Thread: fill optimizations

I have the following recursive implementation of the fill algorithm:

Code:

void fill(int x, int y)
{		
	if ( mat[x][y] == 0 )
        {
	        mat[x][y] = 1;
		fill(x+1, y);
		fill(x-1, y);
		fill(x, y+1);
		fill(x, y-1);
	}
}

I'm wondering, is this efficient? I read that the recursion doesn't slow things down in this case as much as it does in other algorithms, is this true?

Would an interative approach give a significant increase in speed? Is it worth implementing it iteratively, and if so, how would I approach that?