Re: [Simh] OpenBSD on Simh VAX 3.8-1 - Segmentation fault

[Mark Pizzolato]

On Wednesday, November 8, 2017 at 9:43 PM, Gregg Levine wrote:
> Hello!
> Gary which release  of Slackware Linux was this? I've found that SIMH
> properly builds on all releases from 7.2 (Which was only released as a Snap
> Shot Disk from the time of the first shows) all the way the 14.1.
> I only needed to figure out how to update the LibCap library so that
> networking would work properly once that arrived not too long ago. As an
> aside I moved away from trying to build it here on this laptop running 14.2
> 64bit, because networking requires a fixed connection, not WiFi.

His issues might not have much to do with Slackware, but more to do 
with the version of simh.

The github simh code (4.x) has built easily on most systems for a very long 
time.

Simh 3.9 was reasonable as well but not as robust as the current code.

Simh 3.8-x might have needed local customizations relating to the 
local system environment.  The folks putting together the debian and 
other simh packages certainly messed with the makefile.

- Mark

> -
> Gregg C Levine gregg.drw...@gmail.com
> "This signature fought the Time Wars, time and again."
> 
> 
> On Wed, Nov 8, 2017 at 11:11 AM, Gary Lee Phillips 
> wrote:
> > Thanks for the very clear answer, Mark.
> >
> > I followed your instructions and the 4.0-0 beta built in minutes. Last
> > time I tried to build Simh myself, it took several frustrating days of
> > patching the makefile and changing references. That was on Slackware
> > Linux and several versions back of both the environment and Simh. So
> > thanks ever so much to the developers who have continued to enhance and
> tune this package.
> > It is brilliantly done.
> >
> > Now the even better news. The vax3900/vax module in 4.0-0 beta does
> > not have the segmentation fault I encountered with 3.8-1. OpenBSD 5.7
> > is able to ftp and sftp transfer in and out without any complaints. So
> > thank you again to whoever it was that found and fixed that problem.
> >
> > --Gary
> >
> >
> >
> > On Wed, Nov 8, 2017 at 1:50 AM, Mark Pizzolato 
> wrote:
> >>
> >> Hi Gary,
> >>
> >> On Tuesday, November 7, 2017 at 5:29 AM, Gary Lee Phillips wrote:
> >> > I am running VAX 3.8-1 on a 64-bit Linux system. It performs well
> >> > with OpenVMS as the operating system. But when I try to run it with
> >> > OpenBSD 5.7 there are issues.
> >> >
> >> > The base system installs and boots correctly. But in order to bring
> >> > in tools such as language compilers or editors, it is necessary to
> >> > use ftp or sftp. I cannot get these to transfer anything because
> >> > the moment a transfer starts, no matter which mode or direction or
> >> > whether the transfer is initiated from within OpenBSD or from the
> >> > outside, an immediate segmentation fault occurs. This appears to be
> >> > within the SimH code itself, and not on OpenBSD.
> >>
> >> That would certainly seem to be the case.  Simulator crashes
> >> definitely should be fixed.  3.8-1 was released some 8 years ago, and
> >> 3.9 was some 6 years ago.  Both are way behind the current code base
> >> located at https://github.com/simh/simh/ downloadable as
> >> https://github.com/simh/simh/archive/master.zip
> >>
> >> > The same SimH VAX environment running OpenVMS 7.3 can perform
> >> > transfers over ethernet without any apparent issues.
> >>
> >> This is a useful data point.
> >>
> >> > I have tried running SimH in supervisor mode or in user mode, same
> >> > results. Also same results whether the transfer is initiated by
> >> > root or by a normal user account.
> >>
> >> I wouldn't expect you to be able to run the simulator as non-root and
> >> have Ethernet functionality...
> >>
> >> > The ethernet connection itself seems to be working. Telnet
> >> > connections function normally, and ftp or sftp will pass directory
> >> > information and accept commands, but file transfers fail.
> >> >
> >> > I guess I could try another version of OpenBSD. Upgrading SimH is
> >> > also possible but a bit more daunting since 3.8 is what is offered
> >> > as current on my Linux distro. (Linux Mint 18 "Sarah" 64 bit
> >> > generic
> >> > 4.4.0-97)
> >>
> >> It really isn't particularly daunting:
> >>
> >> First, you'll need the following:
> >>
> >> $ sudo apt install libpcap-dev
> >> Then:
> >> $ wget https://github.com/simh/simh/archive/master.zip
> >> $ unzip master.zip
> >> $ cd simh-master
> >> $ make vax
> >> $# Start the newly built simulator:
> >> $ BIN/vax
> >>
> >>
> >> If you see any sort of segmentation fault, there is a problem that needs
> >> fixing, and I'll be glad to address it.   If you have a problem please
> >> create
> >> an issue at https://github.com/simh/simh/issues
> >>
> >> - Mark
> >
> >
> >
> > ___
> > Simh mailing list
> > Simh@trailing-edge.com
> > http://mailman.trailing-edge.com/mailman/listinfo/simh

___
Simh mailing list
Simh@

Re: [Simh] Attach tar to tape device and access it from v7 in simh

[Lars Brinkhoff]

Johnny Billquist  writes:
> I bet that would be written in some ANSI C, for which you won't find a
> C compiler for V7...

GCC has a PDP-11 backend, so maybe you can cross compile it.

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

Re: [Simh] OpenBSD on Simh VAX 3.8-1 - Segmentation fault

[Gregg Levine]

Hello!
Gary which release  of Slackware Linux was this? I've found that SIMH
properly builds on all releases from 7.2 (Which was only released as a
Snap Shot Disk from the time of the first shows) all the way the 14.1.
I only needed to figure out how to update the LibCap library so that
networking would work properly once that arrived not too long ago. As
an aside I moved away from trying to build it here on this laptop
running 14.2 64bit, because networking requires a fixed connection,
not WiFi.
-
Gregg C Levine gregg.drw...@gmail.com
"This signature fought the Time Wars, time and again."


On Wed, Nov 8, 2017 at 11:11 AM, Gary Lee Phillips  wrote:
> Thanks for the very clear answer, Mark.
>
> I followed your instructions and the 4.0-0 beta built in minutes. Last time
> I tried to build Simh myself, it took several frustrating days of patching
> the makefile and changing references. That was on Slackware Linux and
> several versions back of both the environment and Simh. So thanks ever so
> much to the developers who have continued to enhance and tune this package.
> It is brilliantly done.
>
> Now the even better news. The vax3900/vax module in 4.0-0 beta does not have
> the segmentation fault I encountered with 3.8-1. OpenBSD 5.7 is able to ftp
> and sftp transfer in and out without any complaints. So thank you again to
> whoever it was that found and fixed that problem.
>
> --Gary
>
>
>
> On Wed, Nov 8, 2017 at 1:50 AM, Mark Pizzolato  wrote:
>>
>> Hi Gary,
>>
>> On Tuesday, November 7, 2017 at 5:29 AM, Gary Lee Phillips wrote:
>> > I am running VAX 3.8-1 on a 64-bit Linux system. It performs well
>> > with OpenVMS as the operating system. But when I try to run it
>> > with OpenBSD 5.7 there are issues.
>> >
>> > The base system installs and boots correctly. But in order to bring
>> > in tools such as language compilers or editors, it is necessary to
>> > use ftp or sftp. I cannot get these to transfer anything because the
>> > moment a transfer starts, no matter which mode or direction or
>> > whether the transfer is initiated from within OpenBSD or from the
>> > outside, an immediate segmentation fault occurs. This appears to
>> > be within the SimH code itself, and not on OpenBSD.
>>
>> That would certainly seem to be the case.  Simulator crashes
>> definitely should be fixed.  3.8-1 was released some 8 years ago,
>> and 3.9 was some 6 years ago.  Both are way behind the current
>> code base located at https://github.com/simh/simh/ downloadable
>> as https://github.com/simh/simh/archive/master.zip
>>
>> > The same SimH VAX environment running OpenVMS 7.3 can
>> > perform transfers over ethernet without any apparent issues.
>>
>> This is a useful data point.
>>
>> > I have tried running SimH in supervisor mode or in user mode,
>> > same results. Also same results whether the transfer is initiated
>> > by root or by a normal user account.
>>
>> I wouldn't expect you to be able to run the simulator as non-root
>> and have Ethernet functionality...
>>
>> > The ethernet connection itself seems to be working. Telnet
>> > connections function normally, and ftp or sftp will pass directory
>> > information and accept commands, but file transfers fail.
>> >
>> > I guess I could try another version of OpenBSD. Upgrading SimH
>> > is also possible but a bit more daunting since 3.8 is what is offered
>> > as current on my Linux distro. (Linux Mint 18 "Sarah" 64 bit generic
>> > 4.4.0-97)
>>
>> It really isn't particularly daunting:
>>
>> First, you'll need the following:
>>
>> $ sudo apt install libpcap-dev
>> Then:
>> $ wget https://github.com/simh/simh/archive/master.zip
>> $ unzip master.zip
>> $ cd simh-master
>> $ make vax
>> $# Start the newly built simulator:
>> $ BIN/vax
>>
>>
>> If you see any sort of segmentation fault, there is a problem that needs
>> fixing, and I'll be glad to address it.   If you have a problem please
>> create
>> an issue at https://github.com/simh/simh/issues
>>
>> - Mark
>
>
>
> ___
> 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] TE16 CRC Errors

[Johnny Billquist]

On 2017-11-08 15:00, Paul Koning wrote:




On Nov 7, 2017, at 8:01 PM, Johnny Billquist  wrote:

On 2017-11-08 01:54, Paul Koning wrote:

...RSTS definitely did not.  It does support mixed RP/RM massbus, which is more 
usual but not universal either, I think.  But no mixing disk and tape.


Thanks. I was wondering about RSTS/E, but didn't dare saying something 
definitive. That it allows mixing of RM/RP is nice. I think in plain -11M, even 
that might not be allowed, but I would have to go and check and refresh my 
memory.
M+ allows any kind of combinations.


It's possible that the first Massbus support in RSTS didn't allow mixing RP and 
RM, but it certainly appeared soon after, I'm pretty sure by V7.0 at the 
latest.  So as of that point, RSTS supported two Massbus adapters for disks, 
and each of these could have any Massbus disk on it.  Even an RP07, on an 11/70 
at least, though officially that was not supported (DEC didn't sell that device 
on lowly PDP-11s).


Heh. I sortof noticed. I was running an RP07 on an 11/70 for several 
years in the 90s. RSX more or less have support for it, but disk 
formatting tool lacked the module for it, so I had to write that one. 
The rest worked fine. All drivers, error logging, SYSGEN and so on 
supported the RP07 just fine. I assumed that the issue with formatting 
was because they in the end decided to not really care about the RP07 
for PDP-11s, and there was also a formatting tool under XXDP, which I 
assumed people would have to use instead.
And of course, documentation do not mention the RP07 either (it also 
mostly omits the TU78, even though there was a device driver for that too).


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] Attach tar to tape device and access it from v7 in simh

[Warren Young]

On Wed, Nov 8, 2017 at 2:07 PM, Johnny Billquist  wrote:

> I bet that would be written in some ANSI C, for which you won't find a C
> compiler for V7...
>

The latest Less 487  is still K&R C.
Whether it'll build on UNIX V7 is a different question.

I'd start with Less 290 and maybe upgrade from there if that works.
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] Attach tar to tape device and access it from v7 in simh

[Bob Eager]

There's a version for 2.11BSD here; that ought to compile OK...

ftp://minnie.tuhs.org/UnixArchive/Distributions/UCB/2.11BSD_patch277/tape2.tar


On Wed, 8 Nov 2017 22:07:49 +0100
Johnny Billquist  wrote:

> I bet that would be written in some ANSI C, for which you won't find
> a C compiler for V7...
> 
>Johnny
> 
> On 2017-11-08 03:09, khandy21yo wrote:
> > Does V7 require a special version, or can you just download less
> > from gnu.org?
> > 
> > 
> > 
> > Sent from my Galaxy Tab® A
> > 
> >  Original message 
> > From: Mark Abene 
> > Date: 11/7/17 6:46 PM (GMT-07:00)
> > To: Will Senn 
> > Cc: simh@trailing-edge.com
> > Subject: Re: [Simh] Attach tar to tape device and access it from v7
> > in simh
> > 
> > I personally don't recall there ever being a less for v7. more, ex,
> > vi, and csh were on a tape image of UCB tools, if you can find that.
> > 
> > -Mark
> > 
> > 
> > On Tue, Nov 7, 2017 at 2:02 PM, Will Senn  > > wrote:
> > 
> > On 11/7/17 2:45 PM, Clem Cole wrote:  
> >>
> >>
> >> On Tue, Nov 7, 2017 at 3:37 PM, Clem Cole  >> > wrote:
> >>
> >> You need to convert the foo.tar file to a foo.tap before
> >> simh's mag tape reader will understand it.
> >>
> >>
> >> ?1.) ?
> >> gunzip 1bsd.tar.gz
> >> 2
> >> ?.)
> >> enblock <1bsd.tar >
> >> ?1bsd
> >> .tap
> >> ?
> >> ?3.) ?in simh: att mt0 1bsd.tap
> >> ?4.) in unix:  ?tar vxfb /dev/rmt0 ...
> >>  
> > Thanks, Clem. Now I'm getting somewhere!
> > 
> > I converted to a tap file (lo and behold, I wrote a perl script
> > that did this for the silly thing 2 years ago) and ran it. Then
> > attached the tap to simh and in v7:
> > 
> > # tar xvfb /dev/rmt0
> > Invalid blocksize. (Max 20)
> > 
> > hmm... that doesn't sound good, let's try it another way:
> > 
> > # tar xv0
> > tar: ashell/ - cannot create
> > x ashell/cont.a, 114488 bytes, 224 tape blocks
> > x ashell/READ_ME, 825 bytes, 2 tape blocks
> > # ls -ld ashell
> > drwxrwxr-x 2 root   64 Dec 31 22:18 ashell
> > # ls ashell
> > READ_ME
> > cont.a
> > # find . -name "more*" -a -print
> > #
> > ...
> > 
> > It's coming back to me now, but oh so slowly, why is it
> > complaining about cannot create all the while creating the folder?
> > Also, darn! I was hoping more would be in there somewhere. Anybody
> > know where the more or less pager source code might be for v7?
> > 
> > Thanks,
> > 
> > Will
> > 
> > 
> > 
> > -- 
> > GPG Fingerprint: 68F4 B3BD 1730 555A 4462  7D45 3EAA 5B6D A982
> > BAAF
> > 
> > 
> > ___
> > 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
> >   
> 
> 

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

Re: [Simh] Attach tar to tape device and access it from v7 in simh

[Johnny Billquist]

I bet that would be written in some ANSI C, for which you won't find a C 
compiler for V7...


  Johnny

On 2017-11-08 03:09, khandy21yo wrote:
Does V7 require a special version, or can you just download less from 
gnu.org?




Sent from my Galaxy Tab® A

 Original message 
From: Mark Abene 
Date: 11/7/17 6:46 PM (GMT-07:00)
To: Will Senn 
Cc: simh@trailing-edge.com
Subject: Re: [Simh] Attach tar to tape device and access it from v7 in simh

I personally don't recall there ever being a less for v7. more, ex, vi, 
and csh were on a tape image of UCB tools, if you can find that.


-Mark


On Tue, Nov 7, 2017 at 2:02 PM, Will Senn > wrote:


On 11/7/17 2:45 PM, Clem Cole wrote:



On Tue, Nov 7, 2017 at 3:37 PM, Clem Cole mailto:cl...@ccc.com>> wrote:

You need to convert the foo.tar file to a foo.tap before
simh's mag tape reader will understand it.


​1.) ​
gunzip 1bsd.tar.gz
2
​.)
enblock <1bsd.tar >
​1bsd
.tap
​
​3.) ​in simh: att mt0 1bsd.tap
​4.) in unix:  ​tar vxfb /dev/rmt0 ...


Thanks, Clem. Now I'm getting somewhere!

I converted to a tap file (lo and behold, I wrote a perl script that
did this for the silly thing 2 years ago) and ran it. Then attached
the tap to simh and in v7:

# tar xvfb /dev/rmt0
Invalid blocksize. (Max 20)

hmm... that doesn't sound good, let's try it another way:

# tar xv0
tar: ashell/ - cannot create
x ashell/cont.a, 114488 bytes, 224 tape blocks
x ashell/READ_ME, 825 bytes, 2 tape blocks
# ls -ld ashell
drwxrwxr-x 2 root   64 Dec 31 22:18 ashell
# ls ashell
READ_ME
cont.a
# find . -name "more*" -a -print
#
...

It's coming back to me now, but oh so slowly, why is it complaining
about cannot create all the while creating the folder? Also, darn! I
was hoping more would be in there somewhere. Anybody know where the
more or less pager source code might be for v7?

Thanks,

Will



-- 
GPG Fingerprint: 68F4 B3BD 1730 555A 4462  7D45 3EAA 5B6D A982 BAAF



___
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




--
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] EXT :Re: OpenBSD on Simh VAX 3.8-1 - Segmentation fault

[Paul Koning]

> On Nov 8, 2017, at 12:04 PM, Bob Eager  wrote:
> 
> As an aside, I'm the maintainer of the simh package on FreeBSD.
> 
> I've stuck at 3.9, again because of the ongoing development of 4.0. If
> there is a stable version I can grab, I'll do a 4.x package.

It seems that the new model is that V4 is simply what's on Github, and there 
isn't any "stable version" snapshot contemplated.  Given that packagers are 
likely to be expecting such a stable version marking, the current model might 
be problematic for anyone who likes prepackaged stuff.

I may be wrong about the model, but it sure has been a very long time since the 
last "official" release, and V4.0 appears to be stable enough to mark it as 
such.

paul

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

Re: [Simh] EXT :Re: OpenBSD on Simh VAX 3.8-1 - Segmentation fault

[Bob Eager]

As an aside, I'm the maintainer of the simh package on FreeBSD.

I've stuck at 3.9, again because of the ongoing development of 4.0. If
there is a stable version I can grab, I'll do a 4.x package.

On Wed, 8 Nov 2017 10:56:52 -0600
Gary Lee Phillips  wrote:

> A short answer to Dave's question. I am running Linux Mint, but not
> the very latest version. I imagine the most recent version does have
> 3.9 in it. Since 4.0 is still marked "beta" it is probably not
> included.
> 
> Further details: Linux Mint is derived from Ubuntu which in turn is
> derived from Debian. If packages are upgraded at the head of the
> stream, it is at least supposed to "trickle down" to the dependent
> distributions. The differences from Debian to Ubuntu to Mint are
> largely related to user interfaces rather than to actual software
> packages, as I understand it. So if upgraded versions get into
> Debian, they will appear in subsequent releases of the descendents.
> 
> I confess I've grown lazy in my old age here. Mint issues LTS (long
> term support) versions that are supported for up to five years now.
> Since most of what I do is not bleeding edge, I've stopped rushing to
> upgrade the entire OS every time a new version comes out. I'm running
> 17.1 of Mint here, and I believe they are up to 19.x now.
> 
> Including concise "cookbook" style instructions is always useful *if*
> they can be counted on to work. That can be the rub. I fully
> understand that developers can't always test everything on every
> possible hardware and software platform.  Mark's succinct directions
> are great, but would be even better if a list of environments in
> which they have been tested were to be attached.
> 
> As a negative example, I recently tried someone's "concise"
> instructions for building an emulated environment to run/test Android
> apps on Linux. I know Android developers do this regularly, and often
> even do their development on Linux. However, these instructions were
> a disaster. They may have worked on some specific system, but they
> exploded here and I'm still picking out bits of debris that were left
> all over the file system.
> 
> --Gary
> 
> On Wed, Nov 8, 2017 at 10:40 AM, Hittner, David T [US] (MS) <
> david.hitt...@ngc.com> wrote:
> 
> > So two follow on action items, now that the problem seems to be
> > solved:
> >
> >
> >
> > 1) How can we work with this Linux vendor to provide a baseline
> > SIMH 4.0 package instead of the older SIMH 3.8-1 package?
> >
> >
> >
> > 2) Should the instructions for downloading and building the latest
> > SIMH that Mark provided be included in the FAQ (under Linux hosts)?
> >
> >
> >
> > Dave
> >
> >
> >
> >

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

Re: [Simh] EXT :Re: OpenBSD on Simh VAX 3.8-1 - Segmentation fault

[Gary Lee Phillips]

A short answer to Dave's question. I am running Linux Mint, but not the
very latest version. I imagine the most recent version does have 3.9 in it.
Since 4.0 is still marked "beta" it is probably not included.

Further details: Linux Mint is derived from Ubuntu which in turn is derived
from Debian. If packages are upgraded at the head of the stream, it is at
least supposed to "trickle down" to the dependent distributions. The
differences from Debian to Ubuntu to Mint are largely related to user
interfaces rather than to actual software packages, as I understand it. So
if upgraded versions get into Debian, they will appear in subsequent
releases of the descendents.

I confess I've grown lazy in my old age here. Mint issues LTS (long term
support) versions that are supported for up to five years now. Since most
of what I do is not bleeding edge, I've stopped rushing to upgrade the
entire OS every time a new version comes out. I'm running 17.1 of Mint
here, and I believe they are up to 19.x now.

Including concise "cookbook" style instructions is always useful *if* they
can be counted on to work. That can be the rub. I fully understand that
developers can't always test everything on every possible hardware and
software platform.  Mark's succinct directions are great, but would be even
better if a list of environments in which they have been tested were to be
attached.

As a negative example, I recently tried someone's "concise" instructions
for building an emulated environment to run/test Android apps on Linux. I
know Android developers do this regularly, and often even do their
development on Linux. However, these instructions were a disaster. They may
have worked on some specific system, but they exploded here and I'm still
picking out bits of debris that were left all over the file system.

--Gary

On Wed, Nov 8, 2017 at 10:40 AM, Hittner, David T [US] (MS) <
david.hitt...@ngc.com> wrote:

> So two follow on action items, now that the problem seems to be solved:
>
>
>
> 1) How can we work with this Linux vendor to provide a baseline SIMH 4.0
> package instead of the older SIMH 3.8-1 package?
>
>
>
> 2) Should the instructions for downloading and building the latest SIMH
> that Mark provided be included in the FAQ (under Linux hosts)?
>
>
>
> Dave
>
>
>
>
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] EXT :Re: OpenBSD on Simh VAX 3.8-1 - Segmentation fault

[Hittner, David T [US] (MS)]

So two follow on action items, now that the problem seems to be solved:

1) How can we work with this Linux vendor to provide a baseline SIMH 4.0 
package instead of the older SIMH 3.8-1 package?

2) Should the instructions for downloading and building the latest SIMH that 
Mark provided be included in the FAQ (under Linux hosts)?

Dave

From: Simh [mailto:simh-boun...@trailing-edge.com] On Behalf Of Gary Lee 
Phillips
Sent: Wednesday, November 08, 2017 11:12 AM
To: Mark Pizzolato 
Cc: simh@trailing-edge.com
Subject: EXT :Re: [Simh] OpenBSD on Simh VAX 3.8-1 - Segmentation fault

Thanks for the very clear answer, Mark.
I followed your instructions and the 4.0-0 beta built in minutes. Last time I 
tried to build Simh myself, it took several frustrating days of patching the 
makefile and changing references. That was on Slackware Linux and several 
versions back of both the environment and Simh. So thanks ever so much to the 
developers who have continued to enhance and tune this package. It is 
brilliantly done.
Now the even better news. The vax3900/vax module in 4.0-0 beta does not have 
the segmentation fault I encountered with 3.8-1. OpenBSD 5.7 is able to ftp and 
sftp transfer in and out without any complaints. So thank you again to whoever 
it was that found and fixed that problem.
--Gary

On Wed, Nov 8, 2017 at 1:50 AM, Mark Pizzolato 
mailto:m...@infocomm.com>> wrote:
Hi Gary,

On Tuesday, November 7, 2017 at 5:29 AM, Gary Lee Phillips wrote:
> I am running VAX 3.8-1 on a 64-bit Linux system. It performs well
> with OpenVMS as the operating system. But when I try to run it
> with OpenBSD 5.7 there are issues.
>
> The base system installs and boots correctly. But in order to bring
> in tools such as language compilers or editors, it is necessary to
> use ftp or sftp. I cannot get these to transfer anything because the
> moment a transfer starts, no matter which mode or direction or
> whether the transfer is initiated from within OpenBSD or from the
> outside, an immediate segmentation fault occurs. This appears to
> be within the SimH code itself, and not on OpenBSD.

That would certainly seem to be the case.  Simulator crashes
definitely should be fixed.  3.8-1 was released some 8 years ago,
and 3.9 was some 6 years ago.  Both are way behind the current
code base located at https://github.com/simh/simh/ downloadable
as https://github.com/simh/simh/archive/master.zip

> The same SimH VAX environment running OpenVMS 7.3 can
> perform transfers over ethernet without any apparent issues.

This is a useful data point.

> I have tried running SimH in supervisor mode or in user mode,
> same results. Also same results whether the transfer is initiated
> by root or by a normal user account.

I wouldn't expect you to be able to run the simulator as non-root
and have Ethernet functionality...

> The ethernet connection itself seems to be working. Telnet
> connections function normally, and ftp or sftp will pass directory
> information and accept commands, but file transfers fail.
>
> I guess I could try another version of OpenBSD. Upgrading SimH
> is also possible but a bit more daunting since 3.8 is what is offered
> as current on my Linux distro. (Linux Mint 18 "Sarah" 64 bit generic
> 4.4.0-97)

It really isn't particularly daunting:

First, you'll need the following:

$ sudo apt install libpcap-dev
Then:
$ wget https://github.com/simh/simh/archive/master.zip
$ unzip master.zip
$ cd simh-master
$ make vax
$# Start the newly built simulator:
$ BIN/vax


If you see any sort of segmentation fault, there is a problem that needs
fixing, and I'll be glad to address it.   If you have a problem please create
an issue at https://github.com/simh/simh/issues

- Mark

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

Re: [Simh] OpenBSD on Simh VAX 3.8-1 - Segmentation fault

[Gary Lee Phillips]

Thanks for the very clear answer, Mark.

I followed your instructions and the 4.0-0 beta built in minutes. Last time
I tried to build Simh myself, it took several frustrating days of patching
the makefile and changing references. That was on Slackware Linux and
several versions back of both the environment and Simh. So thanks ever so
much to the developers who have continued to enhance and tune this package.
It is brilliantly done.

Now the even better news. The vax3900/vax module in 4.0-0 beta does not
have the segmentation fault I encountered with 3.8-1. OpenBSD 5.7 is able
to ftp and sftp transfer in and out without any complaints. So thank you
again to whoever it was that found and fixed that problem.

--Gary


On Wed, Nov 8, 2017 at 1:50 AM, Mark Pizzolato  wrote:

> Hi Gary,
>
> On Tuesday, November 7, 2017 at 5:29 AM, Gary Lee Phillips wrote:
> > I am running VAX 3.8-1 on a 64-bit Linux system. It performs well
> > with OpenVMS as the operating system. But when I try to run it
> > with OpenBSD 5.7 there are issues.
> >
> > The base system installs and boots correctly. But in order to bring
> > in tools such as language compilers or editors, it is necessary to
> > use ftp or sftp. I cannot get these to transfer anything because the
> > moment a transfer starts, no matter which mode or direction or
> > whether the transfer is initiated from within OpenBSD or from the
> > outside, an immediate segmentation fault occurs. This appears to
> > be within the SimH code itself, and not on OpenBSD.
>
> That would certainly seem to be the case.  Simulator crashes
> definitely should be fixed.  3.8-1 was released some 8 years ago,
> and 3.9 was some 6 years ago.  Both are way behind the current
> code base located at https://github.com/simh/simh/ downloadable
> as https://github.com/simh/simh/archive/master.zip
>
> > The same SimH VAX environment running OpenVMS 7.3 can
> > perform transfers over ethernet without any apparent issues.
>
> This is a useful data point.
>
> > I have tried running SimH in supervisor mode or in user mode,
> > same results. Also same results whether the transfer is initiated
> > by root or by a normal user account.
>
> I wouldn't expect you to be able to run the simulator as non-root
> and have Ethernet functionality...
>
> > The ethernet connection itself seems to be working. Telnet
> > connections function normally, and ftp or sftp will pass directory
> > information and accept commands, but file transfers fail.
> >
> > I guess I could try another version of OpenBSD. Upgrading SimH
> > is also possible but a bit more daunting since 3.8 is what is offered
> > as current on my Linux distro. (Linux Mint 18 "Sarah" 64 bit generic
> > 4.4.0-97)
>
> It really isn't particularly daunting:
>
> First, you'll need the following:
>
> $ sudo apt install libpcap-dev
> Then:
> $ wget https://github.com/simh/simh/archive/master.zip
> $ unzip master.zip
> $ cd simh-master
> $ make vax
> $# Start the newly built simulator:
> $ BIN/vax
>
>
> If you see any sort of segmentation fault, there is a problem that needs
> fixing, and I'll be glad to address it.   If you have a problem please
> create
> an issue at https://github.com/simh/simh/issues
>
> - Mark
>
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] DEC Net for DOS and the product on an emulated PDP-11 OS

[Gregg Levine]

Hello!
Okay. That's one for me. And both won and lost. Now about that example
of Pathworks-32, where did you find it? Now my steps are to get a
proper PDP-11 emulation up and running, and of course installing an OS
there, and even picking one... Oh my!

More later I think. But okay.
-
Gregg C Levine gregg.drw...@gmail.com
"This signature fought the Time Wars, time and again."


On Wed, Nov 8, 2017 at 4:21 AM, Jordi Guillaumes Pons
 wrote:
>
>
>
> VMWare Player is available for free, and will allow you to create
> virtual machines without a problem. It's basically the same as
> Workstation without the problems of an evaluation period ending and
> catching you or the user out. Of course trying to get it to properly
> engage on the laptop when running Linux, is a trip to be certain.
>
> Of course I can't go ahead with this idea until I get more feedback
> from the rest of the group.
> ——
>
>
> Hello,
>
> This is a Windows 98 virtual machine running under VMWare Fusion:
>
> c:\>ver
>
> 4DOS 8,00   (Win98) DOS 7,10
>
> c:\>ncp loop node bitxor
>
> LOOP NODE test started at Wed Nov 08 10:14:40 2017
>
>   Connect complete to node  BITXOR
>   Remote node maximum buffer size for loopback:   184
>
>   Send 1, 46 bytes.
>   Successful send and receive, message  1.
>
> LOOP NODE test finished successfully at Wed Nov 08 10:14:41 2017
>
>
> c:\>
>
>
> It’s not strictly DECNET-DOS; it’s PATHWORKS-32, which includes the DECNET
> stack. I’ve not been able to try the old, “pure” DECNET-DOS because the
> supported LAN cards (DEPCA et all) are not emulated in any VM software I
> know. But this setup works OK:
>
> c:\>nft
>NFT - Network File Transfer - V6.0.004
> NFT>dir bitxor::
>
> Directory of: bitxor::DU1:[DECNET]
> INFO.TXT;1  4  22-NOV-12 12:15:25
> INFO.TXT;9  8  29-DEC-12 01:25:14
> INFO.TXT;10 9  14-MAR-13 13:05:37
> NFT>
>
> Let’s try a remote FAL access:
>
> [BITXOW]$ dir 7.98::
>
> Directory 7.98::C:[PW32]
>
> API.CNT API.HLP ASLAPI.DLL  ASLCZ.DLL
> ASLDE.DLL   ASLFR.DLL   ASLPASS.EXE ASLUS.DLL
> CHARSETS.DATCSM32.DLL   CSM_W3.DLL  DAP32.DLL
> DECIC16.DLL DECITOT.CNT DECITOT.DAT DECITOT.HLP
> DECITOT.TPL DECLOG.CNT  DECLOG.EXE  DECLOG.HLP
> DECTAL.DLL  DECTNET.DLL DEINST.EXE  DEINSTAL.DAT
> DEINSTAL.TXTFAL32.EXE   INDEXFIX.HLPMANDEC.CNT
> MANDEC.HLP  NCP.EXE NCPHELP.BIN NCPTAB.BIN
> NFTW.CNTNFTW.HLPNFTW32.EXE  NMAPI32.DLL
> NML32.EXE   NM_MSG.BIN  PW32.CNTPW32.HLP
> PWASSTCZ.CNTPWASSTCZ.HLPPWASSTDE.CNTPWASSTDE.HLP
> PWASSTFR.CNTPWASSTFR.HLPPWASSTUS.CNTPWASSTUS.GID
> PWASSTUS.HLPPWCTERM.DLL PWENT.DAT   PWINACC.DAT
> PWLICLM.CNT PWLICLM.HLP PWMSG.CNT   PWMSG.HLP
> PWNODE.DAT  PWOBJ.DAT   PWREADME.TXTPWS2DNST.EXE
> PWS2INST.EXEPWSOCK32.DLLPWTELNT.DLL SPAWN32.EXE
> UPGRADE.INI WHATSNEW.CNTWHATSNEW.HLPWHATSNEW.TXT
> nft.exe
>
> Total of 69 files.
> [BITXOW]$
>
>
>
>
>
> ___
> 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] Simh Digest, Vol 166, Issue 21

[Peter Allan]

Thanks for the suggestion Larry, but the way that simh is set up for a VAX
780, there are two massbuses and the disks (RP or RM) go on the first
massbus, while the tape drives go on the second. You have no choice about
that.

Cheers

Peter Allan

On 8 November 2017 at 01:36,  wrote:

> Send Simh mailing list submissions to
> simh@trailing-edge.com
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://mailman.trailing-edge.com/mailman/listinfo/simh
> or, via email, send a message with subject or body 'help' to
> simh-requ...@trailing-edge.com
>
> You can reach the person managing the list at
> simh-ow...@trailing-edge.com
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Simh digest..."
>
>
> Today's Topics:
>
>1. Re:  TE16 CRC Errors (Johnny Billquist)
>2. Re:  TE16 CRC Errors (Paul Koning)
>3. Re:  TE16 CRC Errors (Johnny Billquist)
>4. Re:  TE16 CRC Errors (Larry Baker)
>
>
> --
>
> Message: 1
> Date: Wed, 8 Nov 2017 01:49:05 +0100
> From: Johnny Billquist 
> To: simh@trailing-edge.com
> Subject: Re: [Simh] TE16 CRC Errors
> Message-ID: <050102c6-9d68-cf61-2f5a-7e75a5eb7...@softjar.se>
> Content-Type: text/plain; charset=utf-8; format=flowed
>
> I actually don't think he has a mixed massbus. I don't think you could
> even configure simh for this. Simh have a very simplistic view on
> hardware configurations.
> You don't configure massbuses, and put devices on them. You normally
> just configure the devices, and all the bus and controller
> configurations come implicit from that.
>
> So, if you specify a massbus disk, you'll get a corresponding massbus.
> And for tapes, you'll get another one.
> Both configured appropriately for that specific type of device you have
> hooked to it.
>
> In RSX land, only M+ supported mixed massbus, as you mention. Not sure
> when/if VMS did. Not sure how much of anything else that might have
> supported it either.
>
>Johnny
>
> On 2017-11-08 00:57, Larry Baker wrote:
> > Peter,
> >
> > My reading of your hardware setup sounds like what was called a
> > "Mixed-MASSBUS" configuration, that is, mixing disks and tapes on the
> > same MASSBUS.  PDP-11 RSX-11M-Plus supported Mixed-MASSBUS.  We
> > dual-ported all our disks and used the MASSBUS that came with our tape
> > drives that way.  I think Mixed-MASSBUS support was unusual.  I do not
> > remember if VMS 3.0 supported Mixed-MASSBUS.  If not, what you are
> > seeing might be the result.
> >
> > Larry Baker
> > US Geological Survey
> > 650-329-5608
> > ba...@usgs.gov 
> >
> >
> >
> >> On 7 Nov 2017, at 12:54:40 PM, simh-requ...@trailing-edge.com
> >>  wrote:
> >>
> >> --
> >>
> >> Message: 4
> >> Date: Tue, 7 Nov 2017 20:54:31 +
> >> From: Peter Allan mailto:petermal...@gmail.com>
> >
> >> To: "simh@trailing-edge.com "
> >> mailto:simh@trailing-edge.com>>
> >> Subject: Re: [Simh] Simh Digest, Vol 166, Issue 12
> >> Message-ID:
> >>  >>  mail.gmail.com>>
> >> Content-Type: text/plain; charset="utf-8"
> >>
> >> In order to try to understand why I am having trouble reading a
> simulated
> >> tape with a simulated TE16 tape drive on a simh VAX 780, I have tried
> >> several configurations.
> >>
> >>
> >> Firstly, a point of clarification. The problem that I came across was in
> >> finishing the installation of VMS 3.0. This was necessary in order to be
> >> able to install VMS 1.5 using DSC2, but I have no problem with the
> >> installation of VMS 1.5 itself. VMS 1.5 is a red herring - the problem
> is
> >> with VMS 3.0.
> >>
> >>
> >> I had configured a VAX 780 in simh with three RP06 disk drives and two
> >> TE16
> >> tape drives. I was using TE16 tape drives rather than TS11 really just
> >> because when I was the system manager of a real 780, it had two TE16
> tape
> >> drives.
> >>
> >>
> >> All of this has been done on a host system running CentOS 5 (for simh
> >> 3.8-1) and CentOS 6 (for simh 4.0).
> >>
> >>
> >> To remind you of the problem, I started the installation of VMS 3.0 by
> >> booting stand-alone backup and restoring a saveset from a tape that I
> >> downloaded from the University of Stuttgart. The first phase went fine
> and
> >> when I rebooted the 780, the system asked which device contained the
> >> installation media so that it could automatically complete the
> >> installation
> >> of VMS. I entered MTA0: and this is where things went wrong. Wilm
> >> suggested
> >> I should trying using a TS11 drive (MSA0) rather than a TE16 drive and
> >> this
> >> did indeed work, but this left me wondering why?
> >>
> >>
> >> I have tried to restore a BACKUP saveset from a tape to a disk using
> >> several configurations. I have tried:
> >>
> >> -

Re: [Simh] TE16 CRC Errors

[Paul Koning]

> On Nov 7, 2017, at 8:01 PM, Johnny Billquist  wrote:
> 
> On 2017-11-08 01:54, Paul Koning wrote:
>> ...RSTS definitely did not.  It does support mixed RP/RM massbus, which is 
>> more usual but not universal either, I think.  But no mixing disk and tape.
> 
> Thanks. I was wondering about RSTS/E, but didn't dare saying something 
> definitive. That it allows mixing of RM/RP is nice. I think in plain -11M, 
> even that might not be allowed, but I would have to go and check and refresh 
> my memory.
> M+ allows any kind of combinations.

It's possible that the first Massbus support in RSTS didn't allow mixing RP and 
RM, but it certainly appeared soon after, I'm pretty sure by V7.0 at the 
latest.  So as of that point, RSTS supported two Massbus adapters for disks, 
and each of these could have any Massbus disk on it.  Even an RP07, on an 11/70 
at least, though officially that was not supported (DEC didn't sell that device 
on lowly PDP-11s).

paul


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

Re: [Simh] TE16 errors on VMS 3.0

[Bob Supnik]

The simulated Massbus tape only throws CRC errors if the tape image 
itself is marked with 'record-in-error'. This seems rather unlikely, and 
the TS11 would also choke on the same problem.


For debug, please prepare an archive containing:

1. The VMS 3.0 image you are using, or a pointer to where it can be found.
2. The tape file you are using, or a pointer to where it can be found.
3. A transcript of the session, from simulator startup to the error, 
including the configuration.


Put the archive on a sharing site like Mega or Mediafire and post the 
URL to the mailing list. I'll take a look at what's happening.


/Bob Supnik

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

Re: [Simh] DEC Net for DOS and the product on an emulated PDP-11 OS

[Jordi Guillaumes Pons]

> 
> VMWare Player is available for free, and will allow you to create
> virtual machines without a problem. It's basically the same as
> Workstation without the problems of an evaluation period ending and
> catching you or the user out. Of course trying to get it to properly
> engage on the laptop when running Linux, is a trip to be certain.
> 
> Of course I can't go ahead with this idea until I get more feedback
> from the rest of the group.
> ——

Hello,

This is a Windows 98 virtual machine running under VMWare Fusion:

c:\>ver

4DOS 8,00   (Win98) DOS 7,10

c:\>ncp loop node bitxor

LOOP NODE test started at Wed Nov 08 10:14:40 2017

  Connect complete to node  BITXOR
  Remote node maximum buffer size for loopback:   184

  Send 1, 46 bytes.
  Successful send and receive, message  1.

LOOP NODE test finished successfully at Wed Nov 08 10:14:41 2017


c:\>


It’s not strictly DECNET-DOS; it’s PATHWORKS-32, which includes the DECNET 
stack. I’ve not been able to try the old, “pure” DECNET-DOS because the 
supported LAN cards (DEPCA et all) are not emulated in any VM software I know. 
But this setup works OK:

c:\>nft
   NFT - Network File Transfer - V6.0.004
NFT>dir bitxor::

Directory of: bitxor::DU1:[DECNET]
INFO.TXT;1  4  22-NOV-12 12:15:25
INFO.TXT;9  8  29-DEC-12 01:25:14
INFO.TXT;10 9  14-MAR-13 13:05:37
NFT>

Let’s try a remote FAL access:

[BITXOW]$ dir 7.98::

Directory 7.98::C:[PW32]

API.CNT API.HLP ASLAPI.DLL  ASLCZ.DLL  
ASLDE.DLL   ASLFR.DLL   ASLPASS.EXE ASLUS.DLL  
CHARSETS.DATCSM32.DLL   CSM_W3.DLL  DAP32.DLL  
DECIC16.DLL DECITOT.CNT DECITOT.DAT DECITOT.HLP
DECITOT.TPL DECLOG.CNT  DECLOG.EXE  DECLOG.HLP 
DECTAL.DLL  DECTNET.DLL DEINST.EXE  DEINSTAL.DAT   
DEINSTAL.TXTFAL32.EXE   INDEXFIX.HLPMANDEC.CNT 
MANDEC.HLP  NCP.EXE NCPHELP.BIN NCPTAB.BIN 
NFTW.CNTNFTW.HLPNFTW32.EXE  NMAPI32.DLL
NML32.EXE   NM_MSG.BIN  PW32.CNTPW32.HLP   
PWASSTCZ.CNTPWASSTCZ.HLPPWASSTDE.CNTPWASSTDE.HLP   
PWASSTFR.CNTPWASSTFR.HLPPWASSTUS.CNTPWASSTUS.GID   
PWASSTUS.HLPPWCTERM.DLL PWENT.DAT   PWINACC.DAT
PWLICLM.CNT PWLICLM.HLP PWMSG.CNT   PWMSG.HLP  
PWNODE.DAT  PWOBJ.DAT   PWREADME.TXTPWS2DNST.EXE   
PWS2INST.EXEPWSOCK32.DLLPWTELNT.DLL SPAWN32.EXE
UPGRADE.INI WHATSNEW.CNTWHATSNEW.HLPWHATSNEW.TXT   
nft.exe 

Total of 69 files.
[BITXOW]$ 




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