[HECnet] simh vax version

[Mark Pizzolato]

On Monday, November 18, 2019 at 11:19 AM, Bill Cunningham wrote:
> On 11/18/2019 1:53 PM, Mark Pizzolato wrote:
> > Hi Bill,
> >
> > On Monday, November 18, 2019 at 10:40 AM, Bill Cunningham wrote:
> >> On 11/18/2019 1:10 AM, Mark Pizzolato wrote:
> >>> Where did you end up with this?  Did you use a newly built binary
> >>> from the
> >> latest source and at least get reasonable error messages when it was
> >> not able to access files you mentioned?
> >>> If so, were you able to actually attach the VMS license file and
> >>> then access it
> >> within a running VMS OS?
> >>> - Mark
> >>>
> >>> I can't remember the source code really. I think it might have come
> >>> from the https://simh.trailing-edge.com website. That's all I
> >>> remember pulling any code from other than git. I believe their code
> >>> is older. I have not loaded my license because I would have to have
> >>> done it the old way which is fine with me. Thing is it's been so
> >>> long I forget how to do it and though I'm run into some old notes
> >>> over the last day or so, on paper, They aren't giving detail I need
> >>> now. I have been able to run genisoimage and load the license into a
> >>> ISO but forget vax's device names and such to load the ISO into the
> >>> vms simulated enviornment and change the attribs and run '@' to load
> >>> the license. I am just sure it would work. My memory is just gone :(
> >>       Your TQ method is showing more than anything "//mst" when I
> >> would try the vmst license . It's like it's not reading the first
> >> char.  I know 'set <dev> cdrom' would be used. The <dev> and mount
> >> <dev> and attrb switching I can't remember seems like it was as
> >> simple as stm to stmlf. A simple line feed toggle.
> > It is really frustrating for me not be able to see what you are seeing and
> having you summarize what happened into phrases like "Your TQ method is
> showing more than anything "//mst" when I would try the vmst license".  This
> gives me no information and is at best a rumor of a problem.  It is hard to
> respond to rumors that lack any detail.
> >
> > 1) Please save ALL of the following lines into a file on your Linux system called
> TQ_TEST.ini in the same directory as your vmst file.
> >
> > SET CONSOLE -N LOG=TQ_TEST.log
> > SHOW VERSION
> > DIR vmst
> > ATTACH TQ -FV ANSI-VMS vmst
> > EXIT
> >
> > 2) Please run your vax simulator (in the same directory as your vmst file).
> >
> > Enter this command at the sim> prompt:
> > DO TQ_TEST.ini
> >
> > 3) When it exits, please send back 2 things:
> >        a) The TQ_TEST.log
> >        b) The configuration file you use when run VMS in this simulator.
> >
> > Thanks.
> >
> > - Mark
> 
> OK when I typed "do DQ_TEST.ini"
> 
> I got a file error, file does not exist.

That was your typing error since I asked you enter "DO TQ_TEST.ini" 
instead of "DO DQ_TEST.ini".

> here, is the log file, DQ_TEST.LOG
> 
> Logging to file "TQ_TEST.LOG"
> Logging to file "TQ_TEST.LOG" at Mon Nov 18 14:01:36 2019
> sim> sho version
> MicroVAX 3900 simulator V4.0-0 Current
>      Simulator Framework Capabilities:
>          64b data
>          64b addresses
>          Threaded Ethernet Packet transports:PCAP:TAP:NAT:UDP
>          Idle/Throttling support is available
>          Virtual Hard Disk (VHD) support
>          RAW disk and CD/DVD ROM support
>          Asynchronous I/O support (Lock free asynchronous event queue)
>          Asynchronous Clock support
>          FrontPanel API Version 12
>      Host Platform:
>          Compiler: GCC 8.3.1 20190223 (Red Hat 8.3.1-2)
>          Simulator Compiled as C arch: x64 (Release Build) on Nov 16
> 2019 at 19:46:12
>          Memory Access: Little Endian
>          Memory Pointer Size: 64 bits
>          Large File (>2GB) support
>          SDL Video support: No Video Support
>          RegEx support for EXPECT commands
>          OS clock resolution: 1ms
>          Time taken by msleep(1): 1ms
>          Ethernet packet info: libpcap version 1.9.1 (with TPACKET_V3)
>          Time taken by msleep(1): 1ms
>          OS: Linux local 5.3.6-100.fc29.x86_64 #1 SMP Mon Oct 14
> 21:48:12 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux
>          git commit id: 318b0e32
>          git commit time: 2019-11-16T00:39:01-08:00
> sim> dir vmst
>   Directory of /
> 
> 11/18/2019  02:00 PM            79,508 mst
>                 1 File(s)            79,508 bytes
>                 0 Dir(s)
> sim> att tq -fv ansi-vms vmst
> Can't stat: //mst
> Can't open ANSI-VMS format tape image: vmst
> sim> ex
> Too few arguments
> sim> q 
> Goodbye
> Log file closed

Very interesting.  Of course, I've never sat and had a simulator mess with files in the root directory (/) so I never saw this bug in the file scanning code.  I suspect that if you move the license file to another directory, things will work just fine.  Now that I've actually seen the failure it is easily reproduced (having nothing specific to do with the ANSI-VMS logic - it was demonstrated by the DIR command which uses the same internal file name scanning).  If you pick up the latest code and rebuild the simulator, you can keep your files in the root directory if you want.  :-).

Once you get that far, the:

ATT TQ -F ANSI-VMS vmst

Will work, and then once VMS is running and you've logged in:

$ MOUNT/OVER=ID MUA0:
$ COPY MUA0:VMST. SYS$LOGIN:
$ @SYS$LOGIN:VMST.

This will install the licenses cleanly.  It seems that VMS's DCL execution can't handle command procedures run directly from magtape that have IF THEN ELSE statements in them, hence the need to copy the file to disk first.

- Mark