[Simh] SDS Card Reader
kenr
kdrhoo at yahoo.com
Thu Feb 13 01:38:59 EST 2020
Hi Mark,
At SDS/XDS/Xerox we tried to ignore the 940. We had our own set of
OSes to work on and promote, so you don't find much 940 software amongst
the remnants of SDS.
I'm trying to recall who made our card reader equipment. What I do
recall is that they ate cards like a Lower Elbonian accordian factory.
I too looked at the card reader resources and I think I could at least
get something crude to work.
But I have a easier but cruder short term solution. And I found out
more about the files on the PAL Tapes.
Even though the PAL Catalog says they're absolute card decks, the files
I chose to look at are actually relocatable cards prefaced with a two
card relocating loader. So my plan is to throw away the card loader and
relocate the remainder into an absolute file and put your 940boot loader
on the front of that. It was when I got that working that I found that
the files were really relocatable.
The first program I picked is an unecoder so I might be able to dissect
source programs from the PAL tapes.
I'm still not able to get the loader that comes on the front of these
files to complete it's self loading task up to the point where it needs
the card reader. But I'm sure that problem will be clear eventually
like the other stuff I'v learned so far.
My wife thinks I'm mad. She hasn't seen me in several weeks.
Ken
Mark Emmer wrote on 2/12/20 10:07 PM:
> I'd be very surprised if /any /SDS 940s had card readers or punches.
> They are remnants from the days when we all marched down to the
> machine room and passed boxes of 2,000 cards through the window to the
> operator to run in a batch environment. In 1965, the Berkeley time
> sharing system on the 940 was meant to break that model -- users would
> create programs interactively via remote teletypes. No cards needed
> or wanted.
>
> That said, the SDS 940 Reference Manual describes the card reader and
> punch I/O instructions in detail and provides example code for
> operating both devices in both Hollarith and binary formats. The logic
> was retained from its non-time sharing predecessor, the SDS 930.
>
> Quickly glancing at sim_card.c, it should be straight-forward to wire
> up a card reader and punch driver for the 940 simulator. sim_card
> supports binary mode, which would be used by Ken's boot program.
> Unfortunately, I can't volunteer to do that now as I have too many
> other higher-priority projects on my plate.
>
> I've been in contact with Ken privately about using the 940 simulator
> in "normal" (SDS 930) mode for his PAL tapes. Seems quite feasible if
> he can get around the card reader issue.
>
> Mark Emmer
>
>
>
> Get BlueMail for Android <http://www.bluemail.me/r?b=15774>
> On Feb 12, 2020, at 1:28 PM, Mark Pizzolato <mark at infocomm.com
> <mailto:mark at infocomm.com>> wrote:
>
> Hi Ken,
>
> On Wednesday, February 12, 2020 at 10:51 AM, Kenr wrote:
>
> I'm interested in using the SDS 940 simulator to run programs
> from the XDS PAL tapes found on bitsavers. Most of these
> programs were loaded onto the PAL library tapes from punched
> cards by SDS/XDS in a self loading format for distribution to
> customers. Customers would copy the files from tape onto
> punched cards. The self loading programs assume a card
> reader, not a paper tape or mag tape. The SDS 940 simulator
> does not have a card reader driver so I have a couple of
> questions. 1. Why isn't there a card driver for the 940? 2. I
> notice simulators for similar machines also don't have card
> drivers, why is that? 3. Does anyone know of a (maybe
> unfinished, unofficial) card driver that I might use? 4.
> Where might one find the developer documentation mentioned in
> the simh faq? Anyone care to answer these or offer an opinion
> on driver development?
>
>
> From a simh point of view, the terminology for what you're looking for is a
> card reader/punch device simulator. I can't personally speak to why such a
> device simulator hasn't been implemented for any particular simulator, but
> the likely reasons may be:
>
> 1) Lack of precise documentation for the programming interface to the
> hardware in question. Sometimes this is available in specific hardware
> documentation or even schematics, other times the programming
> interface can be specifically extracted from the pieces of software
> (aka OS drivers) that manipulated a particular device. This is most easily
> observed if this software is available in source rather than binary form.
> 2) Lack of interest for a particular device by the folks who developed or
> added to the existing system simulator.
>
> It seems like your interest is now addressing point #2, so given your
> interest if you can find sufficient documentation to address the how
> the software should interact with this device, you'd be welcome to
> come back here and ask for help implementing it. Myself or others
> will certainly be able to stand up and provide some nuts and bolts
> guidance. Simh has a 'library' that is designed to specifically interface
> with and create virtual card decks on modern systems, so the missing
> pieces are really how the hardware looked to the software that uses
> it. This library was created by Rich Cornwell and is call sim_card.c.
> Bob Supnik wrote the SDS simulator and he and Mark Emmer have
> made somewhat recent changes.
>
>
> Good Luck,
>
> - Mark
>
> ------------------------------------------------------------------------
>
> Simh mailing list
> Simh at trailing-edge.com
> http://mailman.trailing-edge.com/mailman/listinfo/simh
>
>
>
> _______________________________________________
> Simh mailing list
> Simh at trailing-edge.com
> http://mailman.trailing-edge.com/mailman/listinfo/simh
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.trailing-edge.com/pipermail/simh/attachments/20200212/a058929b/attachment-0002.html>
More information about the Simh
mailing list