undocumented AT&T 3.2.2 questions

Bill Mayhew wtm at uhura.neoucom.EDU
Thu Jun 14 00:17:39 AEST 1990

1.  Re: usning a Kraft generic mouse with VP/ix.  Much may depend
on the the mouse driver.  Many mouse drivers use the IRQ 5
interrupt (the lpt2: irq).  It is likely that the kernel programs
the 8259 interrupt controller to not receive interrupts from IRQ 5
wihile the kernel is executing time slices other than VP/ix.  This
means that interrupts from the mouse movement will be lost.  It may
be necessary to obtain an AT&T mouse (maybe a Logitec Bus Mouse
will do) and use the AT&T mouse driver that comes with the Unix.
I do not have VP/ix on my system, so I can not comment on how well
the AT&T mouse works; one would hope(insert smiley) that the AT&T
mouse is well integrated.

2.  Re:  keyboard lock-up.  AT&T mentions in the 3.2.1 manuals that
there are problems with the keyboard MPU chip (8748 or whatever it
is) on some non AT&T systems.  The mention of the keyboard problem
is in the back of the 3.2.1 release notes manual.  There is a part
number for a replacemnt chip.  My manuals are at home and I am at
work, so I can't give the number right now.

3.  Re:  Cursor.  You can muddle through the console(7) section
trying to figure out the ioctl(2) calls to program the display.  I
don't know about the 3.2.2 manuals, but the expamples in the 3.2.1
manual are sort of buggy.  I used it a while back to gain access to
the 8253 channel used to make sounds with the speaker.  They
mention to include console.h, but I belive the definitions have
been moved to sys/kd.h.  I beleive you have to change another
include file too, but it has been a while since I worked on it.
Anyway, one of the ioctl() calls allows you to access the 6845
registers on the video display, so you should be able to program
your cursor.  While on the subject, I found the TAM translation
library to be really buggy too.  I worked on porting some TAM
applications from the 3b1, and found it to be quite a bit of
work.  If you are using a monochrome display, use caution when
programming the 6845.  On some moniotrs, botching the scan
registers in the 6845 can stop the scan on the monitor, which in
turn destorys the monitor becuase the transistor that drives the
flyback transformer remains on solid. If the horizontal drive
transistor stays on, the flyback smokes rather quickly.  This is not
a problem with color monitors.  Peter Norton's DOS programming
book has inforamtion about programming the 6845.


Bill Mayhew  Northeastern Ohio Universities College of Medicine
Rootstown, OH  44272-9995  USA    216-325-2511
wtm at uhura.neoucom.edu   ....!uunet!aablue!neoucom!wtm

More information about the Comp.unix.i386 mailing list