Move KVEventsConfig from config/__init__.py to config/kv_events.py (vllm-project#24433)

Signed-off-by: Harry Mellor <[email protected]>

Author: hmellor

tests/distributed/conftest.py

@@ -8,7 +8,7 @@
 import pytest
 import zmq
 
-from vllm.config import KVEventsConfig
+from vllm.config.kv_events import KVEventsConfig
 from vllm.distributed.kv_events import EventPublisherFactory
 
 from .test_events import SampleBatch

vllm/config/__init__.py

@@ -33,6 +33,7 @@
                                PrefixCachingHashAlgo)
 from vllm.config.compilation import (CompilationConfig, CompilationLevel,
                                      CUDAGraphMode, PassConfig)
+from vllm.config.kv_events import KVEventsConfig
 from vllm.config.parallel import (DistributedExecutorBackend, EPLBConfig,
                                   ParallelConfig)
 from vllm.config.scheduler import SchedulerConfig, SchedulerPolicy
@@ -3310,48 +3311,6 @@ def get_from_extra_config(self, key, default) -> Any:
         return self.kv_connector_extra_config.get(key, default)
 
 
-@config
-@dataclass
-class KVEventsConfig:
-    """Configuration for KV event publishing."""
-
-    enable_kv_cache_events: bool = False
-    """If True, enable KV cache events for tracking block storage and removal.
-    Events can be published externally by zmq using the event publisher config.
-    """
-
-    publisher: str = "null"
-    """The publisher to use for publishing kv events. Can be "null", "zmq".
-    """
-
-    endpoint: str = "tcp://*:5557"
-    """The zmq endpoint to use for publishing kv events.
-    """
-
-    replay_endpoint: Optional[str] = None
-    """The zmq endpoint to use for replaying kv events.
-    """
-
-    buffer_steps: int = 10_000
-    """The number of steps to cache for replay endpoint. Will only save
-    events from the last N steps for the replay endpoint.
-    """
-
-    hwm: int = 100_000
-    """The zmq high water mark for the event publisher. After queueing N events,
-    events will start dropping if the consumer is not keeping up.
-    """
-
-    max_queue_size: int = 100_000
-    """The maximum number of events to queue while waiting for publishing.
-    """
-
-    topic: str = ""
-    """The topic to use for the event publisher. Consumers can subscribe to
-    this topic to receive events.
-    """
-
-
 @config
 @dataclass(config=ConfigDict(arbitrary_types_allowed=True))
 class VllmConfig:

vllm/config/kv_events.py

@@ -0,0 +1,50 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
+
+from typing import Optional
+
+from pydantic.dataclasses import dataclass
+
+from vllm.config.utils import config
+
+
+@config
+@dataclass
+class KVEventsConfig:
+    """Configuration for KV event publishing."""
+
+    enable_kv_cache_events: bool = False
+    """If True, enable KV cache events for tracking block storage and removal.
+    Events can be published externally by zmq using the event publisher config.
+    """
+
+    publisher: str = "null"
+    """The publisher to use for publishing kv events. Can be "null", "zmq".
+    """
+
+    endpoint: str = "tcp://*:5557"
+    """The zmq endpoint to use for publishing kv events.
+    """
+
+    replay_endpoint: Optional[str] = None
+    """The zmq endpoint to use for replaying kv events.
+    """
+
+    buffer_steps: int = 10_000
+    """The number of steps to cache for replay endpoint. Will only save
+    events from the last N steps for the replay endpoint.
+    """
+
+    hwm: int = 100_000
+    """The zmq high water mark for the event publisher. After queueing N events,
+    events will start dropping if the consumer is not keeping up.
+    """
+
+    max_queue_size: int = 100_000
+    """The maximum number of events to queue while waiting for publishing.
+    """
+
+    topic: str = ""
+    """The topic to use for the event publisher. Consumers can subscribe to
+    this topic to receive events.
+    """

vllm/distributed/kv_events.py

@@ -14,7 +14,7 @@
 import msgspec
 import zmq
 
-from vllm.config import KVEventsConfig
+from vllm.config.kv_events import KVEventsConfig
 from vllm.logger import init_logger
 
 logger = init_logger(__name__)