AUTOSTAR INFORMATION Last updated: 18 April 2000

Autostar Proc Traps

Subject:	 Autostar Proc Traps... the List
Sent:	Saturday, April 15, 2000 21:09:07
From:	rseymour@wolfenet.com (richard seymour)
Here are the numbered Proc Traps you should never see when
using your Autostar.  If you get one, there's been a hardware or
computer error... Autostar versions before 2.0 -could- get
Proc Trap 5 (IRQ), but v2.0 and later now handle that gracefully.

What a Proc Trap is:  an interrupt.  The Motorola 68HC11 which
is the heart (well, brain) of the Autostar can be optionally diverted
from its normal path of operation by various hardware signals.
(and a couple of software ones).  These can arrive at any time,
and the disruption of normal flow is called an Interrupt.
The Autostar firmware program uses some of them (clock ticks
to keep track of the time of day, Reset to start thing rolling at
power-up, etc.)
The ones it is -not- using (nor expecting) all cause a message to
appear on the display, and then the Autostar will "freeze".
Since it's a hardware problem, Meade expects us to turn it off,
wait a few seconds, and then turn it back on.  With luck, the
cause will have been truly weird, and should not reappear.

The message you'll see is:  Proc. Trap N,
with "N" being a hexadecimal number.

The numbers mean:
0      Clock Monitor Fail
1      COP timeout failure
2      Illegal Op Code
3      Software Interrupt
4      XIRQ pin
5      IRQ pin
6,7,8   Timer Input Capture 1,2,3
9,A,B,C  Timer Output Compare 1,2,3,4
D     Combined TIC4/TOC5
E     Pulse Accumulator Overflow
F     Pulse Acc Input Edge
10   SPI serial transfer complete

Of the above list, numbers 2 and 3 cannot be blocked
(or "masked").  If their circumstances occur, you'll Trap.

In the months i've been reading Mike's site, i've seen reports
of number 2 the most...even i've experienced it doing "point at
Polaris, do lots of tiny speed-zero motions".  It's what you'd
get if the computer accidently tried to "execute" a byte which
was really a data word.  I suspect cold weather could cause
a mis-fetch of an instruction which would cause it, too.

The Number 5 used to occur (before they programmed around it)
probably due to electrical noise getting picked up by one of the
68hc11's pins.  "IRQ" is Interrupt ReQuest"... the 68hc11 has
two pins you can wire external signals to which may want to
request (or demand) service.

The COP failure is a system the Autostar doesn't use: you
can program a timer in the 68hc11 to kick that one if the
processor stops doing -anything- else for up to a second.
The Clock Monitor triggers if the system clock (the 8 megahertz
crystal in the Autostar's case) slows down too far.

The Timer Inputs and Timer Outputs are for monitoring
the collection of timers, and some input and output pins,
available inside the 68hc11.

The Pulse Accumulator interrupts are for additional
timer-like functions... you can tell a portion of the 68hc11
chip to count up how many times (yet) another pin is
signalled from the outside, and to interrupt normal flow
when a pre-set limit is reached, or if the counter merely
rolls over (overflows) from full-scale to zero.

The SPI is a secondary serial communication channel
(independent of the rs232 channel) the 68hc11 can use
to communicate with other devices.  The Autostar does
not appear to use it.

Full details of all of the above are contained in the 590-page
Motorola 68HC11 Reference Guide, available as a pdf file.
(i'm sorry, i don't have the URL here... start at www.motorola.com,
 or use a search engine)

May you never need the above information.
--dick

Go back to the Autostar Information page.

Go back to the ETX Home Page.