Unverified Commit 3b457143 authored by Simon Mo's avatar Simon Mo Committed by GitHub
Browse files

[Bugfix] Register serializers for V0 MQ Engine (#15009)


Signed-off-by: default avatarsimon-mo <simon.mo@hey.com>
parent ab656f2c
...@@ -29,6 +29,8 @@ from vllm.engine.multiprocessing import (ENGINE_DEAD_ERROR, IPC_DATA_EXT, ...@@ -29,6 +29,8 @@ from vllm.engine.multiprocessing import (ENGINE_DEAD_ERROR, IPC_DATA_EXT,
# yapf: enable # yapf: enable
from vllm.logger import init_logger from vllm.logger import init_logger
from vllm.outputs import RequestOutput from vllm.outputs import RequestOutput
from vllm.transformers_utils.config import (
maybe_register_config_serialize_by_value)
from vllm.usage.usage_lib import UsageContext from vllm.usage.usage_lib import UsageContext
from vllm.worker.model_runner_base import InputProcessingError from vllm.worker.model_runner_base import InputProcessingError
...@@ -428,6 +430,9 @@ def run_mp_engine(vllm_config: VllmConfig, usage_context: UsageContext, ...@@ -428,6 +430,9 @@ def run_mp_engine(vllm_config: VllmConfig, usage_context: UsageContext,
ipc_path: str, disable_log_stats: bool, ipc_path: str, disable_log_stats: bool,
disable_log_requests: bool, engine_alive): disable_log_requests: bool, engine_alive):
try: try:
# Ensure we can serialize transformer config before spawning
maybe_register_config_serialize_by_value()
engine = MQLLMEngine.from_vllm_config( engine = MQLLMEngine.from_vllm_config(
vllm_config=vllm_config, vllm_config=vllm_config,
usage_context=usage_context, usage_context=usage_context,
......
...@@ -82,6 +82,8 @@ from vllm.entrypoints.openai.serving_transcription import ( ...@@ -82,6 +82,8 @@ from vllm.entrypoints.openai.serving_transcription import (
from vllm.entrypoints.openai.tool_parsers import ToolParserManager from vllm.entrypoints.openai.tool_parsers import ToolParserManager
from vllm.entrypoints.utils import load_aware_call, with_cancellation from vllm.entrypoints.utils import load_aware_call, with_cancellation
from vllm.logger import init_logger from vllm.logger import init_logger
from vllm.transformers_utils.config import (
maybe_register_config_serialize_by_value)
from vllm.usage.usage_lib import UsageContext from vllm.usage.usage_lib import UsageContext
from vllm.utils import (FlexibleArgumentParser, get_open_zmq_ipc_path, from vllm.utils import (FlexibleArgumentParser, get_open_zmq_ipc_path,
is_valid_ipv6_address, set_ulimit) is_valid_ipv6_address, set_ulimit)
...@@ -221,6 +223,9 @@ async def build_async_engine_client_from_engine_args( ...@@ -221,6 +223,9 @@ async def build_async_engine_client_from_engine_args(
# so we need to spawn a new process # so we need to spawn a new process
context = multiprocessing.get_context("spawn") context = multiprocessing.get_context("spawn")
# Ensure we can serialize transformer config before spawning
maybe_register_config_serialize_by_value()
# The Process can raise an exception during startup, which may # The Process can raise an exception during startup, which may
# not actually result in an exitcode being reported. As a result # not actually result in an exitcode being reported. As a result
# we use a shared variable to communicate the information. # we use a shared variable to communicate the information.
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment