NetNews Usenet Archive 1992 #16/NN_1992_16.iso/spool/comp/lang/c/11771

Internet Message Format | 1992-07-30 | 2.3 KB

Path: sparky!uunet!crdgw1!rdsunx.crd.ge.com!bart!volpe
From: [e-mail] (Christopher R Volpe)
Newsgroups: comp.lang.c
Subject: Re: Pointers to freed memory
Message-ID: <[e-mail]>
Date: 30 Jul 92 18:09:08 GMT
References: <[e-mail]> <[e-mail]>
Sender: volpe@bart (Christopher R Volpe)
Reply-To: [e-mail]
Organization: GE Corporate Research & Development
Lines: 49
Nntp-Posting-Host: bart.crd.ge.com

In article <[e-mail]>, [e-mail] (Jos Horsmeier) writes:
|> In article <[e-mail]> [e-mail] (M. J. Saltzman) writes:
|> |I know that a pointer to freed memory should never be dereferenced,
|> |but is it standard-conforming to assign or do arithmetic with such
|> |pointers?
|> |
|> |The application is a routine to free the elements of a circular list.
|> |Since I have a count of the elements, I can simply run a pair of pointers
|> |around the list, like so:
|> |
|> |    for ( p = head, i = 0;  i < n;  i++ ) {
|> |        q = p;
|> |        p = p->next;
|> |        free(q);
|> |    }
|> |
|> |The question comes up because on the last iteration, p->next points
|> |to the element freed in the first iteration.  Do I need to break the
|> |circle before I start the loop? Or am I just being paranoid after the
|> |discussion of pointers to invalid array elements?
|>
|> Erm, yes, I hate to say it, but you're getting paranoid ;-)

I'd say that's true for most real compilers, but...

|>
|> As long as you do not dereference such a pointer (as you indicated
|> yourself,) everything is fine.  If pointer q contained a valid pointer

I disagree that this will always be true. There is nothing to prevent
an implementation from trapping assignments of invalid addresses to
pointer variables. I wouldn't be surprised if a debugging C interpreter
(ala Saber) did this someday.

|> value _before_ the call to free(), it'll contain a valid pointer value
|> _after_ the call. In the last iteration, the value p->next points to
|> a location in memory where the first element of the list _would have
|> been_ if it weren't freed in the first iteration. And, again, if you
|> don't dereference that pointer value, everything is safe and sound.
|>
|> kind regards,
|>
|> Jos aka [e-mail]

--
==================
Chris Volpe
G.E. Corporate R&D
[e-mail]