EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[PD] Correct the ip and port env (#2450)

Browse Source 
1. rename `VLLM_LLMDD_RPC_PORT` to `VLLM_ASCEND_LLMDD_RPC_PORT` to make
the prefix the same in vllm-ascend
2. enable `VLLM_ASCEND_LLMDD_RPC_IP` env for PD feature.

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
This commit is contained in:
wangxiyuan
2025-08-20 11:39:05 +08:00
committed by GitHub
parent 7bec1a9b9c
commit 73acdcfc3b
2 changed files with 6 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
vllm_ascend/envs.py
View File

@@ -123,11 +123,11 @@ env_variables: Dict[str, Callable[[], Any]] = {
# remote worker.
"VLLM_ASCEND_LLMDD_RPC_IP":
lambda: os.getenv("VLLM_ASCEND_LLMDD_RPC_IP", "0.0.0.0"),
# `LLMDataDistCMgrConnector` required variable. `VLLM_LLMDD_RPC_PORT` is used as the
# `LLMDataDistCMgrConnector` required variable. `VLLM_ASCEND_LLMDD_RPC_PORT` is used as the
# rpc communication listening port, which will be used to receive the agent metadata from the
# remote worker.
"VLLM_LLMDD_RPC_PORT":
lambda: int(os.getenv("VLLM_LLMDD_RPC_PORT", 5557)),
"VLLM_ASCEND_LLMDD_RPC_PORT":
lambda: int(os.getenv("VLLM_ASCEND_LLMDD_RPC_PORT", 5557)),
# Whether to enable mla_pa for deepseek mla decode, this flag will be removed after its available torch_npu is public accessible
# and the mla_pa will be the default path of deepseek decode path.
"VLLM_ASCEND_MLA_PA":