problem with cudaMallocPitch

My CUDA application has suddenly stopped working

it now hangs/freezes when it gets to the cudaMallocPitch

is there anything i should know about cudaMallocPitch

has anyone had this problem

can anyone suggest a solution?

float *g_matrix=(float*)0;

size_t pitch;

int neq=5;

int ncons=6;

size = ncons*sizeof(float)+neq*sizeof(float)+sizeof(float);

cudaMallocPitch((void**)&g_matrix,&pitch,(size_t)size,(size_t)neq);

can anyone see anything in my code that would make it hang/freeze?

i have encounter this question also, why blocked,too large ??