Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.os.msdos.programmer > #442

Re: Persistent memory during reset

Cross-posted to 3 groups.

Show all headers | View raw

"Adrian Smith" <adrian.maxwell.smith@nospicedham.googlemail.com> wrote in
message
news:de96da1c-b329-4bab-9761-0e8bf36f0686@h3g2000yqa.googlegroups.com...
>
> Is there any standard region of BIOS or x86 platform memory
> that persists after a system reset.
>

In general, not that I'm aware of.  It would "open the door" to persistent
malware.

You may be able to use CMOS.  IIRC, the early PCs had fewer CMOS registers,
like 64, but modern PCs have more, like 128.  So, I think a bunch of them
may be free or useable, i.e., > 64.  RBIL's CMOS.LST file gives background
information on which values are standardized.  Basically, CMOS registers
below 64, except a few are standardized.  So, there are a few below 64, and
possibly 64 above 64.  You probably won't know if a register is in use by
your BIOS unless you change it and cause a problem ...  Personally, instead
of looking for a free register, I would read CMOS.LST looking for a register
that's already in-use or defined, but one that can be changed and has little
or no impact on the system.

If you do a 286 style PM reset with correct settings, all memory used prior
to the reset should be intact.  However, the BIOS reset code for a warm or
cold boot won't be executed.  You'll want to read up on the shutdown code
for CMOS register 0Fh.  Value of zero is a normal reset.  Values of 5 or
0Ah, upon reset, the BIOS code will jump to the RM address in 40:67h.  I'm
not sure if 40:72h reset flag needs to be zero (cold boot) or 1234h (warm
boot) or if it doesn't matter.  There are other values for CMOS 0Fh and
40:72 that I've not mentioned.  RBIL covers this in a very generic way in
MEMORY.LST sections M00400067, M00400072 and in CMOS.LST
section R0F.  Otherwise, there are just a hanful of posts to Usenet on the
topic over the decades, so you may need a good book on programming
for more info.


Rod Pemberton

Back to comp.os.msdos.programmer | Previous | Next — Next in thread | Find similar

Thread

Re: Persistent memory during reset "Rod Pemberton" <do_not_have@nospicedham.noavailemail.cmm> - 2011-11-30 21:03 -0500
  Re: Persistent memory during reset "wolfgang kern" <nowhere@never.at> - 2011-12-03 00:07 +0100
  Re: Persistent memory during reset Ruud Baltissen <rbaltissen@nospicedham.gmail.com> - 2011-12-29 09:44 -0800
    Re: Persistent memory during reset Robert Wessel <robertwessel2@nospicedham.yahoo.com> - 2011-12-30 01:34 -0600

csiph-web