IBM 1130 emulator and OS/1130

[Paul Anagnostopoulos]

I promised to start a thread on my IBM 1130 emulator and OS/1130, the new operating system for the fictitious model 7. I'll start with a screen capture of a basic system with 4K memory, a 1442 card read/punch, and an 1132 printer.

[Paul Anagnostopoulos]

Now here is a screen capture of a large model 7 system with 32K memory and lots of I/O devices. Disk cartridges are loaded on many of the drives. The gizmo below the mode dial is an 8-character LCD display. The rectangular box in the lower-right corner of the cabinet is a "CE panel" that displays information including the cycle time, memory size, and number of cycles executed.

I have also attached a text file with the description of the extra features of the fictitious model 7.

[Paul Anagnostopoulos]

Here is a screen capture showing the model 7 immediately after loading the standalone DCIP utility.

[Paul Anagnostopoulos]

Hey folks. After a year being too busy, I'm back hacking on my IBM 1130 emulator and OS/1130, the operating system for the fictitious model 7. Here is a screen capture of the machine after booting the first model 7 test program.

[Paul Anagnostopoulos]

The attachment is a listing of the IPL test program shown in the previous image. The assembler is a cross-assembler that runs on the PC. It is more or less a superset of the DMS2 macro assembler, although I designed it to be reasonably possible to implement it on an 8K 1130.

~~ Paul

[yvette hirth]

nice looking code! after 2 years of php and javascript, i really miss
assembler...

yvette

[Paul Anagnostopoulos]

You are probably more productive in PHP and Javascript, but it's just not quite the same sort of fun.

~~ Paul

[Joseph Ambrose]

Paul,

 

Is the emulator you working on your creation or SIMH ??

 

On Sun, Aug 19, 2012 at 6:58 PM, Paul Anagnostopoulos <pcanagno...@gmail.com> wrote:

Hey folks. After a year being too busy, I'm back hacking on my IBM 1130 emulator and OS/1130, the operating system for the fictitious model 7. Here is a screen capture of the machine after booting the first model 7 test program.

--

=================================================

Joseph Ambrose

172 Ketcham Avenue

Amityville, NY 11701

Primary: 631-598-0528

  Mobile: 516-380-6047

    Email: jambro...@gmail.com (Job search related)

               joeam...@optonline.net (Personal)

Profile: http://www.linkedin.com/in/jambrose0321

 

[Paul Anagnostopoulos]

It's my own emulator. I love writing emulators, so I had to do my own. Also, it emulates all models, including the fictitious model 7 that has additional instructions and devices. I'll attach the Functional Characteristics for the model 7 enhancements.

~~ Paul

[Bob Flanders]

Paul,

Have you finished it? Are you going to "publish" it?

If so, I recommend github.

[Paul Anagnostopoulos]

Gee, it didn't occur to me to publish it. It's written in my personal programming language, Hearsay, and only runs on a PC. I think the SimH emulator makes more sense for people.

I will, however, keep this community informed of my exciting progress!

~~ Paul

[Paul Anagnostopoulos]

Attached is the terse documentation for the cross-assembler for the Model 7, Assembler E. It assembles assembler source files and produces object files in the format designed for OS/1130.

It is a "minor superset" of the 1130 macro assembler. The primary enhancement is more powerful expressions. Other important enhancements have to do with the macros, both flow of control and substitution. It also has a BLOCK ... BLEND construct that is a poor-man's version of an OS/360 assembler DSECT.

I believe that the enhancements are simple enough that a native assembler would still fit in 8K.

~~ Paul

[Paul Anagnostopoulos]

Today I finished writing the IPLable bootstrap loader test program. The IPLable bootstrap loader loads an 1800-format card deck starting at location 7K (OS/1130 assumes at least 8K). It then transfers control to it by branching indirect through location 7K.

The test program rings the bell, types a message with some interesting information, and enables the clock so it can display the number of seconds since boot. The attached image shows the results.

The next step is to write a card core image loader that can be IPLed and load an arbitrary card core image into location 0. Then I can write some serious code such as the standalone disk utility program.

~~ Paul

[Paul Anagnostopoulos]

I now have enough of the OS/1130 resident supervisor written that I can test the console printer driver. As with DMS2, it is not part of the supervisor, but is linked with the user application. So I have a card core image with the beginnings of the resident supervisor, hacked up system initializer, test harness program, and console printer driver. The attached image shows the output.

This is fun!

~~ Paul

[Paul Anagnostopoulos]

I just finished writing and debugging the 2501 card reader driver for OS/1130. Attached is an image of the 1130 after running both the console printer and the 2501 tests. Note the card count in the status display. Printing the final line is overlapped with reading the cards.

Next: Combined keyboard/console printer driver.

~~ Paul

[Paul Anagnostopoulos]

I've just about finished testing the combined keyboard/console printer driver for OS/1130. You can see a prompt and some keyboard input on the last line of the console paper.

Next task is the disk driver. This is going to be some fun!

~~ Paul

[yvette hirth]

i'm just so impressed with the dedication of this group!

thank you all so much! i have such fond memories of my first 'puter...

yvette hirth

[Paul Anagnostopoulos]

I've now written enough of the OS/1130 resident supervisor and library subroutines that I can link them all together to make SDMU, the standalone disk maintenance utility. This is DMS2's equivalent of DCIP. I've attached a capture of the emulator after booting SDMU. The prompts are given and responses obtained using the qry$qry subroutine, which is a fancy query routine that provides, among other things, the ability to print help when a question mark (?) is entered.

The next big task is the disk driver. Once that is complete, I will be able to format a new disk cartridge.

~~ Paul

[Paul Anagnostopoulos]

Huge progress over the last five weeks. The disk driver, a complex bit of programming, is working quite well. The standalone disk maintenance utility (SDMU) can format a disk cartridge, including the skelton disk directory and the label sector. It also tests every sector and establishes the defective cylinder table in the label.

The system disk builder (SDB) can read a system from cards and write it on a formatted disk. This includes the resident supervisor image, the supervisor overlay file, the command processor state file, and the system configuration sector (sector 1). Additional files will be included as they are developed, of course.

The system boot deck can boot the system disk and transfer control to the system initializer. The initializer is linked with the resident supervisor and acts as a user program, initializing the system by creating the device control blocks (DCB) according to the configuration, the volume information blocks (VIB), initializing overlay areas and buffers, etc. Then the initializer returns to the supervisor, which forces an ss$exit service to transfer control to the command processor.

Next task: Begin work on the interactive command language (ICL) processor.

Attached is an image of the system disk builder as it prompts for the unit number.

~~ Paul