fix : 修复存储层 rglob 死锁、DDP LOCAL_RANK 绑定

This commit is contained in:
yegroup001
2026-06-02 01:01:00 +08:00
parent 01ce1fb9e3
commit 746a1475b2
3 changed files with 33 additions and 8 deletions
+2 -1
View File
@@ -2,6 +2,7 @@
import contextlib
import logging
import os
from contextlib import contextmanager
from typing import Optional, Tuple
@@ -181,7 +182,7 @@ class DDPExecutor(BaseExecutor):
if not self.use_distributed:
logger.warning("DDP backend selected but world_size=1, model not wrapped")
return model
local_rank = get_rank()
local_rank = int(os.environ.get("LOCAL_RANK", get_rank()))
model = DDP(
model,
device_ids=[local_rank],