[sane-devel] Preferences in ieee1394 cards for Athlons?

Steve Nordquist signa@birch.net
Sun, 2 Mar 2003 04:45:29 -0600


On Sunday 02 March 2003 01:35 am, you struggled free to say:
 -- Steve Nordquist <signa@birch.net>

 > I suspect you're running into desiderata concerning how your
 > modules were compiled, and perhaps a bit of data the chipset's
 > contributing during bus enum.
Oops! Correction!  You have no error, only warnings, and just need
to get a SANE backend to connect to your device, i.e. add options in
/etc/modules.conf and/or modprobe the next iitem in the chain towards
mounting the 1394 host and connecting 1394 devices with SANE
and its programs.....


 > On Wednesday 26 February 2003 07:01 pm, you struggled free to say:
 >  Background:
 >
 >     All PC CPUs are x86 compatible...you mean P4s?

 There are places where non-86's have caused people
 problems due to minor differences in the MB's, etc.

   {I think I'm gonna toss this bit of the discussion to
    the ocean, because x86 was from the beginning an 
    AMD standard term for compatibility, and the Centaur
    chips were actually called CenX86 or such.  Instead,
     I'll refer you to the output of gcc and make, which
     definitely note the differences in chipset and CPU.
     Then there's shortening mobo to MB's and adding
      apostrophe-s where posession is not meant and
       the s is only for matching number (not necessary
       since MB or mobo is numberless)} 

 The only difference I can see in these two systems
 is the cpu hardware and one uses a DAC960 for the
 boot drive the other boot's off an LSI SCSI card
 and the CPU/MB vendors. Aside from that they are
 the same kernel compiled with the same ieee options
 with the same verison of gcc with the same command
 line switches, same network cards, built-in sound,
 and brands of SCSI peripherals. That's why I was
 surprised to get a different response from the same
 card on the two systems.

    It took a bit for me to notice 'debian' in all you said....
 
    Besides tracking the BIOS on each system to make
    sure that's good and current with ASUS or whoever
     it was, you might want to note the chipset 
     mentioned for each board so that any kernel
     notices on them won't go unnoticed by you. 

      I think that what you have in particular is a bit of
      module code that's...well actually it's not nagging
       for personal attention, because those
       warnings are common to successes also!
       What's supposed ot follow, however is 
       ieee1394: Host added [excruciating detail] [Debian rul3z]
       ieee1394: Device added [still more] [NEC or whoever]
       ieee1394:: {Sane2-frontend} logging into {details....}



 The x86's look like:
{Maybe I just missed the huge Chicagoland "86!" ad
campaign by Intel.}


 >     Does the module complain upon load, according to dmesg?
 >     Running RH8.x with the -multi kernel loaded, I assume....

 All of the ieee drivers are modules so that I could
 try to reset them, etc. insmod ohci1394 repeatably
 gives me the resource length nastygram on both systems.
 On the Asus MB it appears to be an error; on the I-Will
 it acts like a warning.