Re: Pointer Testing
Re: Pointer Testing
- Subject: Re: Pointer Testing
- From: Jonas Maebe <email@hidden>
- Date: Sat, 15 Feb 2003 18:18:33 +0100 (CET)
On Sat, 15 Feb 2003, Brent Marykuca wrote:
>
Yes, I know that C itself has nothing to help, but surely the
>
information is available to the kernel. The virtual memory system must
>
maintain tables of valid/invalid address ranges and read/write
>
permissions for every task, so it's not inconceivable that such an API
>
could be made available.
I don't think that the information that the kernel has is very useful and
would only give you a false sense of security. The reason is that most
memory allocators request big blocks from the kernel and then give
pointers inside these blocks to the application when it requests memory.
So according to the kernel, a pointer may be perfectly OK, while it could
point to a non-allocated or freed block.
Jonas
_______________________________________________
darwin-kernel mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/darwin-kernel
Do not post admin requests to the list. They will be ignored.