[P/D] force with_prefill true after allreduce in kv producer (#3768)

### What this PR does / why we need it?
force with_prefill true after allreduce in kv producer

- vLLM version: v0.11.0rc3
- vLLM main:
c9461e05a4

---------

Signed-off-by: liziyu <liziyu16@huawei.com>
This commit is contained in:
liziyu
2025-10-29 10:15:38 +08:00
committed by GitHub
parent f57bdb09fc
commit c76db627ab
2 changed files with 5 additions and 5 deletions

View File

@@ -1151,7 +1151,7 @@ class MooncakeLayerwiseConnectorWorker:
connector_metadata: MooncakeLayerwiseConnectorMetadata,
**kwargs) -> None:
"""MooncakeLayerwiseConnector does not save explicitly."""
if self.kv_role == 'kv_producer':
if self.kv_role == 'kv_producer' and connector_metadata.request.keys():
if self.pd_head_ratio != 1:
if self.current_layer != 0:
self.completion_event.wait()