%global __cmake_in_source_build 1 Name: onednn Version: 1.6 Release: 1 Summary: Deep Neural Network Library License: ASL 2.0 and BSD and Boost and MIT URL: https://github.com/oneapi-src/oneDNN/ Source0: %{url}/archive/v%{version}/onednn-%{version}.tar.gz # This package only work in few arches for now ExclusiveArch: x86_64 aarch64 ppc64le BuildRequires: cmake BuildRequires: doxygen BuildRequires: gcc-c++ # Optionals not yet enabled #BuildRequires: pkgconfig(OpenCL) #BuildRequires: pkgconfig(tbb) # Virtual provides mkldnn Provides: mkldnn = %{version}-%{release} Provides: mkl-dnn = %{version}-%{release} Obsoletes: mkl-dnn < 1.3 # Provides oneDNN Provides: oneDNN = %{version}-%{release} %description one-API Deep Neural Network Library (oneDNN) is an open-source performance library for deep learning applications. The library includes basic building blocks for neural networks optimized for Intel Architecture Processors and Intel Processor Graphics. oneDNN is intended for deep learning applications and framework developers interested in improving application performance on Intel CPUs and GPUs. Deep learning practitioners should use one of the applications enabled with oneDNN: %package devel Summary: Development files for %{name} Requires: %{name}%{?_isa} = %{version}-%{release} %description devel The %{name}-devel package contains libraries and header files for developing applications that use %{name}. %prep %autosetup -p1 -n oneDNN-%{version} %build mkdir -p build && cd build %cmake .. %make_build %install cd build %make_install # Remove docs rm -rf %{buildroot}%{_docdir}/dnnl %ldconfig_scriptlets %files %license LICENSE THIRD-PARTY-PROGRAMS %doc README.md CONTRIBUTING.md CODE_OF_CONDUCT.md %{_libdir}/libdnnl.so.1 %{_libdir}/libdnnl.so.1.* %{_libdir}/libmkldnn.so.1 %{_libdir}/libmkldnn.so.1.* %files devel %{_includedir}/mkldnn*.h* %{_includedir}/dnnl*.h* %{_libdir}/libdnnl.so %{_libdir}/libmkldnn.so %dir %{_libdir}/cmake/dnnl %{_libdir}/cmake/dnnl/*.cmake %dir %{_libdir}/cmake/mkldnn %{_libdir}/cmake/mkldnn/*.cmake %changelog * Sun Dec 13 2020 sinever - 1.6-1 - Initial spec file.