ufoym/deepo

cannot dlopen some GPU libraries

ytwas opened this issue · 1 comments

ytwas commented

i can see nvidia-smi, but if i run tensorflow.test.is_gpu_available, it returns false

nvidia-docker run --gpus all --rm ufoym/deepo nvidia-smi    1434  09:57:45 

Thu Dec 26 01:58:21 2019
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.44 Driver Version: 440.44 CUDA Version: 10.2 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce RTX 208... Off | 00000000:02:00.0 Off | N/A |
| 0% 47C P8 20W / 260W | 1107MiB / 11019MiB | 11% Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
+-----------------------------------------------------------------------------+

import tensorflow as tf
tf.test.is_gpu_available()

2019-12-26 02:30:04.430145: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
2019-12-26 02:30:04.479781: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 3600000000 Hz
2019-12-26 02:30:04.484737: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x44956f0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:
2019-12-26 02:30:04.484935: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version
2019-12-26 02:30:04.490946: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1
2019-12-26 02:30:04.671031: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:984] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-12-26 02:30:04.675634: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x44d7bd0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:
2019-12-26 02:30:04.675718: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): GeForce RTX 2080 Ti, Compute Capability 7.5
2019-12-26 02:30:04.676496: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:984] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-12-26 02:30:04.677042: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1551] Found device 0 with properties:
name: GeForce RTX 2080 Ti major: 7 minor: 5 memoryClockRate(GHz): 1.635
pciBusID: 0000:02:00.0
2019-12-26 02:30:04.677932: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcudart.so.10.0'; dlerror: libcudart.so.10.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/nvidia/lib:/usr/local/nvidia/lib64
2019-12-26 02:30:04.678208: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcublas.so.10.0'; dlerror: libcublas.so.10.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/nvidia/lib:/usr/local/nvidia/lib64
2019-12-26 02:30:04.678309: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcufft.so.10.0'; dlerror: libcufft.so.10.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/nvidia/lib:/usr/local/nvidia/lib64
2019-12-26 02:30:04.678411: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcurand.so.10.0'; dlerror: libcurand.so.10.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/nvidia/lib:/usr/local/nvidia/lib64
2019-12-26 02:30:04.678506: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcusolver.so.10.0'; dlerror: libcusolver.so.10.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/nvidia/lib:/usr/local/nvidia/lib64
2019-12-26 02:30:04.678612: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcusparse.so.10.0'; dlerror: libcusparse.so.10.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/nvidia/lib:/usr/local/nvidia/lib64
2019-12-26 02:30:04.689660: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
2019-12-26 02:30:04.689768: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1574] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
2019-12-26 02:30:04.689916: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1092] Device interconnect StreamExecutor with strength 1 edge matrix:
2019-12-26 02:30:04.689948: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1098] 0
2019-12-26 02:30:04.689967: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1111] 0: N
False