Hi,
I am running Fedora 10 on a Dell Latitude D830 (equipped with a Quadro NVS 135M), and while I have successfully installed the nVidia accelerated driver from the distro (v 177.82), my card seems not to be recognized by the CUDA libs (v2.0.)
I tried this code:
#include <stdio.h>
int main()
{
int devs;
cudaDeviceProp dev;
cudaGetDeviceCount(&devs);
cudaGetDeviceProperties(&dev, 0);
printf("total devices: %i | name: %s clock: %i cores: %i major: %i\n",
devs, dev.name, dev.clockRate, dev.multiProcessorCount, dev.major);
return 0;
}
and this is what I got:
[cloud@cloud cuda]$ nvcc test.cu
[cloud@cloud cuda]$ ./a.out
total devices: 1 | name: Device Emulation (CPU) clock: 1350000 cores: 16 major: 9999
[cloud@cloud cuda]$
What am I doing wrong? I checked the release notes, and the card should be supported. 3D acceleration works nicely, so I assume that there is no issue with the nvidia driver itself (except perhaps that it comes from the distro’s repos.)
Thanks in advance for your time.
Claudio A.