Hello Steven,
I met this problem before. There is a GLOBAL TXTLIB statement in the PROFILE EXEC of CMSUSER that sets the libraries for the compilers. The GCCLIB and FORTLIB are in the list but they both contain a SQRT function. Because GCCLIB comes first in the list, the SQRT of that library is the one used with your FORTRAN program, and it is the wrong one. Peter, however, first set the GLOBAL TXTLIB to FORTLIB only which results in the right SQRT being used at execution.
Rene FERLAND, Montreal
?
Now I can run the heron program but there seems to be an anomaly with the SQRT function in that it is always returning zero.?
|
Re: Real and Virtual Printers and Punches and File Transfers
On Mon, 20 Jan 2020 at 08:08, Steven Fosdick <stevenfosdick@...> wrote:
Thinking about how I might be able to transfer a simple text file from within CMS to the Hercules host I had the idea that I may be able to send it to the card punch which would should then appear in the a host file but I then discovered that VM virtualises the card punch so when you use the 'punch' command the output is placed in a spool file. Not a great surprise but the spool file is not marked as held and yet I can't seem to find any way to tell the VM spooling system that it should send that file to what it thinks is the real card punch (i.e. the one virtualised by Hercules, not VM). Exactly the same situation applies to the printer.
Other have explained how best to transfer files. But as an aside, there is nothing to stop you from ATTACHing the "real" card punch (i.e. the one virtualized by Hercules) to your userid, and thus avoiding SPOOLing entirely. Obviously this wouldn't work well in a production environment, but in your private instance of VM no one's going to complain. Tony H.
|
Hi Steven,
And this works well right up to calculating P. Running this for a classic
3,4,5 triangle gives:
Execution begins...
3 4 5
A= 3 B= 4 C= 5 S= 6.00 P= 36.00 AREA= 0.0
SQUARE UNITS
So I am wondering if I am doing something silly here. SQRTF was not
accepted as an intrinsic function and apparently was replaced with a
generic named version. SQRT doesn't start with one of the characters that
would suggest integer arithmetic so it's a bit of a mystery.
It seems to work ok for me: global txtlib fortlib Ready; T=0.01/0.01 23:03:15 fortran heron FORTRAN IV (G) COMPILATION COMPLETE. Ready; T=0.15/0.30 23:03:20 filedef ft05f001 term Ready; T=0.01/0.01 23:03:32 filedef ft06f001 term Ready; T=0.01/0.01 23:03:40 load heron (start Execution begins... 3 4 5 A= 3 B= 4 C= 5 S= 6.00 P= 36.00 AREA= 6.00 SQUARE UNITS Could you have something called SQRT TEXT on your A disk or some other disk which might be getting picked up in preference to the SQRT function in FORTLIB TXTLIB Y where the FORTRAN function library lives? Regards, Peter.
|
On 1/20/20 5:08 PM, Gregg Levine wrote: Hello!
Interesting. I can not imagine why that crowd of <DELETED!> boobs
would want to do that to your end. As for Bitsavers, I usually go to
the site behind (FTP that is.)
Now why is there a big caravan of blue trucks heading your way, Dave? They're bringing more hardware! The latest bit of IBM gear I'm looking at has great big servo-motors. -Dave -- Dave McGuire, AK4HZ New Kensington, PA
|
Hello!
Interesting. I can not imagine why that crowd of <DELETED!> boobs
would want to do that to your end. As for Bitsavers, I usually go to
the site behind (FTP that is.)
Now why is there a big caravan of blue trucks heading your way, Dave?
-----
Gregg C Levine gregg.drwho8@...
"This signature fought the Time Wars, time and again."
toggle quoted message
Show quoted text
On Mon, Jan 20, 2020 at 2:03 PM Dave McGuire <mcguire@...> wrote:
On 1/20/20 1:57 PM, pjfarley3@... wrote:
Does anyone know if there is an archive anywhere of VM/ESA era manuals?
I tried getting to bitsavers.org but it seems to be bandwidth-overloaded
and does not respond. Are there bitsavers mirrors anywhere?
The ¡°Internet Archve¡± interface doesn¡¯t seem to provide the bitsavers
directory structure (or I haven¡¯t figured out how to get it to just let
me browse the directories).
Any help you can provide is appreciated. The primary bitsavers.org site has been having problems for the past
couple of days. It started at about the same time that my own network
started getting hit by a rather substantial denial-of-service attack
originating in Iran, sigh.
There are bitsavers mirrors everywhere, google "bitsavers mirror". I
don't think there's anything as recent as ESA on there, though.
I have a few dozen VM/ESA manuals in PDF format that I can tar up for
you if that would be helpful.
-Dave
|
Thanks, Peter.
Now I can run the heron program but there seems to be an anomaly with the SQRT function in that it is always returning zero.? The latest?incarnation of the program is now:
C AREA OF A TRIANGLE WITH A STANDARD SQUARE ROOT FUNCTION
C INPUT - TAPE READER UNIT 5, INTEGER INPUT
C OUTPUT - LINE PRINTER UNIT 6, REAL OUTPUT
C INPUT ERROR DISPLAY ERROR OUTPUT CODE 1 IN JOB CONTROL LISTING
? ? ? READ (5, 501) IA, IB, IC
? 501 FORMAT (3I5)
C IA, IB, AND IC MAY NOT BE NEGATIVE OR ZERO
C FURTHERMORE, THE SUM OF TWO SIDES OF A TRIANGLE
C MUST BE GREATER THAN THE THIRD SIDE, SO WE CHECK FOR THAT, TOO
? ? ? IF (IA .LE. 0) ? ? STOP 1
? 701 IF (IB .LE. 0) ? ? STOP 1
? 702 IF (IC .LE. 0) ? ? STOP 1
? 703 IF (IA+IB .LT. IC) STOP 1
? 704 IF (IA+IC .LT. IB) STOP 1
? 705 IF (IB+IC .LT. IA) STOP 1
C USING HERON'S FORMULA WE CALCULATE THE
C AREA OF THE TRIANGLE
? ? ? S = FLOAT(IA + IB + IC) / 2.0
? ? ? P = S * (S - FLOAT(IA)) * (S - FLOAT(IB)) * (S - FLOAT(IC))
? ? ? AREA = SQRT (P)
? ? ? WRITE (6, 601) IA, IB, IC, S, P, AREA
? 601 FORMAT (4H A= ,I5,4H B= ,I5,4H C= ,I5,4H S= F10.2,4H P= ,F10.2,
? ? ?+ ? ? ? ?7H AREA= ,F10.2,13H SQUARE UNITS)
? ? ? STOP
? ? ? END
And this works well right up to calculating P.? Running this for a classic 3,4,5 triangle gives:
Execution begins...
? ? 3 ? ?4 ? ?5
A= ? ? 3 B= ? ? 4 C= ? ? 5 S= ? ? ? 6.00 P= ? ? ?36.00 AREA= ? ? ? 0.0 ?SQUARE UNITS
So I am wondering if I am doing something silly here.? SQRTF was not accepted as an intrinsic function and apparently was replaced with a generic named version.? SQRT doesn't start with one of the characters that would suggest integer arithmetic so it's a bit of a mystery.
Running the same program on GNU Fortran on Linux produces sensible looking output:
$ ./heron
? ? 3 ? ?4 ? ?5
?A= ? ? 3 B= ? ? 4 C= ? ? 5 S= ? ? ? 6.00 P= ? ? ?36.00 AREA= ? ? ? 6.00 SQUARE UNITS
Steve.
On Mon, 20 Jan 2020 at 17:42, Peter Coghlan < groups@...> wrote: Hi Steven,
>
> Searching, unit 5 should be keyboard/standard input so I was expecting the
> program to wait for a line at the terminal.? If I work around the input by
> hard coding some values so the the input part becomes:
>
CMS "borrowed" the FORTRAN compiler from MVS.? You will need FILEDEF statements
to associate MVS DDNAMEs and FORTRAN unit numbers with files or devices,
for example (untested):
FILEDEF FT05F001 TERM
FILEDEF FT06F001 DISK OUTPUT FILE A
>
> Now the output seems to be discarded.? It isn't printed to the terminal,
> isn't queued for punch or printer and isn't in a file that I can see.
>
Have a look around for a file called FILE FT06F001 A or something like that.
>
>
> Interestingly the same issue with discarded output seems to happen with the
> supplied hello fortran program.
>
Are they run from an EXEC which contains suitable FILEDEF statements?
Regards,
Peter Coghlan.
|
On Mon, Jan 20, 2020 at 11:23 AM, Dave Wade wrote:
The ESA V2 manuals can still be downloaded from the publications site...
I've found the best way to get the list of VM/ESA manuals is to search for publication number SK2T-5787-04. When you're viewing the publication information, click the link that says "Click here for an overview of the publications that belong to this publication" and you'll get a nice list of the VM/ESA 2.4 manuals. -Matthew
|
Oh loads of bitsavers mirrors.?
Try
? Dave
Does anyone know if there is an archive anywhere of VM/ESA era manuals?? I tried getting to but it seems to be bandwidth-overloaded and does not respond.? Are there bitsavers mirrors anywhere? ? The ¡°Internet Archve¡± interface doesn¡¯t seem to provide the bitsavers directory structure (or I haven¡¯t figured out how to get it to just let me browse the directories). ? Any help you can provide is appreciated. ? Peter --
|
The ESA V2 manuals can still be downloaded from the publications site...
Dave
Does anyone know if there is an archive anywhere of VM/ESA era manuals?? I tried getting to but it seems to be bandwidth-overloaded and does not respond.? Are there bitsavers mirrors anywhere? ? The ¡°Internet Archve¡± interface doesn¡¯t seem to provide the bitsavers directory structure (or I haven¡¯t figured out how to get it to just let me browse the directories). ? Any help you can provide is appreciated. ? Peter --
|
I would love to have a copy of any ESA manuals anyone is willing to share regardless of format.
Beth
|
I found a fair portion of the manuals for r2.3, r2.4 and z/VM on the IBM z/VM website a few years ago. I don't know if they are still there.?
I have a complete set of r2.3 or 2.4, but they are in Bookmanager format.
Gary
Sent from my T-Mobile 4G LTE Device
|
On 1/20/20 1:57 PM, pjfarley3@... wrote: Does anyone know if there is an archive anywhere of VM/ESA era manuals??
I tried getting to bitsavers.org but it seems to be bandwidth-overloaded
and does not respond.? Are there bitsavers mirrors anywhere?
The ¡°Internet Archve¡± interface doesn¡¯t seem to provide the bitsavers
directory structure (or I haven¡¯t figured out how to get it to just let
me browse the directories).
Any help you can provide is appreciated. The primary bitsavers.org site has been having problems for the past couple of days. It started at about the same time that my own network started getting hit by a rather substantial denial-of-service attack originating in Iran, sigh. There are bitsavers mirrors everywhere, google "bitsavers mirror". I don't think there's anything as recent as ESA on there, though. I have a few dozen VM/ESA manuals in PDF format that I can tar up for you if that would be helpful. -Dave -- Dave McGuire, AK4HZ New Kensington, PA
|
Does anyone know if there is an archive anywhere of VM/ESA era manuals?? I tried getting to bitsavers.org but it seems to be bandwidth-overloaded and does not respond.? Are there bitsavers mirrors anywhere? ? The ¡°Internet Archve¡± interface doesn¡¯t seem to provide the bitsavers directory structure (or I haven¡¯t figured out how to get it to just let me browse the directories). ? Any help you can provide is appreciated. ? Peter --
|
Hi Steven,
Searching, unit 5 should be keyboard/standard input so I was expecting the
program to wait for a line at the terminal. If I work around the input by
hard coding some values so the the input part becomes:
CMS "borrowed" the FORTRAN compiler from MVS. You will need FILEDEF statements to associate MVS DDNAMEs and FORTRAN unit numbers with files or devices, for example (untested): FILEDEF FT05F001 TERM FILEDEF FT06F001 DISK OUTPUT FILE A
Now the output seems to be discarded. It isn't printed to the terminal,
isn't queued for punch or printer and isn't in a file that I can see.
Have a look around for a file called FILE FT06F001 A or something like that.
Interestingly the same issue with discarded output seems to happen with the
supplied hello fortran program.
Are they run from an EXEC which contains suitable FILEDEF statements? Regards, Peter Coghlan.
|
Guys, thanks for all the help so far.? I got the card punch, printer and IND$FILE transfers working.
Next I thought I'd try compiling a Fortran program.? I grabbed the example on? area of a triangle by Heron's formula.? After updating the I/O statements to avoid 'TAPE' and changing FLOATF to FLOAT and SQRTF to SQRT it will now compile and load but I/O still does not seem to?work.
This is the first version that would compile/load:
C AREA OF A TRIANGLE WITH A STANDARD SQUARE ROOT FUNCTION
C INPUT - TAPE READER UNIT 5, INTEGER INPUT
C OUTPUT - LINE PRINTER UNIT 6, REAL OUTPUT
C INPUT ERROR DISPLAY ERROR OUTPUT CODE 1 IN JOB CONTROL LISTING
? ? ? READ (5, 501) IA, IB, IC
? 501 FORMAT (3I5)
C IA, IB, AND IC MAY NOT BE NEGATIVE OR ZERO
C FURTHERMORE, THE SUM OF TWO SIDES OF A TRIANGLE
C MUST BE GREATER THAN THE THIRD SIDE, SO WE CHECK FOR THAT, TOO
? ? ? IF (IA) 777, 777, 701
? 701 IF (IB) 777, 777, 702
? 702 IF (IC) 777, 777, 703
? 703 IF (IA+IB-IC) 777, 777, 704
? 704 IF (IA+IC-IB) 777, 777, 705
? 705 IF (IB+IC-IA) 777, 777, 799
? 777 STOP 1
C USING HERON'S FORMULA WE CALCULATE THE
C AREA OF THE TRIANGLE
? 799 S = FLOAT ?(IA + IB + IC) / 2.0
? ? ? AREA = SQRT ( S * (S - FLOAT (IA)) * (S - FLOAT (IB)) *
? ? ?+ ? ? (S - FLOAT (IC)))
? ? ? WRITE (6, 601) ? ? ? ? ? ?IA, IB, IC, AREA
? 601 FORMAT (4H A= ,I5,5H ?B= ,I5,5H ?C= ,I5,8H ?AREA= ,F10.2,
? ? ?+ ? ? ? ?13H SQUARE UNITS)
? ? ? STOP
? ? ? END
When I try to run it with 'start' I get:
IHC218I FIOCS - I/O ERROR ?120S INPUT ?ERROR 001 ON FT05F001,
Searching, unit 5 should be keyboard/standard input so I was expecting the program to wait for a line at the terminal.? If I work around the input by hard coding some values so the the input part becomes:
C ? ? READ (5, 501) IA, IB, IC
C 501 FORMAT (3I5)
? ? ? IA = 3
? ? ? IB = 4
? ? ? IC = 5
Now the output seems to be discarded.? It isn't printed to the terminal, isn't queued for punch or printer and isn't in a file that I can see.? Interestingly the same issue with discarded output seems to happen with the supplied hello fortran program.
|
Re: Real and Virtual Printers and Punches and File Transfers
Steven Well if you are using one of the incarnations of the six-pack it comes with ind$file and you can use that to move files in either direction. If you are using Windows and the standard TN3270 client from the six-pack you may need to tweak the Windows console settings to let you click on the file menu. Then click on ¡°file¡± ¡°transfer¡± and follow the prompts. ? The Punch and reader work well as well. So the first thing is to get your virtual punch pointing to the system spool, and with the right class, so ? ¡°SPOOL PUN SYSTEM CLASS P¡± ? Then punch your file ? ¡°PUNCH fn ft¡± ? Then start the system punch. So from the Operator, which is usually attached to the Hercules console and needs a ¡°/¡± prefix to ? ¡°/cp start d¡± ? And your cards will be placed in the file attached to ¡°D¡± in the config. You can change it by doing ? ¡°devinit d my-punch-file.txt¡± ? On the Hercules console. You don¡¯t need the ¡°/¡± as it¡¯s a hercules command. ? Dave ?
toggle quoted message
Show quoted text
From: [email protected] <[email protected]> On Behalf Of Steven Fosdick
Sent: 20 January 2020 13:09
To: [email protected]
Subject: [h390-vm] Real and Virtual Printers and Punches and File Transfers? Thinking about how I might be able to transfer?a simple text file from within CMS to the Hercules?host I had the idea that I may be able to send it to the card punch which would should then appear in the a host file but I then discovered that VM virtualises the card punch so when you use the 'punch' command the output is placed in a spool file.? Not a great surprise but the spool file is not marked as held and yet I can't seem to find any way to tell the VM spooling system that it should send that file to what it thinks is the real card punch (i.e. the one virtualised by Hercules, not VM).? Exactly the same situation applies to the printer. Searching, I didn't find any solution to for the card punch; most documentation and web resources are more recent and I don't suppose anyone would think that you'd want to use a real card punch. For printing, the only solution I found any mention of involved driving the printer from MVS, either on a completely different machine, or by running MVS in a VM on the same machine and either way using an extra bit of software whose name I have forgotten to transfer the spool files from VM to MVS.? Is there no easier way, especially as I suspect the bit of software concerned may be proprietary. So stepping back a bit, what is the easiest way to transfer text files between CMS and the Hercules host OS?
|
Re: Real and Virtual Printers and Punches and File Transfers
Thinking about how I might be able to transfer a simple text file from
within CMS to the Hercules host I had the idea that I may be able to send
it to the card punch which would should then appear in the a host file but
I then discovered that VM virtualises the card punch so when you use the
'punch' command the output is placed in a spool file. Not a great surprise
but the spool file is not marked as held and yet I can't seem to find any
way to tell the VM spooling system that it should send that file to what it
thinks is the real card punch (i.e. the one virtualised by Hercules, not
VM). Exactly the same situation applies to the printer.
You need to start the "real" card punch or printer, for example, while logged in to OPERATOR or MAINT, use: CP START 00D (00D) is the "real" address of the "real" card punch. Regards, Peter Coghlan.
|
Re: Getting Started with VM/CMS
Steven There are a couple of IBM editors with a prefix area. If you used VM it was most likely you used XEDIT. Superficially it looks similar to the ISPF editor but few, if anyone used that on VM ISPF had to be bought, XEDIT was free and part of VM. The look and feel was very customisable so just because a screen looks different to what you remember on VM is doesn¡¯t mean it¡¯s not the same editor underneath. I actually find FSEDIT frustrating because its so different to XEDIT¡. Many commands are not the same¡ ? Dave ?
toggle quoted message
Show quoted text
From: [email protected] <[email protected]> On Behalf Of Steven Fosdick
Sent: 20 January 2020 12:52
To: [email protected]
Subject: Re: [h390-vm] Getting Started with VM/CMS? Mentioning xedit, that was going to be my next question.? The editor I remember was one where each line was prefixed by an area into which you could type one or two letter commands and block operations were done by typing the command on both the start and end lines.? Looking at the description of xedit it seems like that is probably the editor I remember, though it could have been the editor invoked from ISPF.? Does that give any more clue as to which version I may have been using?? I looked at EDIT from the sixpack and it looks altogether more primitive. I have also had a play with MVS 3.8J and Rob Finn's RPF.? The editor in that looks good and quite similar, at least superficially, to xedit so I immediately started to wonder if that could? be ported to CMS but I am very glad to hear of fsedit. ? On Sun, 19 Jan 2020 at 22:30, adriansutherland67 <adrian@...> wrote: 35 years ago ... 1985 ... VM/ESA ... so a bit more modern to the old six pack (!)
On the six pack 1.3 beta we have fsedit rather than xedit and fslist. Both community created. And also brexx.
Fsedit will bring back memories F7/8, f3 etc.
Adrian
|
Real and Virtual Printers and Punches and File Transfers
Thinking about how I might be able to transfer?a simple text file from within CMS to the Hercules?host I had the idea that I may be able to send it to the card punch which would should then appear in the a host file but I then discovered that VM virtualises the card punch so when you use the 'punch' command the output is placed in a spool file.? Not a great surprise but the spool file is not marked as held and yet I can't seem to find any way to tell the VM spooling system that it should send that file to what it thinks is the real card punch (i.e. the one virtualised by Hercules, not VM).? Exactly the same situation applies to the printer.
Searching, I didn't find any solution to for the card punch; most documentation and web resources are more recent and I don't suppose anyone would think that you'd want to use a real card punch.
For printing, the only solution I found any mention of involved driving the printer from MVS, either on a completely different machine, or by running MVS in a VM on the same machine and either way using an extra bit of software whose name I have forgotten to transfer the spool files from VM to MVS.? Is there no easier way, especially as I suspect the bit of software concerned may be proprietary.
So stepping back a bit, what is the easiest way to transfer text files between CMS and the Hercules host OS?
|
Re: Getting Started with VM/CMS
Mentioning xedit, that was going to be my next question.? The editor I remember was one where each line was prefixed by an area into which you could type one or two letter commands and block operations were done by typing the command on both the start and end lines.? Looking at the description of xedit it seems like that is probably the editor I remember, though it could have been the editor invoked from ISPF.? Does that give any more clue as to which version I may have been using?? I looked at EDIT from the sixpack and it looks altogether more primitive.
I have also had a play with MVS 3.8J and Rob Finn's RPF.? The editor in that looks good and quite similar, at least superficially, to xedit so I immediately started to wonder if that could? be ported to CMS but I am very glad to hear of fsedit.
On Sun, 19 Jan 2020 at 22:30, adriansutherland67 < adrian@...> wrote: 35 years ago ... 1985 ... VM/ESA ... so a bit more modern to the old six pack (!)
On the six pack 1.3 beta we have fsedit rather than xedit and fslist. Both community created. And also brexx.
Fsedit will bring back memories F7/8, f3 etc.
Adrian
|
Ren¨¦ Ferland
Dave McGuire