On Mon, 3 Jan 2005, John Foust wrote:
> If the program binary sent over the serial line didn't include
> any characters that were eaten or rejected by COPY, and it
> ended with a CTRL/Z and COPY stopped there, it would work,
> wouldn't it?
>
> This reminds me of the constraints on today's exploits and code
> injection techniques: "write a series of three progressive
> exploits and loaders that fit in 64, 256 and 512 bytes,
> respectively, and do not contain any zeroes."
That sounds a bit much for the task at hand... !
> At 02:37 PM 1/3/2005, Tom Jennings wrote:
>> Hell, with debug you could WRITE the program to input the file
>> in binary! We're talking MSDOS, right? Not Windows?
>> You can type this crap directly into debug, "aXXXX".
>
> Go for it, Skippy! When it's debugged and tested, ship it! :-)
> As the hex to enter into DEBUG.
debug will assemble for you! It jsut won't do symbols.
I am surprised that no one pointed out that the input, store, incremnt
could more easily be done by:
mov dx, DATAPORT
...
es:insb
djnz label
>
> You're right - the program for this would be small enough to
> fit in the margin of a FAQ. "Load and run from the serial port."
>
> - John
>
Received on Wed Jan 05 2005 - 00:07:15 GMT
This archive was generated by hypermail 2.3.0
: Fri Oct 10 2014 - 23:37:43 BST