[V1] Exception Handling when Loading KV Cache from Remote Store (#21534)

Signed-off-by: liuyumoye <adeline_ly2023@outlook.com> Co-authored-by: liuyumoye <adeline_ly2023@outlook.com>
2025-07-28 11:34:17 +08:00
parent 04ff4be310
commit 15a72ac478
10 changed files with 229 additions and 5 deletions
--- a/vllm/sequence.py
+++ b/vllm/sequence.py
@@ -1167,6 +1167,8 @@ class IntermediateTensors:
    # [req_ids]
    finished_sending: Optional[set[str]] = None
    finished_recving: Optional[set[str]] = None
+    #req_id -> num_actual_load_tokens
+    finished_loading_dict: Optional[dict[str, int]] = None

    def __init__(self, tensors):
        # manually define this function, so that