NetNews Usenet Archive 1992 #27/NN_1992_27.iso/spool/comp/std/c/3032

Internet Message Format | 1992-11-18 | 1.7 KB

Path: sparky!uunet!ferkel.ucsb.edu!taco!rock!stanford.edu!agate!doc.ic.ac.uk!uknet!acorn!ixi!clive
From: [e-mail] (Clive Feather)
Newsgroups: comp.std.c
Subject: Re: fwrite+fread of pointer
Message-ID: <[e-mail]>
Date: 18 Nov 92 13:18:09 GMT
References: <[e-mail]> <[e-mail]> <[e-mail]>
Organization: IXI Limited, Cambridge, UK
Lines: 24

In article <[e-mail]> [e-mail] (Richard A. O'Keefe) writes:
> The question can now be paraphrased:  is a pointer like a tensor, or like
> the components of a tensor?  I would regard a compacting garbage collector
> as simply changing the co\"ordinate system.  I have used hardware and
> software systems where stacks could be moved (in order to "stretch" them)
> while the owning thread was running.

An "object" is defined (ISO 3.14) as a "region of storage". I would read
this as meaning that, for the lifetime of that object, a given object
maps to a given region of storage. Therefore, since the address of an
object is a pointer to it, the address must also be a constant. I agree
that the ice is thin and an RFI would be a good idea. You might also ask
whether &external_object_or_function has to be the same in all the
translation units of a program.

This is all, of course, modulo the "as-if" rule. Mark Brader has shown a
way to have objects move in memory whilst keeping constant addresses;
this doesn't alter what I have said.

--
Clive D.W. Feather     | IXI Limited         | If you lie to the compiler,
[e-mail]          | Vision Park         | it will get its revenge.
Phone: [phone] | Cambridge   CB4 4RZ |   - Henry Spencer
Fax:   [phone] | United Kingdom      |