Re: [Simh] Operating Systems with Sources

[Richard Cornwell]

Hi,

> On 23 October 2016 at 21:55, Bob Supnik  wrote:
> > Not in assembly language, but quite compact, is an early version of
> > Burroughs MCP for the B5500, which runs on the various B5500
> > simulators. 
> This distribution?
>  As an
> aside, how would one run it under SIMH?

   Pretty well actually. I wrote a simulator last year for it. 

   MCP XIII and XV run very well. Hopefully we will have XVI in a
   couple of weeks. 

   Also adding to the list of OS's TOPS 10 also was distributed by
   Source.

> Also, I'm not one hundred percent sure, but isn't OS/360 also
> distributed with assembly sources?
> Tape images can be found on Jay Maynard's site:
>  The "k360s-mb1xx-aws.zip" archive
> has the tapes needed to bring up OS/360 from scratch. (You, of course,
> need Hercules. And the initial restore wants 2311s as your DASD.)

   Yes it was. 

Rich

-- 
==
Richard Cornwell
sky...@sky-visions.com
http://sky-visions.com
LinkedIn:   https://www.linkedin.com/in/richard-cornwell-991076107
==

___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] Operating Systems with Sources

[Christian Gauger-Cosgrove]

On 23 October 2016 at 21:55, Bob Supnik  wrote:
> Not in assembly language, but quite compact, is an early version of
> Burroughs MCP for the B5500, which runs on the various B5500 simulators.
>
This distribution? 
As an aside, how would one run it under SIMH?


Also, I'm not one hundred percent sure, but isn't OS/360 also
distributed with assembly sources?
Tape images can be found on Jay Maynard's site:
 The "k360s-mb1xx-aws.zip" archive
has the tapes needed to bring up OS/360 from scratch. (You, of course,
need Hercules. And the initial restore wants 2311s as your DASD.)


Regards,
Christian
-- 
Christian M. Gauger-Cosgrove
STCKON08DS0
Contact information available upon request.
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

[Simh] Operating Systems with Sources

[Bob Supnik]

I know of at least the following additional assembly language operating 
systems that have sources, mostly on Bitsavers:


1. OS/8 - PDP-8 assembly language
2. XVM/DOS-15 - PDP-15 assembly language
3. CAPS-11 - PDP-11 assembly language
4. TSS/8 - PDP-8 assembly language
5. ADSS-9/15 - PDP-9/15 assembly language
6. RSX11C - PDP-11 assembly language
7. XVM/RSX-15 - PDP-15 assembly language
8. IBSYS - IBM 7094 assembly language
9. CTSS - IBM 7094 assembly language
10. OS/32 - Interdata 32b assembly language
11. SDS940 timesharing - SDS 940 assembly language
12. ITS - PDP-10 assembly language

Not in assembly language, but quite compact, is an early version of 
Burroughs MCP for the B5500, which runs on the various B5500 simulators.


/Bob



___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] RT-11 source

[Dave Wade]

PDP-11 OS’s or OS’s in general? 

IBM’s S/370 VM/370R6 (a very old version) has two OS’s CP and CMS and both can 
be built from source..

 

Dave

G4UGM

 

From: Simh [mailto:simh-boun...@trailing-edge.com] On Behalf Of khandy21yo
Sent: 23 October 2016 21:12
To: Al Kossow ; simh@trailing-edge.com
Subject: Re: [Simh] RT-11 source

 

Is there a list of OS that have sufficient sources to rebuild it?

Would be intresting to those who want to play.research OS designs.

I suspect a lot of the old OS have lost their source code.

 

--

___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] RT-11 source

[khandy21yo]

Is there a list of OS that have sufficient sources to rebuild it?Would be 
intresting to those who want to play.research OS designs.I suspect a lot of the 
old OS have lost their source code.
--___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] RT-11 source

[Bob Eager]

On Sun, 23 Oct 2016 14:07:33 -0400
Paul Koning  wrote:

> I've never seen DOS sources.  Also, DOS/Batch is a later version with
> more stuff in it.  DOS V4 might be better.

I couldn't agree more. We had DOS/BATCH on an 11/20 in about 1974 (we'd
had it for a year or two before I got my hands one it). 

I know someone had to patch it as it used 12 bits for the date:

  date = (year-1970)*1000 + day in month

They called it a Julian date (which it wasn't, and of course it was
incredibly bit-wasteful). It 'ran out' in March 1974, and the patch was
regularly applied to fix dates so that we worked in the middle of a
four year window.

We also put it on RK05s, and I remember reading the (badly xeroxed)
sources that DEC gave us, and working out how to patch it to fix a
fairly nasty bug in the RK05 driver.

But it was amazingly small - 4kB resident, with hundreds of tiny
overlays. I think running a program used four 128kB overlays.

> On the other hand, DOS is a truly evil operating system.  I don't
> know its internals, but just from using it you can tell it should
> never be used as a model.

Indeed! I did fire it up on SIMH a while ago.

> RT-11 is very clean.  I've worked with V2A (the FB version when
> possible, SJ when I didn't have enough memory).  They are very simple
> and compact; the UI is the old TOPS-10 style, not the newer bloated
> "DCL" interface.  So a V2 era edition would be a good way to go.

Yes, it's very nice. Someone once gave me electronic copies of some
sources (circa 1975) and I read them with glee. Lovely comments, too:
the system call dispatcher had the comment "What's it going to be then,
eh?" from the Clockwork Orange. I think those comments were only in the
FB monitor.

I assembled it on DOS/BATCH, and wrote programs in MACRO-11 to convert
the binaries to RT-11 format. I also wrote stuff to make a bootable
DECtape, got it running, and used that to transfer it to disk!
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] RT-11 source

[Al Kossow]

On 10/23/16 11:07 AM, Paul Koning wrote:

> I've never seen DOS sources.  Also, DOS/Batch is a later version with more 
> stuff in it.  DOS V4 might be better.
> 
> On the other hand, DOS is a truly evil operating system.  I don't know its 
> internals, but just from using it you can tell it should never be used as a 
> model.
> 
> RT-11 is very clean.  I've worked with V2A (the FB version when possible, SJ 
> when I didn't have enough memory).  They are very simple and compact; the UI 
> is the old TOPS-10 style, not the newer bloated "DCL" interface.  So a V2 era 
> edition would be a good way to go.
> 


DOS and RT V2C sources are on the (pretty much disorganized) dectape images 
under
www.bitsavers.org/bits/DEC/pdp11/dectape/BS/

V2CT1   DEC-11-ORTSA-E-UA1  3/16/76
RT-11 V02C SOURCE TAPE 1 OF 6
V2CT2   RT-11 V02C SOURCE TAPE 2 OF 6
V2CT3   RT-11 V02C SOURCE TAPE 3 OF 6
V2CT4   RT-11 V02C SOURCE TAPE 4 OF 6
V2CT5   RT-11 V02C SOURCE TAPE 5 OF 6
V2CT6   RT-11 V02C SOURCE TAPE 6 OF 6

___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] RT-11 source

[Paul Koning]

> On Oct 22, 2016, at 10:19 PM, Ray Jewhurst  wrote:
> 
> Thank you Johnny. I may ruffle some feathers, but I hate C.  I am mildly 
> autistic and the way my mind works I actually prefer assembly over high level 
> languages. I would really like a blueprint to see what I am doing. Is there 
> commented code for DOS/BATCH? Or even CAPS-11. I would like love to see a 
> fully commented kernel to see what I am up against.

I've never seen DOS sources.  Also, DOS/Batch is a later version with more 
stuff in it.  DOS V4 might be better.

On the other hand, DOS is a truly evil operating system.  I don't know its 
internals, but just from using it you can tell it should never be used as a 
model.

RT-11 is very clean.  I've worked with V2A (the FB version when possible, SJ 
when I didn't have enough memory).  They are very simple and compact; the UI is 
the old TOPS-10 style, not the newer bloated "DCL" interface.  So a V2 era 
edition would be a good way to go.

paul

___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] RT-11 source

[Paul Koning]

> On Oct 23, 2016, at 9:41 AM, billdeg  wrote:
> 
> Earlier I claimed that a copy of rt11 source was on a page about Project 
> Delta, I was wrong, it was RSTS/E, the link here:
> 
> https://www.eecis.udel.edu/~mader/delta/downloadrsts.html
> 
> Bill

Those are binaries (distribution kits), not sources.

paul


___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] RT-11 source

[Tom Morris]

On Sun, Oct 23, 2016 at 9:47 AM, Jacob Goense  wrote:

> On 2016-10-22 17:46, Jacob Goense wrote:
>
>> I vaguely recall lore about 4.x BSD stuff being ported to the PDP-11
>> ending
>> with the machine being thrown out of the window. Citations are welcome, or
>> maybe this should be crossed over to TUHS.
>>
>
> Found it. https://www.youtube.com/watch?v=bVSXXeiFLgk (31:50 - 33:33)


Link with timecode: https://youtu.be/bVSXXeiFLgk?t=31m50s
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] RT-11 source

[Davis Johnson]

You might think that the Tanenbaum book would be narrowly focused on 
Minix. It isn't -- it covers operating system concepts in general.


I recommend the book more than the OS. Minix is written to be an 
understandable classroom example. It does things that a serious OS 
probably would optimize better. For example a generic hardware interrupt 
handler adds a device specific interrupt handler to the process queue 
that is scheduled as a high priority process by the process scheduler. 
Unibus interrupts on a 780 weren't that bad.



On 10/23/2016 12:23 PM, Dave Wade wrote:


Ray,

I really suggest that you look at the Tanenbaum book. It goes into 
each component of an operating system and explains the structures that 
it uses and how the parts fit together. Looking at the code often does 
not explain this. Even were there are comments the usually don’t 
explain how things mesh, which is all important.


Dave

*From:*Simh [mailto:simh-boun...@trailing-edge.com] *On Behalf Of *Ray 
Jewhurst

*Sent:* 23 October 2016 03:20
*To:* Johnny Billquist 
*Cc:* simh@trailing-edge.com
*Subject:* Re: [Simh] RT-11 source

Thank you Johnny. I may ruffle some feathers, but I hate C.  I am 
mildly autistic and the way my mind works I actually prefer assembly 
over high level languages. I would really like a blueprint to see what 
I am doing. Is there commented code for DOS/BATCH? Or even CAPS-11. I 
would like love to see a fully commented kernel to see what I am up 
against.


Thanks
Ray

On Oct 22, 2016 9:11 PM, "Johnny Billquist" > wrote:


While we're at it then... Ray asked for RT-11, since he felt that
it was smaller and simpler than most other operating systems
available, and also because he felt more comfortable with
assembler than some other language.

Both those points are missed with any Unix-like OS, even if the
intention is good.

I could just as well offer up RSX, since it actually comes with
source where the comments are still in place, and it's actually
written in assembler for the most part as well. However, it is a
much more complex system than RT-11, and in some ways probably
more complex than Unix as well. So I don't think it might be a
good choice if you just want to understand how an OS works.

In fact, I would probably suggest Ray start with just writing some
code to do some simple things without looking at existing code.
The first thing needed would be to just have something that can
load programs from a device, and run them. This will require some
simple device driver, some simple file system, and a simple
command line interpreter. Then you can go on an expand from there.
You'll soon realize things you want to abstract away, and deal
with in a somewhat coherent way.
I wouldn't bother with interrupt system, MMU, or any more fancy
stuff to start with. A plain 64K PDP-11, with the program loader
just located in one end, and then go from there. Do system calls
through TRAP, EMT or some other instruction, and then have a
vector installed. If the user program overwrites that, tough luck.

Johnny



On 2016-10-23 02:45, Nelson H. F. Beebe wrote:

Ray Jewhurst > asks today for documented
operating system source code for the PDP-11. Besides the
Lions' Unix
v6 code, there is also Doug Comer's Xinu project about which
he wrote
several books.  Current versions are targeted at x86 and ARM CPUs,

http://www.xinu.cs.purdue.edu/

but he still provides code for older systems (PDP-11, SPARC, VAX):

ftp://ftp.cs.purdue.edu/pub/comer/

There is more about him here, including links to his books Web
site:

https://en.wikipedia.org/wiki/Douglas_Comer


---
- Nelson H. F. BeebeTel: +1 801 581 5254
 -
- University of UtahFAX: +1 801 581 4148
 -
- Department of Mathematics, 110 LCBInternet e-mail:
be...@math.utah.edu   -
- 155 S 1400 E RM 233 be...@acm.org 
be...@computer.org  -
- Salt Lake City, UT 84112-0090, USAURL:
http://www.math.utah.edu/~beebe/
 -

---
___
Simh mailing list
Simh@trailing-edge.com 
http://mailman.trailing-edge.com/mailman/listinfo/simh

-- 
Johnny Billquist  || "I'm on a 

Re: [Simh] RT-11 source

[Dave Wade]

Ray,

 

I really suggest that you look at the Tanenbaum book. It goes into each 
component of an operating system and explains the structures that it uses and 
how the parts fit together. Looking at the code often does not explain this. 
Even were there are comments the usually don’t explain how things mesh, which 
is all important.

 

Dave

 

From: Simh [mailto:simh-boun...@trailing-edge.com] On Behalf Of Ray Jewhurst
Sent: 23 October 2016 03:20
To: Johnny Billquist 
Cc: simh@trailing-edge.com
Subject: Re: [Simh] RT-11 source

 

Thank you Johnny. I may ruffle some feathers, but I hate C.  I am mildly 
autistic and the way my mind works I actually prefer assembly over high level 
languages. I would really like a blueprint to see what I am doing. Is there 
commented code for DOS/BATCH? Or even CAPS-11. I would like love to see a fully 
commented kernel to see what I am up against. 

Thanks 
Ray 

 

On Oct 22, 2016 9:11 PM, "Johnny Billquist"  > wrote:

While we're at it then... Ray asked for RT-11, since he felt that it was 
smaller and simpler than most other operating systems available, and also 
because he felt more comfortable with assembler than some other language.

Both those points are missed with any Unix-like OS, even if the intention is 
good.

I could just as well offer up RSX, since it actually comes with source where 
the comments are still in place, and it's actually written in assembler for the 
most part as well. However, it is a much more complex system than RT-11, and in 
some ways probably more complex than Unix as well. So I don't think it might be 
a good choice if you just want to understand how an OS works.

In fact, I would probably suggest Ray start with just writing some code to do 
some simple things without looking at existing code. The first thing needed 
would be to just have something that can load programs from a device, and run 
them. This will require some simple device driver, some simple file system, and 
a simple command line interpreter. Then you can go on an expand from there. 
You'll soon realize things you want to abstract away, and deal with in a 
somewhat coherent way.
I wouldn't bother with interrupt system, MMU, or any more fancy stuff to start 
with. A plain 64K PDP-11, with the program loader just located in one end, and 
then go from there. Do system calls through TRAP, EMT or some other 
instruction, and then have a vector installed. If the user program overwrites 
that, tough luck.

Johnny



On 2016-10-23 02:45, Nelson H. F. Beebe wrote:

Ray Jewhurst  > asks 
today for documented
operating system source code for the PDP-11.  Besides the Lions' Unix
v6 code, there is also Doug Comer's Xinu project about which he wrote
several books.  Current versions are targeted at x86 and ARM CPUs,

http://www.xinu.cs.purdue.edu/

but he still provides code for older systems (PDP-11, SPARC, VAX):

ftp://ftp.cs.purdue.edu/pub/comer/

There is more about him here, including links to his books Web site:

https://en.wikipedia.org/wiki/Douglas_Comer

---
- Nelson H. F. BeebeTel: +1 801 581 5254 
   -
- University of UtahFAX: +1 801 581 4148 
   -
- Department of Mathematics, 110 LCBInternet e-mail: be...@math.utah.edu 
   -
- 155 S 1400 E RM 233   be...@acm.org 
   be...@computer.org   -
- Salt Lake City, UT 84112-0090, USAURL: http://www.math.utah.edu/~beebe/ -
---
___
Simh mailing list
Simh@trailing-edge.com  
http://mailman.trailing-edge.com/mailman/listinfo/simh

 

-- 
Johnny Billquist  || "I'm on a bus
  ||  on a psychedelic trip
email: b...@softjar.se   ||  Reading murder 
books
pdp is alive! ||  tryin' to stay hip" - B. Idol

___
Simh mailing list
Simh@trailing-edge.com  
http://mailman.trailing-edge.com/mailman/listinfo/simh

 

___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] RT-11 source

[Dave Wade]

> -Original Message-
> From: Simh [mailto:simh-boun...@trailing-edge.com] On Behalf Of Al
> Kossow
> Sent: 22 October 2016 23:53
> To: simh@trailing-edge.com
> Subject: Re: [Simh] RT-11 source
> 
> 
> 
> On 10/22/16 2:46 PM, Jacob Goense wrote:
> > On 2016-10-22 12:55, Al Kossow wrote:
> >> On 10/22/16 7:44 AM, Ray Jewhurst wrote:
> >>> I have an idea for project to teach my self rudimentary OS design and I
> would like to use RT-11 as an example
> >>>
> >>
> >> Did anyone ever port MINIX to the PDP-11?
> >
> > No way.
> >
> 
> It ran on an 8088
> 

I think Minix and the Tannenbaum book are a great start for anyone wanting to 
learn about operating systems. The latest Minix 3 compiles with a public domain 
compiler. The book is available on Abe Books for under $10.

Dave
G4UGM

> 
> ___
> Simh mailing list
> Simh@trailing-edge.com
> http://mailman.trailing-edge.com/mailman/listinfo/simh

___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] RT-11 source

[Al Kossow]

On 10/22/16 6:11 PM, Johnny Billquist wrote:
> While we're at it then... Ray asked for RT-11, since he felt that it was 
> smaller and simpler than most other operating
> systems available, and also because he felt more comfortable with assembler 
> than some other language.
> 

If he's looking for something small, I'll try to find my copy of the RT-11 V2C 
sources today.

DOS-11 was a horror. RT-11 is much more in the spirit of OS-8


___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] RT-11 source

[Jacob Goense]

On 2016-10-22 17:46, Jacob Goense wrote:
I vaguely recall lore about 4.x BSD stuff being ported to the PDP-11 
ending
with the machine being thrown out of the window. Citations are welcome, 
or

maybe this should be crossed over to TUHS.


Found it. https://www.youtube.com/watch?v=bVSXXeiFLgk (31:50 - 33:33)
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

[Simh] Fwd: Re: RT-11 source

[billdeg]

Sent from my Digital PDP 1105 

Earlier I claimed that a copy of rt11 source was on a page about Project Delta, 
I was wrong, it was RSTS/E, the link here:
https://www.eecis.udel.edu/~mader/delta/downloadrsts.html
Bill
Sent from my Digital PDP 1105 ___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] RT-11 source

[Johnny Billquist]

On 2016-10-23 14:45, Rhialto wrote:

On Sat 22 Oct 2016 at 23:03:41 +0200, Johnny Billquist wrote:

On 2016-10-22 18:55, Al Kossow wrote:



On 10/22/16 7:44 AM, Ray Jewhurst wrote:

I have an idea for project to teach my self rudimentary OS design and I would 
like to use RT-11 as an example



Did anyone ever port MINIX to the PDP-11?


I don't think so. I suspect it would be non-trivial, as the memory
limitations of the PDP-11 require that you sometimes need to design with
that in mind from the start, or else face serious problems.


But the original 8086 MINIX (which I used in university in the course on
operating system design) used "small memory model". That means it sets
the segment registers for each process to a fixed value and therefore
its processes (including kernel server processes) were limited to 64KB
of text and 64KB of data. That would fit with PDP-11s with sepid just
fine.


Possible. I haven't looked at that code, so I can only guess. However, 
the segmentation registers essentially mean that you have more than 64K, 
as you note (64K+64K). With separate I/D space you accomplish the same 
on the -11, but it works in a different way, so depending on how the 
code is written, it could be easy, or it could turn out to be a real 
headache.
Like I said, you sometimes needs to design things with the hardware in 
mind, as the solutions might be very different, even with similar 
constraints.


And of course, this then requires that you handle the MMU on the -11. 
How did Minix work? I don't think the 8086 had much of a proper MMU. Did 
you have any memory protection, or did the kernel somehow live in your 
address space? If not, how did the remapping changes needed on 
interrupts work? (Yes, I do not know the 8086 that well...)



At some point during inter-process communication it no doubt uses the
segmentation registers more actively (I don't remember the details) but
that can be done in some other ways on the -11.


Everything is always possible. It's just a question of how much work, 
and how slow will the result be.


Johnny

--
Johnny Billquist  || "I'm on a bus
  ||  on a psychedelic trip
email: b...@softjar.se ||  Reading murder books
pdp is alive! ||  tryin' to stay hip" - B. Idol
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] RT-11 source

[Rhialto]

On Sat 22 Oct 2016 at 23:03:41 +0200, Johnny Billquist wrote:
> On 2016-10-22 18:55, Al Kossow wrote:
> > 
> > 
> > On 10/22/16 7:44 AM, Ray Jewhurst wrote:
> > > I have an idea for project to teach my self rudimentary OS design and I 
> > > would like to use RT-11 as an example
> > > 
> > 
> > Did anyone ever port MINIX to the PDP-11?
> 
> I don't think so. I suspect it would be non-trivial, as the memory
> limitations of the PDP-11 require that you sometimes need to design with
> that in mind from the start, or else face serious problems.

But the original 8086 MINIX (which I used in university in the course on
operating system design) used "small memory model". That means it sets
the segment registers for each process to a fixed value and therefore
its processes (including kernel server processes) were limited to 64KB
of text and 64KB of data. That would fit with PDP-11s with sepid just
fine.

At some point during inter-process communication it no doubt uses the
segmentation registers more actively (I don't remember the details) but
that can be done in some other ways on the -11.

>   Johnny
-Olaf.
-- 
___ Olaf 'Rhialto' Seibert  -- Wayland: Those who don't understand X
\X/ rhialto/at/xs4all.nl-- are condemned to reinvent it. Poorly.


signature.asc
Description: PGP signature
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh