unable to create StreamExecutor for CUDA:0: failed initializing StreamExecutor for CUDA device...

环境

• Ubuntu: 16.04
• CUDA: 10.0
• CUDNN: 7.6.5

执行情况

$ python3 test.py
2020-10-13 15:20:14.602211: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1
2020-10-13 15:20:14.605874: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1006] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-10-13 15:20:14.606291: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1618] Found device 0 with properties: 
name: Tesla V100-SXM2-32GB major: 7 minor: 0 memoryClockRate(GHz): 1.53
pciBusID: 0000:00:08.0
2020-10-13 15:20:14.606480: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0
2020-10-13 15:20:14.607907: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10.0
2020-10-13 15:20:14.609115: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10.0
2020-10-13 15:20:14.609426: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10.0
2020-10-13 15:20:14.611090: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10.0
2020-10-13 15:20:14.612343: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10.0
2020-10-13 15:20:14.616370: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
2020-10-13 15:20:14.616475: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1006] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-10-13 15:20:14.616929: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1006] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-10-13 15:20:14.617299: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1746] Adding visible gpu devices: 0
2020-10-13 15:20:14.617549: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
2020-10-13 15:20:14.625251: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 2499995000 Hz
2020-10-13 15:20:14.625774: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x560b01298670 executing computations on platform Host. Devices:
2020-10-13 15:20:14.625802: I tensorflow/compiler