From c8ad253ceb2a73b6f7208fc85c6794824c8b11c9e462e5262426142e3e3da5e6 Mon Sep 17 00:00:00 2001 From: Guillaume GARDET Date: Tue, 13 Apr 2021 08:43:15 +0000 Subject: [PATCH 1/2] Accepting request 884885 from home:Guillaume_G:branches:science:machinelearning - Update to 2.2.1 OBS-URL: https://build.opensuse.org/request/show/884885 OBS-URL: https://build.opensuse.org/package/show/science:machinelearning/onednn?expand=0&rev=5 --- onednn-2.1.tar.gz | 3 --- onednn-2.2.1.tar.gz | 3 +++ onednn.changes | 5 +++++ onednn.spec | 2 +- 4 files changed, 9 insertions(+), 4 deletions(-) delete mode 100644 onednn-2.1.tar.gz create mode 100644 onednn-2.2.1.tar.gz diff --git a/onednn-2.1.tar.gz b/onednn-2.1.tar.gz deleted file mode 100644 index 17d7cef..0000000 --- a/onednn-2.1.tar.gz +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:5f7fd92e2d0bf83580656695d4404e2cd1390ecad36496fd8ba10b5adc905f70 -size 9273493 diff --git a/onednn-2.2.1.tar.gz b/onednn-2.2.1.tar.gz new file mode 100644 index 0000000..36f9a0b --- /dev/null +++ b/onednn-2.2.1.tar.gz @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3faf3b7811dd37835169a9d5c57f17fd591f062029851186f9b11db117b9d1d9 +size 9533694 diff --git a/onednn.changes b/onednn.changes index 1dd4655..eb9ec91 100644 --- a/onednn.changes +++ b/onednn.changes @@ -1,3 +1,8 @@ +------------------------------------------------------------------- +Tue Apr 13 07:53:16 UTC 2021 - Guillaume GARDET + +- Update to 2.2.1 + ------------------------------------------------------------------- Wed Feb 17 14:17:47 UTC 2021 - Guillaume GARDET diff --git a/onednn.spec b/onednn.spec index 77d7119..2a2e683 100644 --- a/onednn.spec +++ b/onednn.spec @@ -31,7 +31,7 @@ %define libname libdnnl2 Name: onednn -Version: 2.1 +Version: 2.2.1 Release: 0 Summary: Intel(R) Math Kernel Library for Deep Neural Networks License: Apache-2.0 From 6531f118270734bc1e75c770d06cd3574b0f4dfe1ed452d59ad54b87958d5df3 Mon Sep 17 00:00:00 2001 From: Guillaume GARDET Date: Tue, 13 Apr 2021 13:59:51 +0000 Subject: [PATCH 2/2] OBS-URL: https://build.opensuse.org/package/show/science:machinelearning/onednn?expand=0&rev=6 --- onednn.changes | 35 ++++++++++++++++++++++++++++++++++- 1 file changed, 34 insertions(+), 1 deletion(-) diff --git a/onednn.changes b/onednn.changes index eb9ec91..3fe412d 100644 --- a/onednn.changes +++ b/onednn.changes @@ -1,7 +1,40 @@ ------------------------------------------------------------------- Tue Apr 13 07:53:16 UTC 2021 - Guillaume GARDET -- Update to 2.2.1 +- Update to 2.2.1, changes: + * From 2.2: + Fixed segfault for cases when primitive descriptor or attributed contain NaN (e6d05ec, dbca1e9, 0326b09, 0326b09) + Fixed engine creation failure for GPU subdevices (4c3a114) + Fixed long lines clipping in verbose output (70d70a8) + Fixed segfault in bfloat16 convolution weight gradient implementation on processors with Intel AMX support (a3a73a3) + Fixed performance regression in binary primitive with per_oc broadcast strategy (9ac85d8) + Worked around a bug with Microsoft Visual C++ compiler version detection in CMake 3.19 (2f39155) + Removed -std=c++11 build flag for DPC++ code to align with SYCL standard (1b026f5) + * Changes between 2.1 and 2.2: + Performance Optimizations + Intel Architecture processors + Improved performance of int8 compute functionality for future Intel Xeon Scalable processor (code name Sapphire Rapids). The functionality is disabled by default and should be enabled via CPU dispatcher control. + Improved performance of compute functionality for future Intel Core processor with Intel AVX2 and Intel DL Boost instructions support (code name Alder Lake). + Improved fp32 inner product forward propagation performance for processors with Intel AVX-512 support. + Improved dnnl_gemm performance for cases with n=1 on all supported processors. + Intel Graphics products + Introduced NHWC format support for activations for int8 primitives. + AArch64-based processors + Improved performance of fp32 and int8 convolution, and softmax primitives for processors with SVE 512 support. + Improved performance of fp32 convolution via Arm Compute Library (ACL). + Improved performance of convolution with a combination of sum and relu post-ops via ACL. + Functionality + Extended eltwise primitive with support for mish and hardswish algorithms. + Extended binary primitive with support for comparison operators. + Introduced support for post-ops in GPU resampling implementation. + Introduced asymmetric quantization support for int8 deconvolution. + Introduced binary post-ops support for matmul primitive. + Usability + Improved presentation of oneDNN primitives in VTune Amplifier. + Introduced Linux perf support for AArch64. + Introduced support for Fujitsu C++ compiler. + Introduced a build time check for minimal supported ACL version. Currently oneDNN requires ACL 21.02 or later. + Added support for cuDNN 8.x ------------------------------------------------------------------- Wed Feb 17 14:17:47 UTC 2021 - Guillaume GARDET