one more note

[br...@quarterbyte.com]

The re-use of the buffer was definitely understood. The comments at the top of the file note:

              EX000 MOVES A SMALL PROGRAM TO THE  * J0903570

*    1132 SCAN AREA THAT READS IN PHASE 0/1 OF    * J0903580

*    THE CORE LOAD BUILDER.                       * J0903590

So, the expectation had to be that this would never overlap with printing on a 1132.

The implication for modern hardware- or software-emulated 1130 and 1800 machines is that you have to do one of the following

• use an emulated 1403 printer, not an 1132
• never use STORECI
  
• slow your disk(s) down
• patch this module so that STORECI will work.
  

(Probably in descending order of desirability)

Might there be other places where these eight words are used for purposes other than printing? Scary thought.

brian

[John McKee]

I thought of a use for those words once, while at school.  During FORTRAN class, it was too common for another student to fail to include a terminator card, such as /* behind their card data.  CARDZ immediately exits if it reads a // card.  This caused problems for the next student.

A simple fix would be to modify CARDZ to place a small move sequence in the 1132 scan field that moved the contents of the just read monitor card to the area used by MCRA.  Then, set the exit code indicating that a monitor control record had been read.  I wrote a test program to that effect.  I built a monitor control record, set the flag and exited.  The built record was then processed.  Perhaps CARDZ did not do this to prevent issues with the 1132.

John McKee

--
You received this message because you are subscribed to the Google Groups "IBM1130" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ibm1130+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Carl Claunch]

There are certain to be quite a few potential race hazards strewn through the software. Emulating at 1130 speeds will avoid the problem, but most people want the speed of modern circuits so they are not waiting a long time for even smallish jobs to run. 

Carl

PS - Even the hardware in the 1130 contained extraneous gates simply to delay signals a bit, since they act when they arrive, asynchronously, and not interlocked with other signals nor to the system clock. A common way to do things back then. 

[John R Pierce]

On 7/31/2016 11:27 PM, br...@quarterbyte.com wrote:

The implication for modern hardware- or software-emulated 1130 and 1800 machines is that you have to do one of the following

• use an emulated 1403 printer, not an 1132

I really can't imagine why one would want to emulate the 1132.    I do believe if I was writing an emulator, that wouldn't even be an option.

-- 
john r pierce, recycling bits in santa cruz

[Richard Stofer]

When I built my FPGA version, the 1403/2501 were the peripherals of choice because they handled the data transfers through DMA.

 

But here’s the thing:  I can’t do a STORECI on my FPGA version even with the 1403.  It is very probably some kind of timing issue but I don’t even have the 1132 to blame it on.

 

I’ll wait and see what the next full release contains because, for some reason, I can’t build the dms.dsk with the version I just downloaded from IBM1130.org.  More specifically ‘ibm1130 loaddms’ is broken after I change the CONFIG cards to remove peripherals I don’t have and replace them with cards for the 1403 and 2501.  I know it used to work, I have done this before.  I must have broken something.  BTW, it works the way it is shipped.  It will build the dms.dsk for a configuration I don’t have.

 

OTOH, Carl sent me his loaddeck and he had used it to build a correct image for the EALOG problem.  I haven’t tried it but our deck changes are identical.  I’ll just have to wait for the official release before spending any more time on these issues.  As I said, Carl’s image resolves the EALOG issue and STORECI just doesn’t matter all that much.

--

[Bob Flanders]

Fellow eleventhirtyers...

Would it be worthwhile to have a github project for the 1130 OS and start doing change management on it.... pull requests, etc?

Maybe we are starting a Disk Monitor V2 R21 (21 for the 21st century?)

What do you all think?

Bob

[John R Pierce]

On 8/1/2016 9:59 AM, Bob Flanders wrote:
> Fellow eleventhirtyers...
>
> Would it be worthwhile to have a github project for the 1130 OS and
> start doing change management on it.... pull requests, etc?
>
> Maybe we are starting a Disk Monitor V2 R21 (21 for the 21st century?)
>

we always called our enhanced R12's, R13 :) but I'm not sure we really
would want to skip version numbers to R21.

[Steve Zoppi]

I “second” John’s proposal and I’d be happy to set it up unless someone (perhaps Brian?) feels strongly otherwise … I already have a local git repo to keep my changes “clear” from the distro.

 

-Steve

[Yvette Seifert Hirth, CCP, CDP]

> -----Original Message-----
> From: ibm...@googlegroups.com [mailto:ibm...@googlegroups.com] On
> Behalf Of John R Pierce
> Sent: Monday, August 01, 2016 10:03 AM
> To: ibm...@googlegroups.com
> Subject: Re: [IBM1130] one more note
>

> On 8/1/2016 9:59 AM, Bob Flanders wrote:
> > Fellow eleventhirtyers...
> >
> > Would it be worthwhile to have a github project for the 1130 OS and
> > start doing change management on it.... pull requests, etc?
> >
> > Maybe we are starting a Disk Monitor V2 R21 (21 for the 21st
> century?)
> >
>
> we always called our enhanced R12's, R13 :) but I'm not sure we
> really
> would want to skip version numbers to R21.

R21 is just R12 in little-endian notation...

-y-

[Bob Flanders]

Hehe...

Sorry... my proposal was less about the revision # and more about getting a repository started where the current source code can live and have a current version. 

This would allow us to share changes and have a comment "current version"/

I suggest github.

Bob

[John R Pierce]

On 8/1/2016 10:41 AM, Bob Flanders wrote:
> Sorry... my proposal was less about the revision # and more about
> getting a repository started where the current source code can live
> and have a current version.
>
> This would allow us to share changes and have a comment "current version"/
>
> I suggest github.
>

has IBM put the 1130 DM2 sources in the public domain? or are they
still nominally copyright IBM ?

[Bob Flanders]

I am pretty sure it's like the OS/MVT DOS/VS VM/370 MVS 3.8j and other earlier IBM OS's -- not copyrighted. 

[Yvette Seifert Hirth, CCP, CDP]

Iirc copyright lasts 26 years.  We’ve lasted 50+ - almost double.  Hah!

 

So 1130 copyrights, if any, probably expired.  S/360, maybe that’s another issue.

 

-y-

 

From: ibm...@googlegroups.com [mailto:ibm...@googlegroups.com] On Behalf Of Bob Flanders
Sent: Monday, August 01, 2016 11:07 AM
To: ibm...@googlegroups.com
Subject: Re: [IBM1130] one more note

 

I am pretty sure it's like the OS/MVT DOS/VS VM/370 MVS 3.8j and other earlier IBM OS's -- not copyrighted. 

[eddy]

I don't think it is, I was into that code a lot in the late 60's even as far as to make changes to Fortran and never saw a copyright notice.

Sent via the Samsung Galaxy S® 6, an AT&T 4G LTE smartphone

[Carl Claunch]

The COBOL for the 1130 is clearly distinguished as a licensed product which may make it the only part of the DMS2 environment that might have legal entanglements. 

Carl

[Eddy Quicksall]

>> There are certain to be quite a few potential race hazards strewn through the software.

 

Actually you would be suppressed. There were much fewer than on Windows … that is one of the reasons Windows crashes so much. The only race that I know of is still there, I think, which occurs if you punch STOP on the 1403 in between the time you check for READY and issue the XIO. We corrected that in TSO for the 1130. I think the 1800 drivers had that fix.

 

Eddy

--

[Eddy Quicksall]

Suppressed == surprised

[Eddy Quicksall]

>> There are certain to be quite a few potential race hazards strewn through the software. Emulating at 1130 speeds will avoid the problem, but most people want the speed of modern circuits so they are not waiting a long time for even smallish jobs to run. 

 

I take acceptation to this.

 

I used the 1132 from the release of the 1130 to the introduction of the 1403; a colleague and I made changes to the 1132 driver to always print at the full speed of “numeric only” printing (it was simply a matter of reading the character wheel to determine where to start). I tested it in a very tight loop so there was very little delay between printed lines and did not have any problems.

 

My customer has been running DM2 now for over 30 years in every day operation supporting general ledger, pay checks, accounts payable, etc. with multiple clients. I run compiles, links, edits, etc. I write the assembly and his programmer writes the Fortran, COBOL and RPG. He is currently on the latest PC’s with gobs of speed and I use an I7. I never did anything to slow down the emulation. We have never run into a race condition over those years.

 

As I said in an earlier email, the only race in DM2 that we ever ran into over 48 years was the one that occurs if you press STOP on the device between the time the driver checks for BUSY and the time it issues the XIO. On the IBM 1800 the driver tested BUSY correctly. Don Nichols at DNA, whom I worked for in the 70’s, fixed that for the 1130 in the early 70’s. The correct procedure is to issue the XIO then check for BUSY. After the XIO, if the device does not show BUSY then it means it did not accept the XIO and it is NOT READY … that way there is no race.

 

Aside from the very simple IBM 1401, the IBM 1130 running DM2 is the most solid system I have ever used. Note: version 1, which was strictly card based with an attached disk, did have some race conditions but IBM fixed all of those for version 2.

 

My hat’s off to IBM, they wrote solid software in those days.

 

Eddy

 

From: ibm...@googlegroups.com [mailto:ibm...@googlegroups.com] On Behalf Of Carl Claunch
Sent: Monday, August 01, 2016 12:17 PM
To: IBM1130
Cc: br...@quarterbyte.com
Subject: [IBM1130] Re: one more note

 

There are certain to be quite a few potential race hazards strewn through the software. Emulating at 1130 speeds will avoid the problem, but most people want the speed of modern circuits so they are not waiting a long time for even smallish jobs to run. 

 

Carl