zzh/vllm-npu-plugin
1
0
Fork 0
mirror of https://github.com/handsomezhuzhu/vllm-npu-plugin.git synced 2026-02-20 19:50:15 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix: Defer SOC version initialization to avoid NPU lazy init issues

Browse Source
This commit is contained in:
handsomezhuzhu
2026-02-10 23:19:34 +08:00
parent f49538ea8d
commit ae10ce68f0
2 changed files with 10 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
vllm_npu/worker/worker_v1.py
View File

@@ -87,9 +87,8 @@ class NPUWorker(WorkerBase):
ops.register_dummy_fusion_op()
_register_atb_extensions()
register_ascend_customop(vllm_config)
# init ascend config and soc version
# init ascend config (soc version deferred to init_device)
init_ascend_config(vllm_config)
init_ascend_soc_version()
use_sparse = False
if vllm_config.model_config is not None:
use_sparse = hasattr(vllm_config.model_config.hf_config,
@@ -213,6 +212,8 @@ class NPUWorker(WorkerBase):
def init_device(self):
device = self._init_device()
# Now that NPU device is initialized, get soc version
init_ascend_soc_version()
# Init ModelRunner here, so that we have access to self.device.
self.model_runner = NPUModelRunner(self.vllm_config, device)