Why do these two pointer subtractions give different results?

Question

Consider the following code:

char* p = new char[2];
long* pi = (long*) p;
assert(p == pi);         // OK

char* p1 = &p[1];
long* pi1 = (long*) p1;
asser         


        

Answer 1

As others have pointed, this is undefined behavior. However, there is a very simple explanation for what you are seeing.

The difference between pointers is the number of elements, not the number of bytes between them.

pi and pi1 both point to longs, but the address pointed to by pi1 is only one byte further than pi. Presuming longs are 4 bytes long, the difference in the addresses, 1, divided by the size of the element, 4, is 0.

Another way of thinking of this is you could imagine the compiler would generate code equivalent to this for calculating d1:

int d1 = ((BYTE*)pi1 - (BYTE*)pi)/sizeof(long).