[DLIGHT][GPU] Enhance opencl thread limit for schedules#16972
Merged
Hzfengsy merged 4 commits intoapache:mainfrom May 21, 2024
Merged
[DLIGHT][GPU] Enhance opencl thread limit for schedules#16972Hzfengsy merged 4 commits intoapache:mainfrom
Hzfengsy merged 4 commits intoapache:mainfrom
Conversation
Enhanced the opencl thread limit and improved the gpu schedules for opencl targets. It improves decode performance 20 % for few set of models.
Contributor
Author
|
@srkreddy1238 @tqchen : Please take a look to this PR and let me know your advise. |
tqchen
approved these changes
May 7, 2024
tqchen
reviewed
May 8, 2024
Member
|
cc @mengshyu would be nice to confirm metal improvement and see if we want it for webgpu/metal |
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Enhanced the opencl thread limit and improved the gpu schedules for opencl targets.
It improves decode performance 20 % for few set of MLC llm models.
LLM model --- baseline --- improved
gemma-2b-it --- 22.4 tok/sec --- 28.2 tok/sec
Qwen-7b-chat --- 11 tok/sec --- 11.8 tok/sec