[Worker] Register mindie_turbo while initializing NPUWorker (#13)

Add `try_register_lib` and import mindie-turbo when init.

---------

Signed-off-by: hw_whx <wanghexiang7@huawei.com>
Co-authored-by: hw_whx <wanghexiang7@huawei.com>
This commit is contained in:
whx
2025-02-07 16:47:17 +08:00
committed by GitHub
parent 4495fc6838
commit 8fc5dc966a
2 changed files with 41 additions and 0 deletions

35
vllm_ascend/utils.py Normal file
View File

@@ -0,0 +1,35 @@
#
# Copyright (c) 2025 Huawei Technologies Co., Ltd. All Rights Reserved.
# This file is a part of the vllm-ascend project.
# Adapted from vllm-project/vllm/vllm/worker/worker.py
# Copyright 2023 The vLLM team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
from vllm.logger import init_logger
logger = init_logger(__name__)
def try_register_lib(lib_name: str, lib_info: str = ""):
import importlib
import importlib.util
try:
module_spec = importlib.util.find_spec(lib_name)
if module_spec is not None:
importlib.import_module(lib_name)
if lib_info:
logger.info(lib_info)
except Exception:
pass

View File

@@ -47,6 +47,7 @@ from vllm.worker.worker_base import (LocalOrDistributedWorkerBase, WorkerBase,
WorkerInput)
from vllm_ascend.model_runner import NPUModelRunner
from vllm_ascend.utils import try_register_lib
logger = init_logger(__name__)
@@ -69,6 +70,11 @@ class NPUWorker(LocalOrDistributedWorkerBase):
) -> None:
WorkerBase.__init__(self, vllm_config=vllm_config)
# Try to import mindie_turbo to accelerate vLLM inference.
try_register_lib(
"mindie_turbo",
"MindIE Turbo is installed. vLLM inference will be accelerated with MindIE Turbo."
)
# distribute related config
self.parallel_config.rank = rank
self.local_rank = local_rank