Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!mips!spool.mu.edu!snorkelwacker.mit.edu!ai-lab!rice-chex!bson
From: bson@rice-chex.ai.mit.edu (Jan Brittenson)
Newsgroups: comp.sys.handhelds
Subject: Re: M/L Interrupts/Concurrent Routines
Keywords: ml, interrupts, concurrent routines
Message-ID: <15232@life.ai.mit.edu>
Date: 24 Apr 91 07:15:47 GMT
References: <28106e62: 2851comp.sys.handhelds@hpcvbbs.UUCP> <1991Apr23.115645.9250@corpane.uucp>
Sender: news@ai.mit.edu
Organization: nil
Lines: 33

In a posting of [23 Apr 91 11:56:45 GMT]
   herman@corpane.uucp (Harry Herman) writes:

 > In <28106e62:2851comp.sys.handhelds@hpcvbbs.UUCP> akcs.scotty@hpcvbbs.UUCP (SCOTTY THOMPSON) writes:

 >> I would be very interested in being able to have interrupt-driven
 >> subroutines

 > Are you sure you can do this?  I thought all objects in HP48 memory
 > were subject to being shifted to new addresses whenever garbage
 > collection occured.

   Actually, allocating memory that's not reclaimable by GC is not
difficult. A program can allocate such a chunk of memory and copy
itself into it, and when it decides it's time to terminate, it can
copy itself back to reclaimable space and finish off by deallocating
the non-reclaimable space. The stack-menu-PICT GROBs are examples of
objects not in the reclaimable space, which can be thought of as an
extension of the static system data. It resides between the system
variables in low memory and the dynamic memory where commonly
allocated object types (numbers, strings, arrays, lists, etc) go.
(Other object types, such as programs and directories, which are much
less commonly allocated, appear to go at the top of memory, above the
stack, with the HOME directory at the absolute top.)

   The part appearing less trivial is that of intercepting interrupts.
They all seem to vector to location #0000F, regardless of origin,
which determined in software by the interrupt handler. It doesn't
vector via RAM, so unless there is a vector somewhere in the I/O page,
I doubt interrupt-driven programs can be written at all on the HP-48.

						-- Jan Brittenson
						   bson@ai.mit.edu