OpenCL: add initial FA support #14987
Merged
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
github-actions
bot
added
ggml
lhez
reviewed
Aug 1, 2025
|
@rmatif Very cool, thank you! |
|
Sorry, got distracted during the past week. Will come back to this asap. |
lhez
reviewed
Aug 14, 2025
|
It seems to help small models like qwen2.5-0.5b, qwen2.5-0.5b-Q4_0A750ggml_opencl: OpenCL driver: OpenCL 3.0 QUALCOMM build: commit unknown Compiler E031.45.02.16
A830ggml_opencl: OpenCL driver: OpenCL 3.0 QUALCOMM build: 0800.35 Compiler E031.47.18.28
|
lhez
reviewed
Aug 15, 2025
lhez
approved these changes
Aug 16, 2025
|
Current implementation works well for small models (e.g., qwen2.5-0.5B), significantly improving pp performance. For larger models, larger configs (e.g., We will use this implementation as the baseline and do further investigations and improvements. |
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Outdated suggestions cannot be applied.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 introduces F16/F32 FA support for the OpenCL backend. It has been extremely challenging to achieve good performance on this kind of hardware, but I believe it is now decent enough to serve as a baseline that we can further iterate on. I also believe there is room for improvement for tg
Results on Adreno 830:
Adreno 750: