biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Core] Scheduler: Publish connector events after output (#25875)

Browse Source 
Signed-off-by: Or Ozeri <oro@il.ibm.com>
This commit is contained in:
Or Ozeri
2025-10-28 23:01:33 +02:00
committed by GitHub
parent 6afc28a9ba
commit 111faf1118
2 changed files with 135 additions and 43 deletions
Download Patch File Download Diff File Expand all files Collapse all files

34
vllm/v1/core/sched/scheduler.py
View File

@@ -646,23 +646,6 @@ class Scheduler(SchedulerInterface):
meta = self.connector.build_connector_meta(scheduler_output)
scheduler_output.kv_connector_metadata = meta
# collect KV cache events from KV cache manager
events = self.kv_cache_manager.take_events()
# collect KV cache events from connector
if self.connector is not None:
connector_events = self.connector.take_events()
if connector_events:
if events is None:
events = list(connector_events)
else:
events.extend(connector_events)
# publish collected KV cache events
if events:
batch = KVEventBatch(ts=time.time(), events=events)
self.kv_event_publisher.publish(batch)
self._update_after_schedule(scheduler_output)
return scheduler_output
@@ -1057,6 +1040,23 @@ class Scheduler(SchedulerInterface):
if kv_connector_output:
self._update_from_kv_xfer_finished(kv_connector_output)
# collect KV cache events from KV cache manager
events = self.kv_cache_manager.take_events()
# collect KV cache events from connector
if self.connector is not None:
connector_events = self.connector.take_events()
if connector_events:
if events is None:
events = list(connector_events)
else:
events.extend(connector_events)
# publish collected KV cache events
if events:
batch = KVEventBatch(ts=time.time(), events=events)
self.kv_event_publisher.publish(batch)
# Create EngineCoreOutputs for all clients that have requests with
# outputs in this step.
engine_core_outputs = {