From numpy matrix to C array. Segmentation fault (memory corruption) on 64bit architecture -

i'm trying build python c extension in order pass numpy matrix c array. following suggestions reported here:

http://wiki.scipy.org/cookbook/c_extensions/numpy_arrays

but when python tries run c line:

v=(float **)malloc((size_t) (n*sizeof(float))); 

of follow portion of code:

float **_ptrvector(long n) {      float **v;     v=(float **)malloc((size_t) (n*sizeof(float)));     if (!v)   {         printf("in **ptrvector. allocation of memory array failed.");         exit(0); }     return v; }  float **pymatrix_to_carray(pyarrayobject *arrayin)  {     float **c, *a;     int i,n,m;      n = pyarray_dim(arrayin, 0);     m = pyarray_dim(arrayin, 1);     c=_ptrvector(n);     = (float*) pyarray_data(arrayin);     ( i=0; i<n; i++)  {         c[i]=a+i*m;  }       return c; } 

i segmentation fault on linux 64 bit machine. problem code works on 32 bit machines (both windows , linux). sizeof(size_t) returns correctly 4 on 32bit machines , 8 on 64bit machines. i'm using python 2.7.1 [epd 7.0-2 (64-bit)] , gcc 4.1.2 on red hat linux 4.2.1-44 kernel version 2.6.18.

i tried npy_intp , py_ssize_t instead of size_t no positive effect.

here want allocate memory pointers float:

float **v; v=(float **)malloc((size_t) (n*sizeof(float))); 

but allocated memory floats themselfes. on 32bit system pointers need 4 bytes, works.

on 64bit system pointers need 8 byte, might change above line be:

float ** v = malloc(n * sizeof(float*)); 

or more secure

float ** v = malloc(n * sizeof(*v)); 

---

btw: in c not needed caste result of malloc/calloc/realloc nor recommended. more dangerous might hide awful mistakes.