Skip to content

[perf] fix perf regression from #12253 #12380

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
merged 2 commits into from
Jan 24, 2025
Merged

[perf] fix perf regression from #12253 #12380

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b900f08
fix perf
youkaichao Jan 24, 2025
9ace57c
add comments
youkaichao Jan 24, 2025
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
5 changes: 4 additions & 1 deletion vllm/worker/model_runner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -455,7 +455,6 @@ def __init__(self,
self.enable_prompt_adapter = (self.runner.prompt_adapter_config
is not None)
self.multi_modal_input_mapper = self.runner.multi_modal_input_mapper
self.decode_only = True

# Attention metadata inputs.
if self.attn_backend is not None:
Expand All @@ -477,6 +476,10 @@ def prepare(self,
finished_requests_ids: Optional[List[str]] = None) -> None:
self.finished_requests_ids = finished_requests_ids

# if the current batch is decode-only.
# will be set to False if there is any non-decode request.
self.decode_only = True

# Intermediate data (data in CPU before going to GPU) for
# the current sequence group.
self.inter_data_list: List[
Expand Down
Loading