cudaGetDeviceProperties is showing memory as 0

Accelerated Computing CUDA CUDA Programming and Performance
1

hi,
Firstly i started using cuda beta 0.8 in Fedora 6. i am using a nvidia geforce 8800 gtx card. after installing both cuda and sdk, i tried to compile the sample programs. but it was giving some error. am using linux driver 1.0-9755.

Now i am using RHEL ES release 4. now the programs are compiled without any errors… but when i run them most of the programs fails…

i have also tried my hand on device managemet programs… it looks like below-----

#include<stdio.h>

int main(int argc, char** argv)
{
int ct,dev;
struct cudaDeviceProp prop;

cudaGetDeviceCount(&ct);

if(ct == 0)
{
   printf("\nNo Device Found\n");
   exit(0);
}
else
{
   printf("\n%d Device Found\n",ct);

   cudaGetDevice(&dev);
   printf("\nThe Device ID is %d\n",dev);

   cudaGetDeviceProperties(&prop,dev);
   printf("\nThe Properties of the Device with ID %d are\n",dev);
   printf("\tDevice Name : %s",prop.name);
   printf("\n\tDevice Memory Size (in bytes) : %u",prop.bytes);
   printf("\n\tDevice Major Revision Numbers : %d",prop.major);
   printf("\n\tDevice Minor Revision Numbers : %d",prop.minor);
}
printf("\n");

}

and the ouput is as below…

1 Device Found

The Device ID is 0

The Properties of the Device with ID 0 are
Device Name : GeForce 8800 GTX
Device Memory Size (in bytes) : 0
Device Major Revision Numbers : 1
Device Minor Revision Numbers : 0
~

        Here all other functions give the correct results.... only the device memory is showing 0.

can any body guide me in this regard …

thanks…
farid khan

2

Try to use the official beta display driver (97.51) and see if the problems are still there.

3

Hi ,

Thanks…a lot… It worked… Even the memory allocation was not working… now its working…

farid…