apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 08:37:06 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

Unify server-side and model-side Config(Part-5) (#3497)
Some checks failed
CE Compile Job / ce_job_pre_check (push) Has been cancelled
Details
CE Compile Job / print_ce_job_pre_check_outputs (push) Has been cancelled
Details
CE Compile Job / FD-Clone-Linux (push) Has been cancelled
Details
CE Compile Job / Show Code Archive Output (push) Has been cancelled
Details
CE Compile Job / BUILD_SM8090 (push) Has been cancelled
Details
CE Compile Job / BUILD_SM8689 (push) Has been cancelled
Details
CE Compile Job / CE_UPLOAD (push) Has been cancelled
Details
Deploy GitHub Pages / deploy (push) Has been cancelled
Details
Publish Job / publish_pre_check (push) Has been cancelled
Details
Publish Job / print_publish_pre_check_outputs (push) Has been cancelled
Details
Publish Job / FD-Clone-Linux (push) Has been cancelled
Details
Publish Job / Show Code Archive Output (push) Has been cancelled
Details
Publish Job / BUILD_SM8090 (push) Has been cancelled
Details
Publish Job / BUILD_SM8689 (push) Has been cancelled
Details
Publish Job / PADDLE_PYPI_UPLOAD_8090 (push) Has been cancelled
Details
Publish Job / PADDLE_PYPI_UPLOAD_8689 (push) Has been cancelled
Details
Publish Job / Run FastDeploy Unit Tests and Coverage (push) Has been cancelled
Details
Publish Job / Run FastDeploy LogProb Tests (push) Has been cancelled
Details
Publish Job / Extracted partial CE model tasks to run in CI. (push) Has been cancelled
Details
Publish Job / Run Base Tests (push) Has been cancelled
Details
Publish Job / Run Accuracy Tests (push) Has been cancelled
Details

Browse Source 
* move config

* fix xpu

* fix

* fix vl

* fix vl

* fix unitest

* fix args

* add unitest

* fix test
This commit is contained in:
YuanRisheng
2025-08-21 19:00:21 +08:00
committed by GitHub
parent e5aa7087db
commit c389a4013c
15 changed files with 480 additions and 499 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
fastdeploy/engine/expert_service.py
View File

@@ -50,8 +50,8 @@ class ExpertService:
cfg (Config): Config object containing all the configuration parameters.
"""
self.cfg = cfg
start_pos = (local_data_parallel_id * self.cfg.tensor_parallel_size) % cfg.worker_num_per_node
end_pos = start_pos + self.cfg.tensor_parallel_size
start_pos = (local_data_parallel_id * self.cfg.parallel_config.tensor_parallel_size) % cfg.worker_num_per_node
end_pos = start_pos + self.cfg.parallel_config.tensor_parallel_size
if cfg.splitwise_role != "mixed":
self.cfg.cache_config.rdma_comm_ports = self.cfg.cache_config.rdma_comm_ports[start_pos:end_pos]
self.cfg.local_device_ids = self.cfg.device_ids.split(",")[start_pos:end_pos]
@@ -69,13 +69,13 @@ class ExpertService:
address=address,
is_server=False,
client_id=0,
num_client=cfg.tensor_parallel_size,
num_client=cfg.parallel_config.tensor_parallel_size,
local_data_parallel_id=local_data_parallel_id,
)
self.resource_manager = ResourceManager(
cfg.max_num_seqs,
cfg,
cfg.tensor_parallel_size,
cfg.parallel_config.tensor_parallel_size,
cfg.splitwise_role,
local_data_parallel_id,
)
@@ -125,7 +125,7 @@ class ExpertService:
if self.cfg.splitwise_role != "mixed":
self.cache_manager_processes = self.resource_manager.cache_manager.launch_cache_manager(
cache_config=self.cfg.cache_config,
tensor_parallel_size=self.cfg.tensor_parallel_size,
tensor_parallel_size=self.cfg.parallel_config.tensor_parallel_size,
device_ids=self.cfg.local_device_ids,
pod_ip=self.cfg.master_ip,
engine_worker_queue_port=self.cfg.engine_worker_queue_port,
@@ -343,7 +343,7 @@ class ExpertService:
if not is_decode:
llm_logger.info(f"Tasks are sent to engine, req_ids={req_ids}")
if not is_prefill and self.cfg.cache_config.enable_chunked_prefill:
if not self.cfg.enable_mm:
if not self.cfg.model_config.enable_mm:
self.update_requests_chunk_size(tasks)
else:
self.update_mm_requests_chunk_size(tasks)