reading ROM BIOS

From: Allan Adler (spamtrap_at_crayne.org)
Date: 09/28/04 

Date: Tue, 28 Sep 2004 19:43:15 +0000 (UTC)

I have a Dell Latitude CsX laptop running RedHat 7.1 Linux and I'm trying
to read the ROM BIOS. I copied the BIOS to a file and diassembled it
using nasm-0.98.38 and now I'm trying to read it. My goal is to understand
more about the hardware of the PC by observing the "conversation" that takes
place between the CPU and the other hardware on bootup. I've already posted
to comp.os.linux.misc and to a laptop group and gotten some helpful
suggestions, but this newsgroup may be a more appropriate place to pursue
this topic. Following one suggestion, I got van Gilluwe's book, The
Undocumented PC, 2nd ed. I'm rather handicapped by not really knowing
a lot about programming in assembler or machine language on this CPU.
However, the nasm documentation has some information and I also downloaded
the manuals on the Pentium III from Intel, since the laptop has a Pentium III.

I assumed that the first instruction executed on bootup is at FFFF0.
That lead after a couple of jumps to a sequence of about 60 instructions,
the last of which is a HLT and the one before which is a write to port 92h
which sets a bit that, according to Gilluwe's book, causes the machine
to reset. I don't understand why the machine doesn't therefore go into
an infinite loop, going back to FFFF0, the couple of jumps and the 60
instructions again and then the write to port 92h and the reset.

Does the machine in fact start at FFFF0?
If so, does the machine return to FFFF0 when the bit is set at port 92h?
If not, where does it go?
If so, why isn't there an infinite loop?

Some respondents on other groups have used terms such as "saving state"
and "interrupt vector table". These may be the right things to talk about,
but I don't know enough to use the vague suggestion that these things
explain everything. If someone has an elementary explanation, it will
be greatly appreciated.

Alternatively, if there is a free BIOS, with fully documented source code,
for a system involving a Pentium III, I'd be willing to study that instead
just to get more experience with BIOSes before reading the BIOS for my
laptop. 

-- 
Ignorantly,
Allan Adler <ara@zurich.csail.mit.edu>
* Disclaimer: I am a guest and *not* a member of the MIT CSAIL. My actions and
* comments do not reflect in any way on MIT. Also, I am nowhere near Boston.

Relevant Pages

  • Re: Infa Red Port
    ... If it's a laptop, infrared options can be ... enabled in the BIOS. ... All you need to do is just associate the infrared port ...
    (microsoft.public.windowsxp.hardware)
  • Re: IR Devices
    ... IR ports must be enabled in the laptop's BIOS before XP can be able to use ... the IR port shown be listed in Device Manager. ... but I couldn't see how to turn the IR on the laptop on. ...
    (microsoft.public.windowsxp.hardware)
  • Re: Port assignments under Windows XP Pro - Svc Pack 2
    ... Your PC or Laptop should have at least one serial COM port. ... ON?, Check the BIOS. ... > or get Windows to automatically assign a port would be helpful. ...
    (microsoft.public.windowsxp.hardware)
  • Re: Failed Flash BIOS after upgrade
    ... Do not trust ACER Engineers - they have created a mess of the ... BIOS Flashing process, which doesn't ever work in Windows correctly. ... So, Always create a DOS Boot Floppy and then Flash the BIOS, its ... If you have an Acer 4152 model laptop, ...
    (microsoft.public.windowsxp.general)
  • Re: Tecra failure
    ... I have boot disk but does not respond on booting. ... I have now managed to download Bios u/g but again did not work, ... Have you tried hold down the 'ESC' key when powering on the laptop this ... You can download a boot disk from www.bootdisk.com. ...
    (uk.comp.sys.laptops)