Skip to content

[CB] fix scheduler constraint max tokens #332

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

Draft
wants to merge 1 commit into
base: main
Choose a base branch
from
Draft

[CB] fix scheduler constraint max tokens #332

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
5 changes: 3 additions & 2 deletions vllm_spyre/v1/core/scheduler.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -224,8 +224,9 @@ def can_schedule(self, request) -> bool:
cond2 = len(self.waiting) < max_prompt_batch_size
# check that the prompt length does not exceed the current tkv
cond3 = request.num_prompt_tokens <= self.tkv
# check that the number of requested tokens can be served
cond4 = request.max_tokens <= (max_context_len - self.tkv)
# check that the number of requested tokens can be served (-1 for free
# prefill token)
cond4 = request.max_tokens - 1 <= (max_context_len - self.tkv)
# check that there are enough free blocks/pages remaining
# Note: we only have to do check in case of a running batches
# (not start_new_batch), because the minimal number of blocks covers
Expand Down
Loading