Skip to content

Fix crash on prefill-after-preemption: conditionally serialize token_ids #103

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

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
+10 −6
Open

Fix crash on prefill-after-preemption: conditionally serialize token_ids #103

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
16 changes: 10 additions & 6 deletions nanovllm/engine/sequence.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -72,12 +72,16 @@ def append_token(self, token_id: int):
self.num_tokens += 1

def __getstate__(self):
return (self.num_tokens, self.num_prompt_tokens, self.num_cached_tokens, self.block_table,
self.token_ids if self.num_completion_tokens == 0 else self.last_token)
needs_prefill = self.num_cached_tokens < self.num_tokens
payload = self.token_ids if needs_prefill else self.last_token
return (self.num_tokens, self.num_prompt_tokens, self.num_cached_tokens,
self.block_table, payload)

def __setstate__(self, state):
self.num_tokens, self.num_prompt_tokens, self.num_cached_tokens, self.block_table = state[:-1]
if self.num_completion_tokens == 0:
self.token_ids = state[-1]
(self.num_tokens, self.num_prompt_tokens, self.num_cached_tokens, self.block_table, payload) = state

if isinstance(payload, list):
self.token_ids = payload
self.last_token = payload[-1] if payload else -1
else:
self.last_token = state[-1]
self.last_token = payload