Current Status - Githubissues

PashPaw commented 4 years ago

So, I got it to compile. Sort of.

It's not willing to run, however. There are still too many missing symbols for it to be functional. We might want to get a hold of RAND to see if they have anything to help us like more listings.

If not, it's going to take some serious reverse engineering to get it to work. It looks like KMON covers both the supervisor and distributor.

larsbrinkhoff commented 4 years ago

I'm working on RAND, but so far I haven't got anything beyond those two additional documents.

PashPaw commented 4 years ago

Okay. Thanks. The CHM has some documents of historical significance (as you likely know) but trying to get anything out of them is going to be impossible.

Things born digital should be made easily digitally accessible.

larsbrinkhoff commented 4 years ago

Do you have a list? I may have a favor to call in at the CHM.

PashPaw commented 4 years ago

https://www.computerhistory.org/collections/catalog/102734547 https://www.computerhistory.org/collections/catalog/102679426

larsbrinkhoff commented 4 years ago

One thing that remains to be done, is to compare the addresses in the new listing with the old. I don't think they match yet.

PashPaw commented 4 years ago

True. Do you think the extra comments could be causing that or something else?

PashPaw commented 4 years ago

This is peculiar. Pg. 41 doesn't seem to be missing any instructions and most of the page's addresses match except for the last few instructions.

dabridgham commented 4 years ago

I don't have the combined file you're assembling but it looks like in the file I typed in I missed one instruction. There's supposed to be a "jsr cvtl1" just before the "aos tint(c)".

PashPaw commented 4 years ago

Thanks. Go ahead and correct it.

I've just been just joining different files together with cat and using a tool included with simh called asc to process the line endings so TOPS-10 can read them properly after using the simulated KA10's paper tape reader (it's a stupid trick that also works with the PDP-11 sims).

I don't know what @larsbrinkhoff is doing to get them in. He's probably using a file he's created on his own and we're coming up with the same results.

Unprocessed (jos_su.mac): jos_su.pt.txt . Processed (jos_su.pt): jos_su.txt

PashPaw commented 4 years ago

Never mind. Added.

PashPaw commented 4 years ago

On TOPS-10, using + instead of commas seems to produce fewer errors. I might need to see why this is.

I might have to see what the <> notation does.

OH! It tells the compiler and/or loader that it's one program more easily.

larsbrinkhoff commented 4 years ago

All I did was cat page*.mac > joss.mac and then edited joss.mac to have CRLF line endings.

Maybe it's time to go from the separate pageNNN files to a single su.mac?

PashPaw commented 4 years ago

Maybe but the separate files are easier to edit.

Here’s a compromise: produce a single file that will be updated when the pages get fixed.

larsbrinkhoff commented 4 years ago

My CHM agent says he's working on it. Especially the first item looks interesting because the description includes "program listing".

PashPaw commented 4 years ago

Cool! Thanks.

So, my hunch was likely correct?

larsbrinkhoff commented 4 years ago

They say it's 30 cents per scanned page.

PashPaw commented 4 years ago

And how many pages?

larsbrinkhoff commented 4 years ago

I don't know, that's all I heard so far.

larsbrinkhoff commented 3 years ago

CHM quoted $6000 for scanning their documents.

PashPaw commented 3 years ago

They’re out of their mind. That reads like they don’t want to do it.

Sent from my iPhone

On Aug 24, 2021, at 3:22 AM, Lars Brinkhoff @.***> wrote: CHM quoted $6000 for scanning their documents.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

larsbrinkhoff commented 3 years ago

I contacted RAND Archives again through their online form and got this back:

Thank you for contacting the RAND Corporation Archives again. Since you're interested in taking at look at more public material on RAND's JOSS program, I've attached a custom bibliography of all of RAND's public papers and reports on JOSS to this email.

For all of the papers and reports not currently available online for free on our www.rand.org website, you can order them through our Customer Services department at order@rand.org. If you're only looking for a few papers that aren't online yet, I can scan them for you as a courtesy. Just let me know which ones you're most interested in.

JOSS Bibliography.docx

larsbrinkhoff commented 3 years ago

Going through that bibliography, I see some things from around 1967 that seem of interest:

JOSS: User Scheduling and Resource Allocation. Santa Monica, CA: RAND Corporation, RM-5216-PR
JOSS: Accounting and Performance Measurement. Santa Monica, CA: RAND Corporation, RM-5217-PR
JOSS: Introduction to the System Implementation. Santa Monica, CA: RAND Corporation, P-3486

Everything else is either already online, or is not related to the PDP-6 implementation.

PashPaw commented 3 years ago

Going through that bibliography, I see some things from around 1967 that seem of interest:

JOSS: User Scheduling and Resource Allocation. Santa Monica, CA: RAND Corporation, RM-5216-PR

JOSS: Accounting and Performance Measurement. Santa Monica, CA: RAND Corporation, RM-5217-PR

JOSS: Introduction to the System Implementation. Santa Monica, CA: RAND Corporation, P-3486

Everything else is either already online, or is not related to the PDP-6 implementation.

Correct. In fact, those are the bits we likely wanted.

Also, the Disc File System and Distributor documents contain a weird mix of pseudocode and machine code in them. It'll be an undertaking but we might be able to resurrect them.

There may be some other difficulties but we'll get to that when we get there. In the meantime, it's probably time for me to start studying MACRO-10 again with whatever else I'm studying ATM.

larsbrinkhoff commented 3 years ago

So I'll ask for those three documents. After that, I guess we'll have to roll our own or get code from AID for the components we don't have original JOSS-II source code for.

larsbrinkhoff commented 3 years ago

Hey all,

I received those three documents. They are now on a branch called lars/rand.

larsbrinkhoff commented 3 years ago

Note, the documents are not available for free from the Rand website so please don't distribute.

PashPaw commented 3 years ago

Cool! I’ll merge them later.

Sent from my iPhone

On Sep 13, 2021, at 10:03 PM, Lars Brinkhoff @.***> wrote:

Hey all,

I received those three documents. They are now on a branch called lars/rand.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

PashPaw commented 3 years ago

Or not. But okay.

Sent from my iPhone

On Sep 13, 2021, at 10:42 PM, Lars Brinkhoff @.***> wrote:

Note, the documents are not available for free from the Rand website so please don't distribute.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

PashPaw commented 3 years ago

Oh. And thanks to both you and RAND, Lars. I don’t think I would have gotten as far without help.

So, from a cursory look in the JOSS: User Scheduling and Allocation, there is a single piece of non-DEC hardware mentioned because of the line concentrator. It’s apparently custom and based around the Stowager switch. It then inputs into a line the 630 if I’m reading things right.

Sent from my iPhone

On Sep 14, 2021, at 3:50 AM, Paula Ash @.***> wrote:

Cool! I’ll merge them later.

Sent from my iPhone

On Sep 13, 2021, at 10:03 PM, Lars Brinkhoff @.***> wrote:

Hey all,

I received those three documents. They are now on a branch called lars/rand.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

PashPaw commented 3 years ago

My current assumption is that kmon was a shim for the monitor. I could be wrong but the AID code does mention some weird things. I should figure out how to extract the code from the TOPS-10 7.03/7.04 tapes and upload it since segments were removed.

There's a mention of DDT in it, which, even for Monitor 5.03, would have been odd. AID returned to the Monitor level and never to DDT on Monitor/TOPS-10 and TOPS-20. I should run diff on the known files and find what changed in the code. I'm also tempted to rename the repo to JOSS-II, ATM.

PDP-6 / JOSS-II

Current Status #33