Argument 13 is invalid. culaSgemm

[cs10b026]

Hi,

I am trying to multiply two very small matrices using culaSgemm. But I am getting the following error

Argument 13 is invalid (see the Reference Manual for more information)

Here is the code I am using.
1 #include <stdio.h>
2 #include <stdlib.h>
3 #include <string.h>
4
5 #include <cula_lapack_device.h>
6
7 #include <cuda_runtime.h>
8 #ifdef _MSC_VER
9 # pragma comment(lib, "cudart.lib")
10 #endif
11
12 void checkCudaError(cudaError_t err)
13 {
14 if(!err)
15 return;
16
17 printf("%s\n", cudaGetErrorString(err));
18
19 culaShutdown();
20 exit(EXIT_FAILURE);
21 }
22
23 int main()
24 {
25 culaStatus status;
26 culaFloat* a = NULL;
27 culaFloat* b = NULL;
28 culaFloat* c = NULL;
29
30 a = (culaFloat*)malloc(9 * sizeof(culaFloat));
31 b = (culaFloat*)malloc(9 * sizeof(culaFloat));
32 c = (culaFloat*)malloc(9 * sizeof(culaFloat));
33
34 int i;
35
36 for(i = 0; i < 9; i++)
37 a[i] = b[i] = c[i] = 1;
38
39 status = culaInitialize();
40
41 checkCudaError(status);
42
43 status = culaSgemm('N', 'N', 3, 3, 3, 1, a, 3, b, 3, 1, c, 3);
44
45 checkCudaError(status);
46
47 return 0;
48 }

Can someone tell me what I am doing wrong?

Thank you.