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

[P/D] Remove restrictions on mooncake for IPv6 (#5946)

Browse Source 
### What this PR does / why we need it?
Remove restrictions on mooncake for IPv6
Dependencies: cann8.5、mooncake v0.3.8.post1

- vLLM version: v0.13.0
- vLLM main:
2c24bc6996

---------

Signed-off-by: liziyu <liziyu16@huawei.com>
This commit is contained in:
liziyu
2026-01-24 11:30:22 +08:00
committed by GitHub
parent 019a2fe6e6
commit 14bef9af6f
2 changed files with 1 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docs/source/tutorials/pd_disaggregation_mooncake_multi_node.md
View File

@@ -177,7 +177,7 @@ Mooncake is the serving platform for Kimi, a leading LLM service provided by Moo
First, we need to obtain the Mooncake project. Refer to the following command: First, we need to obtain the Mooncake project. Refer to the following command:
```shell ```shell
git clone -b v0.3.7.post2 --depth 1 https://github.com/kvcache-ai/Mooncake.git git clone -b v0.3.8.post1 --depth 1 https://github.com/kvcache-ai/Mooncake.git
``` ```
(Optional) Replace go install url if the network is poor (Optional) Replace go install url if the network is poor

8
vllm_ascend/distributed/kv_transfer/utils/mooncake_transfer_engine.py
View File

@@ -14,14 +14,6 @@ class GlobalTE():
self.register_buffer_lock = threading.Lock() self.register_buffer_lock = threading.Lock()
def get_transfer_engine(self, hostname: str, device_name: Optional[str]): def get_transfer_engine(self, hostname: str, device_name: Optional[str]):
try:
ip = ipaddress.ip_address(hostname)
if isinstance(ip, ipaddress.IPv6Address):
raise RuntimeError(
"The backend of mooncake's Ascend Direct Xfer Library currently does not support IPv6."
)
except ValueError:
pass
if self.transfer_engine is None: if self.transfer_engine is None:
with self.transfer_engine_lock: with self.transfer_engine_lock:
# Double-Checked Locking # Double-Checked Locking