Re: [PARPORT] fingerprint reader

From: Laurent Duchesne (parport@oemprince.net)
Date: Sun Nov 10 2002 - 15:36:21 EST

  • Next message: Thomas Bodine: "Re: [PARPORT] re: HP printer and 2595xDVD"

    Thanks for the info. I was going to go on a more "software" road.
    I don't know if it's been done, but I thought about
    adding a system call to the linux kernel, which imitate
    the outb function, plus some logging features. My goal
    was to actually use vmware (in which my fingerprinting
    device works perfectly), and change in it's binary the call
    to the outb function by an interruption to my new system call,
    so the data could be collected.
     
    Like you suggested, this might be illegal, but I heard that if
    you just "sniff" the protocol to obtain interoperability with an
    OS the company don't want to support, it's okay to do so
    (if, of course, you don't disassemble their software or use
    their algorithm).
     
    The vmware part was going to be illegal, I think.
     
    Fortunately, a technician at an optical analisys lab in my
    university offered me his help to analyse the protocol
    used by my device. He told me he was doing this kind of
    thing all the time and it was going to be easy to obtain
    all the information I need on the protocol. His tools can measure
    up to 183 data line at the same time, so the mere 18 or so needed
    for the parallel port should be no problem :)
     
    Thanks again.
     
    Laurent
     
     
    p.s. in case someone whant's to know, here's how the system call would
    have looked like, without the logging part :
     
    asmlinkage void sys_outb_hack(unsigned char value, unsigned short port)
    {
        __asm__ __volatile__ ("out" "b" " %" "b" "0,%" "w" "1" : : "a"
    (value), "Nd" (port));
    }
     
    and a page to implement it :
    http://fossil.wpi.edu/docs/howto_add_systemcall.html
     
    I put it in linux/fs/outbhack.c and added it in the
    fs/Makefile so it would be compiled easily in the kernel.
     
    This HASN'T been tested but maybe it will help someone someday..
     
     
     
    Selon Simon Krix <kinsei@netspeed.com.au>:
     
    > Hello,
    >
    > On Fri, 2002-11-08 at 05:08, parport@oemprince.net wrote:
    > > I read the archive of all posts about reverse engineering on this mailing
    > list
    > > and I can't seem to find where to begin.. any suggestions?
    >
    > I used a small home-made circuit to reverse-engineer my scanner. You can
    > see the details of my project here:
    > http://www.netspeed.com.au/kinsei/scannerpage/
    >
    > This uses an ISA card based on a 8255C chip to read the signals on the
    > parallel port while using the scanner on a slow Windows PC. I hear some
    > people are using WINE to do the same thing more reliably. Be careful
    > that you are not doing anything illegal, as some countries have more
    > restrictive laws about reverse-engineering than others.
    >
    > Good luck.
    >
    > -- Simon Krix
    >
    >
    >
     
     

    -- To unsubscribe, send mail to: linux-parport-request@torque.net --
    -- with the single word "unsubscribe" in the body of the message. --



    This archive was generated by hypermail 2b29 : Sun Nov 10 2002 - 15:47:59 EST