xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-04-06 14:47:05 +08:00
Code Issues Packages Projects Releases Wiki Activity

minor

Browse Source 
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
This commit is contained in:
Woosuk Kwon 2025-03-12 23:44:46 -07:00
parent 1e7bf7970a
commit a7facf98d9
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/simple_scheduler.py
View File

@ -103,7 +103,7 @@ class SimpleScheduler(SchedulerInterface):
# For logging.
scheduled_timestamp = time.monotonic()
# First, schedule the WAITING requests.
# First, schedule prefill requests.
while self.waiting and token_budget > 0:
if len(self.running) == self.max_num_running_reqs:
break
@ -165,9 +165,9 @@ class SimpleScheduler(SchedulerInterface):
# NOTE(woosuk): Check that the request is "fully" scheduled.
assert num_computed_tokens + num_new_tokens == request.num_tokens
# If no prefill was scheduled, schedule the RUNNING requests.
num_prefill_reqs = len(scheduled_new_reqs) + len(
scheduled_resumed_reqs)
# If no prefill was scheduled, schedule decode requests.
num_prefill_reqs = (len(scheduled_new_reqs) +
len(scheduled_resumed_reqs))
if num_prefill_reqs == 0:
req_index = 0
while req_index < len(self.running) and token_budget > 0: