matrix class mpi

[andrix]

Hi everyone,

i have implemented my matrix class as follows:

class matrix{
private:
unsigned rows;
unsigned cols;
bool **m;
void allocate(bool ***,unsigned, unsigned){

*ptr=new bool*[r];
(*ptr)[0]=new bool[r*c];
for(unsigned i=1;i<r;++i)
(*ptr)[i]=(*ptr)[0]+i*c;

}


.....
....
};

now i should parallel it with MPI, for example to compute the matrix vector product;
In particular i would like to share it row-wise between the processors, but I don't know how to do the send.

Can anybody help me?