diff --git a/_service b/_service new file mode 100644 index 0000000..6df2763 --- /dev/null +++ b/_service @@ -0,0 +1,3 @@ + + + diff --git a/oneDNN-3.6.2.tar.gz b/oneDNN-3.6.2.tar.gz new file mode 100644 index 0000000..25befeb --- /dev/null +++ b/oneDNN-3.6.2.tar.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e79db0484dcefe2c7ff6604c295d1de2830c828941898878c80dfb062eb344d1 +size 13788211 diff --git a/oneDNN-3.6.tar.gz b/oneDNN-3.6.tar.gz deleted file mode 100644 index ec33ffe..0000000 --- a/oneDNN-3.6.tar.gz +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:20c4a92cc0ae0dc19d3d2beca0e357b1d13a5a3af9890a2cc3e41a880e4a0302 -size 13782760 diff --git a/onednn.changes b/onednn.changes index a132008..63ee567 100644 --- a/onednn.changes +++ b/onednn.changes @@ -1,3 +1,9 @@ +------------------------------------------------------------------- +Fri Jan 3 23:23:08 UTC 2025 - Eyad Issa + +- Update to 3.6.2: + * https://github.com/oneapi-src/oneDNN/releases/tag/v3.6.2 + ------------------------------------------------------------------- Thu Oct 17 11:42:58 UTC 2024 - Guillaume GARDET @@ -24,27 +30,27 @@ Mon Sep 23 10:04:43 UTC 2024 - Guillaume GARDET Fri Apr 19 17:27:48 UTC 2024 - Alessandro de Oliveira Faria - Update to 3.4.1: - * Fixed an issue with caching and serialization of primitives in + * Fixed an issue with caching and serialization of primitives in deterministic mode (7ed604a) - * Introduced memory descriptor serialization API + * Introduced memory descriptor serialization API (4cad420, 929a27a, 9b848c8) - * Fixed incorrect results in fp64 convolution and deconvolution - on Intel GPUs based on Xe-LPG architecture (ebe77b5, 0b399ac, + * Fixed incorrect results in fp64 convolution and deconvolution + on Intel GPUs based on Xe-LPG architecture (ebe77b5, 0b399ac, d748d64, 9f4f3d5, 21a8cae) - * Fixed incorrect results in reorder with large sizes on + * Fixed incorrect results in reorder with large sizes on Intel CPUs and GPUs (69a111e, 4b72361, 74a343b) - * Reduced creation time for deconvolution primitive on + * Reduced creation time for deconvolution primitive on Intel CPUs (bec487e, 1eab005) - * Fixed performance regression in deconvolution on + * Fixed performance regression in deconvolution on Intel CPUs (fbe5b97, 1dd3c6a) - * Removed dangling symblols from static builds + * Removed dangling symblols from static builds (e92c404, 6f5621a) - * Fixed crash during platform detection on some + * Fixed crash during platform detection on some AArch64-based systems (406a079) - * Fixed performance regression in int8 deconvolution on + * Fixed performance regression in int8 deconvolution on Intel CPUs (7e50e15) - * Fixed handling of zero points for matmul in verbose - logs converter (15c7916) + * Fixed handling of zero points for matmul in verbose + logs converter (15c7916) ------------------------------------------------------------------- Fri Dec 1 04:33:49 UTC 2023 - Alessandro de Oliveira Faria @@ -279,38 +285,38 @@ Wed Feb 26 10:36:26 UTC 2020 - Tomáš Chvátal ------------------------------------------------------------------- Thu Feb 20 10:26:52 UTC 2020 - Christian Goll -- enabled tests +- enabled tests ------------------------------------------------------------------- Thu Jan 30 14:20:22 UTC 2020 - Christian Goll - packaged separate benchnn packae with its input files - updated to v1.1.3 which includes - * Fixed the mean and variance memory descriptors in layer + * Fixed the mean and variance memory descriptors in layer normalization (65f1908) * Fixed the layer normalization formula (c176ceb) ------------------------------------------------------------------- Wed Jan 8 15:21:54 UTC 2020 - Christian Goll -- updated to v1.1.2 +- updated to v1.1.2 * Fixed threading over the spatial in bfloat16 batched normalization (017b6c9) * Fixed read past end-of-buffer error for int8 convolution (7d6f45e) - * Fixed condition for dispatching optimized channel blocking in + * Fixed condition for dispatching optimized channel blocking in fp32 backward convolution on Intel Xeon Phi(TM) processor (846eba1) - * Fixed fp32 backward convolution for shapes with spatial strides + * Fixed fp32 backward convolution for shapes with spatial strides over the depth dimension (002e3ab) * Fixed softmax with zero sizes on GPU (936bff4) * Fixed int8 deconvolution with dilation when ih <= dh (3e3bacb) * Enabled back fp32 -> u8 reorder for RNN (a2c2507) - * Fixed segmentation fault in bfloat16 backward convolution from + * Fixed segmentation fault in bfloat16 backward convolution from kd_padding=0 computation (52d476c) - * Fixed segmentation fault in bfloat16 forward convolution due + * Fixed segmentation fault in bfloat16 forward convolution due to push/pop imbalance (4f6e3d5) * Fixed library version for OS X build (0d85005) * Fixed padding by channels in concat (a265c7d) - * Added full text of third party licenses and + * Added full text of third party licenses and copyright notices to LICENSE file (79f204c) * Added separate README for binary packages (28f4c96) * Fixed computing per-oc mask in RNN (ff3ffab) @@ -324,7 +330,7 @@ Mon Feb 11 16:35:48 UTC 2019 - cgoll@suse.com ------------------------------------------------------------------- Tue Feb 5 07:45:53 UTC 2019 - Christian Goll -- Initial checking of the Intel(R) Math Kernel Library for +- Initial checking of the Intel(R) Math Kernel Library for Deep Neural Networks which can be used by: * tensorflow * Caffee diff --git a/onednn.spec b/onednn.spec index 488cc32..11cc20e 100644 --- a/onednn.spec +++ b/onednn.spec @@ -1,7 +1,7 @@ # # spec file for package onednn # -# Copyright (c) 2024 SUSE LLC +# Copyright (c) 2025 SUSE LLC # # All modifications and additions to the file contributed by third parties # remain the property of their copyright owners, unless otherwise agreed @@ -16,27 +16,25 @@ # +%define libname libdnnl3 %ifarch x86_64 %bcond_without opencl %else # Build broken on non-x86, with openCL %bcond_with opencl %endif - %ifarch aarch64 # Disable ACL until fixed upstream - https://github.com/oneapi-src/oneDNN/issues/2137 %bcond_with acl %else %bcond_with acl %endif - -%define libname libdnnl3 Name: onednn -Version: 3.6 +Version: 3.6.2 Release: 0 -Summary: Intel Math Kernel Library for Deep Neural Networks +Summary: oneAPI Deep Neural Network Library (oneDNN) License: Apache-2.0 -URL: https://01.org/onednn +URL: https://github.com/oneapi-src/oneDNN Source0: https://github.com/oneapi-src/oneDNN/archive/v%{version}/oneDNN-%{version}.tar.gz BuildRequires: chrpath BuildRequires: cmake @@ -44,7 +42,12 @@ BuildRequires: doxygen BuildRequires: fdupes BuildRequires: gcc-c++ BuildRequires: graphviz +BuildRequires: ninja BuildRequires: texlive-dvips-bin +Provides: mkl-dnn = %{version} +Obsoletes: mkl-dnn <= %{version} +Provides: oneDNN = %{version} +ExclusiveArch: x86_64 aarch64 ppc64le %if %{with acl} BuildRequires: ComputeLibrary-devel >= 24.08.1 %endif @@ -53,17 +56,13 @@ BuildRequires: opencl-headers BuildRequires: pkgconfig BuildRequires: pkgconfig(OpenCL) %endif -ExclusiveArch: x86_64 aarch64 ppc64le -Provides: mkl-dnn = %{version} -Obsoletes: mkl-dnn <= %{version} -Provides: oneDNN = %{version} %description -Intel Math Kernel Library for Deep Neural Networks (Intel MKL-DNN) is an -open-source performance library for deep-learning applications. The library -accelerates deep-learning applications and frameworks on Intel architecture. -Intel MKL-DNN contains vectorized and threaded building blocks that you can use -to implement deep neural networks (DNN) with C and C++ interfaces. +oneAPI Deep Neural Network Library (oneDNN) is an open-source cross-platform +performance library of basic building blocks for deep learning applications. + +oneDNN project is part of the UXL Foundation and is an implementation of the +oneAPI specification for oneDNN component. %package -n benchdnn Summary: Header files of Intel Math Kernel Library @@ -81,13 +80,13 @@ This package only includes the benchmark utility including its input files. %package devel Summary: Header files of Intel Math Kernel Library Requires: %{libname} = %{version} +Provides: mkl-dnn-devel = %{version} +Obsoletes: mkl-dnn-devel <= %{version} +Provides: oneDNN-devel = %{version} %if %{with opencl} Requires: opencl-headers Requires: pkgconfig(OpenCL) %endif -Provides: mkl-dnn-devel = %{version} -Obsoletes: mkl-dnn-devel <= %{version} -Provides: oneDNN-devel = %{version} %description devel Intel Math Kernel Library for Deep Neural Networks (Intel MKL-DNN) is an @@ -121,6 +120,7 @@ to implement deep neural networks (DNN) with C and C++ interfaces. %autosetup -p1 -n oneDNN-%{version} %build +%define __builder ninja %cmake \ -DCMAKE_INSTALL_LIBDIR=%{_lib} \ -DMKLDNN_ARCH_OPT_FLAGS="" \ @@ -164,7 +164,7 @@ chrpath -d %{buildroot}/%{_bindir}/benchdnn # do not use macro so we can exclude all gpu and cross (gpu and cpu) tests (they need gpu set up) pushd build export LD_LIBRARY_PATH=%{buildroot}%{_libdir} -ctest --output-on-failure --force-new-ctest-process %{_smp_mflags} -E '(gpu|cross)' +ctest --output-on-failure --force-new-ctest-process %{?_smp_mflags} -E '(gpu|cross)' popd %post -n %{libname} -p /sbin/ldconfig