Hi. everyone.
I installed cuda toolkit and SDK. But SDK projects failed when executing cudaMalloc function.
“out of memory”… in nvidia 8800 gtx. ok. i know it has 768mega memory. ;;
first. i’m using ubuntu desktop 6.10(x86 version).
second. i installed nividia x86 1.0 9751 driver and SDK 0.8.1 and cuda 0.8
third. and i, intel core2duo desktop with 2 gigabyte main memory.
i have many reasons that this small code will be running smooth. but another SDK projects has same problem. How can i get memory… :(
this is a test code and an error message.
$ cat test.cu
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <cutil.h>
int
main(int argc, char** argv)
{
CUT_CHECK_DEVICE();
// allocate device memory
float* d_A;
CUDA_SAFE_CALL(cudaMalloc((void **)&d_A, 256)); // line 12
// clean up memory
CUDA_SAFE_CALL(cudaFree(d_A));
CUT_EXIT(argc, argv);
}
$ nvcc test2.cu -D_DEBUG -I ~/NVIDIA_CUDA_SDK/common/inc -L ~/NVIDIA_CUDA_SDK/common/lib -L ~/NVIDIA_CUDA_SDK/lib -l cutil -l GL -l GLU
$ ./a.out
Cuda error in file 'test2.cu' in line 12 : out of memory.