delorie.com/archives/browse.cgi   search  
Mail Archives: djgpp/1999/05/15/11:51:43.2

From: Endlisnis <s257m AT unb DOT ca>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: ? array[n] or array[n+1] ?
Date: Sat, 15 May 1999 00:31:39 -0400
Organization: BrunNet
Lines: 40
Message-ID: <373CF8AB.A4CFA442@unb.ca>
References: <37367f71 DOT 6271443 AT news DOT wanadoo DOT fr> <3736A098 DOT 55E268BE AT softhome DOT net> <373c0195 DOT 3902906 AT news DOT wanadoo DOT fr> <3737f542 DOT 0 AT nnrp1 DOT news DOT uk DOT psi DOT net> <3739F455 DOT 1818FDD7 AT unb DOT ca> <373a1915 DOT 3102234 AT noticias DOT iies DOT es> <7hfenm$k6m$1 AT news4 DOT svr DOT pol DOT co DOT uk> <373c4f0a DOT 7518034 AT noticias DOT iies DOT es> <373B63CB DOT EBFAE296 AT unb DOT ca> <373c6ff0 DOT 1828476 AT noticias DOT iies DOT es>
NNTP-Posting-Host: ftnts2c9.brunnet.net
Mime-Version: 1.0
X-Mailer: Mozilla 4.51 [en] (Win95; U)
X-Accept-Language: en
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

Guillermo Rodriguez Garcia wrote:

> >    You aren't writting to the hard-drive, you are writing to RAM.  Now this
> >situation is VERY UNLIKELY, but it is still possible:
> [...]
> Are you sure that you can write to CMOS memory by writing to RAM? I
> think this is not possible.

   It isn't a direct approach.  We know that you can write code that writes to CMOS.
If you were to overwrite the return address of your function, and it were to jump into
memory that had instructions to overwrite CMOS (which could also have been put there
by jumping your array bound).  That is how by just writting into RAM can cause (a few
clock-cycles later) your CMOS to be overwritten.  I'm not arguing this is likely or
easy to produce, but it is possible.

> >    It's not only the configuration you may loose, but the configuration PROGRAM
> >which BIOS loads.  That could be very bad.  Most boards these days allow upgrades
> >to the CMOS program to fix bugs, so this is a viable option.
> But this program is in flash ROM, not RAM, AFAIK.
>
> Anyway, I still ask: what hardware is being damaged here? It is all a
> mater of software / data / configuration. Nothing is broken. No need
> to buy new parts or replace any IC or component.

    It has to be on reprogrammable chip if it wants to be updated by the vendor.  If
the vendor can update it, that means that you data could be machine code to overwrite
the "flash ROM" or whatever.
    You are right about that not causing any direct hardware damage, but as I
mentioned before, my CMOS controls my voltage to my processor, so if it screwed up, I
would smell smoke very soon.

--
     (\/) Endlisnis (\/)
          s257m AT unb DOT ca
          Endlisnis AT HotMail DOT com
          ICQ: 32959047




- Raw text -


  webmaster     delorie software   privacy  
  Copyright © 2019   by DJ Delorie     Updated Jul 2019