View Issue Details

IDProjectCategoryView StatusLast Update
0005354Kali Linux[All Projects] Kali Package Improvementpublic2019-04-08 22:29
ReporterrcogdenAssigned To 
PrioritynormalSeveritymajorReproducibilityalways
Status newResolutionopen 
Product Version2019.1 
Target VersionFixed in Version 
Summary0005354: Unable to run Tensorflow-gpu in Kali-2019.1
DescriptionTensorflow-gpu 1.12 has Nvidia driver runtime error in Kali-2019.1. Need Nvidia driver 396.26 or newer for CUDA Toolkit 9.2. See table 1 in https://docs.nvidia.com/deeplearning/sdk/pdf/cuDNN-Support-Matrix.pdf. Thanks
Steps To ReproduceBuild tensorflow-gpu 1.12 from source. Run simple Tensorflow "hello world" program. Get error:

$ python
Python 2.7.15+ (default, Feb 3 2019, 13:13:16)
[GCC 8.2.0] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow as tf
>>> hello = tf.constant('Hello Tensorflow!')
>>> sess = tf.Session()
2019-03-18 00:02:17.807344: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:964] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-03-18 00:02:17.808150: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1432] Found device 0 with properties:
name: GeForce GTX 1060 major: 6 minor: 1 memoryClockRate(GHz): 1.6705
pciBusID: 0000:01:00.0
totalMemory: 5.93GiB freeMemory: 5.70GiB
2019-03-18 00:02:17.808191: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1511] Adding visible gpu devices: 0
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 1551, in __init__
    super(Session, self).__init__(target, graph, config=config)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 676, in __init__
    self._session = tf_session.TF_NewSessionRef(self._graph._c_graph, opts)
tensorflow.python.framework.errors_impl.InternalError: cudaGetDevice() failed. Status: CUDA driver version is insufficient for CUDA runtime version
>>>

Activities

steev

2019-04-08 22:29

developer   ~0010491

Which version of nvidia-driver do you have installed? We have version 410.104 in the repos, which should be new enough.

Issue History

Date Modified Username Field Change
2019-03-18 07:06 rcogden New Issue
2019-04-08 22:29 steev Note Added: 0010491