xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-01-07 21:48:44 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Misc] Fix wrong comment in scheduler (#28880)

Browse Source 
Signed-off-by: Zhuohan Li <zhuohan123@gmail.com>
This commit is contained in:
Zhuohan Li 2025-11-17 15:32:22 -08:00 committed by GitHub
parent 61485844fc
commit 552cac95b5
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 4 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
vllm/v1/core/sched/scheduler.py
View File

@ -472,9 +472,9 @@ class Scheduler(SchedulerInterface):
num_computed_tokens = (
num_new_local_computed_tokens + num_external_computed_tokens
)
# KVTransfer: WAITING reqs have num_computed_tokens > 0
# after async KV recvs are completed.
else:
# KVTransfer: WAITING reqs have num_computed_tokens > 0
# after async KV recvs are completed.
new_computed_blocks = self.kv_cache_manager.empty_kv_cache_blocks
num_new_local_computed_tokens = 0
num_computed_tokens = request.num_computed_tokens
@ -483,12 +483,12 @@ class Scheduler(SchedulerInterface):
external_load_encoder_input = []
new_encoder_compute_budget = encoder_compute_budget
# KVTransfer: loading remote KV, do not allocate for new work.
if load_kv_async:
# KVTransfer: loading remote KV, do not allocate for new work.
assert num_external_computed_tokens > 0
num_new_tokens = 0
# Number of tokens to be scheduled.
else:
# Number of tokens to be scheduled.
# We use `request.num_tokens` instead of
# `request.num_prompt_tokens` to consider the resumed
# requests, which have output tokens.