Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

SRT refactor #844

Closed
wants to merge 35 commits into from
Closed

SRT refactor #844

wants to merge 35 commits into from

Conversation

hnyls2002
Copy link
Collaborator

@hnyls2002 hnyls2002 commented Jul 31, 2024 •
edited
Loading

This PR is to clarify the logic of the SRT code.

  • Move InputMetaData and ForwardMode into ForwardInfo
  • Add PrefillAdder
  • Organize prepare_for_extend
  • Move some preparation into InputMetadata

This PR has been split into #960, #967, #968, #977, #980, #984, #981

@hnyls2002 hnyls2002 marked this pull request as draft July 31, 2024 07:46
min-xu-et
min-xu-et reviewed Aug 1, 2024
View reviewed changes
Copy link
Contributor

@min-xu-et min-xu-et left a comment

Choose a reason for hiding this comment

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

I am new to the code base. This seems like a code movement change? No logical change other than the removal of "sampler" in qwen model code?

python/sglang/srt/models/qwen2_moe.py Outdated Show resolved Hide resolved
@hnyls2002 hnyls2002 marked this pull request as ready for review August 5, 2024 02:26
@hnyls2002 hnyls2002 changed the title Split schedule batch and forward batch info SRT refactor Aug 5, 2024
@hnyls2002 hnyls2002 force-pushed the shedule-batch branch 2 times, most recently from b48fe4c to dc50933 Compare August 7, 2024 05:04
@hnyls2002 hnyls2002 requested review from Ying1123, merrymercy and zhyncs and removed request for Ying1123, merrymercy and zhyncs August 7, 2024 09:05
@zhyncs
Copy link
Member

zhyncs commented Aug 7, 2024

I'll review asap 😂

@Ying1123 Ying1123 closed this Aug 8, 2024
@hnyls2002 hnyls2002 reopened this Aug 8, 2024
@hnyls2002 hnyls2002 closed this Aug 8, 2024
@zhyncs zhyncs deleted the shedule-batch branch August 13, 2024 11:45
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@min-xu-et min-xu-et min-xu-et left review comments

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@hnyls2002 @zhyncs @min-xu-et @Ying1123