We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent 5942a62 commit 809f25fCopy full SHA for 809f25f
lightllm/server/router/model_infer/mode_backend/pd_nixl/prefill_node_impl/prefill_impl.py
@@ -65,6 +65,10 @@ def _prefill_chuncked_handle_func(
65
"""
66
在每一步chuncked prefill 后,尝试生成chuncked 传输任务,发个 kv_move_manager 进行处理。
67
68
+ # 系统内部的 health 请求不创建 kv 传输任务。
69
+ if req_obj.req_id < 0:
70
+ return
71
+
72
# 传输的 kv 要少一个,不然decode 无法有下一个输入推理出下一个token
73
input_len = req_obj.shm_req.input_len - 1
74
page_size = self.args.nixl_pd_kv_page_size
0 commit comments