31 Dec
2018
31 Dec
'18
2:25 a.m.
Dear Unix Enthusiasts,
We are seriously considering upgrading our PDP 11/40 clone (SIMH), to a PDP 11/45
(preferably another SIMH) for our Unix v6 installation. Our CEO was traveling and met a
techie in first class (seriously, first class?) who told him that we needed one. I thought
I had better ask some folks who have gone before about it before we jumped on the
bandwagon. By way of background, Our install is pretty small with a few rk’s and 256K of
ram along with a few standard peripherals, and some stuff our oldtimers refuse to part
with (papertape, card punch, etc). It has fairly low utilization - a developer logs in and
writes code every few days and the oldtimers hunt the wumpus and play this weird Brit game
about cows. It could be considered a casual development and test environment and an
occasional gaming console.
Here is what I would like to know that I think y’all might be particularly equipped to
answer:
1. Are there any v6 specific concerns about upgrading?
2. Why should we consider taking the leap to the 11/45? Everything seems to work fine now.
3. If we jump in and do the upgrade, how can we immediately recognize what has changed in
the environment? I.e what are some things that we can now do that we couldn’t do before?
4. If we just insert our current diskpacks into the new system, will it just boot and
work? Or what do we need to before/after booting to prepare/respond to the new system?
5. Is 256K enough memory or what configuration do y’all recommend?
6. Is there anything else we need to know about?
Regards,
Will
Sent from my iPhone
31 Dec
31 Dec
2:43 a.m.
The primary difference between the 11/40 class and 11/45 class is separate I/D space which
I sometimes refer to as the 17th address bit because it allows you a full 64k of data
space as well as a 64k of instructions space.
After you are booted, a 45 class machine will run 40 class binaries unchanged. 40 class
machines can not run a.outs that are seperate I/D
You’ll probably want to configure a kernel for the 45 class machine. Look at the
differences in the *.s files in the kernel. IIRC there is a different file for 40 class
and 45 class systems
That said if you running the simh I would recommend going all the way to an 11/70
configuration because you can set it up for 4M of main memory.
Clem
Sent from my PDP-7 Running UNIX V0 expect things to be almost but not quite.
3:31 a.m.
Do you know of some commonly used at the time v6 programs that needed that much space?
After you are booted, a 45 class machine will run 40
class binaries unchanged. 40 class machines can not run a.outs that are seperate I/D.
Good to know.
You’ll probably want to configure a kernel for the 45
class machine. Look at the differences in the *.s files in the kernel. IIRC there is a
different file for 40 class and 45 class systems
I read about this in ’Setting up Unix Sixth Edition” and I see the source comments. It
looks pretty straightforward to configure the system for separate I/D. Is there any
material difference between doing it at install time vs having run on 11/40 for a while
and moving the disk over to the 11/45 later?
On a related note, how difficult is it to copy the system from rk to hp? I know I can
rebuild, but I’m sure there’s a quicker/easier method...
That said if you running the simh I would recommend
going all the way to an 11/70 configuration because you can set it up for 4M of main
memory.
Maybe this will be the next upgrade :).
2:55 p.m.
On Sun, Dec 30, 2018 at 10:31 PM Will Senn <will.senn(a)gmail.com> wrote:
Besides everything that Noel pointed to you too, look at 1BSD - the pascal
system needs to be seperate I/D for sure. I believe ex is linked seperate
I/D. The C compiler itself can be, the reason to do that it allow the
tables to be larger, so you don't run into errors where you run out of
space.
My experience from the old days, was that once you had seperate I/D, we
tended to link most programs that way if we could, as we slowly deprecated
the 11/40 class systems. The original Tek Labs machine was an 11/60
(which is a 40 class), and was quickly replaced with an 11/70. The 60
is what I used for the original Able Enable work that Noel referenced, so
it have 4M in it for a short period (we borrowed the memory for the
development).
But within 2 years we had a 11/44 to replace the 11/60, which was seperate
I/D system.
Making a system work on both could be done, but it chews up precious
address space in code that will not be executed.
Remember - what seems 'natural' for the modern user of cramming everything
into a single binary, or keeping lots of copies of things, was not done.
You lack address space, main memory or disk space.
Easiest method is probably grabing the v6tar binary that you described how
to make in your v6 for sim6 directionions, then use and dual tar
programs***. Other wise, find(1) is your friend. That said, PWD (1.0) was
v6 based, so there is a version of cpio on the spencer_pwb.tar.gz tape.
That should work.
Clem
*** Note to Warren. It might be a wise to put copies of v6tar (both
seperate I/D and not) binaries and maybe cpio(v6) on the TUHS we site in
the V6 directory; maybe, a 'collected_tools' directory. Noel's tools would
probably make sense to add there also. I bet people that are downloading
and playing might find them helpful.
3:05 p.m.
Yep, and notwithstanding split I/D, we switched to -n (410 magic) read-only code space.
Of course, the kludge nargs() function didn’t work at all in split-I/D mode unless you
made a hack to the processor to change the behavior of MFPI.
3:53 p.m.
On Mon, Dec 31, 2018 at 10:05 AM <ron(a)ronnatalie.com> wrote:
Agreed, that was pretty standard and used the sticky bit as needed. IIRC
this was to help sharing, but I admit those bits in my brain had not been
refreshed in years.
Ah yes, I've forgotten that hack. Does simh or any of the other
simulators support it? It would need to be an option of course.
ᐧ
5:30 p.m.
Yeah, ISVTX only worked on 410 and 411 files. It locked the text segment in swap. It
kind of got obsoleted by later versions that could load stuff direct to memory from the
disk image.
From: Clem Cole <clemc(a)ccc.com>
Sent: Monday, December 31, 2018 10:54 AM
To: Ronald Natalie <ron(a)ronnatalie.com>
Cc: Will Senn <will.senn(a)gmail.com>; The Eunuchs Hysterical Society
<tuhs(a)tuhs.org>
Subject: Re: [TUHS] Upgrading from 11/40 to 11/45 in Unix v6
On Mon, Dec 31, 2018 at 10:05 AM <ron(a)ronnatalie.com <mailto:ron@ronnatalie.com>
> wrote:
Yep, and notwithstanding split I/D, we switched to -n (410 magic) read-only code space.
Agreed, that was pretty standard and used the sticky bit as needed. IIRC this was to help
sharing, but I admit those bits in my brain had not been refreshed in years.
Of course, the kludge nargs() function didn’t work at all in split-I/D mode unless you
made a hack to the processor to change the behavior of MFPI.
Ah yes, I've forgotten that hack. Does simh or any of the other simulators support
it? It would need to be an option of course.
<https://mailfoogae.appspot.com/t?sender=aY2xlbWNAY2NjLmNvbQ%3D%3D&type=zerocontent&guid=0488a3b9-8b85-4fae-afd5-ff3bcec06dd4>
ᐧ
11:36 p.m.
New subject: Useful Unix tools + Usenix tapes
On Mon, Dec 31, 2018 at 09:55:31AM -0500, Clem Cole wrote:
In the Unix Archive, there is this location:
https://www.tuhs.org/Archive/Tools/
It's separate from the distributions. Pro: it keeps the original files
separate from 3rd party things; con: it's a bit harder to find things
when you need them.
If anybody has other tools or useful utilities to add in here, let me know!
There are some Usenix tapes in the archive here:
https://www.tuhs.org/Archive/Applications/
Look in Shoppa_Tapes, Spencer_Tapes and Spencer_Tapes. If there are
other tape images out there that I could add, let me know as well.
Happy New Year, everyone.
Cheers, Warren
1 Jan
1 Jan
1:30 a.m.
New subject: Useful Unix tools + Usenix tapes
Warren
I understand and thank you. That is surely a possible place to put things. I was thinking
something a little different. The directory you have are more general tools that really
apply across releases and specific targets.
I was thinking when you have tools like the set I mentioned previously that are system
specific and you probable want to supply target binaries that you try to keep them in a
directory next the system that they relate. Thus in your Research directory - create
a v6 specific contributed tool directory.
Sent from my PDP-7 Running UNIX V0 expect things to be almost but not quite.
1:58 a.m.
New subject: Useful Unix tools + Usenix tapes
Warren There are a number of usenix tapes missing in your archives. The first three were
called 1 2 and 3 from the mid 70s but where included in the Harvard tape*** from about 76
or 77 which I sort of consider the first usenix tape. Note that This is an stp format
distribution. Which IIRC the v6 tp could read or at least read it enough to pull the
stp binary off the tape which would then allow you read the whole thing. You will
need the v6 ar because the directories inside the tape were archived as files called
cont.a And I think I remember that some of those were compressed with pack/unpack tools
which were in the wild in those days — probably also on the Harvard tape. As I
mentioned the other day the 1BSD tape you have seems to be a conversion from stp to tar.
FYI: one of the issues with tp and stp is that the directory for the tape is at the
beginning of the tape itself and is fixed in size [this is have DECtape worked]. Because
it was fixed in size folks archived directories together so the tp directory needed only
the folder and a single file it. FWIW One of the big enhancements tar provided over tp
was the threading the directory throughout the archive which eliminated tat issue, but of
course if there is a tape error recovery is more difficult. IIRC Harvard had added a
second directory to the end of tp in the stp format to help reliability. Ie if you had
errors in the tp directory at the front of the tape, you had a chance to recover by using
the second copy of it.
*** the Harvard tape takes it name from the meeting at Harvard of the Unix News readers.
This would become USENIX as an org shortly there after. The earlier tapes (1 2 and 3)
were what files had been collected at earlier meetings.
Sent from my PDP-7 Running UNIX V0 expect things to be almost but not quite.
2:08 a.m.
New subject: Useful Unix tools + Usenix tapes
Regarding these aggregates (tapes, zips, tars etc) can we consider the
option please of having them unpacked into their tree structure so a
search-engine can index the content?
2:17 a.m.
New subject: Useful Unix tools + Usenix tapes
On Tue, Jan 01, 2019 at 01:08:19PM +1100, Nigel Williams wrote:
Argh! I'm mindful of the disk space on my system and also on the
mirrors.
How about I write a script that unpacks and builds a table of contents
list for all the tarball and Zip files in the archive. It would run
regularly and leave a text file in the Archive root.
Cheers, Warren
8:18 a.m.
New subject: Useful Unix tools + Usenix tapes
On Tue, Jan 01, 2019 at 12:17:29PM +1000, Warren Toomey wrote:
Done. It seems to work, but let me know if there are small issues.
The file is generated daily and is at:
https://www.tuhs.org/Archive/tarball_tocs.txt.gz
Cheers, Warren
31 Dec
31 Dec
2:47 a.m.
One other thought. The basic v6 has and RP driver in it. I’ve forgotten if RP06 was in
there but the RP04 certainly was by then. Check the driver rp.c to see what is
configured. But either way you should configure the system to use the largest drive v6
has.
Sent from my PDP-7 Running UNIX V0 expect things to be almost but not quite.
2326
days inactive
2327
days old
17 comments
7 participants
participants (7)
-
Clem cole -
Clem Cole
-
Nigel Williams -
Paul Winalski -
ron＠ronnatalie.com -
Warren Toomey -
Will Senn