Skip to content

Conversation

@qti-kromero
Copy link
Contributor

Description

  • Adds Genie libraries to onnxruntime-qnn for future GenAI improvements accessible through Genie API

Motivation and Context

  • Lays groundwork for eventual integration of Genie API with onnxruntime-qnn

@edgchen1 edgchen1 added the ep:QNN issues related to QNN exeution provider label Dec 12, 2025
@edgchen1
Copy link
Contributor

why not add the additional libraries at the time they are used?

Copy link
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

This PR adds Genie library dependencies to the QNN (Qualcomm Neural Network) execution provider to enable future GenAI improvements through the Genie API. The changes ensure that Genie libraries are properly included in both the Python package distribution and CMake build configurations across Linux and Windows platforms.

Key changes:

  • Added Genie library references to QNN dependency lists for both Linux (.so) and Windows (.dll) platforms
  • Updated CMake glob patterns to include Genie libraries when copying QNN dependencies

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File Description
setup.py Added libGenie.so (Linux) and Genie.dll (Windows) to QNN dependency lists for Python package distribution
cmake/CMakeLists.txt Added glob patterns for libGenie*.so and Genie*.dll to include Genie libraries in QNN build artifacts

The changes look good and follow the existing patterns for QNN library dependencies. The implementation is consistent across both platforms and aligns with how other QNN libraries are handled in the codebase. No issues were identified.


💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

@qti-kromero
Copy link
Contributor Author

why not add the additional libraries at the time they are used?

admittedly thought the above was going to be more work than it was so had separated this into its own unit of work - we could defer merging this till other work is ready if that's the preference

@edgchen1
Copy link
Contributor

why not add the additional libraries at the time they are used?

admittedly thought the above was going to be more work than it was so had separated this into its own unit of work - we could defer merging this till other work is ready if that's the preference

I see, I think it would be preferable to add them together with the code that uses them.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ep:QNN issues related to QNN exeution provider

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants