Upgrade submodule oneDNN to v3.7 #147498
Closed
Upgrade submodule oneDNN to v3.7 #147498
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/147498
Note: Links to docs will display an error until the docs builds have been completed. ❌ 2 New Failures, 1 Unrelated FailureAs of commit 1e329e1 with merge base 90e3a3d ( NEW FAILURES - The following jobs have failed:
FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Closed
8000
|
@atalman has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
@pytorchmergebot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
|
@pytorchbot rebase -b main |
|
@pytorchbot started a rebase job onto refs/remotes/origin/main. Check the current status here |
|
Rebase failed due to Raised by https://github.com/pytorch/pytorch/actions/runs/13523596472 |
|
@pytorchbot rebase -b main |
|
@pytorchbot started a rebase job onto refs/remotes/origin/main. Check the current status here |
|
Rebase failed due to Raised by https://github.com/pytorch/pytorch/actions/runs/13524908106 |
a024a93 to
1e329e1
Compare
|
@xuhancn I have rebased to main manually. |
xuhancn
pushed a commit
to xuhancn/pytorch
that referenced
this pull request
Feb 26, 2025
This PR is to upgrade submodule oneDNN to v3.7. ## Improvements - Improved performance of convolution and matmul primitives on Intel Xeon processors with Intel AMX instruction set support (formerly Sapphire Rapids and Granite Rapids). - Improved performance of int8 and fp32 forward convolution primitive on processors with Intel AVX2 instruction set support. - Improved performance of fp8 matmul primitives with bf16 and fp16 bias data type on Intel Xeon processors with Intel AMX instruction set support (formerly Sapphire Rapids and Granite Rapids). - Introduced initial optimizations for Intel GPUs based on Xe3 architecture. - Added bfloat16 support for SDPA, implemented fp16 and bf16 gemm kernel in SDPA. - Fixed f16 matmul accuracy, the issue of SDPA cannot dispatched to ukernel, bf16/fp16/fp32 conv performance, INT8 Kernel trigger page fault, deconvolution precision issue on complex128 and fp64 and gemm correctness issue in float16 issues. - Improved bf16 matmul performance with fp32 destination with Arm Compute Library (ACL). - Improved bf16 to fp32 reorder performance. - Improved bf16 reorder performance. - Improved bf16 convolution with ACL. Fixes pytorch#136348. ## Validation results on CPU 1. NLP models accuracy/inference/training   2. Torchbench cpu userbenchmark inference & training  3. Inductor quantization  4. Dynamo benchmarks         ## Validation results on XPU Accuracy is same as baseline. Performance is shown below.  ## Validation results on ARM   Pull Request resolved: pytorch#147498 Approved by: https://github.com/fadara01, https://github.com/mingfeima, https://github.com/atalman
DDEle
added a commit
to DDEle/pytorch
that referenced
this pull request
Feb 26, 2025
This reverts commit e72b4c6.
DDEle
added a commit
to DDEle/pytorch
that referenced
this pull request
Feb 27, 2025
This reverts commit e72b4c6.
aditew01
pushed a commit
that referenced
this pull request
Feb 28, 2025
This PR is to upgrade submodule oneDNN to v3.7. ## Improvements - Improved performance of convolution and matmul primitives on Intel Xeon processors with Intel AMX instruction set support (formerly Sapphire Rapids and Granite Rapids). - Improved performance of int8 and fp32 forward convolution primitive on processors with Intel AVX2 instruction set support. - Improved performance of fp8 matmul primitives with bf16 and fp16 bias data type on Intel Xeon processors with Intel AMX instruction set support (formerly Sapphire Rapids and Granite Rapids). - Introduced initial optimizations for Intel GPUs based on Xe3 architecture. - Added bfloat16 support for SDPA, implemented fp16 and bf16 gemm kernel in SDPA. - Fixed f16 matmul accuracy, the issue of SDPA cannot dispatched to ukernel, bf16/fp16/fp32 conv performance, INT8 Kernel trigger page fault, deconvolution precision issue on complex128 and fp64 and gemm correctness issue in float16 issues. - Improved bf16 matmul performance with fp32 destination with Arm Compute Library (ACL). - Improved bf16 to fp32 reorder performance. - Improved bf16 reorder performance. - Improved bf16 convolution with ACL. Fixes #136348. ## Validation results on CPU 1. NLP models accuracy/inference/training   2. Torchbench cpu userbenchmark inference & training  3. Inductor quantization  4. Dynamo benchmarks         ## Validation results on XPU Accuracy is same as baseline. Performance is shown below.  ## Validation results on ARM   Pull Request resolved: #147498 Approved by: https://github.com/fadara01, https://github.com/mingfeima, https://github.com/atalman
aditew01
pushed a commit
that referenced
this pull request
Feb 28, 2025
This reverts commit 576ed1e. Reverted #147498 on behalf of https://github.com/wdvr due to failing some tests on trunk - see below ([comment](#147498 (comment)))
aditew01
pushed a commit
that referenced
this pull request
Feb 28, 2025
This PR is to upgrade submodule oneDNN to v3.7. ## Improvements - Improved performance of convolution and matmul primitives on Intel Xeon processors with Intel AMX instruction set support (formerly Sapphire Rapids and Granite Rapids). - Improved performance of int8 and fp32 forward convolution primitive on processors with Intel AVX2 instruction set support. - Improved performance of fp8 matmul primitives with bf16 and fp16 bias data type on Intel Xeon processors with Intel AMX instruction set support (formerly Sapphire Rapids and Granite Rapids). - Introduced initial optimizations for Intel GPUs based on Xe3 architecture. - Added bfloat16 support for SDPA, implemented fp16 and bf16 gemm kernel in SDPA. - Fixed f16 matmul accuracy, the issue of SDPA cannot dispatched to ukernel, bf16/fp16/fp32 conv performance, INT8 Kernel trigger page fault, deconvolution precision issue on complex128 and fp64 and gemm correctness issue in float16 issues. - Improved bf16 matmul performance with fp32 destination with Arm Compute Library (ACL). - Improved bf16 to fp32 reorder performance. - Improved bf16 reorder performance. - Improved bf16 convolution with ACL. Fixes #136348. ## Validation results on CPU 1. NLP models accuracy/inference/training   2. Torchbench cpu userbenchmark inference & training  3. Inductor quantization  4. Dynamo benchmarks         ## Validation results on XPU Accuracy is same as baseline. Performance is shown below.  ## Validation results on ARM   Pull Request resolved: #147498 Approved by: https://github.com/fadara01, https://github.com/mingfeima, https://github.com/atalman
|
Windows CI fail is root caused to MSVC 19.29.30158.0 bug that breaks binary primitive constructor in C++17 mode. PR uxlfoundation/oneDNN#2783 with a workaround is in validation. We'll have oneDNN v3.7.1 tagged as soon as it lands. |
pytorch-bot bot
pushed a commit
that referenced
this pull request
Mar 3, 2025
This PR is to upgrade submodule oneDNN to v3.7. ## Improvements - Improved performance of convolution and matmul primitives on Intel Xeon processors with Intel AMX instruction set support (formerly Sapphire Rapids and Granite Rapids). - Improved performance of int8 and fp32 forward convolution primitive on processors with Intel AVX2 instruction set support. - Improved performance of fp8 matmul primitives with bf16 and fp16 bias data type on Intel Xeon processors with Intel AMX instruction set support (formerly Sapphire Rapids and Granite Rapids). - Introduced initial optimizations for Intel GPUs based on Xe3 architecture. - Added bfloat16 support for SDPA, implemented fp16 and bf16 gemm kernel in SDPA. - Fixed f16 matmul accuracy, the issue of SDPA cannot dispatched to ukernel, bf16/fp16/fp32 conv performance, INT8 Kernel trigger page fault, deconvolution precision issue on complex128 and fp64 and gemm correctness issue in float16 issues. - Improved bf16 matmul performance with fp32 destination with Arm Compute Library (ACL). - Improved bf16 to fp32 reorder performance. - Improved bf16 reorder performance. - Improved bf16 convolution with ACL. Fixes #136348. ## Validation results on CPU 1. NLP models accuracy/inference/training   2. Torchbench cpu userbenchmark inference & training  3. Inductor quantization  4. Dynamo benchmarks         ## Validation results on XPU Accuracy is same as baseline. Performance is shown below.  ## Validation results on ARM   Pull Request resolved: #147498 Approved by: https://github.com/fadara01, https://github.com/mingfeima, https://github.com/atalman
pytorchmergebot
pushed a commit
that referenced
this pull request
Mar 4, 2025
Motivation === This PR is part of the plan of OneDNN Upstreaming, as #114848 [(comment)](#114848 (comment)) stated. The support of SDPA is via the overridable variance on XPU backend. Beside the added `Attention.cpp` file, `Graph.h` is added to hold utils for OneDNN graph including those for kernel/compile graph caching. In addition, a selection of testcases in `test/test_transformers.py` are copied into the new `test/xpu/test_transformers.py` and modified accordingly to provide additional tests beyond `./third_party/torch-xpu-ops/test/xpu/test_ops_xpu.py`. Depends on OneDNN version v3.7 upgrade in #147498 Depends on BUILD_GRAPH switch in #147608 Pull Request resolved: #147614 Approved by: https://github.com/jansel, https://github.com/EikanWang
pytorchmergebot
pushed a commit
to min-jean-cho/pytorch
that referenced
this pull request
Mar 5, 2025
Motivation === This PR is part of the plan of OneDNN Upstreaming, as pytorch#114848 [(comment)](pytorch#114848 (comment)) stated. The support of SDPA is via the overridable variance on XPU backend. Beside the added `Attention.cpp` file, `Graph.h` is added to hold utils for OneDNN graph including those for kernel/compile graph caching. In addition, a selection of testcases in `test/test_transformers.py` are copied into the new `test/xpu/test_transformers.py` and modified accordingly to provide additional tests beyond `./third_party/torch-xpu-ops/test/xpu/test_ops_xpu.py`. Depends on OneDNN version v3.7 upgrade in pytorch#147498 Depends on BUILD_GRAPH switch in pytorch#147608 Pull Request resolved: pytorch#147614 Approved by: https://github.com/jansel, https://github.com/EikanWang
|
Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR is to upgrade submodule oneDNN to v3.7.
Improvements
Fixes #136348.
Validation results on CPU
Validation results on XPU
Accuracy is same as baseline. Performance is shown below.
Validation results on ARM
cc @gujinghui @PenghuiCheng @XiaobingSuper @jianyuh @jgong5 @mingfeima @sanchitintel @ashokei @jingxu10 @min-jean-cho @Guobing-Chen @Xia-Weiwen @snadampal @malfet @milpuz01 @aditew01 @nikhil-arm @fadara01 @voznesenskym @penguinwu @EikanWang @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @chenyang78 @kadeng @muchulee8 @amjames @chauhang @aakhundov @yf225