<div dir='auto'><div>Hi Brian,</div><div dir="auto"><br></div><div dir="auto">Since I changed something that broke this, and James's problem report includes enough detail to demonstrate the problem I'll take a crack at this first. I'll let you know directly.</div><div dir="auto"><br></div><div dir="auto">- Mark<br><div class="gmail_extra" dir="auto"><br><div class="gmail_quote">On May 4, 2017 11:29 AM, brian@quarterbyte.com wrote:<br type="attribution"><blockquote class="quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div>
<div dir="ltr">Since the GUI code itself hasn't changed, it's possible that the simh scp code changed the way it handles the commands injected by the GUI. I may have time this weekend to look at it.
<div><br>
<div>thanks for the heads up.</div>
</div>
<div><br>
</div>
<div>brian</div>
<div><br>
</div>
</div>
<div><br>
<div class="elided-text">On Wed, May 3, 2017 at 2:15 PM, Rich Alderson <span dir="ltr">
<<a href="mailto:simh@alderson.users.panix.com">simh@alderson.users.panix.com</a>></span> wrote:<br>
<blockquote style="margin:0 0 0 0.8ex;border-left:1px #ccc solid;padding-left:1ex">
JAMES FEHLINGER <<a href="mailto:jfehlinger@comcast.net">jfehlinger@comcast.net</a>> posted the following to the ClassicCmp<br>
mailing list. I don't see anyone forwarding it (yet).<br>
<br>
Rich<br>
<br>
-----Original Message-----<br>
From: cctalk On Behalf Of JAMES FEHLINGER<br>
Sent: Wednesday, May 03, 2017 9:41 AM<br>
To: <a href="mailto:cctalk@classiccmp.org">cctalk@classiccmp.org</a><br>
Subject: SimH IBM1130 GUI appears broken<br>
<br>
I originally attempted to post this to <a href="mailto:simh@trailing-edge.com">simh@trailing-edge.com</a>,<br>
but they have a policy of automatically rejecting any messages from<br>
non-subscribers, and as I only read that list via the Web<br>
interface, it doesn't seem worth it to subscribe just in order to<br>
post one message. However, I suspect that Brian Knittel and/or<br>
Mark Pizzolato probably see stuff that's posted here too.<br>
<br>
---<br>
<br>
Once upon a time (i.e., prior to early January, 2016), the<br>
SimH IBM1130 GUI worked like this (on Windows):<br>
<br>
You start the simulator in a console window and issue the<br>
commands (either manually or via a command file argument):<br>
<br>
-------------------------<br>
reset<br>
detach prt<br>
delete printer.txt<br>
<br>
att dsk0 dms.dsk<br>
<br>
att prt printer.txt<br>
boot dsk<br>
-------------------------<br>
<br>
At this point, the simulator, after having booted the DMS<br>
operating system, enters a "Wait state" and drops<br>
back to the SimH prompt. The console window shows<br>
(if you've started the simulator with the command file<br>
argument "guijob" containing the above commands):<br>
<br>
-------------------------<br>
IBM 1130 Simulator V4.0-0 Beta git commit id: e8ea427d<br>
guijob-2> detach prt<br>
Not attached<br>
PRT: creating new file<br>
Loaded DMS V2M12 cold start card<br>
<br>
Wait, IAR: 0000002A (4c80 BSC I ,00028 )<br>
sim><br>
-------------------------<br>
<br>
The line printer icon in the GUI now "shows paper" which,<br>
if you view it (thereby "tearing it off" and causing<br>
the creation of a new printer file) will show the boot<br>
message from DMS.<br>
<br>
At this point, you can submit a "job deck" by dragging a<br>
file to the GUI's card reader icon -- e.g., one of the<br>
decks from the software kit at <a href="http://ibm1130.org">
ibm1130.org</a><br>
(such as a job to print the LET [Location Equivalence Table] list.job,<br>
a sample Fortran program for.job or one of the more substantial<br>
Fortran programs like csort.job or swave.job, etc.)<br>
<br>
Each time you drag the "job deck" file over the GUI's<br>
card reader icon, the depicted input tray "fills up" and<br>
the console shows, e.g.<br>
<br>
sim> attach cr "C:\ibm1130\dms\list.job"<br>
sim><br>
<br>
You then click the green PROGRAM START button, the blinkenlights<br>
flash briefly, the card reader "empties", the printer "shows paper"<br>
(if it was empty beforehand) and, most significantly here,<br>
**the simulator enters Wait state and drops back to its prompt<br>
when the program has finished**. The console window now shows (e.g.)<br>
<br>
sim> attach cr "C:\ibm1130\dms\list.job"<br>
sim> cont<br>
Wait, IAR: 0000002A (4c80 BSC I ,00028 )<br>
sim><br>
<br>
You can continue to submit jobs, without rebooting DMS, by dragging<br>
a new deck to the card reader icon, and clicking PROGRAM START.<br>
Each time, the simulator swallows the "cards", flashes the blinkenlights,<br>
adds output to the line printer file, and enters Wait and drops<br>
back to its prompt.<br>
<br>
The "git commit id" in the IBM1130.exe used for the above example is e8ea427d from<br>
<a href="https://github.com/simh/Win32-Development-Binaries">https://github.com/simh/Win32-<wbr>Development-Binaries</a> archive<br>
simh-4.0-Beta--2016-01-07-<wbr>e8ea427d.zip (January 7, 2016).<br>
<br>
If I use the IBM1130.exe from the very next archive<br>
simh-4.0-Beta--2016-01-29-<wbr>b8049645.zip (January 29, 2016), I get<br>
a different result (using the same file of startup commands):<br>
<br>
-------------------------<br>
IBM 1130 Simulator V4.0-0 Beta git commit id: b8049645<br>
guijob-2> detach prt<br>
Not attached<br>
PRT: creating new file<br>
Loaded DMS V2M12 cold start card<br>
-------------------------<br>
<br>
Here, the simulator has not entered a Wait state and dropped back<br>
to a prompt. If I try to drag a "job deck" file to the card reader<br>
icon, the attempt is rejected with a Windows error gong and<br>
the card reader remains empty.<br>
<br>
I can get around this by performing the following:<br>
<br>
Click the red "IMM STOP" button.<br>
<br>
Drag the "job deck" file to the card reader icon. The reader<br>
"fills up".<br>
<br>
In the console window, type "boot dsk" to re-boot DMS<br>
(clicking PROGRAM START doesn't work here!),<br>
The card reader empties, the blinkenlights flash and apparently<br>
the program runs. However, the simulator does not enter<br>
Wait state when the program is finished. (Nor can I click on<br>
the lineprinter icon to "tear off" and view the results at this point.)<br>
<br>
However, if I again click the red "IMM STOP" button and then<br>
click the lineprinter icon, I can see the results of the<br>
program that just ran. Or, I can click IMM STOP, submit a<br>
new job deck to the card reader, type "boot dsk" (you have<br>
to re-boot DMS; PROGRAM START won't work!),<br>
and the new program will run and add its results to the lineprinter output.<br>
<br>
The very latest Win32 build of IBM1130.exe, from<br>
simh-4.0-Beta--2017-05-02-<wbr>e9dea63b.zip (May 2, 2017)<br>
exhibits slightly different (but still apparently broken)<br>
behavior.<br>
<br>
When the startup command file executes, the initial result looks<br>
correct (i.e., the same as the January 7, 2016 result above):<br>
<br>
-------------------------<br>
IBM 1130 Simulator V4.0-0 Beta git commit id: e9dea63b<br>
guijob-2> detach prt<br>
Not attached<br>
PRT: creating new file<br>
Loaded DMS V2M12 cold start card<br>
<br>
Wait, IAR: 0000002A (4c80 BSC I ,00028 )<br>
sim><br>
-------------------------<br>
<br>
Note that the simulator has entered the Wait state and dropped back<br>
to its prompt, and you can now drag a job deck to the card reader<br>
and start the program normally by clicking PROGRAM START.<br>
<br>
However, at this point the simulator reverts to the broken behavior<br>
exhibited by the January 29, 2016 example above -- when the program is<br>
finished running, the simulator does not enter Wait state and<br>
drop back to a "sim>" prompt. The console window looks like:<br>
<br>
-------------------------<br>
IBM 1130 Simulator V4.0-0 Beta git commit id: e9dea63b<br>
guijob-2> detach prt<br>
Not attached<br>
PRT: creating new file<br>
Loaded DMS V2M12 cold start card<br>
<br>
Wait, IAR: 0000002A (4c80 BSC I ,00028 )<br>
sim> cont<br>
-------------------------<br>
<br>
and the simulator remains unresponsive until the IMM STOP button<br>
is pushed. Then we see:<br>
<br>
-------------------------<br>
IBM 1130 Simulator V4.0-0 Beta git commit id: e9dea63b<br>
guijob-2> detach prt<br>
Not attached<br>
PRT: creating new file<br>
Loaded DMS V2M12 cold start card<br>
<br>
Wait, IAR: 0000002A (4c80 BSC I ,00028 )<br>
sim> cont<br>
<br>
Immediate Stop key requested, IAR (4c80 BSC I ,00028 )<br>
-------------------------<br>
<br>
At this point, the GUI becomes responsive again, and the<br>
line printer output can be "torn off" and viewed, or a<br>
new job can be dragged to the card reader.<br>
<br>
However, you cannot then use the PROGRAM START button to<br>
start the new program. If you do, the simulator will issue<br>
a "cont" command, but the card reader will not empty and<br>
the new program will not run.<br>
<br>
You have to type "boot dsk" in the console window to reboot DMS.<br>
At this point, the loaded card reader will empty by itself,<br>
the new program will run, and the program's output will be added<br>
to the lineprinter listing.<br>
<br>
For each additional job deck you want to run, you have to use<br>
the IMM STOP, load card reader, "boot dsk" (reboot DMS)<br>
sequence.<br>
<br>
So it appears something happened to break the normal functioning<br>
of the 1130 GUI back in January of last year, which has not yet<br>
been completely repaired.<br>
<br>
<br>
<br>
______________________________<wbr>_________________<br>
Simh mailing list<br>
<a href="mailto:Simh@trailing-edge.com">Simh@trailing-edge.com</a><br>
<a href="http://mailman.trailing-edge.com/mailman/listinfo/simh">http://mailman.trailing-edge.<wbr>com/mailman/listinfo/simh</a></blockquote>
</div>
<br>
</div>
</div>
</blockquote></div><br></div></div></div>