[GitHub] BELZHANG opened a new issue #11125: mxnet 1.2.0 GPU error

[GitBox <gi...@apache.org>]

BELZHANG opened a new issue #11125: mxnet 1.2.0 GPU error
URL: https://github.com/apache/incubator-mxnet/issues/11125
 
 
   Note: Providing complete information in the most concise form is the best way to get help. This issue template serves as the checklist for essential information to most of the technical issues and bug reports. For non-technical issues and feature requests, feel free to present the information in what you believe is the best form.
   
   For Q & A and discussion, please start a discussion thread at https://discuss.mxnet.io 
   
   ## Description
   HI ALL, I'm getting this error when try to run this code  a = mx.nd.ones((2, 3), mx.gpu())
   
   terminate called after throwing an instance of 'dmlc::Error'
     what():  [18:12:18] /home/travis/build/dmlc/mxnet-distro/mxnet-build/3rdparty/mshadow/mshadow/./stream_gpu-inl.h:182: Check failed: e == cudaSuccess CUDA: unknown error
   
   Stack trace returned 10 entries:
   [bt] (0) /home/chenxi/anaconda2/envs/mxnet/lib/python2.7/site-packages/mxnet/libmxnet.so(+0x30cbe2) [0x7fb294d46be2]
   [bt] (1) /home/chenxi/anaconda2/envs/mxnet/lib/python2.7/site-packages/mxnet/libmxnet.so(+0x30d1b8) [0x7fb294d471b8]
   [bt] (2) /home/chenxi/anaconda2/envs/mxnet/lib/python2.7/site-packages/mxnet/libmxnet.so(+0x293b9c8) [0x7fb2973759c8]
   [bt] (3) /home/chenxi/anaconda2/envs/mxnet/lib/python2.7/site-packages/mxnet/libmxnet.so(+0x293c394) [0x7fb297376394]
   [bt] (4) /home/chenxi/anaconda2/envs/mxnet/lib/python2.7/site-packages/mxnet/libmxnet.so(+0x294ea9f) [0x7fb297388a9f]
   [bt] (5) /home/chenxi/anaconda2/envs/mxnet/lib/python2.7/site-packages/mxnet/libmxnet.so(+0x294ec6e) [0x7fb297388c6e]
   [bt] (6) /home/chenxi/anaconda2/envs/mxnet/lib/python2.7/site-packages/mxnet/libmxnet.so(+0x294b46b) [0x7fb29738546b]
   [bt] (7) /home/chenxi/anaconda2/envs/mxnet/bin/../lib/libstdc++.so.6(+0xafc5c) [0x7fb2d9701c5c]
   [bt] (8) /lib/x86_64-linux-gnu/libpthread.so.0(+0x76ba) [0x7fb2e00bc6ba]
   [bt] (9) /lib/x86_64-linux-gnu/libc.so.6(clone+0x6d) [0x7fb2df6e241d]
   
   I have serached  this problem online, and it looks like most of the cases is casued by wrong cuda driver or graphic card driver, i have tried different version of cuda(9.0, 8.0(i install with mxnet-cu80)), different nvidia driver and all of them give me same error.
   so what is the correct version of cuda ,cudnn and nvidia driver should i use for mxnet 1.2.0?
   By the way, i have already verify the correctness of cuda installation before doing this.
   
   Thank you!
   ## Environment info (Required)
   ```
   What to do:
   1.Ubuntu 16.04
   2.I'm using conda and use pip to install mxnet in the conda enviroment
   3.cuda version 9.0 with cudnn 7
   4.nvidia driver 384.130
   ```
   
   Package used (Python/R/Scala/Julia):
   mxnet-cu90
   
   
   
   
   ## What have you tried to solve it?
   
   1.try different verison of cuda driver and graphic card driver
   
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services