Upgrade xnnpack to latest #25275
Merged
Upgrade xnnpack to latest #25275
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
Contributor
Author
|
@fs-eire , There seems to be quite a lot of download failures from vcpkg during the build. Do you have any suggestions on how to address those issues? |
Member
|
@fanchenkong1 , I have uploaded the files to the internal mirror. Now please check the remaining errors. |
Contributor
Author
Thanks, that’s very helpful! |
Contributor
|
/azp run Linux QNN CI Pipeline, Win_TRT_Minimal_CUDA_Test_CI, Windows ARM64 QNN CI Pipeline, Windows GPU Doc Gen CI Pipeline, Windows x64 QNN CI Pipeline |
|
Azure Pipelines successfully started running 5 pipeline(s). |
fanchenkong1
force-pushed
the
upgrade-xnnpack
branch
from
b0d7146 to
477cdf9
Compare
Contributor
|
/azp run Linux QNN CI Pipeline, Win_TRT_Minimal_CUDA_Test_CI, Windows ARM64 QNN CI Pipeline, Windows GPU Doc Gen CI Pipeline, Windows x64 QNN CI Pipeline |
|
Azure Pipelines successfully started running 5 pipeline(s). |
Member
|
Please format the code with clang-format |
Contributor
|
There are QNN CI pipeline failures: Build Logs |
Contributor
Author
|
Thanks, I will take a look. |
Contributor
Author
I reproduced the ARM64 QNN CI run on my local device and find that the same “unknown opcode” errors appear as soon as I roll MSVC back to 19.39. It seems that the older toolset rejects the SVE/SVE2/SME assembly used in Kleidiai. Kleidiai does not appear to expose a switch for selecting the kernel set. Could you let me know which way you’d prefer to resolve the failure?
|
Contributor
|
@HectorSVC is it OK to upgrade minimal MSVC version requirement to 19.40 for QNN build? |
Contributor
|
/azp run Linux QNN CI Pipeline, Win_TRT_Minimal_CUDA_Test_CI, Windows ARM64 QNN CI Pipeline, Windows GPU Doc Gen CI Pipeline, Windows x64 QNN CI Pipeline |
|
Azure Pipelines successfully started running 5 pipeline(s). |
fanchenkong1
force-pushed
the
upgrade-xnnpack
branch
from
0286414 to
c680ef4
Compare
Contributor
Author
|
The macos xnnpack CI has started failing on a vcpkg download issue when fetching googletest. |
Contributor
|
/azp run Linux QNN CI Pipeline, Win_TRT_Minimal_CUDA_Test_CI, Windows ARM64 QNN CI Pipeline, Windows GPU Doc Gen CI Pipeline, Windows x64 QNN CI Pipeline |
|
Azure Pipelines successfully started running 5 pipeline(s). |
Contributor
Author
|
The failed CI are still using msvc 19.34. Will there be another toolchain upgrade on Windows Arm64 bot? |
Contributor
|
@fanchenkong1 Upgrading the QNN CI pipeline is being worked on. Will keep you updated. |
Contributor
Author
Hi @fs-eire , just checking in to see if there’s any update on the QNN CI pipeline upgrade. Appreciate your efforts—thanks! |
Member
|
I updated the compiler. Would you please resolve the merge conflict and let us try it again? |
fanchenkong1
force-pushed
the
upgrade-xnnpack
branch
from
c680ef4 to
4ce96e3
Compare
fs-eire
approved these changes
Jul 28, 2025
sanketkaleoss
pushed a commit
to sanketkaleoss/onnxruntime
that referenced
this pull request
Aug 11, 2025
### Description This change is based on microsoft#25135. Upgrade xnnpack and several related third-party dependencies, including pthreadpool, cpuinfo, and kleidiai. This change also updates the xnnpack execution provider code to accommodate changes in the xnnpack api. Average pooling qu8 is removed as the corresponding microkernel seems no longer exist in xnnpack.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
Description
This change is based on #25135.
Upgrade xnnpack and several related third-party dependencies, including pthreadpool, cpuinfo, and kleidiai. This change also updates the xnnpack execution provider code to accommodate changes in the xnnpack api. Average pooling qu8 is removed as the corresponding microkernel seems no longer exist in xnnpack.