docker 启动报错get device count of NPU failed
收藏回复举报
docker 启动报错get device count of NPU failed
t('forum.solved') 已解决
新人帖
发表于2025-03-25 09:57:25
0 查看

python -c "import torch;import torch_npu;print(torch.cuda.is_available());print(torch.npu.is_available());print(torch_npu.npu.is_available());"

[EVENT] PROFILING(277,python):2025-03-25-01:36:07.839.746 [msprof_callback_impl.cpp:334] >>> (tid:277) Started to register profiling ctrl callback.

[EVENT] PROFILING(277,python):2025-03-25-01:36:07.840.783 [msprof_callback_impl.cpp:341] >>> (tid:277) Started to register profiling hash id callback.

[INFO] PROFILING(277,python):2025-03-25-01:36:07.841.068 [prof_atls_plugin.cpp:83] >>> (tid:277) RegisterProfileCallback, callback type is 7

[EVENT] PROFILING(277,python):2025-03-25-01:36:07.841.277 [msprof_callback_impl.cpp:348] >>> (tid:277) Started to register profiling enable host freq callback.

[INFO] PROFILING(277,python):2025-03-25-01:36:07.841.478 [prof_atls_plugin.cpp:83] >>> (tid:277) RegisterProfileCallback, callback type is 8

[ERROR] ATRACE(277,python):2025-03-25-01:36:07.845.996 [trace_driver_api.c:58](tid:277) get platform info failed, drvErr=87.

[ERROR] RUNTIME(277,python):2025-03-25-01:36:08.059.786 [runtime.cc:1682]277 CheckHaveDevice:Call halGetDeviceInfo failed: drvRet=87, module type=0, info type=1.

[INFO] PROFILING(277,python):2025-03-25-01:36:08.060.319 [prof_atls_plugin.cpp:160] >>> (tid:277) Module[7] register callback of ctrl handle.

[ERROR] RUNTIME(277,python):2025-03-25-01:36:08.062.681 [driver.cc:65]277 GetDeviceCount:Call drvGetDevNum, drvRetCode=87.

[ERROR] RUNTIME(277,python):2025-03-25-01:36:08.062.904 [api_c.cc:1714]277 rtGetDeviceCount:ErrCode=507899, desc=[driver error:internal error], InnerCode=0x7020010

[ERROR] RUNTIME(277,python):2025-03-25-01:36:08.063.003 [error_message_manage.cc:53]277 FuncErrorReason:report error module_type=3, module_name=EE8888

[ERROR] RUNTIME(277,python):2025-03-25-01:36:08.063.107 [error_message_manage.cc:53]277 FuncErrorReason:rtGetDeviceCount execute failed, reason=[driver error:internal error]

[ERROR] ASCENDCL(277,python):2025-03-25-01:36:08.063.317 [device.cpp:343]277 aclrtGetDeviceCount: get device count failed, runtime result = 507899.

[ERROR] APP(277,python):2025-03-25-01:36:08.063.464 [log_inner.cpp:76]277 compiler_depend.ts:device_count:21: "[PTA]:"get device count of NPU failed""

[ERROR] RUNTIME(277,python):2025-03-25-01:36:08.263.604 [driver.cc:65]277 GetDeviceCount:Call drvGetDevNum, drvRetCode=87.

[ERROR] RUNTIME(277,python):2025-03-25-01:36:08.263.774 [api_c.cc:1714]277 rtGetDeviceCount:ErrCode=507899, desc=[driver error:internal error], InnerCode=0x7020010

[ERROR] RUNTIME(277,python):2025-03-25-01:36:08.263.865 [error_message_manage.cc:53]277 FuncErrorReason:report error module_type=3, module_name=EE8888

[ERROR] RUNTIME(277,python):2025-03-25-01:36:08.263.953 [error_message_manage.cc:53]277 FuncErrorReason:rtGetDeviceCount execute failed, reason=[driver error:internal error]

[ERROR] ASCENDCL(277,python):2025-03-25-01:36:08.264.065 [device.cpp:343]277 aclrtGetDeviceCount: get device count failed, runtime result = 507899.

[ERROR] APP(277,python):2025-03-25-01:36:08.264.168 [log_inner.cpp:76]277 compiler_depend.ts:device_count:21: "[PTA]:"get device count of NPU failed""

False

False

False

[INFO] RUNTIME(277,python):2025-03-25-01:36:09.022.527 [runtime.cc:1831] 277 ~Runtime: deconstruct runtime

[INFO] RUNTIME(277,python):2025-03-25-01:36:09.038.450 [runtime.cc:1838] 277 ~Runtime: wait monitor success, use=0.

我要发帖子