xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-04-01 15:37:05 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'fix-connector-agg' into debug-logging

Browse Source
This commit is contained in:
Robert Shaw 2025-07-23 18:20:39 +00:00
commit 10abfaf309
1 changed files with 10 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
vllm/v1/worker/gpu_worker.py
View File

@ -15,7 +15,8 @@ from vllm.config import VllmConfig
from vllm.distributed import (ensure_model_parallel_initialized,
init_distributed_environment,
set_custom_all_reduce)
from vllm.distributed.kv_transfer import ensure_kv_transfer_initialized
from vllm.distributed.kv_transfer import (ensure_kv_transfer_initialized,
has_kv_transfer_group)
from vllm.distributed.parallel_state import get_pp_group, get_tp_group
from vllm.logger import init_logger
from vllm.lora.request import LoRARequest
@ -333,19 +334,20 @@ class Worker(WorkerBase):
assert isinstance(output, IntermediateTensors)
get_pp_group().send_tensor_dict(output.tensors,
all_gather_group=get_tp_group())
if not has_kv_transfer_group():
return None
# In case of PP with kv transfer, we need to pass through the
# finished_sending and finished_recving buffers.
empty_output = EMPTY_MODEL_RUNNER_OUTPUT
new_output = EMPTY_MODEL_RUNNER_OUTPUT
if output.finished_sending or output.finished_recving:
empty_output = copy.copy(empty_output)
empty_output.finished_sending = output.finished_sending
empty_output.finished_recving = output.finished_recving
output = empty_output
new_output = copy.copy(new_output)
new_output.finished_sending = output.finished_sending
new_output.finished_recving = output.finished_recving
output = new_output
assert isinstance(output, ModelRunnerOutput)
# return output only from the driver worker
return output if self.is_driver_worker else None
return output
def profile(self, is_start: bool = True):
if self.profiler is None: