Hijacking oem hardware

Webb cwebb at polarnet.com
Tue Sep 23 20:17:41 GMT 1997


Greetings

I have been sorta "lurking" on the sidelines for a long time,
evaluating the projects, and contemplating doing something
myself.

I have a number of OEM units, and went as far as to lool at the
332 project closely. I also bought an 80552 board, and worked for
a while at implimenting the algorythms I have encountered.

I'm an EE, with considerable experience with microcontrollers. It
strikes me that the ultimate solution to the problem of cheap,
reliable hardware platform to experiment with  is to "hijack"
some newer OEM platform that is cheap and common at the
boneyards.

The OEM stuff is not standard hardware - some of it anyway. It
has been reworked for the intended purpose, and hardened for the
hostile environment to a far greater degree than anything that
could be cobbled together in my garage.

The problem is that the hardware is proprietary. No memory maps,
no hardware configuration, no documentation for the I/O chips, no
way of disassembling the code, and no assemblers, or compilers
available. Furthermore, most units are potted, so you can't even
get to the stuff without melting the potting.

If some industrious individual were to do a comprehensive job of
reverse engineering one of these units, it would create a
beautiful, cheap, rugged platform for any wierd idea anyone
wanted to try.

If one wanted to do that, one would need to

1) Identify a candidate platform
        - as much known about it already as possible
        - with the latest technology available

2) Obtain one
        - Melt the potting
        - Copy the code in all the ROMS
        - Attempt to identify all the hardware
        - Trace out a schematic, and put it on a schematic
capture package
        - Connect a logic analyzer, and watch the operation in
progress
        - Using this info, attempt to devine the I/O memory map,
and put
           together a package of published documentation where
possible

This strikes me as more of a project than any one person's
talents, spare time, and energy are likely to take to fruition. I
am equally certain that between us, we could carry off such a
project.

Is anyone interested? Who can do what?





More information about the Diy_efi mailing list