[PerfFix] Avoid separate thread for MP executor shm spin

njhill · njhill · commit 3c65afeb0e60 · 2025-11-03T16:25:43.000-08:00
Signed-off-by: Nick Hill &lt;nhill@redhat.com&gt;
diff --git a/tests/v1/executor/test_executor.py b/tests/v1/executor/test_executor.py
@@ -4,6 +4,7 @@
 import asyncio
 import os
 from collections.abc import Callable
+from concurrent.futures import Future
 from typing import Any
 
 import pytest
@@ -27,7 +28,7 @@ def collective_rpc(
         kwargs: dict | None = None,
         non_block: bool = False,
         unique_reply_rank: int | None = None,
-    ) -> list[Any]:
+    ) -> Any | list[Any] | Future[Any | list[Any]]:
         # Drop marker to show that this was run
         with open(".marker", "w"):
             ...
diff --git a/vllm/distributed/kv_transfer/kv_connector/utils.py b/vllm/distributed/kv_transfer/kv_connector/utils.py
@@ -221,39 +221,24 @@ def update_finished_set(
 
     def async_aggregate(
         self,
-        output_futures: Sequence[Future[ModelRunnerOutput | None]],
+        output_future: Future[Sequence[ModelRunnerOutput | None]],
         output_rank: int = 0,
     ) -> Future[ModelRunnerOutput | None]:
-        """Takes a list of futures and returns a single future which resolves
-        to the respective list of outputs."""
+        """Takes a future that resolves to a list of outputs and returns a future
+        which resolves to a single aggregated output."""
         result_future: Future[ModelRunnerOutput | None] = Future()
 
-        outputs: list[ModelRunnerOutput | None] = [None] * len(output_futures)
-        remaining = len(output_futures)
-
-        def make_callback(idx):
-            def callback(fut):
-                if result_future.done():
-                    return
-
-                try:
-                    outputs[idx] = fut.result()
-                except CancelledError:
-                    result_future.cancel()
-                except Exception as e:
-                    result_future.set_exception(e)
-
-                # this check assumes io_thread_pool uses a single thread
-                nonlocal remaining
-                remaining -= 1
-                if not remaining:
-                    result_future.set_result(self.aggregate(outputs, output_rank))
-
-            return callback
-
-        for i, output_future in enumerate(output_futures):
-            output_future.add_done_callback(make_callback(i))
-
+        def callback(fut):
+            if result_future.done():
+                return
+            try:
+                result_future.set_result(self.aggregate(fut.result(), output_rank))
+            except CancelledError:
+                result_future.cancel()
+            except Exception as e:
+                result_future.set_exception(e)
+
+        output_future.add_done_callback(callback)
         return result_future
 
 
diff --git a/vllm/v1/executor/abstract.py b/vllm/v1/executor/abstract.py
@@ -171,7 +171,7 @@ def collective_rpc(
         args: tuple = (),
         kwargs: dict | None = None,
         non_block: Literal[True] = True,
-    ) -> list[Future[_R]]:
+    ) -> Future[list[_R]]:
         pass
 
     @abstractmethod
@@ -219,7 +219,7 @@ def sample_tokens(
 
     def sample_tokens(
         self, grammar_output: GrammarOutput | None, non_block: bool = False
-    ) -> ModelRunnerOutput | Future[ModelRunnerOutput]:
+    ) -> ModelRunnerOutput | None | Future[ModelRunnerOutput | None]:
         output = self.collective_rpc(  # type: ignore[call-overload]
             "sample_tokens", args=(grammar_output,), non_block=non_block
         )
diff --git a/vllm/v1/executor/multiproc_executor.py b/vllm/v1/executor/multiproc_executor.py
@@ -9,8 +9,9 @@
 import time
 import traceback
 import weakref
+from collections import deque
 from collections.abc import Callable
-from concurrent.futures import Future, ThreadPoolExecutor
+from concurrent.futures import Future
 from dataclasses import dataclass
 from enum import Enum, auto
 from functools import cached_property, partial
@@ -54,6 +55,28 @@
 logger = init_logger(__name__)
 
 
+class FutureWrapper(Future):
+    def __init__(self, futures_queue: deque[tuple["FutureWrapper", Callable]]):
+        self.futures_queue = futures_queue
+        super().__init__()
+
+    def result(self, timeout=None):
+        if timeout is not None:
+            raise RuntimeError("timeout not implemented")
+        # Drain any futures ahead of us in the queue.
+        while not self.done():
+            future, get_response = self.futures_queue.pop()
+            future.update_with_response(get_response)
+        return super().result()
+
+    def update_with_response(self, get_response: Callable):
+        try:
+            response = get_response()
+            self.set_result(response)
+        except Exception as e:
+            self.set_exception(e)
+
+
 class MultiprocExecutor(Executor):
     supports_pp: bool = True
 
@@ -64,7 +87,6 @@ def _init_executor(self) -> None:
         self.is_failed = False
         self.shutdown_event = threading.Event()
         self.failure_callback: FailureCallback | None = None
-        self.io_thread_pool: ThreadPoolExecutor | None = None
 
         self.world_size = self.parallel_config.world_size
         tensor_parallel_size = self.parallel_config.tensor_parallel_size
@@ -132,12 +154,7 @@ def _init_executor(self) -> None:
                         uw.death_writer.close()
                 self._ensure_worker_termination([uw.proc for uw in unready_workers])
 
-        # Note: must use only 1 IO thread to keep dequeue sequence
-        # from the response queue.
-        # _async_aggregate_workers_output also assumes a single IO thread.
-        self.io_thread_pool = ThreadPoolExecutor(
-            max_workers=1, thread_name_prefix="mp_exec_io"
-        )
+        self.futures_queue = deque[tuple[FutureWrapper, Callable]]()
 
         self.output_rank = self._get_output_rank()
         self.has_connector = self.vllm_config.kv_transfer_config is not None
@@ -195,14 +212,13 @@ def _execute_with_aggregation(
     ) -> ModelRunnerOutput | None | Future[ModelRunnerOutput | None]:
         if not self.has_connector:
             # get output only from a single worker (output_rank)
-            (output,) = self.collective_rpc(
+            return self.collective_rpc(
                 method,
                 args=args,
                 unique_reply_rank=self.output_rank,
                 non_block=non_block,
                 timeout=envs.VLLM_EXECUTE_MODEL_TIMEOUT_SECONDS,
             )
-            return output
 
         # get output from all workers
         outputs = self.collective_rpc(
@@ -223,20 +239,21 @@ def execute_dummy_batch(self) -> None:
 
     def take_draft_token_ids(self) -> DraftTokenIds | None:
         # OPTIMIZATION: Get output only from a single worker (output_rank)
-        outputs = self.collective_rpc(
+        return self.collective_rpc(
             "take_draft_token_ids", unique_reply_rank=self.output_rank
         )
-        return outputs[0]
 
-    def collective_rpc(
+    def collective_rpc(  # type: ignore[override]
         self,
         method: str | Callable,
         timeout: float | None = None,
         args: tuple = (),
         kwargs: dict | None = None,
         non_block: bool = False,
         unique_reply_rank: int | None = None,
-    ) -> list[Any]:
+    ) -> Any | list[Any] | Future[Any | list[Any]]:
+        """Returns single result if unique_reply_rank is provided, otherwise list."""
+
         if self.is_failed:
             raise RuntimeError("Executor failed.")
 
@@ -246,63 +263,52 @@ def collective_rpc(
         # NOTE: If the args are heterogeneous, then we pack them into a list,
         # and unpack them in the method of every worker, because every worker
         # knows their own rank.
-        try:
-            if isinstance(method, str):
-                send_method = method
-            else:
-                send_method = cloudpickle.dumps(
-                    method, protocol=pickle.HIGHEST_PROTOCOL
-                )
-            self.rpc_broadcast_mq.enqueue(
-                (send_method, args, kwargs, unique_reply_rank)
-            )
 
-            workers = (
-                (self.workers[unique_reply_rank],)
-                if unique_reply_rank is not None
-                else self.workers
-            )
-            responses = []
+        if isinstance(method, str):
+            send_method = method
+        else:
+            send_method = cloudpickle.dumps(method, protocol=pickle.HIGHEST_PROTOCOL)
+        self.rpc_broadcast_mq.enqueue((send_method, args, kwargs, unique_reply_rank))
 
-            def get_response(
-                w: WorkerProcHandle,
-                dequeue_timeout: float | None = None,
-                cancel_event: threading.Event | None = None,
-            ):
-                status, result = w.worker_response_mq.dequeue(
-                    timeout=dequeue_timeout, cancel=cancel_event
-                )
+        workers = (
+            (self.workers[unique_reply_rank],)
+            if unique_reply_rank is not None
+            else self.workers
+        )
 
-                if status != WorkerProc.ResponseStatus.SUCCESS:
-                    raise RuntimeError(
-                        f"Worker failed with error '{result}', please check the"
-                        " stack trace above for the root cause"
-                    )
-                return result
+        shutdown_event = self.shutdown_event
 
+        def get_response():
+            responses = []
             for w in workers:
                 dequeue_timeout = (
                     None if deadline is None else (deadline - time.monotonic())
                 )
-
-                if self.io_thread_pool is not None:
-                    # We must consume worker_response_mq from a single thread.
-                    result = self.io_thread_pool.submit(  # type: ignore
-                        get_response, w, dequeue_timeout, self.shutdown_event
+                try:
+                    status, result = w.worker_response_mq.dequeue(
+                        timeout=dequeue_timeout, cancel=shutdown_event
                     )
-                    if not non_block:
-                        result = result.result()
-                elif not non_block:
-                    result = get_response(w, dequeue_timeout, self.shutdown_event)
-                else:
+                except TimeoutError as e:
+                    raise TimeoutError(f"RPC call to {method} timed out.") from e
+                if status != WorkerProc.ResponseStatus.SUCCESS:
                     raise RuntimeError(
-                        "non_block can only be used when max_concurrent_batches > 1"
+                        f"Worker failed with error '{result}', please check the"
+                        " stack trace above for the root cause"
                     )
                 responses.append(result)
+            return responses[0] if unique_reply_rank is not None else responses
+
+        if non_block:
+            future = FutureWrapper(self.futures_queue)
+            self.futures_queue.appendleft((future, get_response))
+            return future
+
+        # First drain any pending futures in the queue.
+        while self.futures_queue:
+            future, get_fut_response = self.futures_queue.pop()
+            future.update_with_response(get_fut_response)
 
-            return responses
-        except TimeoutError as e:
-            raise TimeoutError(f"RPC call to {method} timed out.") from e
+        return get_response()
 
     @staticmethod
     def _ensure_worker_termination(worker_procs: list[BaseProcess]):
@@ -348,9 +354,6 @@ def shutdown(self):
                 self._ensure_worker_termination([w.proc for w in workers])
 
             self.shutdown_event.set()
-            if self.io_thread_pool is not None:
-                self.io_thread_pool.shutdown(wait=False, cancel_futures=True)
-                del self.io_thread_pool
 
         self.rpc_broadcast_mq = None
 
diff --git a/vllm/v1/executor/ray_executor.py b/vllm/v1/executor/ray_executor.py
@@ -441,20 +441,19 @@ def sample_tokens(  # type: ignore[override]
         assert self.kv_output_aggregator is not None
         if not non_block:
             # Block and get results from all workers
-            outputs = [ref.get() for ref in refs]
-            return self.kv_output_aggregator.aggregate(outputs)
+            return self.kv_output_aggregator.aggregate(ray.get(refs))
 
         # Return a future that will aggregate outputs from all workers
         return FutureWrapper(refs, self.kv_output_aggregator)
 
-    def collective_rpc(
+    def collective_rpc(  # type: ignore[override]
         self,
         method: str | Callable,
         timeout: float | None = None,
         args: tuple = (),
         kwargs: dict[str, Any] | None = None,
         non_block: bool = False,
-    ) -> list[Any]:
+    ) -> list[Any] | Future[list[Any]]:
         """Runs the given method on all workers."""
         sent_method = method if isinstance(method, str) else cloudpickle.dumps(method)
         del method
@@ -470,7 +469,7 @@ def collective_rpc(
 
         # Get the results of the ray workers.
         if non_block:
-            return [FutureWrapper((output,)) for output in ray_worker_outputs]
+            return FutureWrapper(ray_worker_outputs)
 
         return ray.get(ray_worker_outputs, timeout=timeout)
 
diff --git a/vllm/v1/executor/ray_utils.py b/vllm/v1/executor/ray_utils.py
@@ -141,19 +141,19 @@ class FutureWrapper(Future):
     the result() call. If not only the first worker's output is returned.
     """
 
-    def __init__(self, refs, aggregator: KVOutputAggregator | None = None):
+    def __init__(self, ref, aggregator: KVOutputAggregator | None = None):
         super().__init__()
-        self.refs = refs
+        self.ref = ref
         self.aggregator = aggregator
 
     def result(self, timeout=None):
         if timeout is not None:
             raise NotImplementedError("timeout is not supported")
 
+        outputs = ray.get(self.ref, timeout=timeout)
         if self.aggregator is None:
-            return self.refs[0].get()
+            return outputs[0]
 
-        outputs = [ref.get() for ref in self.refs]
         return self.aggregator.aggregate(outputs, output_rank=0)
 
 
diff --git a/vllm/v1/executor/uniproc_executor.py b/vllm/v1/executor/uniproc_executor.py
diff --git a/vllm/v1/worker/gpu_worker.py b/vllm/v1/worker/gpu_worker.py
