5") 
-- Removing -DNDEBUG from compile flags
-- TH_LIBRARIES: TH
-- MAGMA not found. Compiling without MAGMA support
-- Autodetected CUDA architecture(s):  6.1
-- got cuda version 10.1
-- Found CUDA with FP16 support, compiling with torch.CudaHalfTensor
-- CUDA_NVCC_FLAGS: -DONNX_NAMESPACE=onnx_c2;-gencode;arch=compute_61,code=sm_61;-Xcudafe;--diag_suppress=cc_clobber_ignored;-Xcudafe;--diag_suppress=integer_sign_change;-Xcudafe;--diag_suppress=useless_using_declaration;-Xcudafe;--diag_suppress=set_but_not_used;-std=c++11;-Xcompiler;-fPIC;--expt-relaxed-constexpr;--expt-extended-lambda;-D__CUDA_NO_HALF_OPERATORS__;-gencode;arch=compute_61,code=sm_61;-DCUDA_HAS_FP16=1
-- THC_SO_VERSION: 0
-- Performing Test HAS_LUAL_SETFUNCS
-- Performing Test HAS_LUAL_SETFUNCS - Success
CMake Error at CMakeLists.txt:43 (ADD_TORCH_WRAP):
  Unknown CMake command "ADD_TORCH_WRAP".


-- Configuring incomplete, errors occurred!
See also "<FOLDER>/aur-torch7-cutorch-git/src/torch7-cutorch-git/CMakeFiles/CMakeOutput.log".
See also "<FOLDER>/aur-torch7-cutorch-git/src/torch7-cutorch-git/CMakeFiles/CMakeError.log".
==> ERROR: A failure occurred in build().
    Aborting...
==> ERROR: Makepkg was unable to build torch7-cutorch-git.
==> Restart building torch7-cutorch-git ? [y/N]
==> -------------------------------------------
==> 

chrhasse commented on 2019-04-11 04:59 (UTC)

Looks like the latest version of cuda supports gcc-8 and gcc-7 has been removed from community into the AUR so you don't need -DCMAKE_C_COMPILER=gcc-7 anymore.

ei-grad commented on 2018-11-07 23:29 (UTC) (edited on 2018-11-07 23:30 (UTC) by ei-grad)

Getting error with CUDA 10:

lib/THC/THCAtomics.cuh(97): error: cannot overload functions distinguished by return type alone

Patch is needed https://github.com/torch/cutorch/issues/834:

diff --git a/lib/THC/THCAtomics.cuh b/lib/THC/THCAtomics.cuh
index 400875c..ccb7a1c 100644
--- a/lib/THC/THCAtomics.cuh
+++ b/lib/THC/THCAtomics.cuh
@@ -94,6 +94,7 @@ static inline __device__ void atomicAdd(long *address, long val) {
 }

 #ifdef CUDA_HALF_TENSOR
+#if !(__CUDA_ARCH__ >= 700 || !defined(__CUDA_ARCH__) )
 static inline  __device__ void atomicAdd(half *address, half val) {
   unsigned int * address_as_ui =
       (unsigned int *) ((char *)address - ((size_t)address & 2));
@@ -117,6 +118,7 @@ static inline  __device__ void atomicAdd(half *address, half val) {
    } while (assumed != old);
 }
 #endif
+#endif

fuan_k commented on 2018-09-16 19:30 (UTC) (edited on 2018-09-16 21:28 (UTC) by fuan_k)

gcc-5 doesn't exist anymore in Arch. Cannot compile:

If I edit PKGBUILD for gcc version 8.2.0 (currently), it yields an error:

/opt/cuda/include/crt/host_config.h:119:2: error: #error -- unsupported GNU version! gcc versions later than 7 are not supported! #error -- unsupported GNU version! gcc versions later than 7 are not supported!

Do we really need gcc-5 in order to use cuda? It should at least be marked as a dependency.

Found a pre-built binary package on the manjaro repos for gcc 4.9 (http://kibo.remi.lu/stable/extra/x86_64/gcc49-4.9.3-2-x86_64.pkg.tar.xz) , and used that instead of gcc5, seems to have compiled fine. Had to trick the linker for gcc with ln -s /usr/lib/libmpfr.so /usr/lib/libmpfr.so.4 because we're now at version 6 for libmpfr, and luckily it compiled.

To point to your custom gcc-4.9 edit this line in the PKGBUIL: cmake . -DCMAKE_INSTALL_PREFIX=/usr -DCMAKE_C_COMPILER=/path/to/gcc49/usr/bin/gcc-4.9 -DCMAKE_BUILD_TYPE=Release

Alternatively, one could use the gcc shipped with cuda in /opt/cuda/bin/gcc I guess...

eigengrau commented on 2017-09-25 14:12 (UTC)

I’m also seeing implicit function declaration errors when building this. The patch posted below resolved this. Maybe it would make sense to apply this here until it is taken up upstream.