Hi, I am trying to install the tensorflow gpu framework and I got some troubles with the cudnn. I run the mnistCUDNN sample to verify mu installation and I got the below output:
cudnnGetVersion() : 7605 , CUDNN_VERSION from cudnn.h : 7605 (7.6.5)
Host compiler version : GCC 7.5.0
There are 1 CUDA capable devices on your machine :
device 0 : sms 22 Capabilities 7.5, SmClock 1815.0 Mhz, MemSize (Mb) 5911, MemClock 4001.0 Mhz, Ecc=0, boardGroupID=0
Using device 0
Testing single precision
Loading image data/one_28x28.pgm
Performing forward propagation ...
Testing cudnnGetConvolutionForwardAlgorithm ...
Fastest algorithm is Algo 0
Testing cudnnFindConvolutionForwardAlgorithm ...
^^^^ CUDNN_STATUS_SUCCESS for Algo 0: 0.030720 time requiring 0 memory
^^^^ CUDNN_STATUS_SUCCESS for Algo 1: 0.069536 time requiring 3464 memory
^^^^ CUDNN_STATUS_SUCCESS for Algo 2: 0.071744 time requiring 57600 memory
^^^^ CUDNN_STATUS_SUCCESS for Algo 5: 0.103584 time requiring 203008 memory
^^^^ CUDNN_STATUS_SUCCESS for Algo 7: 0.118624 time requiring 2057744 memory
Cublas failure
Error code 0
gemv.h:77
Aborting...
Some more information:
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 418.87.00 Driver Version: 418.87.00 CUDA Version: 10.1 |
|-------------------------------+----------------------+----------------------+
| 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 GTX 1660 Off | 00000000:01:00.0 On | N/A |
| 27% 41C P8 9W / 120W | 211MiB / 5911MiB | 1% Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 989 G /usr/lib/xorg/Xorg 78MiB |
| 0 1438 G cinnamon 31MiB |
| 0 8622 G ...uest-channel-token=16736224539216711033 99MiB |
+-----------------------------------------------------------------------------+
CUDNN_VERSION from cudnn.h : 7605 (7.6.5)
Host compiler version : GCC 7.5.0
Tensorflow: 2.1.0-rc0;
CUDNN lib is in my LD_LIBRARY_PATH
What did I do wrong? How can I pass the mnistCUDNN test?