[bugfix] layerwise D first plan (#3866)

### What this PR does / why we need it?
Refactored the layerwise code to send to the D node first, preventing
P-node hangs due to communication timeouts when DP > 1.
### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
By ci

- vLLM version: v0.11.0
- vLLM main:
83f478bb19

---------

Signed-off-by: wangxiaoteng <wangxiaoteng@huawei.com>
Signed-off-by: liziyu <liziyu16@huawei.com>
Co-authored-by: liziyu <liziyu16@huawei.com>

examples/disaggregated_prefill_v1/load_balance_proxy_layerwise_server_example.py

@@ -88,18 +88,17 @@ import argparse
 import asyncio
 import functools
 import heapq
-import json
 import os
 import sys
 import threading
 import uuid
 from contextlib import asynccontextmanager
-from dataclasses import dataclass
+from typing import List
-from typing import Any, List
 
 import httpx
 from fastapi import FastAPI, Request
 from fastapi.responses import StreamingResponse
+from transformers import AutoTokenizer
 from vllm.logger import init_logger
 
 logger = init_logger(__name__)
@@ -107,7 +106,6 @@ logger = init_logger(__name__)
 # Add uvloop for faster event loop if available
 try:
     import uvloop
-
     asyncio.set_event_loop_policy(uvloop.EventLoopPolicy())
 except ImportError:
     pass
@@ -154,6 +152,9 @@ class ProxyState:
         heapq.heapify(self.prefiller_heap)
         heapq.heapify(self.decoder_heap)
         self.req_id_future = {}
+        self.req_data_dict = {}
+        self.tokenizer = AutoTokenizer.from_pretrained(
+            global_args.tokenizer_dir)
 
     def _update_prefiller_priority(self, server_idx: int):
         """Update the priority of a prefiller server in the heap."""
@@ -280,6 +281,10 @@ def parse_args():
                         nargs="+",
                         default=["localhost"])
     parser.add_argument("--decoder-ports", type=int, nargs="+", default=[8002])
+    parser.add_argument("--tokenizer-dir",
+                        type=str,
+                        default="/mnt/weight/Qwen3-235B-A22B-W8A8",
+                        help="Maximum number of retries for HTTP requests")
     parser.add_argument("--max-retries",
                         type=int,
                         default=3,
@@ -356,17 +361,6 @@ async def send_request_to_service(client: httpx.AsyncClient,
     aborted_requests = proxy_state.aquire_aborted_prefiller_requests(
         prefiller_id)
     req_data = req_data.copy()
-    req_data['kv_transfer_params'] = {
-        "do_remote_decode": True,
-        "do_remote_prefill": False,
-        "remote_engine_id": None,
-        "remote_block_ids": None,
-        "remote_host": None,
-        "remote_port": None,
-        "aborted_request": list(aborted_requests),
-        "metaserver":
-        f"http://{global_args.host}:{global_args.port}/v1/metaserver"
-    }
     req_data["stream"] = False
     req_data["max_tokens"] = 1
     if "stream_options" in req_data:
@@ -458,59 +452,11 @@ def get_api_request_id(api, req_id):
         return "chatcmpl-" + req_id
 
 
-async def _handle_select_instance(api: str, req_data: Any,
+def get_origin_request_id(api, req_id):
-                                  request_length: int):
+    if api == "/completions":
-    prefiller_score = proxy_state.calculate_prefill_scores(request_length)
+        return req_id.replace("cmpl-", "").replace("-0", "")
-    logger.debug(
+    elif api == "/chat/completions":
-        f"Request length: {request_length}, Prefiller score: {prefiller_score}"
+        return req_id.replace("chatcmpl-", "")
-    )
-    request_id = await proxy_state.next_req_id()
-    # Select prefiller
-    prefiller_idx = proxy_state.select_prefiller(prefiller_score)
-    prefiller = proxy_state.prefillers[prefiller_idx]
-    result_future = asyncio.Future()  # type: ignore
-    request_id_api = get_api_request_id(api, request_id)
-    proxy_state.req_id_future[request_id_api] = result_future
-    # Send request to prefiller
-    asyncio.get_running_loop().create_task(
-        send_request_to_service(prefiller.client,
-                                prefiller_idx,
-                                api,
-                                req_data,
-                                request_id,
-                                max_retries=global_args.max_retries,
-                                base_delay=global_args.retry_delay))
-    proxy_state.release_prefiller(prefiller_idx, prefiller_score)
-
-    response = await result_future
-    del proxy_state.req_id_future[request_id_api]
-    req_data["kv_transfer_params"] = response
-
-    # Select decoder
-    decoder_score = proxy_state.calculate_decode_scores(request_length)
-    logger.debug("Decoder score: %f", decoder_score)
-    # Use the prefiller's kv_transfer_params to select decoder
-    decoder_idx = proxy_state.select_decoder(decoder_score)
-    decoder = proxy_state.decoders[decoder_idx]
-    logger.debug("Using %s %s", prefiller.url, decoder.url)
-    return InstanceInfo(request_id=request_id,
-                        prefiller_idx=prefiller_idx,
-                        prefiller_score=prefiller_score,
-                        prefiller=prefiller,
-                        decoder=decoder,
-                        decoder_idx=decoder_idx,
-                        decoder_score=decoder_score)
-
-
-@dataclass
-class InstanceInfo:
-    request_id: str
-    prefiller_idx: int
-    prefiller_score: float
-    prefiller: ServerState
-    decoder_idx: int
-    decoder_score: float
-    decoder: ServerState
 
 
 async def _handle_completions(api: str, request: Request):
@@ -518,120 +464,47 @@ async def _handle_completions(api: str, request: Request):
         req_data = await request.json()
         req_body = await request.body()
         request_length = len(req_body)
-        instance_info = await _handle_select_instance(api, req_data,
+        request_id = await proxy_state.next_req_id()
-                                                      request_length)
+        request_id_api = get_api_request_id(api, request_id)
-        stream_flag = bool(req_data.get("stream", False))
+        proxy_state.req_data_dict[request_id_api] = (req_data, request_length,
-        chat_flag = "messages" in req_data
+                                                     api)
-
+        req_data['kv_transfer_params'] = {
-        if "prompt" in req_data:
+            "do_remote_decode":
-            origin_prompt = req_data["prompt"]
+            False,
-        elif chat_flag:
+            "do_remote_prefill":
-            messages = req_data["messages"]
+            True,
-            origin_prompt = messages[0].get("content", "")
+            "metaserver":
-        else:
+            f"http://{global_args.host}:{global_args.port}/v1/metaserver"
-            origin_prompt = ""
+        }
-        # refer to vLLM sampling_params: max_token default value
+        # Select decoder
-        origin_max_tokens = req_data.get("max_tokens", 16)
+        decoder_score = proxy_state.calculate_decode_scores(request_length)
+        logger.debug("Decoder score: %f", decoder_score)
+        # Use the prefiller's kv_transfer_params to select decoder
+        decoder_idx = proxy_state.select_decoder(decoder_score)
+        decoder = proxy_state.decoders[decoder_idx]
+        # logger.debug("Using %s %s", prefiller.url, decoder.url)
+        # Stream response from decoder
+        released_kv = False
 
         async def generate_stream():
-            nonlocal instance_info
+            nonlocal released_kv
-            generated_token = ""
-            released_kv = False
-            retry_count = 0
-            retry = True
-            completion_tokens = 0
             # Only one await per chunk, minimal logic in loop
             try:
-                while retry:
+                async for chunk in stream_service_response_with_retry(
-                    retry = False
+                        decoder.client,
-                    async for chunk in stream_service_response_with_retry(
+                        api,
-                            instance_info.decoder.client,
+                        req_data,
-                            api,
+                        request_id=request_id,
-                            req_data,
+                        max_retries=global_args.max_retries,
-                            request_id=instance_info.request_id,
+                        base_delay=global_args.retry_delay):
-                            max_retries=global_args.max_retries,
+                    yield chunk
-                            base_delay=global_args.retry_delay):
-                        if not released_kv and chunk:
-                            proxy_state.release_prefiller_kv(
-                                instance_info.prefiller_idx,
-                                instance_info.prefiller_score)
-                            released_kv = True
-                        try:
-                            chunk_str = chunk.decode("utf-8").strip()
-                        except UnicodeDecodeError:
-                            logger.debug(
-                                f"Skipping chunk: {chunk}")
-                            yield chunk
-                            continue
-                        if not chunk_str:
-                            continue
-                        if chunk_str.startswith("data: "):
-                            chunk_str = chunk_str[len("data: "):]
-                        try:
-                            chunk_json = json.loads(chunk_str)
-                        except json.JSONDecodeError:
-                            # if chunk is [done], skip it.
-                            logger.debug(
-                                f"Skipping chunk: {chunk_str}")
-                            yield chunk
-                            continue
-                        choices = chunk_json.get("choices", [])
-                        if not choices:
-                            yield chunk
-                            continue
-
-                        choice = choices[0]
-                        delta = choice.get("delta") or {}
-                        message = choice.get("message") or {}
-                        content = (
-                                delta.get("content")
-                                or message.get("content")
-                                or choice.get("text")
-                                or ""
-                                )
-                        generated_token += content
-
-                        stop_reason = choice.get(
-                            "stop_reason")
-                        usage = chunk_json.get("usage", {})
-                        completion_tokens = (completion_tokens + 1) if stream_flag else \
-                            (completion_tokens + usage.get("completion_tokens"))
-                        if stop_reason == "recomputed":
-                            retry = True
-                            retry_count += 1
-                            if chat_flag:
-                                messages[0][
-                                    "content"] = origin_prompt + generated_token
-                            else:
-                                req_data[
-                                    "prompt"] = origin_prompt + generated_token
-                            req_data[
-                                "max_tokens"] = origin_max_tokens - completion_tokens + retry_count
-                            tmp_request_length = len(
-                                json.dumps(req_data).encode("utf-8"))
-                            instance_info = await _handle_select_instance(
-                                api, req_data, tmp_request_length)
-                            break
-                        if retry_count > 0 and not stream_flag:
-                            if chat_flag:
-                                choices[0]["message"][
-                                    "content"] = generated_token
-                            else:
-                                choices[0]["text"] = generated_token
-                            chunk = json.dumps(chunk_json).encode("utf-8")
-                        yield chunk
             except Exception as e:
                 logger.error(
-                    f"Error during streaming from decoder {instance_info.decoder.url}: {str(e)} the aborted request {instance_info.request_id} will be routing to the target prefiller when new request is ready to dispatch to it"
+                    f"Error during streaming from decoder {decoder.url}: {str(e)} the aborted request {request_id} will be routing to the target prefiller when new request is ready to dispatch to it"
                 )
-                proxy_state.abort_prefiller_request(
-                    instance_info.prefiller_idx, instance_info.request_id)
-                proxy_state.release_prefiller_kv(instance_info.prefiller_idx,
-                                                 instance_info.prefiller_score)
 
             # After streaming done, release tokens
-            proxy_state.release_decoder(instance_info.decoder_idx,
+            proxy_state.release_decoder(decoder_idx, decoder_score)
-                                        instance_info.decoder_score)
 
         return StreamingResponse(generate_stream(),
                                  media_type="application/json")
@@ -669,11 +542,33 @@ async def healthcheck():
 @app.post("/v1/metaserver")
 async def metaserver(request: Request):
     try:
-        req_data = await request.json()
+        kv_transfer_params = await request.json()
-        request_id = req_data.pop("request_id", None)
+
-        if request_id in proxy_state.req_id_future:
+        request_id = kv_transfer_params["request_id"]
-            result_future = proxy_state.req_id_future[request_id]
+        assert request_id in proxy_state.req_data_dict
-            result_future.set_result(req_data)
+        req_data, request_length, api = proxy_state.req_data_dict[request_id]
+        request_id = get_origin_request_id(api, request_id)
+        req_data["kv_transfer_params"] = kv_transfer_params
+        prefiller_score = proxy_state.calculate_prefill_scores(request_length)
+        logger.debug(
+            f"Request length: {request_length}, Prefiller score: {prefiller_score}"
+        )
+
+        # Select prefiller
+        prefiller_idx = proxy_state.select_prefiller(prefiller_score)
+        prefiller = proxy_state.prefillers[prefiller_idx]
+        logger.debug(f"Using prefill {prefiller.url=} {req_data=}")
+        # Send request to prefiller
+        response = await send_request_to_service(
+            prefiller.client,
+            prefiller_idx,
+            api,
+            req_data,
+            request_id,
+            max_retries=global_args.max_retries,
+            base_delay=global_args.retry_delay)
+        proxy_state.release_prefiller(prefiller_idx, prefiller_score)
+
     except Exception as e:
         logger.error(f"Post metaserver failed with: {str(e)}")
 
@@ -682,5 +577,4 @@ if __name__ == '__main__':
     global global_args
     global_args = parse_args()
     import uvicorn
-
     uvicorn.run(app, host=global_args.host, port=global_args.port)

vllm_ascend/distributed/parallel_state.py

@@ -96,7 +96,8 @@ def init_ascend_model_parallel(parallel_config: ParallelConfig, ):
                 parallel_config.data_parallel_size, num_head_replica, -1,
                 alltoall_group_size
             )  # [DP_size, num_head_replica, num_alltoall_group, alltoall_group_size]
-            group_ranks = group_ranks.view(-1, alltoall_group_size).unbind(0)
+            group_ranks = group_ranks.reshape(-1,
+                                              alltoall_group_size).unbind(0)
         group_ranks = [x.tolist() for x in group_ranks]
         local_rank = get_world_group().local_rank
         num = next(