jsvp1@tnofel.fel.tno.nl (J S de Vries) (01/07/91)
I want to use A/UX 2.0 with my old Kinetics EtherPort II card, but, sofar, I
have not been able to get it up and running on a MacII at work or my MacIIx at
home. Because I only WAS a casual UNIX user and now, suddenly, an A/UX System
Administrator, I need experienced help!!
I have tried to use the old A/UX 1.0 (Rev A) Kinetics EtherPort II driver
software to configure the A/UX kernel for BNet or NFS. However, during
linking the kernel one undefined symbol is found ('mcIget') in the binary
file /etc/boot.d/if_ep, which is one of the driver S/W files, supplied by
Kinetics. This kills the linking process.
The following is an excerpt from the listing:
# autoconfig -u -I -o unix
Warning cannot find driver(s) for device ID 5 Version 0.0
undefined first referenced
symbol in file
mclget /etc/boot.d/if_ep
ld fatal: Symbol referencing errors. No output written to /tmp/kernBAAa00161
Autoconfig error: Kernel link failed
mv unix /unix
mv: cannot access unix
I have tried to get an update for the A/UX driver S/W. But, as it
turns out, Kinetics (Dayna Systems (?), nowadays..) is no longer supporting
the old (long) cards. These are replaced by shorter NUBUS cards with improved
ROM firmware, which are autoconfigured into the kernel, if one so wishes.
I have been hearing rumors of people using the old Kinetics EtherPort II card
and driver with A/UX 2.0. So, I hope to find information on the installation
of the old Kinetics A/UX 1.0 driver here on this forum, before buying a new
Ethernet card.
My hardware consists of the following:
5 megs of memory and the old 8 bit Apple display card in both Macs. The MacII
has a 320 MB Micronet harddisk and the IIx an 80MB Quantum.
--- Jan
P.S.
If I do the MacOS installation of the EtherPort II card using the Installer
on the MacPartition, A/UX 2.0 hangs during startup and produces the message
after switching screens during loading 'panic: no slot 11 interrupt handler'.
Does this actually mean, that the two installations are mutually exclusive?
I.e., if the EtherPort card is installed in MacOS, I cannot use A/UX 2.0,
and, when installed in A/UX, I cannot use it in MacOS?
----------------------------------------------------------------------------
Jan S. de Vries Internet: jsvp1@fel.tno.nl
Physics and Electronics Laboratory TNO Phone: ++31-70-3264221 ext209
P.O.Box 96864 Fax: ++31-70-3280961
NL-2509 JG THE HAGUEksand@Apple.COM (Kent Sandvik) (01/08/91)
In article <1991Jan7.134806.10592@tnofel.fel.tno.nl> jsvp1@tnofel.fel.tno.nl (J S de Vries) writes: >I want to use A/UX 2.0 with my old Kinetics EtherPort II card, but, sofar, I >have not been able to get it up and running on a MacII at work or my MacIIx at >home. Because I only WAS a casual UNIX user and now, suddenly, an A/UX System >Administrator, I need experienced help!! >I have tried to use the old A/UX 1.0 (Rev A) Kinetics EtherPort II driver >software to configure the A/UX kernel for BNet or NFS. However, during >linking the kernel one undefined symbol is found ('mcIget') in the binary >file /etc/boot.d/if_ep, which is one of the driver S/W files, supplied by >Kinetics. This kills the linking process. Device drivers written under A/UX 1.1 won't work with A/UX 2.0, because a couple of kernel structures are different. I.e. the company/persons behind the driver should simply recompile the driver(s) under A/UX 2.0 with the new A/UX 2.0 device driver kit files. Regards, Kent Sandvik -- Kent Sandvik, Apple Computer Inc, Developer Technical Support NET:ksand@apple.com, AppleLink: KSAND DISCLAIMER: Private mumbo-jumbo Zippy++ says: "C++, anything less is BCPL..."
liam@cs.qmw.ac.uk (William Roberts;) (01/08/91)
In <47816@apple.Apple.COM> ksand@Apple.COM (Kent Sandvik) writes: >In article <1991Jan7.134806.10592@tnofel.fel.tno.nl> jsvp1@tnofel.fel.tno.nl (J S de Vries) writes: >>I want to use A/UX 2.0 with my old Kinetics EtherPort II card >>I have tried to use the old A/UX 1.0 (Rev A) Kinetics EtherPort II driver >>software to configure the A/UX kernel for BNet or NFS. However, during >>linking the kernel one undefined symbol is found ('mcIget') in the binary >>file /etc/boot.d/if_ep, which is one of the driver S/W files, supplied by >>Kinetics. This kills the linking process. Isn't Monaco a stupid font! The actual problem is "mclget" as you can see if Courier is used. >Device drivers written under A/UX 1.1 won't work with A/UX 2.0, because a couple >of kernel structures are different. I.e. the company/persons behind the >driver should simply recompile the driver(s) under A/UX 2.0 with the new >A/UX 2.0 device driver kit files. The A/UX 1.1 version of the driver works with A/UX 2.0 but doesn't support AppleTalk - it's fine for NFS, TCP/IP and so on. I'd be happy to send a copy of the 1.1 driver binary to apple.com for people to fetch as needed, but only if the present owners of the rights to that code give their consent: you could try asking them for the 1.1 drivers anyway. -- William Roberts ARPA: liam@cs.qmw.ac.uk Queen Mary & Westfield College UUCP: liam@qmw-cs.UUCP Mile End Road AppleLink: UK0087 LONDON, E1 4NS, UK Tel: 071-975 5250 (Fax: 081-980 6533)
ksand@Apple.COM (Kent Sandvik) (01/10/91)
In article <2813@redstar.cs.qmw.ac.uk> liam@cs.qmw.ac.uk (William Roberts;) writes: >>Device drivers written under A/UX 1.1 won't work with A/UX 2.0, because a >couple >>of kernel structures are different. I.e. the company/persons behind the >>driver should simply recompile the driver(s) under A/UX 2.0 with the new >>A/UX 2.0 device driver kit files. > >The A/UX 1.1 version of the driver works with A/UX 2.0 but doesn't support >AppleTalk - it's fine for NFS, TCP/IP and so on. I'd be happy to send a copy >of the 1.1 driver binary to apple.com for people to fetch as needed, but only >if the present owners of the rights to that code give their consent: you could >try asking them for the 1.1 drivers anyway. OK, I'm quoting the A/UX 2.0 device driver manual: If you wrote a device driver for an earlier relase of the A/UX operating system, you should recompile it under A/UX release 2.0. Recompilation is necessary because *kernel data structures*, such as the user structure and the proc structure, have changed size and location within the kernel. Configuring a driver object that has not been recompiled under A/UX 2.0 into a 2.0 kernel may cause the system to crash when the driver is called. Just to add to this, there are some other important issues: Concering the fx IOPs, there are some issues with the serial device drivers that assume they talk with an SCC. A/UX 2.0 supports the serial Manager (with the exception of asynchronous I/O !). The virtual memory code is rewritten, the call sptalloc() is no longer available. Some of the old slot manager driver routines are no longer available in 2.0 where the Toolbox Slot Manager is supported. The line discipline routine ttopen() is modified in order to comform to the POSIX Standard concerning the oflag from open(2). The memory maps are modified, there's a separate user and kernel address space. Anyway, I recommend to purchase the new A/UX 2.0 device driver manual/kit for those working with drivers for A/UX. I would personally recompile any drivers from 1.1 and run them for a while before I would sleep during nights. Eventually there are 1.1 drivers that work without problems under 2.0 as well, but it's better to be sure and just recompile the code. BTW the manual should be available from APDA. Hope this helps, Kent Sandvik, DTS -- Kent Sandvik, Apple Computer Inc, Developer Technical Support NET:ksand@apple.com, AppleLink: KSAND DISCLAIMER: Private mumbo-jumbo Zippy++ says: "C++, anything less is BCPL..."
qfhca81@memqa.uucp (Henry Melton) (01/10/91)
In article <47902@apple.Apple.COM>, ksand@Apple.COM (Kent Sandvik) writes: > In article <2813@redstar.cs.qmw.ac.uk> liam@cs.qmw.ac.uk (William Roberts;) writes: > >>>Device drivers written under A/UX 1.1 won't work with A/UX 2.0, because a >>couple >>>of kernel structures are different. I.e. the company/persons behind the >>>driver should simply recompile the driver(s) under A/UX 2.0 with the new >>>A/UX 2.0 device driver kit files. >> >>The A/UX 1.1 version of the driver works with A/UX 2.0 but doesn't support >>AppleTalk - it's fine for NFS, TCP/IP and so on. I'd be happy to send a copy >>of the 1.1 driver binary to apple.com for people to fetch as needed, but only >>if the present owners of the rights to that code give their consent: you could >>try asking them for the 1.1 drivers anyway. > > > OK, I'm quoting the A/UX 2.0 device driver manual: > > If you wrote a device driver for an earlier relase of the A/UX operating > system, you should recompile it under A/UX release 2.0. Recompilation is > necessary because *kernel data structures*, such as the user structure > and the proc structure, have changed size and location within the kernel. > I have been running with the 1.1 drivers and an old EtherPort card for a couple of months now. While everything works, I can also suffer kernal crashes/system lockups whenever I attempt heavy usage that uses lots of memory. When I do get a panic message, it always points to a tcp/ip task as the bad guy. I am doing all of my appletalk stuff over the localtalk wire. The bottom line is that the 1.1 drivers will get your running, but will make your system flakey. If you aren't a heavy memory user, it might work fine. Personally, I am trying to get a different ethernet card. -- Henry Melton qfhca81@memrqa.sps.mot.com {slow} qfhca81@memqa ..!cs.utexas.edu!execu!sequoia!memqa!qfhca81 {home} henry@hutto ..!emx.utexas.edu!hutto!henry
jim@jagubox.gsfc.nasa.gov (Jim Jagielski) (01/14/91)
I have drivers for the "old" EtherPortII card available via anon-FTP on my
mac (jagubox 128.183.44.1).
These drivers were for A/UX ver 1.1 and 1.1.1 and they DO work under 2.0.
They do NOT support EtherTalk, LocalTalk over EtherNet. They DO support all
TCP/IP communication.
Note that these drivers are for the older version of the EPII card... it's a
full size Nubus card. Drivers for the NEWER card (smaller) are available via
Dayna (I will supply a "loaner" driver until you get yours). The two drivers
are NOT interchangeable :(
--
=======================================================================
#include <std/disclaimer.h>
=:^)
Jim Jagielski NASA/GSFC, Code 711.1
jim@jagubox.gsfc.nasa.gov Greenbelt, MD 20771
"Exploding is a perfectly normal medical phenomenon. In many fields of
medicine nowadays, a dose of dynamite can do a world of good."