[comp.sys.att] Kernel rebuild 386 UNIX

bill@ssbn.WLK.COM (Bill Kennedy) (07/31/88)

I am trying to install a Computone Intelliport AT-4 in an AT clone
running AT&T 386 UNIX.  The device drivers that come with the board
support only ubug, ISC, or SCO Xenix.  I'm quite sure that the ubug
driver will work because that driver has worked with this system
before I pitched it and got AT&T 386 UNIX.  I have gotten through
idcheck and idinstall by using my PC 6300 PLUS documentation (darned
near identical) but idbuild blows me away with

LINE: ip  Y    1  195    1  15    0    0    0    0
ERROR:  ipl out of range
FATAL ERROR:  Errors encounteres.  Configuration terminated.

I am pretty sure that it doesn't like interrupt vector 15 but I can not
find anything anywhere that lets me change it.  There's a header file,
/usr/include/sys/ipl.h that talks about something similar but it suggests
that it's there to allow the spl functions to correctly mask lower level
interrupts.  Also there doesn't appear to be anything out on the second
interrupt controller, SPLTTY, the highest number is 7.

Has anyone successfully installed a Computone Intelliport board under
AT&T 386 UNIX (Computone hasn't...)?  Has anyone raised IPLHI beyond 8
where it is now?  I'd sure like to hear about it if you have.  Thanks,
-- 
Bill Kennedy  usenet      {killer,att-cb,rutgers,uunet!bigtex}!ssbn!bill
              internet    bill@ssbn.WLK.COM

mmengel@cuuxb.ATT.COM (~XT4103000~Marc Mengel~C25~G25~6184~) (08/02/88)

In article <205@ssbn.WLK.COM> bill@ssbn.WLK.COM (Bill Kennedy) writes:
>I am trying to install a Computone Intelliport AT-4 in an AT clone
>running AT&T 386 UNIX.  The device drivers that come with the board
>support only ubug, ISC, or SCO Xenix.  I'm quite sure that the ubug
>driver will work because that driver has worked with this system
>before I pitched it and got AT&T 386 UNIX.  I have gotten through
>idcheck and idinstall by using my PC 6300 PLUS documentation (darned
>near identical) but idbuild blows me away with
>
>LINE: ip  Y    1  195    1  15    0    0    0    0
>ERROR:  ipl out of range
>FATAL ERROR:  Errors encounteres.  Configuration terminated.
>
>I am pretty sure that it doesn't like interrupt vector 15 but I can not
>find anything anywhere that lets me change it.  There's a header file,
>/usr/include/sys/ipl.h that talks about something similar but it suggests
>that it's there to allow the spl functions to correctly mask lower level
>interrupts.  Also there doesn't appear to be anything out on the second
>interrupt controller, SPLTTY, the highest number is 7.

"ipl" is NOT the interrupt vector, its the kernel internal priority.
You have it listed as 195...  Here is the description of the
columns in the sdevice file
1 - Device name -- must match a name in mdevice
2 - Configure -- Y or N
3 - Unit -- unit number, within range listed in mdevice
4 - ipl -- kernel ipl level, serial boards probably want to be a 5
5 - type -- 0==no interrupt, 1==exclusive interrupt, 2==shared interrupt
6 - Vector -- interrupt vector for device (your 15 goes here)
7 - SIOA -- start io address
8 - EIOA -- end io address
9 - SCMA -- start memory address
10- ECMA -- ending memory address

You might want to get a copy of the Integrated Software Developers
Guide -- (call the CIC at 1-800-432-6600 and order it, the "comcode"
is 105389001) It documents the new, improved ID scheme fairly 
thoroughly.

>-- 
>Bill Kennedy  usenet      {killer,att-cb,rutgers,uunet!bigtex}!ssbn!bill
>              internet    bill@ssbn.WLK.COM


-- 
 Marc Mengel			       
				
 attmail!mmengel	
 {lll-crg|mtune|ihnp4}!cuuxb!mmengel