[arch-commits] Commit in python-pytorch/trunk (PKGBUILD)

[Sven-Hendrik Haase]

    Date: Sunday, February 16, 2020 @ 21:31:02
  Author: svenstaro
Revision: 572896

upgpkg: python-pytorch 1.4.0-5: Enable intel-dnnl and intel-mkl support

Modified:
  python-pytorch/trunk/PKGBUILD

----------+
 PKGBUILD |   12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

Modified: PKGBUILD
===================================================================
--- PKGBUILD	2020-02-16 21:30:17 UTC (rev 572895)
+++ PKGBUILD	2020-02-16 21:31:02 UTC (rev 572896)
@@ -5,13 +5,15 @@
 pkgname=("python-pytorch" "python-pytorch-opt" "python-pytorch-cuda" "python-pytorch-opt-cuda")
 _pkgname="pytorch"
 pkgver=1.4.0
-pkgrel=4
+pkgrel=5
 pkgdesc="Tensors and Dynamic neural networks in Python with strong GPU acceleration"
 arch=('x86_64')
 url="https://pytorch.org"
 license=('BSD')
-depends=('google-glog' 'gflags' 'opencv' 'openmp' 'nccl' 'pybind11' 'python' 'python-yaml' 'python-numpy' 'protobuf' 'ffmpeg' 'python-future' 'qt5-base')
-makedepends=('python' 'python-setuptools' 'python-yaml' 'python-numpy' 'cmake' 'cuda' 'cudnn' 'git' 'magma' 'ninja' 'pkgconfig')
+depends=('google-glog' 'gflags' 'opencv' 'openmp' 'nccl' 'pybind11' 'python' 'python-yaml'
+         'python-numpy' 'protobuf' 'ffmpeg' 'python-future' 'qt5-base' 'intel-dnnl' 'intel-mkl')
+makedepends=('python' 'python-setuptools' 'python-yaml' 'python-numpy' 'cmake' 'cuda'
+             'cudnn' 'git' 'magma' 'ninja' 'pkgconfig')
 source=("${_pkgname}-${pkgver}::git+https://github.com/pytorch/pytorch.git#tag=v$pkgver"
         fix_include_system.patch
         nccl_version.patch
@@ -42,6 +44,8 @@
   # https://github.com/pytorch/pytorch/issues/26555
   sed -i 's#^  ${CMAKE_CURRENT_SOURCE_DIR}/tensor_iterator_test.cpp##g' aten/src/ATen/test/CMakeLists.txt
 
+  sed -i "s/intel64//g" cmake/Modules/FindMKL.cmake
+
   # https://bugs.archlinux.org/task/64981
   patch -N torch/utils/cpp_extension.py "${srcdir}"/fix_include_system.patch
 
@@ -69,7 +73,7 @@
   export PYTORCH_BUILD_NUMBER=1
 
   # Check tools/setup_helpers/cmake.py, setup.py and CMakeLists.txt for a list of flags that can be set via env vars.
-  export USE_MKLDNN=OFF
+  export USE_MKLDNN=ON
   # export BUILD_CUSTOM_PROTOBUF=OFF
   # export BUILD_SHARED_LIBS=OFF
   export USE_FFMPEG=ON