¿ªÔÆÌåÓý

  1. H390-Vm
  2. Messages
Expanded Topics Messages Polls
Date
Re: A question about vm/370 print output #VMCE

 

Bob Bolch wrote:

Are you saying that the printer CCW opcodes like:

09 Write space 1
11 Write space 2
89 Write and skip to channel 1
01 Write w/o space

are being translated by the 819/1047 translate table
to get bytes that end up in the ASCII print file that
Hercules creates?
Um, er, uh ... Not any more I'm not!

Yeah, that was a definite brain fart. Sorry. :(

Most carriage control characters are translated into a series of LF (or CRLF) characters, and the actual TEXT of the line is translated from EBCDIC to ASCII.

For those carriage control characters which cause a form eject (i.e. skip to the top of the next page) to occur however, Hercules outputs a "form-feed" character, expressed as '\f' in ASCII (which just so happens to be the exact same hexadecimal value in both ASCII and EBCDIC: hexadecimal 0C).

Sorry for the confusion. :(

But Hercules definitely doesn't write out 0xff characters on its own. The 0xff that's getting written MUST be some unusual data byte that the guest (VM/370) is actually printing/sending to the printer that just so happens to get translated by Hercules (via its CODEPAGE tables) into an ASCII 0xff character.

But whatever is causing the 0xff to get written to Mike's printer output file, it's NOT Hercules that's doing it. It's the guest (VM/370). Why it does that I don't know. z/VM doesn't do it. Why does VM/370?

--
"Fish" (David B. Trout)
Software Development Laboratories

mail: fish@...
Re: A question about vm/370 print output #VMCE

 

Yes, I am sure about 0xff. As a matter of fact, first 5 bytes are 0xff 0x0d 0x0a 0x0d 0x0a and 0x0c which is a form feed. You can see this in files 5664167MEMO.txt and?5664147MEMO.txt which are available in the files area of this group. This character is found even in pdf files created from these text files.?
Re: A question about vm/370 print output #VMCE

 

Hi Fish,
I am confused.?

Are you saying that the printer CCW opcodes like:
09 Write space 1
11 Write space 2
89 Write and skip to channel 1
01 Write w/o space

are being translated by the 819/1047 translate table to get bytes that
end up in the ASCII print file that?Hercules creates??

Bob Bolch

On Fri, Feb 18, 2022 at 2:28 PM Fish Fish <david.b.trout@...> wrote:
mikeci@... wrote:

> I noticed that first byte in the output is 0xff.
> Is there a reason for this character to be there?

I have no idea. Are you sure it's 0xff and not something else? Did you examine it using a hex editor?

I do know that when the guest (VM/370 in this case) issues a skip to channel-1 command to the printer for example (i.e. "skip to the top of the next page"), that command code normally gets translated from EBCDIC to its ASCII equivalent by Hercules according to whatever translation CODEPAGE is in effect in Hercules, so if you're using some unusual CODEPAGE, then the resulting printer file that Hercules produces will obviously appear to contain "unusual" characters from the host system's (i.e. Windows's or Linux's, etc) point of view. Why yours is 0xff I have no idea.

Are ALL of your printer output files like that? Or only the very first one? It might be something VM/370 is doing only on its very first printout. Do a "devinit" on your printer file (or otherwise empty it or clear it out) and then print something else. Does it also contain a 0xff? Mine contains 0x0c, which is an ASCII "FF" ("Form Feed").


> I also noticed that there is no end banner.

The original VM/370 as delivered never produced any end-of-print-job banners, only begin-of-print-job.


> Can this be added?

I'm sure it probably could by customizing some file somewhere, but I don't know how to do so myself. Maybe someone else knows?

--
"Fish" (David B. Trout)
Software Development Laboratories

mail: fish@...
Re: A question about vm/370 print output #VMCE

 

mikeci@... wrote:

I noticed that first byte in the output is 0xff.
Is there a reason for this character to be there?
I have no idea. Are you sure it's 0xff and not something else? Did you examine it using a hex editor?

I do know that when the guest (VM/370 in this case) issues a skip to channel-1 command to the printer for example (i.e. "skip to the top of the next page"), that command code normally gets translated from EBCDIC to its ASCII equivalent by Hercules according to whatever translation CODEPAGE is in effect in Hercules, so if you're using some unusual CODEPAGE, then the resulting printer file that Hercules produces will obviously appear to contain "unusual" characters from the host system's (i.e. Windows's or Linux's, etc) point of view. Why yours is 0xff I have no idea.

Are ALL of your printer output files like that? Or only the very first one? It might be something VM/370 is doing only on its very first printout. Do a "devinit" on your printer file (or otherwise empty it or clear it out) and then print something else. Does it also contain a 0xff? Mine contains 0x0c, which is an ASCII "FF" ("Form Feed").


I also noticed that there is no end banner.
The original VM/370 as delivered never produced any end-of-print-job banners, only begin-of-print-job.


Can this be added?
I'm sure it probably could by customizing some file somewhere, but I don't know how to do so myself. Maybe someone else knows?

--
"Fish" (David B. Trout)
Software Development Laboratories

mail: fish@...
Re: A question about vm/370 print output #VMCE

 

¿ªÔÆÌåÓý

Folks,

?

The first character is 09f on mine. However, remember the host knows nothing of ASCII, its printing in EBCDIC and sending hardware printer control characters to the printer.

Hercules is then converting these to ASCII, so I think it¡¯s a Hercules question. You might want to print in EBCDIC to see what its really sending.

It may also be that this character is used for handshaking with some of the tools that convert these listings to PDF etc.

?

As for an end banner, well by default VM was delivered with no end banner. We left it that way. Its stored in DMKBOX. Fell free to add one, but note there are updates already applied to DMKBOX.

?

Dave

?

?

From: [email protected] <[email protected]> On Behalf Of Bob Bolch
Sent: 18 February 2022 11:42
To: [email protected]
Subject: Re: [h390-vm] A question about vm/370 print output #VMCE

?

The mainframe printer interface produces channel commands and data which is turned into the file on the host platform.?

The specifics of the host operating system interface in Hercules controls what ends up in the host printer file.

I suggest that you ask this question on the hercules forum:?

Bob Bolch

?

On Thu, Feb 17, 2022 at 12:59 AM <mikeci@...> wrote:

I noticed that first byte in the output is 0xff. Is there a reason for this character to be there? I also noticed that there is no end banner. Can this be added??

Re: A question about vm/370 print output #VMCE

 

The mainframe printer interface produces channel commands and data which is turned into the file on the host platform.?
The specifics of the host operating system interface in Hercules controls what ends up in the host printer file.
I suggest that you ask this question on the hercules forum:?
Bob Bolch

On Thu, Feb 17, 2022 at 12:59 AM <mikeci@...> wrote:
I noticed that first byte in the output is 0xff. Is there a reason for this character to be there? I also noticed that there is no end banner. Can this be added??

A question about vm/370 print output #VMCE

 

I noticed that first byte in the output is 0xff. Is there a reason for this character to be there? I also noticed that there is no end banner. Can this be added??
Re: WAKEUP - does a emulation of this module exist

 

On Sat, Feb 12, 2022 at 09:33 AM, Anthony Smith wrote:
MACLIB MAP tells me the SCHMAC MACLIB is not a library. I have had a look and it is full of macros that I can read by eye including the ??EQU that defines the registers. Cannot work out what could be the issue with SCHMAC other than perhaps it originates from a later VM?
Sorry. I did not actually attempt to use it before I uploaded it. It is from a later version of VM.

SCHMAC?? MACLIB?? A1? F??? 80?? 1757??? 176? 1986-04-22 05:45

Somewhere around SP 4/SP 5. Give me a few days, and I'll get it converted to the older format.

?... Mark S.
Re: WAKEUP - does a emulation of this module exist

 

MACLIB MAP tells me the SCHMAC MACLIB is not a library. I have had a look and it is full of macros that I can read by eye including the ??EQU that defines the registers. Cannot work out what could be the issue with SCHMAC other than perhaps it originates from a later VM?

Ant
Re: How do I load a windows folder containing 100+ assemble/copy/macro files onto my 'A' disk under VM #sixpack

 

"James Oswald via groups.io" <oswaldjb@...> writes:

Greetings, I'm running the VM370sixpack-1_3 Beta system, and I'm wondering how I
can load all the assemble / copy / macro files I have in a windows folder onto my 'A'
disk so that I can attempt to assemble and rebuild a series/1 emulator I wrote
almost 40 years ago.
I did successfully read one file I pointed 00C to in the configuration file, but I can't do
that one by one for 100+ membs.
I'm sure there is a way - I just haven't stumbled across it yet.
I'm running the Sixpack system under Windows.
Thanks for any help.
Get a copy of the program vma. One place it looks likes it is
available is ".

This program is used like zip/unzip but with vmarc archives (the
the archive file type used by VM), then trnsfer the resulting
archive to VM run vmarc on the VM system to extract your source.

There may be other ways, but this seems to be the simplest. But
then again, I'm not a VM expert, just fooled around with some
of the vm/370 systems such as the various Sixpack or Community
Editions.

--

Bill Doughty, N2OCM

MVS Sysprog & UNIX/Linux Sysadmin (retired)
Re: How do I load a windows folder containing 100+ assemble/copy/macro files onto my 'A' disk under VM #sixpack

 

¿ªÔÆÌåÓý

James,

?

Lots of ways. If you have a REXX that¡¯s supported by Hercules you could write REXX exec to load all the files via 00C.

If you are using one of Fish¡¯s Hercules you can use his VMFPLC2 utility to create and AWS tape which you can load with VMFPLC2 => you will need a control file

There are some PC utilities in the files section of the group that let you build CMS TAPE tapes and DISKDUMP card decks.

You can use an OMA/TDF tape. You need a Tape Descriptor FILE (.TDF) that lists the disk files you want to be on a tape. See the documentation for your Hercules.

You can ZIP the files up with no compression and use MVSUNZIP on VM

There are PC versions of VMARC¡­¡­

?

Dave

?

?

From: [email protected] <[email protected]> On Behalf Of James Oswald via groups.io
Sent: 09 February 2022 16:00
To: [email protected]
Subject: [h390-vm] How do I load a windows folder containing 100+ assemble/copy/macro files onto my 'A' disk under VM #sixpack

?

Greetings,? I'm running the VM370sixpack-1_3 Beta system, and I'm wondering how I can load all the assemble / copy / macro files I have in a windows folder onto my 'A' disk so that I can attempt to assemble and rebuild a series/1 emulator I wrote almost 40 years ago.?
I did successfully read one file I pointed 00C to in the configuration file, but I can't do that one by one for 100+ members.
I'm sure there is a way - I just haven't stumbled across it yet.
I'm running the Sixpack system under Windows.
Thanks for any help.

Re: How do I load a windows folder containing 100+ assemble/copy/macro files onto my 'A' disk under VM #sixpack

 

¿ªÔÆÌåÓý

Hi James,

you can make a VMARC file of that directory and upload that using IND$FILE, using the upload facility of most 3270 emulators. You can get a windows version of VMA (that can make a VMARC file) on?
If that does not work, send me a zip file and I¡¯ll send you a VMARC file of it back.

Then VMARC UNPK it on VM/370.?

best regards,

¸é±ð²Ô¨¦.

On 9 Feb 2022, at 17:00, James Oswald via <oswaldjb@...> wrote:

Greetings,? I'm running the VM370sixpack-1_3 Beta system, and I'm wondering how I can load all the assemble / copy / macro files I have in a windows folder onto my 'A' disk so that I can attempt to assemble and rebuild a series/1 emulator I wrote almost 40 years ago.?
I did successfully read one file I pointed 00C to in the configuration file, but I can't do that one by one for 100+ members.
I'm sure there is a way - I just haven't stumbled across it yet.
I'm running the Sixpack system under Windows.
Thanks for any help.

How do I load a windows folder containing 100+ assemble/copy/macro files onto my 'A' disk under VM #sixpack

 

Greetings,? I'm running the VM370sixpack-1_3 Beta system, and I'm wondering how I can load all the assemble / copy / macro files I have in a windows folder onto my 'A' disk so that I can attempt to assemble and rebuild a series/1 emulator I wrote almost 40 years ago.?
I did successfully read one file I pointed 00C to in the configuration file, but I can't do that one by one for 100+ members.
I'm sure there is a way - I just haven't stumbled across it yet.
I'm running the Sixpack system under Windows.
Thanks for any help.
Re: Stupid z/VM 7.1 assembler question

 

HLASM is included with z/OS but I believe it is still an additional cost.

Dennis

On 2/6/22 22:55, Fish Fish wrote:
Tony Harminc wrote:

[...]
I think it's separately priced. :-(
:(


It's included in z/OS because it's needed for various things
like sysgens.
I would think the same would be true for z/VM too!


[...]
Of course if you have access to a z/OS system you can use that
and just send the object deck over. There is obviously nothing
in your program that depends on an operating system. :-)
Or I can just use Harold Grovesteen's excellent ASMA that comes with his SATK product:

*

and then just upload the resulting binary .core file to z/VM and load and run it via "cpvload" instead (which is what I usually do and what I did to resolve my problem this time too). It's just that I thought it might be faster and simpler to just write it directly on z/VM itself since it was such a small/simple program rather than having to mess with "vmfplc2"ing it. So I was quite surprised when that didn't work.


I also ran it through HLASM here at work, and it fails because
your TBEGIN[C]s have no operands. Yeah, you'd think that comma
would do, but I changed them to 0,0 and that makes it happy.
Thanks. Fixed. ASMA is more forgiving in that regard though. For any missing arguments it just presumes/uses 0.


(Well it then complains about no base register in the D1(B1)
expression, but that sounds bogus to me.
That sounds completely bogus to me too. The use of a TDB is entirely optional for TBEGIN and is not even allowed for TBEGINC, and the way you indicate that (according to the Principles of Operation) is by specifying the operand as zero.


Regardless, I turned off the complaint (NOPAGE0 option) and it
assembles cleanly. I can send you the listing and object deck
if you like, but I can't do this as an ongoing service.
Nah. I got around it using ASMA as I said. (I just *love* ASMA!)

But thanks anyway.
Re: Stupid z/VM 7.1 assembler question

 

Mike,
Not sure how long ago you got involved, but I feel this goes back to when IBM released VM/ESA and started shipping some modules as OCO.
They wanted to stop customers from tampering with the code, and removing the ASSEMBLER made this simpler.
It was much harder to justify a system modification of you had to buy the assembler to implement it.
You might want to browse some of the VMSAHRE files on OCO here:-



Dave

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Mike Ward
Sent: 07 February 2022 12:02
To: [email protected]
Subject: Re: [h390-vm] Stupid z/VM 7.1 assembler question

It's all about money......

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Fish Fish
Sent: Sunday, February 6, 2022 9:31 PM
To: [email protected]
Subject: Re: [h390-vm] Stupid z/VM 7.1 assembler question

Dave Wade wrote:

[...]
There is no HLASM in the base. It remains a chargeable, installable
product.
!! (me: shocked)


[...]
Not sure but be aware that as well as having no modern OP CODE support
from what I remember it won't generate code that the linker considers
to be 64-bit. It is in fact exactly the same assembler that there is
in VM/370 (well I think it has a couple of fixes) but it still has the
same limits and needs patching to assemble.
I find it both shocking and disappointing that IBM now chooses to charge for
such a product. :(

You would think an assembler that supports the very architecture their very
operating system runs in (as well as what's used in all of their code
samples/examples and what they recommend/encourage all their customers
to use too) would be an integral REQUIRED component of the operating
system itself, and not a chargeable "add-on" product. It seems silly to me
(and somewhat insulting as well) to only provide an assembler that is by
design INCAPABLE of generating the very code that the operating system
itself uses.

Boy, IBM sure has changed over the years.

And not for the better either.

(sigh)

:(

--
"Fish" (David B. Trout)
Software Development Laboratories

mail: fish@...
Re: Stupid z/VM 7.1 assembler question

 

It's all about money......

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Fish Fish
Sent: Sunday, February 6, 2022 9:31 PM
To: [email protected]
Subject: Re: [h390-vm] Stupid z/VM 7.1 assembler question

Dave Wade wrote:

[...]
There is no HLASM in the base. It remains a chargeable, installable
product.
!! (me: shocked)


[...]
Not sure but be aware that as well as having no modern OP CODE support
from what I remember it won't generate code that the linker considers
to be 64-bit. It is in fact exactly the same assembler that there is
in VM/370 (well I think it has a couple of fixes) but it still has the
same limits and needs patching to assemble.
I find it both shocking and disappointing that IBM now chooses to charge for such a product. :(

You would think an assembler that supports the very architecture their very operating system runs in (as well as what's used in all of their code samples/examples and what they recommend/encourage all their customers to use too) would be an integral REQUIRED component of the operating system itself, and not a chargeable "add-on" product. It seems silly to me (and somewhat insulting as well) to only provide an assembler that is by design INCAPABLE of generating the very code that the operating system itself uses.

Boy, IBM sure has changed over the years.

And not for the better either.

(sigh)

:(

--
"Fish" (David B. Trout)
Software Development Laboratories

mail: fish@...
Re: Stupid z/VM 7.1 assembler question

 

Tony Harminc wrote:

[...]
I think it's separately priced. :-(
:(


It's included in z/OS because it's needed for various things
like sysgens.
I would think the same would be true for z/VM too!


[...]
Of course if you have access to a z/OS system you can use that
and just send the object deck over. There is obviously nothing
in your program that depends on an operating system. :-)
Or I can just use Harold Grovesteen's excellent ASMA that comes with his SATK product:

*

and then just upload the resulting binary .core file to z/VM and load and run it via "cpvload" instead (which is what I usually do and what I did to resolve my problem this time too). It's just that I thought it might be faster and simpler to just write it directly on z/VM itself since it was such a small/simple program rather than having to mess with "vmfplc2"ing it. So I was quite surprised when that didn't work.


I also ran it through HLASM here at work, and it fails because
your TBEGIN[C]s have no operands. Yeah, you'd think that comma
would do, but I changed them to 0,0 and that makes it happy.
Thanks. Fixed. ASMA is more forgiving in that regard though. For any missing arguments it just presumes/uses 0.


(Well it then complains about no base register in the D1(B1)
expression, but that sounds bogus to me.
That sounds completely bogus to me too. The use of a TDB is entirely optional for TBEGIN and is not even allowed for TBEGINC, and the way you indicate that (according to the Principles of Operation) is by specifying the operand as zero.


Regardless, I turned off the complaint (NOPAGE0 option) and it
assembles cleanly. I can send you the listing and object deck
if you like, but I can't do this as an ongoing service.
Nah. I got around it using ASMA as I said. (I just *love* ASMA!)

But thanks anyway.

--
"Fish" (David B. Trout)
Software Development Laboratories

mail: fish@...
Re: Stupid z/VM 7.1 assembler question

 

Dave Wade wrote:

[...]
There is no HLASM in the base. It remains a chargeable,
installable product.
!! (me: shocked)


[...]
Not sure but be aware that as well as having no modern OP CODE
support from what I remember it won't generate code that the
linker considers to be 64-bit. It is in fact exactly the same
assembler that there is in VM/370 (well I think it has a couple
of fixes) but it still has the same limits and needs patching
to assemble.
I find it both shocking and disappointing that IBM now chooses to charge for such a product. :(

You would think an assembler that supports the very architecture their very operating system runs in (as well as what's used in all of their code samples/examples and what they recommend/encourage all their customers to use too) would be an integral REQUIRED component of the operating system itself, and not a chargeable "add-on" product. It seems silly to me (and somewhat insulting as well) to only provide an assembler that is by design INCAPABLE of generating the very code that the operating system itself uses.

Boy, IBM sure has changed over the years.

And not for the better either.

(sigh)

:(

--
"Fish" (David B. Trout)
Software Development Laboratories

mail: fish@...
Re: Stupid z/VM 7.1 assembler question

 

Joe Monk wrote:

I think what you are looking for can be found at File 177
of the CBT tape. This is Jan Jaeger's MNEMAC package which
implements the z/arch instructions as macros to assembler XF.
Thanks, Joe!

--
"Fish" (David B. Trout)
Software Development Laboratories

mail: fish@...
Re: Stupid z/VM 7.1 assembler question

 

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Fish Fish
Sent: 06 February 2022 20:25
To: [email protected]
Subject: Re: [h390-vm] Stupid z/VM 7.1 assembler question

Ren¨¦ wrote:

Is HLASM now in the base?
I have no idea! How do I tell? I told you I am inexperienced with modern
z/VM! Help!
There is no HLASM in the base. It remains a chargeable, installable product.


In 6.4 it still wasn¡¯t. If you only have assembler XF, Fish, look for
the downloadable OPCODES package.
Where do I find it? Where can I download it from? How to I install it once it's
downloaded? Help!
Not sure but be aware that as well as having no modern OP CODE support from what I remember it won't generate code that the linker considers to be 64-bit.
It is in fact exactly the same assembler that there is in VM/370 (well I think it has a couple of fixes) but it still has the same limits and needs patching to assemble.


--
"Fish" (David B. Trout)
Software Development Laboratories

mail: fish@...
Dave
G4UGM
2561 - 2580 of 6546