Skip to content

scheduler: add prefill-only update in merge batch #21840

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

Merged
hnyls2002 merged 1 commit into from
Apr 2, 2026
Merged

scheduler: add prefill-only update in merge batch #21840

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions python/sglang/srt/managers/schedule_batch.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2266,6 +2266,7 @@ def merge_batch(self, other: "ScheduleBatch"):
self.has_stream |= other.has_stream
self.has_grammar |= other.has_grammar
self.return_hidden_states |= other.return_hidden_states
self.is_prefill_only = self.is_prefill_only and other.is_prefill_only
Copy link
Copy Markdown
Contributor

@gemini-code-assist gemini-code-assist bot Apr 1, 2026

Choose a reason for hiding this comment

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

medium

Consider using the &= operator for consistency with the other boolean flag updates in this method (lines 2265-2268).

Additionally, the is_prefill_only flag (and potentially return_hidden_states) should be re-evaluated in the filter_batch method (around line 2206). Currently, filter_batch re-evaluates flags like has_stream and has_grammar but omits is_prefill_only, which could lead to stale state if generation requests are filtered out of a mixed batch, causing the scheduler to incorrectly attempt decode steps for prefill-only requests.

Suggested change
self.is_prefill_only = self.is_prefill_only and other.is_prefill_only
self.is_prefill_only &= other.is_prefill_only


if self.spec_info:
self.spec_info.merge_batch(other.spec_info)
Expand Down
Loading