Bot releases are hidden (Show)
This is a patch release containing the following changes to v3.4:
Published by vpirogov 7 months ago
This is a patch release containing the following changes to v3.3.5:
Published by vpirogov 8 months ago
Intel Architecture Processors:
1
and 14
.matmul
and add
operations and mixed int8 and bfloat16 data types with Graph API.reduction
, softmax
and layernorm
operations with experimental Graph Compiler backend.Intel Graphics Products:
AArch64-based Processors:
-mcpu=generic
to improve portability.--num-streams
knob in benchdnn to support benchmarking in multi-stream scenarios.This release contains contributions from the project core team as well as Alexander Grund @Flamefire, David Svantesson @davsva01, Fadi Arafeh @fadara01, Hugh Delaney @hdelan, Ilya Lavrenov @ilya-lavrenov, Jacob Kahn @jacobkahn, Nathan John Sircombe @nSircombe, Renato Barros Arantes @renato-arantes, Sergey Shalnov @shssf, Sunita Nadampalli @snadampal, and Svetlozar Georgiev @sgeor255. We would also like to thank everyone who asked questions and reported issues.
Published by vpirogov 8 months ago
This is a patch release containing the following changes to v3.3.4:
SEGFAULT
in int8 convolution on processors with Intel AMX support (2a8e122b63b55f897c470d23f21003bb70f0e839)Published by harrymao2022 8 months ago
Intel Architecture Processors:
1
and 14
.matmul
and add
operations and mixed int8 and bfloat16 data types with Graph API.reduction
, softmax
and layernorm
operations with experimental Graph Compiler backend.Intel Graphics Products:
AArch64-based Processors:
-mcpu=generic
to improve portability.--num-streams
knob in benchdnn to support benchmarking in multi-stream scenarios.This release contains contributions from the project core team as well as Alexander Grund @Flamefire, David Svantesson @davsva01, Fadi Arafeh @fadara01, Hugh Delaney @hdelan, Ilya Lavrenov @ilya-lavrenov, Jacob Kahn @jacobkahn, Nathan John Sircombe @nSircombe, Renato Barros Arantes @renato-arantes, Sergey Shalnov @shssf, Sunita Nadampalli @snadampal, and Svetlozar Georgiev @sgeor255. We would also like to thank everyone who asked questions and reported issues.
Published by vpirogov 10 months ago
This is a patch release containing the following changes to v3.3.3:
segfault
in 3D convolutions with different h
and w
parameters on Intel CPUs (b5f916ec068f783dbba2cd4f04a673e996f9efba)Published by vpirogov 10 months ago
This is a patch release containing the following changes to v3.3.2:
Published by vpirogov 11 months ago
This is a patch release containing the following changes to v3.3.1:
Published by vpirogov 11 months ago
This is a patch release containing the following changes to v3.3:
avgpool_bwd
operation in Graph API (d025ef6620b131f3487bb748866ddd9d7225c09f, 9e0602ad37afa18d46f407cb52577f1afead238b, e0dc1b3d070313052f5fd6ac739778d45b57859c)SEGFAULT
in experimental Graph Compiler for fp32 MLP subgraph (42071057abb2fcbbca6ed67117bdb1a5ee3dc0cd)unimplemented
on Intel GPUs (bf12207b0312c0174f0c47ae0d3abd70edc31957, 800b5e9613bd0994af82706ef024ad2b453be2b6, ec7054a2c79ae33d3db4ff04ce11360c2c896d56)Published by harrymao2022 about 1 year ago
any
memory format tag.dnnl::graph::set_constant_tensor_cache()
call.This release contains contributions from the project core team as well as Amy Wignall @AmyWignall-arm, @baibeta, Benjamin Taylor @bentaylorhk-arm, Ilya Lavrenov @ilya-lavrenov, Kentaro Kawakami @kawakami-k, Milos Puzovic @milpuz01, Renato Barros Arantes @renato-arantes, @snadampal, @sparkyrider, and Thomas Köppe @tkoeppe. We would also like to thank everyone who asked questions and reported issues.
Published by harrymao2022 about 1 year ago
any
memory format tag.dnnl::graph::set_constant_tensor_cache()
call.This release contains contributions from the project core team as well as Amy Wignall @AmyWignall-arm, @baibeta, Benjamin Taylor @bentaylorhk-arm, Kentaro Kawakami @kawakami-k, Milos Puzovic @milpuz01, @snadampal, @sparkyrider, Thomas Köppe @tkoeppe. We would also like to thank everyone who asked questions and reported issues.
Published by vpirogov about 1 year ago
This is a patch release containing the following changes to v2.7.4:
Published by vpirogov about 1 year ago
This is a patch release containing the following changes to v3.2:
SEGFAULT
when oneDNN primitives created in parallel (0a6202f5000cf347995ab744c25aa26cabf2482d)get_pointer
with get_multi_ptr
(fdbff4591f952d02a0c934f854a9b225a7097a21, 51ed43bb5cb08f38b0b652255a13bb4072b2ee57)ONEDNN_EXPERIMENTAL_PROFILING
build knob. This behavior manifests in apparent memory leak when oneDNN primitives are executed on a queue with enabled profiling (8d796efb609c33ecdd31e3e7b26d94d959dd83b9, 51a8f7ad892b1174d32cba8358804fad09b58f76, 2ca29381eeb5dde64d90468e440f87b6f9ad01d9)Published by harrymao2022 over 1 year ago
Intel Architecture Processors:
Intel Graphics Products:
AArch64-based Processors:
bf16
or any
with ACL.IBM Z Platform:
[experimental] Introduced Graph Compiler backend for Graph API. Graph Compiler improves performance of composite operations like multi-head attention (MHA), multi-level perceptron (MLP), and convolution residual blocks for processors with Intel AVX-512 and Intel AMX instructions support.
Extended Graph API with boolean data type, select, and pow operations.
Introduced support for binary and eltwise post-ops in softmax primitives.
Introduced reference SYCL implementations of batch normalization, layer normalization, linear response normalization (LRN), binary, softmax, eltwise, pooling, PReLU, shuffle, and resampling primitives. These implementations address functional gaps on NVIDIA and AMD GPUs where support is missing in native libraries.
Intel Graphics Products:
NVIDIA GPUs:
AMD GPUs:
--mode=F
) in benchdnn. Testing speed is improved by initializing oneDNN objects in parallel and avoiding use of host memory when benchmarking GPU primitives.This release contains contributions from the project core team as well as Abdelrauf @quickwritereader, Alexey Vishnyakov @SweetVishnya, Annop Wongwathanarat @annop-w, Anthony Roberts @anthony-linaro, Crefeda Rodrigues @cfRod, David Svantesson @davsva01, Fadi Arafeh @fadara01, Ilya Lavrenov @ilya-lavrenov, Jonathan Deakin @jondea, Kentaro Kawakami @kawakami-k, Milos Puzovic @milpuz01, RambabuSwargam @RambabuSwargam, Sai Teja @saiteja13427, Taiju Tsuiki @tzik. We would also like to thank everyone who asked questions and reported issues.
Published by vpirogov over 1 year ago
This is a patch release containing the following changes to v3.1:
bfloat16
convolution on processors with Intel AMX support (461d55e65f2bc0f45fcdfc3405493226218d22ee)int8
convolution primitive with scales (7fa3b6f335893270cdd079f4f8aadd36cf8f490b, bb3ecc460605eae3ca8a8ee79a8d9122f195730b)int8
convolution primitive with zero points on processors with Intel AVX2 and Intel DL Boost support (d721767a554f9a4da70bd6bc1c27c00b1ea80cc2, f6365b1b2c6e6d79e59207dad090b9643224f147)int8
inner product on processors with Intel AVX-512 and Intel DL Boost or Intel AMX support (2ede31e834a25ca14c648e8617b972148c94554c)fp32
convolution primitive weight gradient on Intel GPUs (ff209f967c2bdfa1139779cf59dced374e2064c5, 87108392da71b06594356a18232ac1378e28adfc)int8
convolution with zero points on Intel GPUs (cb9169397ceee206fece71f73b5d627ee9eea33f, 85e58af6b5cb1a9cd42cd602832c035a3b3a660f)fp32
convolution with Winograd algorithm on Intel GPUs (97ac88509bf8799fd03eb768faec302d44ce38dc)any
on Intel GPUs (ab2041d39862de747535037eb5a73c675d93d323, f2c457d72896d6c86245a6c6e80539b842aec369)gelu_erf
algorithm on Intel64 CPUs (e67abefadbb4fd73ea6a4d3981965bc56eb77b97)int8
matmul and inner product primitives on Intel GPUs based on Xe-HPG and Xe-HPC architecture (36aa6224ebae1413a6badd43ffc96d3412c8f8ec)bfloat16
convolution weight gradient on processors with Intel AMX support (c93e673bba299fdc62733f22d65d91f4dbc300dd, 8da108375bc02b08a385b167a49aa8d1189b66d6, f7acf9877b368a5f704dcc9efcb913345b477bbc)fp32
inner product implementation on processors with Intel AVX2 and Intel DL Boost supprt (f27dedbfc093f51032a4580198bb80579440dc15, f8d7c2e40a965fc52521d4ba9c793d8adc2be4e1)Published by harrymao2022 over 1 year ago
Intel Architecture Processors:
Intel Graphics Products:
AArch64-based Processors:
bf16
or any
with ACL.IBM Z Platform:
[experimental] Introduced Graph Compiler backend for Graph API. Graph Compiler improves performance of composite operations like multi-head attention (MHA), multi-level perceptron (MLP), and convolution residual blocks for processors with Intel AVX-512 and Intel AMX instructions support.
Extended Graph API with boolean data type, select, and pow operations.
Introduced support for binary and eltwise post-ops in softmax primitives.
Introduced reference SYCL implementations of batch normalization, layer normalization, linear response normalization (LRN), binary, softmax, eltwise, pooling, PReLU, shuffle, and resampling primitives. These implementations address functional gaps on NVIDIA and AMD GPUs where support is missing in native libraries.
Intel Graphics Products:
NVIDIA GPUs:
AMD GPUs:
--mode=F
) in benchdnn. Testing speed is improved by initializing oneDNN objects in parallel and avoiding use of host memory when benchmarking GPU primitives.This release contains contributions from the project core team as well as Abdelrauf @quickwritereader, Alexey Vishnyakov @SweetVishnya, Annop Wongwathanarat @annop-w, Anthony Roberts @anthony-linaro, Crefeda Rodrigues @cfRod, David Svantesson @davsva01, Fadi Arafeh @fadara01, Ilya Lavrenov @ilya-lavrenov, Jonathan Deakin @jondea, Kentaro Kawakami @kawakami-k, Milos Puzovic @milpuz01, RambabuSwargam @RambabuSwargam, Sai Teja @saiteja13427, Taiju Tsuiki @tzik. We would also like to thank everyone who asked questions and reported issues.
Published by vpirogov over 1 year ago
This is a patch release containing the following changes to v2.7.3:
NaN
issue in convolution weight gradient on Intel CPUs (6d80bb48714f8f8d030f055435f5bfde3a382f15, 4c34f89653259b2e15e277ff0663d6705f093e1b, 017950a16168640764d17558e41010d0ae038377, 796a600c3de2993b5d5819995ad13eb70d097496)Published by harrymao2022 over 1 year ago
Intel Architecture Processors:
Intel Graphics Products:
AArch64-based Processors:
AMD GPUs:
RISC-V-based Processors:
ONEDNN_BUILD_GRAPH=OFF
.This release contains contributions from the project core team as well as Amy Wignall @AmyWignall-arm, Annop Wongwathanarat @annop-w, @arlesniak, @bdmoore1, Crefeda Rodrigues @cfRod, David Svantesson @davsva01, Fadi Arafeh @fadara01, Jonathan Deakin @jondea, Kentaro Kawakami @kawakami-k, Pavel Zamelin @pazamelin, Pawel Piotrowicz @pawelpiotrowicz, Peter Caday @petercad, @ranzhejiang, and Sanchit Grover @sanchit-grover-intel. We would also like to thank everyone who asked questions and reported issues.
Published by vpirogov over 1 year ago
This is the Beta Update 3 release of oneDNN Graph API based on oneDNN v3.0.1.
This release contains contributions from the project core teams as well as Jiong Gong, Chunyuan Wu, Sanchit Jain, Yiqiang Li, Yunfei Mao, Kiefer Kuah and others.
Published by harrymao2022 over 1 year ago
This is a release candidate for oneDNN v3.1. Please provide feedback and submit defect reports via Github issues.
Intel Architecture Processors:
Intel Graphics Products:
AArch64-based Processors:
AMD GPUs:
RISC-V-based Processors:
ONEDNN_BUILD_GRAPH=OFF
.This release contains contributions from the project core team as well as Amy Wignall @AmyWignall-arm, Annop Wongwathanarat @annop-w, @arlesniak, @bdmoore1, Crefeda Rodrigues @cfRod, David Svantesson @davsva01, Fadi Arafeh @fadara01, Jonathan Deakin @jondea, Kentaro Kawakami @kawakami-k, Pavel Zamelin @pazamelin, Pawel Piotrowicz @pawelpiotrowicz, Peter Caday @petercad, @ranzhejiang, and Sanchit Grover @sanchit-grover-intel. We would also like to thank everyone who asked questions and reported issues.