MindIE镜像中推理：

cd /urs/local/Ascend/llm_model

bash examples/models/qwen/run_pa.sh -m /data/models/Qwen1.5-7B/weight -c true

报错：

Traceback (most recent call last): 

  File "/usr/local/python3.10.2/lib/python3.10/runpy.py", line 196, in _run_module_as_main 

    return _run_code(code, main_globals, None, 

  File "/usr/local/python3.10.2/lib/python3.10/runpy.py", line 86, in _run_code 

    exec(code, run_globals) 

  File "/usr/local/Ascend/llm_model/examples/run_pa.py", line 345, in <module> 

    pa_runner = PARunner(**input_dict) 

  File "/usr/local/Ascend/llm_model/examples/run_pa.py", line 40, in __init__ 

    self.model = ModelRunner( 

  File "/usr/local/Ascend/llm_model/atb_llm/runner/model_runner.py", line 71, in __init__ 

    self.process_group, self.device = initialize_distributed(self.rank, self.npu_id, world_size) 

  File "/usr/local/Ascend/llm_model/atb_llm/utils/dist.py", line 93, in initialize_distributed 

    torch.npu.set_device(device) 

  File "/usr/local/python3.10.2/lib/python3.10/site-packages/torch_npu/npu/utils.py", line 55, in set_device 

    torch_npu._C._npu_setDevice(device_id) 

RuntimeError: GetDevice:build/CMakeFiles/torch_npu.dir/compiler_depend.ts:57 NPU function error: aclrtGetCurrentContext(&used_devices[local_device]), error code is 107002 

[ERROR] 2024-11-01-06:53:19 (PID:138986, Device:0, RankID:6) ERR00100 PTA call acl api failed 

[Error]: The context is empty. 

        Check whether acl.rt.set_context or acl.rt.set_device is called. 

EL0005: 2024-11-01-06:53:16.445.069 The resources are busy. 

        Possible Cause: 1. The resources have been occupied. 2. The device is being reset. 3. Software is not ready. 

        Solution: 1. Close applications not in use. 2. Wait for a while and try again. 

        TraceBack (most recent call last): 

        rtGetDeviceCount execute failed, reason=[driver error:internal error][FUNC:FuncErrorReason][FILE:error_message_manage.cc][LINE:53] 

        get device count failed, runtime result = 507899.[FUNC:ReportCallError][FILE:log_inner.cpp][LINE:161] 

        Failed to drvGetPlatformInfo, ret=87[FUNC:DrvGetPlatformInfo][FILE:ai_drv_dev_api.cpp][LINE:79] 

        halGetDeviceInfo get device type version failed , ret:87[FUNC:Init][FILE:config_manager.cpp][LINE:61] 

        [Init][Version]init soc version failed, ret = 507008[FUNC:ReportInnerError][FILE:log_inner.cpp][LINE:145] 

        ctx is NULL![FUNC:GetDevErrMsg][FILE:api_impl.cc][LINE:4916] 

        The argument is invalid.Reason: rtGetDevMsg execute failed, reason=[context pointer null]