[MISC] fix logger (#515)

logger in vllm-ascend doesn't work. This PR fix the issue.

Fix: https://github.com/vllm-project/vllm-ascend/issues/431

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>

tests/conftest.py

@@ -29,7 +29,6 @@ from vllm.config import TaskOption
 from vllm.distributed.parallel_state import (destroy_distributed_environment,
                                              destroy_model_parallel)
 from vllm.inputs import ExplicitEncoderDecoderPrompt, TextPrompt, TokensPrompt
-from vllm.logger import init_logger
 from vllm.outputs import RequestOutput
 from vllm.sampling_params import BeamSearchParams
 from vllm.utils import is_list_of
@@ -37,8 +36,6 @@ from vllm.utils import is_list_of
 from tests.model_utils import (TokensTextLogprobs,
                                TokensTextLogprobsPromptLogprobs)
 
-logger = init_logger(__name__)
-
 _M = TypeVar("_M")
 
 _PromptMultiModalInput = Union[List[_M], List[List[_M]]]

vllm_ascend/platform.py

@@ -22,7 +22,7 @@ from typing import TYPE_CHECKING, Optional, Tuple
 import torch
 import torch_npu  # noqa: F401
 import vllm.envs as envs
-from vllm.logger import init_logger
+from vllm.logger import logger
 
 try:
     # register custom ops into torch_library here
@@ -48,8 +48,6 @@ else:
 
 os.environ["RAY_EXPERIMENTAL_NOSET_ASCEND_RT_VISIBLE_DEVICES"] = "1"
 
-logger = init_logger(__name__)
-
 
 class NPUPlatform(Platform):
 

vllm_ascend/quantization/quant_config.py

@@ -21,7 +21,6 @@ from typing import Any, Callable, Dict, List, Mapping, Optional
 import torch
 import torch_npu  # noqa: F401
 from vllm.distributed import get_tensor_model_parallel_rank
-from vllm.logger import init_logger
 from vllm.model_executor.layers.fused_moe import (FusedMoE, FusedMoEMethodBase,
                                                   FusedMoeWeightScaleSupported)
 from vllm.model_executor.layers.fused_moe.layer import \
@@ -41,8 +40,6 @@ from vllm.model_executor.utils import set_weight_attrs
 
 from .quantizer import AscendQuantizer
 
-logger = init_logger(__name__)
-
 
 @register_quantization_config("ascend")
 class AscendQuantConfig(QuantizationConfig):

vllm_ascend/utils.py

@@ -17,9 +17,7 @@
 # limitations under the License.
 #
 import torch
-from vllm.logger import init_logger
+from vllm.logger import logger
-
-logger = init_logger(__name__)
 
 
 def try_register_lib(lib_name: str, lib_info: str = ""):

vllm_ascend/worker/model_runner.py

@@ -35,7 +35,7 @@ from vllm.core.scheduler import SchedulerOutputs
 from vllm.distributed import get_kv_transfer_group, get_pp_group
 from vllm.forward_context import set_forward_context
 from vllm.inputs import INPUT_REGISTRY, InputRegistry
-from vllm.logger import init_logger
+from vllm.logger import logger
 from vllm.lora.layers import LoRAMapping
 from vllm.lora.request import LoRARequest
 from vllm.model_executor import SamplingMetadata, SamplingMetadataCache
@@ -63,8 +63,6 @@ from vllm.worker.model_runner_base import (
 if TYPE_CHECKING:
     from vllm.attention.backends.abstract import AttentionBackend
 
-logger = init_logger(__name__)
-
 TModelInputForNPU = TypeVar('TModelInputForNPU', bound="ModelInputForNPU")
 
 

vllm_ascend/worker/model_runner_v1.py

@@ -32,7 +32,7 @@ from vllm.config import VllmConfig
 from vllm.distributed.parallel_state import get_pp_group
 from vllm.forward_context import set_forward_context
 from vllm.inputs import INPUT_REGISTRY
-from vllm.logger import init_logger
+from vllm.logger import logger
 from vllm.model_executor.layers.fused_moe import FusedMoE
 from vllm.model_executor.model_loader import get_model
 from vllm.multimodal import MULTIMODAL_REGISTRY, MultiModalKwargs
@@ -56,8 +56,6 @@ if TYPE_CHECKING:
 
 NPU_PAGED_ATTENTION_MASK_VALUE = -10000
 
-logger = init_logger(__name__)
-
 
 class NPUModelRunner:
 

vllm_ascend/worker/multi_step_runner.py

@@ -7,7 +7,6 @@ from typing import (TYPE_CHECKING, Any, Callable, Dict, List, Optional, Tuple,
 import torch
 from torch import nn
 from vllm.distributed import get_pp_group
-from vllm.logger import init_logger
 from vllm.model_executor.layers.sampler import (PromptLogprobs, SampleLogprobs,
                                                 SamplerOutput,
                                                 SamplingMetadata, get_logprobs,
@@ -30,8 +29,6 @@ from vllm_ascend.worker.model_runner import (
 if TYPE_CHECKING:
     from vllm.attention.backends.abstract import AttentionBackend
 
-logger = init_logger(__name__)
-
 
 @dataclass(frozen=False)
 class StatefulModelInputForNPU(StatefulModelInput):

vllm_ascend/worker/worker.py

@@ -28,7 +28,7 @@ from vllm.config import ParallelConfig, VllmConfig
 from vllm.distributed import (ensure_model_parallel_initialized,
                               init_distributed_environment,
                               set_custom_all_reduce)
-from vllm.logger import init_logger
+from vllm.logger import logger
 from vllm.lora.request import LoRARequest
 from vllm.model_executor import set_random_seed
 from vllm.model_executor.layers.sampler import SamplerOutput
@@ -48,8 +48,6 @@ from vllm_ascend.utils import try_register_lib
 from vllm_ascend.worker.model_runner import NPUModelRunner
 from vllm_ascend.worker.pooling_model_runner import NPUPoolingModelRunner
 
-logger = init_logger(__name__)
-
 
 class NPUWorker(LocalOrDistributedWorkerBase):
     """A worker class that executes (a partition of) the model on a NPU.

vllm_ascend/worker/worker_v1.py

@@ -29,7 +29,7 @@ from vllm.config import ParallelConfig, VllmConfig
 from vllm.distributed import (ensure_model_parallel_initialized,
                               init_distributed_environment,
                               set_custom_all_reduce)
-from vllm.logger import init_logger
+from vllm.logger import logger
 from vllm.model_executor import set_random_seed
 from vllm.platforms import current_platform
 from vllm.utils import STR_DTYPE_TO_TORCH_DTYPE
@@ -42,8 +42,6 @@ from vllm.v1.worker.worker_base import WorkerBase
 
 from vllm_ascend.worker.model_runner_v1 import NPUModelRunner
 
-logger = init_logger(__name__)
-
 
 class NPUWorker(WorkerBase):