mindIE 测试 qwen 模型报错
测试 qwen代码如下：
./tests/modeltest/run.sh pa_fp16 performance [[512,512]] 4 qwen ~/Models/Qwen1_5_0_5_chat 4
已经按照文档：MindIE集成部署问题定位基本流程（atb-models版) 排查故障，目前是排查到了 4 b步骤
报错如下：
dist.py-94: initialize_distributed has been Set
Traceback (most recent call last):
File "/home/HwHiAiUser/atb-models/tests/modeltest/core/qwen_test.py", line 82, in <module>
File "/home/HwHiAiUser/miniconda3/envs/llm/lib/python3.10/site-packages/torch/_ops.py", line 933, in load_library
ctypes.CDLL(path)
File "/home/HwHiAiUser/miniconda3/envs/llm/lib/python3.10/ctypes/__init__.py", line 374, in __init__
self._handle = _dlopen(self._name, mode)
OSError: /home/HwHiAiUser/atb-models/lib/libatb_speed_torch.so: undefined symbol: _ZN3c108ListType3getESsNS_4Type24SingletonOrSharedTypePtrIS1_EE
EL0003: 2024-10-25-02:24:17.202.869 The argument is invalid.
Solution: Try again with a valid argument.
TraceBack (most recent call last):
Failed to open device, retCode=0x7020014, deviceId=1.[FUNC:InitRawDriver][FILE:device.cc][LINE:277]
Failed to init RawDriver, device_id=1, retCode=0x7020014[FUNC:Init][FILE:device.cc][LINE:320]
Check param failed, dev can not be NULL![FUNC:DeviceRetain][FILE:runtime.cc][LINE:3825]
Check param failed, dev can not be NULL![FUNC:PrimaryContextRetain][FILE:runtime.cc][LINE:3539]
Check param failed, ctx can not be NULL![FUNC:PrimaryContextRetain][FILE:runtime.cc][LINE:3566]
Check param failed, context can not be null.[FUNC:NewDevice][FILE:api_impl.cc][LINE:2278]
New device failed, retCode=0x7010006[FUNC:SetDevice][FILE:api_impl.cc][LINE:2300]
rtSetDevice execute failed, reason=[device retain error][FUNC:FuncErrorReason][FILE:error_message_manage.cc][LINE:53]
open device 1 failed, runtime result = 507033.[FUNC:ReportCallError][FILE:log_inner.cpp][LINE:161]
ctx is NULL![FUNC:GetDevErrMsg][FILE:api_impl.cc][LINE:4916]
The argument is invalid.Reason: rtGetDevMsg execute failed, reason=[context pointer null]
请协助，谢谢！
点赞
收藏
回复
分享
举报
已解决
mindIE 测试 qwen 模型报错
测试 qwen代码如下：
./tests/modeltest/run.sh pa_fp16 performance [[512,512]] 4 qwen ~/Models/Qwen1_5_0_5_chat 4
已经按照文档：MindIE集成部署问题定位基本流程（atb-models版) 排查故障，目前是排查到了 4 b步骤
报错如下：
dist.py-94: initialize_distributed has been Set
Traceback (most recent call last):
File "/home/HwHiAiUser/atb-models/tests/modeltest/core/qwen_test.py", line 82, in <module>
File "/home/HwHiAiUser/miniconda3/envs/llm/lib/python3.10/site-packages/torch/_ops.py", line 933, in load_library
ctypes.CDLL(path)
File "/home/HwHiAiUser/miniconda3/envs/llm/lib/python3.10/ctypes/__init__.py", line 374, in __init__
self._handle = _dlopen(self._name, mode)
OSError: /home/HwHiAiUser/atb-models/lib/libatb_speed_torch.so: undefined symbol: _ZN3c108ListType3getESsNS_4Type24SingletonOrSharedTypePtrIS1_EE
EL0003: 2024-10-25-02:24:17.202.869 The argument is invalid.
Solution: Try again with a valid argument.
TraceBack (most recent call last):
Failed to open device, retCode=0x7020014, deviceId=1.[FUNC:InitRawDriver][FILE:device.cc][LINE:277]
Failed to init RawDriver, device_id=1, retCode=0x7020014[FUNC:Init][FILE:device.cc][LINE:320]
Check param failed, dev can not be NULL![FUNC:DeviceRetain][FILE:runtime.cc][LINE:3825]
Check param failed, dev can not be NULL![FUNC:PrimaryContextRetain][FILE:runtime.cc][LINE:3539]
Check param failed, ctx can not be NULL![FUNC:PrimaryContextRetain][FILE:runtime.cc][LINE:3566]
Check param failed, context can not be null.[FUNC:NewDevice][FILE:api_impl.cc][LINE:2278]
New device failed, retCode=0x7010006[FUNC:SetDevice][FILE:api_impl.cc][LINE:2300]
rtSetDevice execute failed, reason=[device retain error][FUNC:FuncErrorReason][FILE:error_message_manage.cc][LINE:53]
open device 1 failed, runtime result = 507033.[FUNC:ReportCallError][FILE:log_inner.cpp][LINE:161]
ctx is NULL![FUNC:GetDevErrMsg][FILE:api_impl.cc][LINE:4916]
The argument is invalid.Reason: rtGetDevMsg execute failed, reason=[context pointer null]
请协助，谢谢！