apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-12-24 13:28:13 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[BugFix]fix dp&ep&tp and muti node infer (#3629)

Browse Source 
* rm log

* fix bug

* fix bug

* fix dp&ep&tp and muti node infer

* fix

---------

Co-authored-by: Yuanle Liu <yuanlehome@163.com>
This commit is contained in:
gaoziyuan
2025-08-28 19:09:10 +08:00
committed by GitHub
parent 17731a8acd
commit fc635acc47
7 changed files with 48 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
fastdeploy/config.py
View File

@@ -341,7 +341,8 @@ class ParallelConfig:
def set_tp_group(self):
# different tp group id
# prevent different tp_groups using the same group_id
dist.collective._set_custom_gid(self.data_parallel_rank + 100)
tp_gid_offset = envs.FD_TP_GROUP_GID_OFFSET
dist.collective._set_custom_gid(self.data_parallel_rank + tp_gid_offset)
self.tp_group = dist.new_group(
range(
self.data_parallel_rank * self.tensor_parallel_size,
@@ -349,7 +350,8 @@ class ParallelConfig:
)
)
# same ep group id
dist.collective._set_custom_gid(self.data_parallel_size + 100)
# (TODO:gaoziyuan move this gid config to ep.py)
dist.collective._set_custom_gid(self.data_parallel_size + tp_gid_offset)
logger.info(
f"data_parallel_size: {self.data_parallel_size}, tensor_parallel_size: {self.tensor_parallel_size}, expert_parallel_size: {self.expert_parallel_size}, data_parallel_rank: {self.data_parallel_rank}, tensor_parallel_rank: {self.tensor_parallel_rank}, expert_parallel_rank: {self.expert_parallel_rank}, tp_group: {self.tp_group}."
)