f10acddb78
Ascend scheduler was added for non chunk prefill case before, since that the npu ops didn't work well with chunked prefill. Now the ops with chunked prefill work better, it's time to remove the ascend scheduler to use vLLM default scheduler. - vLLM version: v0.11.2 --------- Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
203 lines
7.2 KiB
Python
203 lines
7.2 KiB
Python
#
|
|
# Copyright (c) 2025 Huawei Technologies Co., Ltd. All Rights Reserved.
|
|
#
|
|
# 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.
|
|
# This file is a part of the vllm-ascend project.
|
|
#
|
|
"""
|
|
Service profiling configuration generator module.
|
|
|
|
This module generates the service_profiling_symbols.yaml configuration file
|
|
to ~/.config/vllm_ascend/ directory.
|
|
"""
|
|
|
|
import tempfile
|
|
from pathlib import Path
|
|
from typing import Optional
|
|
|
|
import vllm
|
|
from vllm.logger import logger
|
|
|
|
VLLM_VERSION = vllm.__version__
|
|
# Configuration file name
|
|
CONFIG_FILENAME = f"service_profiling_symbols.{VLLM_VERSION}.yaml"
|
|
|
|
# Hard-coded YAML content, default symbols changed by user can be added here.
|
|
SERVICE_PROFILING_SYMBOLS_YAML = """
|
|
# ===== Batch / Scheduler =====
|
|
- symbol: vllm.v1.engine.processor:Processor.process_inputs
|
|
min_version: "0.9.1"
|
|
handler: msserviceprofiler.vllm_profiler.vllm_v1.batch_hookers:process_inputs
|
|
|
|
- symbol: vllm.v1.core.sched.scheduler:Scheduler.schedule
|
|
min_version: "0.9.1"
|
|
handler: msserviceprofiler.vllm_profiler.vllm_v1.batch_hookers:schedule
|
|
name: batchFrameworkProcessing
|
|
|
|
- symbol: vllm.v1.core.sched.scheduler:Scheduler._free_request
|
|
min_version: "0.9.1"
|
|
handler: msserviceprofiler.vllm_profiler.vllm_v1.batch_hookers:free_request
|
|
|
|
- symbol: vllm.v1.core.sched.scheduler:Scheduler.add_request
|
|
min_version: "0.9.1"
|
|
handler: msserviceprofiler.vllm_profiler.vllm_v1.batch_hookers:add_request
|
|
|
|
# ===== KV Cache =====
|
|
- symbol: vllm.v1.core.kv_cache_manager:KVCacheManager.allocate_slots
|
|
min_version: "0.9.1"
|
|
handler: msserviceprofiler.vllm_profiler.vllm_v1.kvcache_hookers:allocate_slots
|
|
|
|
- symbol: vllm.v1.core.kv_cache_manager:KVCacheManager.free
|
|
min_version: "0.9.1"
|
|
handler: msserviceprofiler.vllm_profiler.vllm_v1.kvcache_hookers:free
|
|
|
|
- symbol: vllm.v1.core.kv_cache_manager:KVCacheManager.get_computed_blocks
|
|
min_version: "0.9.1"
|
|
handler: msserviceprofiler.vllm_profiler.vllm_v1.kvcache_hookers:get_computed_blocks
|
|
|
|
# ===== Model Execute =====
|
|
- symbol: vllm.model_executor.layers.logits_processor:LogitsProcessor.forward
|
|
min_version: "0.9.1"
|
|
handler: msserviceprofiler.vllm_profiler.vllm_v1.model_hookers:compute_logits
|
|
name: computing_logits
|
|
|
|
- symbol: vllm.v1.sample.sampler:Sampler.forward
|
|
min_version: "0.9.1"
|
|
handler: msserviceprofiler.vllm_profiler.vllm_v1.model_hookers:sampler_forward
|
|
name: sample
|
|
|
|
- symbol: vllm.v1.executor.abstract:Executor.execute_model
|
|
min_version: "0.9.1"
|
|
handler: msserviceprofiler.vllm_profiler.vllm_v1.model_hookers:execute_model
|
|
name: modelExec
|
|
|
|
- symbol: vllm.v1.executor.multiproc_executor:MultiprocExecutor.execute_model
|
|
min_version: "0.9.1"
|
|
handler: msserviceprofiler.vllm_profiler.vllm_v1.model_hookers:execute_model
|
|
name: modelExec
|
|
|
|
- symbol: vllm_ascend.worker.model_runner_v1:NPUModelRunner.execute_model
|
|
name: modelRunnerExec
|
|
domain: ModelExecute
|
|
|
|
- symbol: vllm_ascend.worker.model_runner_v1:NPUModelRunner._update_states
|
|
name: _update_states
|
|
domain: ModelExecute
|
|
|
|
- symbol: vllm_ascend.worker.model_runner_v1:NPUModelRunner._prepare_inputs
|
|
name: _prepare_inputs
|
|
domain: ModelExecute
|
|
|
|
- symbol: vllm_ascend.utils:ProfileExecuteDuration.capture_async
|
|
min_version: "0.9.1"
|
|
handler: msserviceprofiler.vllm_profiler.vllm_v1.model_hookers:capture_async
|
|
|
|
# ===== Request Lifecycle =====
|
|
- symbol: vllm.v1.engine.async_llm:AsyncLLM.add_request
|
|
min_version: "0.9.1"
|
|
handler: msserviceprofiler.vllm_profiler.vllm_v1.request_hookers:add_request_async
|
|
|
|
- symbol: vllm.engine.async_llm_engine:AsyncLLMEngine.add_request
|
|
min_version: "0.9.1"
|
|
handler: msserviceprofiler.vllm_profiler.vllm_v1.request_hookers:add_request_async
|
|
|
|
- symbol: vllm.v1.engine.output_processor:OutputProcessor.process_outputs
|
|
min_version: "0.9.1"
|
|
handler: msserviceprofiler.vllm_profiler.vllm_v1.request_hookers:process_outputs
|
|
"""
|
|
|
|
|
|
def get_config_dir() -> Path:
|
|
"""
|
|
Get the vllm_ascend configuration directory path.
|
|
|
|
Returns:
|
|
Path: The path to ~/.config/vllm_ascend/ directory.
|
|
"""
|
|
home_dir = Path.home()
|
|
config_dir = home_dir / ".config" / "vllm_ascend"
|
|
return config_dir
|
|
|
|
|
|
def _cleanup_temp_file(tmp_path: Optional[Path]) -> None:
|
|
"""
|
|
Clean up a temporary file if it exists.
|
|
|
|
Args:
|
|
tmp_path: Path to the temporary file to clean up.
|
|
"""
|
|
if tmp_path is not None and tmp_path.exists():
|
|
try:
|
|
tmp_path.unlink()
|
|
except OSError:
|
|
pass # Ignore cleanup errors
|
|
|
|
|
|
def generate_service_profiling_config() -> Optional[Path]:
|
|
"""
|
|
Generate the service_profiling_symbols.yaml configuration file
|
|
to ~/.config/vllm_ascend/ directory.
|
|
|
|
If the configuration file already exists, this function will skip
|
|
creating it and return the existing file path.
|
|
|
|
If any error occurs during file creation, it will be logged but
|
|
will not interrupt the execution. The function will return None
|
|
to indicate that the file could not be created.
|
|
|
|
Returns:
|
|
Optional[Path]: The path to the generated (or existing) configuration file.
|
|
Returns None if file creation failed.
|
|
"""
|
|
config_dir = get_config_dir()
|
|
config_file = config_dir / CONFIG_FILENAME
|
|
|
|
# Check if the configuration file already exists
|
|
if config_file.exists():
|
|
return config_file
|
|
|
|
# Create the configuration directory if it doesn't exist
|
|
try:
|
|
config_dir.mkdir(parents=True, exist_ok=True)
|
|
except (OSError, PermissionError) as e:
|
|
logger.error(
|
|
f"Failed to create configuration directory {config_dir}: {e}",
|
|
exc_info=True)
|
|
return None
|
|
|
|
# Write the configuration file atomically using a temporary file
|
|
# This ensures the file is only written if the write succeeds completely
|
|
tmp_path = None
|
|
try:
|
|
# Create a temporary file in the same directory for atomic write
|
|
with tempfile.NamedTemporaryFile(mode='w',
|
|
encoding='utf-8',
|
|
dir=config_dir,
|
|
delete=False,
|
|
suffix='.tmp',
|
|
prefix=CONFIG_FILENAME +
|
|
'.') as tmp_file:
|
|
tmp_file.write(SERVICE_PROFILING_SYMBOLS_YAML)
|
|
tmp_path = Path(tmp_file.name)
|
|
|
|
# Atomically replace the target file with the temporary file
|
|
tmp_path.replace(config_file)
|
|
return config_file
|
|
except (OSError, PermissionError) as e:
|
|
logger.error(f"Failed to write configuration file {config_file}: {e}",
|
|
exc_info=True)
|
|
return None
|
|
finally:
|
|
# Clean up the temporary file if it wasn't successfully replaced
|
|
_cleanup_temp_file(tmp_path)
|