Re: whats wrong with analysizing pointers this way ?

=?ISO-8859-1?Q?Jan_Vorbr=FCggen?= <jvorbrueggen-not@mediasec.de>
24 Feb 2006 09:38:14 -0500

          From comp.compilers

Related articles
whats wrong with analysizing pointers this way ? drizzle76@gmail.com (dz) (2006-02-19)
Re: whats wrong with analysizing pointers this way ? DrDiettrich@compuserve.de (Hans-Peter Diettrich) (2006-02-20)
Re: whats wrong with analysizing pointers this way ? jvorbrueggen-not@mediasec.de (=?ISO-8859-1?Q?Jan_Vorbr=FCggen?=) (2006-02-24)
Re: whats wrong with analysizing pointers this way ? shreyas76@gmail.com (shrey) (2006-02-24)
Re: whats wrong with analysizing pointers this way ? david.thompson1@worldnet.att.net (Dave Thompson) (2006-03-05)
Re: whats wrong with analysizing pointers this way ? jvorbrueggen@mediasec.de (=?ISO-8859-1?Q?Jan_Vorbr=FCggen?=) (2006-03-11)
| List of all articles for this month |
From: =?ISO-8859-1?Q?Jan_Vorbr=FCggen?= <jvorbrueggen-not@mediasec.de>
Newsgroups: comp.compilers
Date: 24 Feb 2006 09:38:14 -0500
Organization: MediaSec Technologies GmbH
References: 06-02-133 06-02-147
Keywords: analysis
Posted-Date: 24 Feb 2006 09:38:14 EST

> Where "pointer" does not only mean, that such variables contain
> addresses, that's just a convention. You also must open the can of
> worms, by allowing access to memory locations, using the *values* stored
> in pointer variables, and by introducing pointer arithmetic, so that
> references can end up in some variable near the variable pointed to, or
> in unallocated memory.


That additional functionality depends, of course, on the language those
pointers are used in. Fortran, for instance, has pointers that just don't
allow such things to be done; thus, the chance of worm-holes occuring is
quite low - actually, if you only write standard-conforming programs, there
are none.


Jan



Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.