squall/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

Fix log message in scheduler (#652)

Browse Source
This commit is contained in:
Lily Liu 2023-08-02 13:35:10 -07:00 committed by GitHub
parent 64f23c2900
commit 20044cab7a
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
vllm/core/scheduler.py
View File

@ -190,13 +190,13 @@ class Scheduler:
break break
num_prompt_tokens = seq_group.get_seqs()[0].get_len() num_prompt_tokens = seq_group.get_seqs()[0].get_len()
if num_prompt_tokens > min( prompt_limit = min(
self.scheduler_config.max_model_len, self.scheduler_config.max_model_len,
self.scheduler_config.max_num_batched_tokens): self.scheduler_config.max_num_batched_tokens)
if num_prompt_tokens > prompt_limit:
logger.warning( logger.warning(
f"Input prompt ({num_prompt_tokens} tokens) is too long" f"Input prompt ({num_prompt_tokens} tokens) is too long"
" and exceeds limit of " f" and exceeds limit of {prompt_limit}")
f"{self.scheduler_config.max_model_len}")
for seq in seq_group.get_seqs(): for seq in seq_group.get_seqs():
seq.status = SequenceStatus.FINISHED_IGNORED seq.status = SequenceStatus.FINISHED_IGNORED
ignored_seq_groups.append(seq_group) ignored_seq_groups.append(seq_group)