Hello community, here is the log from the commit of package mkl-dnn for openSUSE:Factory checked in at 2020-09-29 19:05:46 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Comparing /work/SRC/openSUSE:Factory/mkl-dnn (Old) and /work/SRC/openSUSE:Factory/.mkl-dnn.new.4249 (New) ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Package is "mkl-dnn" Tue Sep 29 19:05:46 2020 rev:5 rq:838546 version:1.6.3 Changes: -------- --- /work/SRC/openSUSE:Factory/mkl-dnn/mkl-dnn.changes 2020-05-05 19:04:31.686579811 +0200 +++ /work/SRC/openSUSE:Factory/.mkl-dnn.new.4249/mkl-dnn.changes 2020-09-29 19:05:51.238056896 +0200 @@ -1,0 +2,13 @@ +Wed Sep 23 13:36:02 UTC 2020 - Guillaume GARDET <guillaume.gardet@opensuse.org> + +- Update to 1.6.3 +- Drop upstream patch: + * cmake-no-install-ocl-cmake.patch + +------------------------------------------------------------------- +Wed Sep 23 13:16:39 UTC 2020 - Guillaume GARDET <guillaume.gardet@opensuse.org> + +- Build on aarch64 and ppc64le which are now also supported +- Provide oneDNN and oneDNN-devel as it is the new official name + +------------------------------------------------------------------- Old: ---- cmake-no-install-ocl-cmake.patch mkl-dnn-1.4.tar.gz New: ---- onednn-1.6.3.tar.gz ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Other differences: ------------------ ++++++ mkl-dnn.spec ++++++ --- /var/tmp/diff_new_pack.OG1aPX/_old 2020-09-29 19:05:52.602058273 +0200 +++ /var/tmp/diff_new_pack.OG1aPX/_new 2020-09-29 19:05:52.602058273 +0200 @@ -15,26 +15,34 @@ # Please submit bugfixes or comments via https://bugs.opensuse.org/ # +%ifarch x86_64 +%bcond_without opencl +%else +# Build broken on non-x86, with openCL +%bcond_with opencl +%endif %define libname libdnnl1 Name: mkl-dnn -Version: 1.4 +Version: 1.6.3 Release: 0 Summary: Intel(R) Math Kernel Library for Deep Neural Networks License: Apache-2.0 -URL: https://01.org/mkl-dnn -Source0: https://github.com/intel/mkl-dnn/archive/v%{version}.tar.gz#/%{name}-%{version}.tar.gz -Patch0: cmake-no-install-ocl-cmake.patch +URL: https://01.org/onednn +Source0: https://github.com/oneapi-src/oneDNN/archive/v%{version}/onednn-%{version}.tar.gz BuildRequires: cmake BuildRequires: doxygen BuildRequires: fdupes BuildRequires: gcc-c++ BuildRequires: graphviz +BuildRequires: texlive-dvips-bin +%if %{with opencl} BuildRequires: opencl-headers BuildRequires: pkgconfig -BuildRequires: texlive-dvips-bin BuildRequires: pkgconfig(OpenCL) -ExclusiveArch: x86_64 +%endif +ExclusiveArch: x86_64 aarch64 ppc64le +Provides: oneDNN = %{version} %description Intel(R) Math Kernel Library for Deep Neural Networks (Intel(R) MKL-DNN) is an @@ -59,6 +67,7 @@ %package devel Summary: Header files of Intel(R) Math Kernel Library Requires: %{libname} = %{version} +Provides: oneDNN-devel = %{version} %description devel Intel(R) Math Kernel Library for Deep Neural Networks (Intel(R) MKL-DNN) is an @@ -97,7 +106,9 @@ -DCMAKE_INSTALL_LIBDIR=%{_lib} \ -DMKLDNN_ARCH_OPT_FLAGS="" \ -DDNNL_CPU_RUNTIME=OMP \ +%if %{with opencl} -DDNNL_GPU_RUNTIME=OCL \ +%endif -DDNNL_INSTALL_MODE=DEFAULT \ -DDNNL_BUILD_TESTS=ON \ -DDNNL_WERROR=OFF @@ -139,6 +150,7 @@ %{_includedir}/%{name} %{_libdir}/*.so %{_libdir}/cmake/dnnl +%{_libdir}/cmake/mkldnn %files doc %{_docdir}/%{name}