customerize operater new to detect memory underrun and overruns

Robertlzw (103)
Hi folks,

I am looking at the following pseudocode:
1
2
3
4
5
6
7
8
9
10
11
12
13
 
static const int signature = 0xDEADBEFF;
typedef unsigned char Byte;
void* operator new(size_t size) throw (std::bad_alloc)
{
	using namespace std;
	size_t realSize = size + 2 * sizeof(int);
	void* pMem = malloc(realSize);
	if (!pMem) throw bad_alloc();

	*(static_cast<int*>(pMem) ) = signature;
	*(reinterpret_cast<int*>(static_cast<Byte*>(pMem) + realSize - sizeof(int))) = signature;
	return static_cast<Byte*>(pMem) + sizeof(int);
}

Since the memory has been allocated in terms of std::size_t (the malloc statement on line 7), can I simply NOT typedef unsigned char as Byte and use std::size_t instead? So my pseudocode looks like this:
1
2
3
4
5
6
7
8
9
10
11
12
 
static const int signature = 0xDEADBEFF;
void* operator new(size_t size) throw (std::bad_alloc)
{
	using namespace std;
	size_t realSize = size + 2 * sizeof(int);
	void* pMem = malloc(realSize);
	if (!pMem) throw bad_alloc();

	*(static_cast<int*>(pMem) ) = signature;
	*(reinterpret_cast<int*>(static_cast<std::size_t*>(pMem) + realSize - sizeof(int))) = signature;
	return static_cast<std::size_t*>(pMem) + sizeof(int);
}

Cheers,
Robert
Last edited on
jsmith (5804)
some_ptr + sizeof( int )

yields different addresses, depending upon the type of some_ptr, so I would say that lines 11 and 12 in the original
code do something different than lines 10 and 11 in the new code.
Topic archived. No new replies allowed.