I've just updated ?with the results of my work to create a Rexx system-level function package for the various VM/370 R6s.? I have specifically tested it with both bREXX 1.0.1 (as shipped in VM/370 CE 1.1) and the first beta of bREXX 1.1.0 (as I announced last month).? In combination with a small change to the bREXX DATE() function (to add BASEDATE), this brings bREXX fully up to the Rexx 3.40 level (with considerable additional function from higher levels).
This package provides the DIAG() and DIAGRC() functions that VM/SP (and later) had, with the exception of C8 and CC (because VM/370 CP doesn't have those DIAGnoses).
I intend to package this for inclusion in the next release of VM/370 CE, but it should work on any VM/370 system that can already run bREXX 1.0.1.? If you want to use it before that, or with another VM/370 system, you can download the ZIPped VMARC file from the GitHub issue.? It contains the full source and MODULE, including a few CMS macro changes that were necessary (and which are teed up for the next VM/370 CE release).
Ross
|
Re: Updated Wiki Page: Reserved CMS Modification Numbers
#wiki-notice
William Denton wrote in that file:
> I would also propose that new files added to the CMS nucleus or CMSSEG be named HDSxxx since CP seems to be using HRCxxx and it would probably be good to avoid any conflicts. Exceptions might be something like a major Assembler-based component that VM/SP would have included in the resident code. I have noticed that some VM/SP-like things have been added with the SP names using DMSxxx pattern.... are we all comfortable with the implications of this?
- Let's do discussions like this on the list, where more folks will see them.
- I know some of the VM/370 CE DMSxxx additions so far have been deliberate clean-room replications of VM/SP functions (e.g., DMSXIO / EXECIO, DMSNXD / NUCXDROP, DMSNXM / NUCXMAP). I'm about to add another - DMSRSF / RXSYSFN, to provide the DIAG() and DIAGRC() functions for bREXX.
- I don't know what implications of naming are of concern to you. The only one that comes to my mind is confusion about whether code was part of the VM/370 R6 package or not.
|
Re: Reserving Modification Numbers
Is there any recommended set of modification numbers for new work to be delivered as part of VM/370 CE post-1.2? I don't mind filling in some of the low-numbered holes, if that's appropriate, or taking from high up, likewise.
Ross
|
Re: #VMCE #rexx EE goes XEDIT - compiling a wish list
#VMCE
#rexx
Hi, Mark,
?
Mark,
?
I hope you and yours are doing well, and preparing for the Thanksgiving holiday.
?
I just saw this message from September.? (replying to an older message)
?
I did manage to install SEDIT, but it was a good number of years ago, and I have slept too many? times since then to remember exactly how long ago that was.
?
But, I did use SEDIT for a while back then.? IIRC, this was before the advent of "EE".
?
First, you also need to install "The PROC System" from the same Waterloo tapes, from the same place that SEDIT came from, because SEDIT is written using "The PROC System" macros.? ?
?
The other thing you need to be aware of is that there are certain "features" of SEDIT that will not work "out of the box", for example, there are many extended commands that have a file type of SEDITMOD.? Those are relocatable modules, but not using the standard from VM/SP and above, because SEDIT was all developed way before then, starting on VM Rel. 3 (maybe even earlier?)
?
The problem is, they used a proprietary relocating loader called RSIFETCH that no one seems to have a copy of, and that no one has "reverse engineered."? In theory, you could just use CMS TEXT decks and then use the OS LOAD macro to load them dynamically.? But, I am not sure how much work this would entail, to retrofit that to SEDIT.
?
Next, one of those SEDITMODs is called "FS" or "FSEDIT" -- a "full-screen editor" front-end for SEDIT.? But, you need to know that they were not using 3270s; this only worked with a very obscure async. "glass teletype" terminals made by a company called "Delta Data" (long since out of business), who sold their "TEMPEST approved" (hardened) terminals to the DoD.? And, they also had somehow integrated their own special software to manage the Delta Datas, so that meant mods deep inside VM/370 CP.? ?So, you can forget about that "FS" for now.
?
?The third thing you need to know is, some "commands" in SEDIT were actually "EDEXECs" -- they took the source code for the CP-67/CMS version of EXEC, and modified it to integrate it into SEDIT, so it would allow you to write edit macros for SEDIT.? Very powerful, but also somewhat cumbersome, as this was way before "EXEC II" appeared, and so it used a kludge to work around the 8-byte tokenization issues.
?
SEDIT, like The PROC System, is a rather large collection of many CSECTs or ASSEMBLE source files.
?
Let me know if you really want to dig into this, and I can try to spend some time to resurrect my knowledge of this artifact, having worked with it for a while back in the day.
?
All the best,
?
Mark S. Waterbury
|
Re: DMSCIT171T Permanent console error; re-IPL CMS after TERM CONMODE 3270
Strange indeed that 3270 programs work.? But that's an addition not available in the beginning.
CMS only speaks linemode, 3215.
But, CMS programs can use 3270 full screen since VM got "DIAGNOSE 58".? A CP feature for CMS.? Editors like EDGAR and XEDIT? use it; just like FLIST and BROWSE to name things common in VM/SP.
The old, public area CP doesn't have DIAG58 support, but "the community" added this to CP too, and now there is a full screen editor too: EE
?
P.S. when I say CMS has no full screen support, I lie a bit: I think that in VM/SP Rel 5 CMS got it, with VSCREENs and WINDOWs? so you can make your own in REXX for example.? And there can the command SET FULLSCREEN ON to make CMS itself use VSCREENs and WINDOWs, with scrolling features etc.? SET FULLSCREEN ON isn't used a lot.?
Some of my tools on VM's download library ( ) include my SELFS EXEC (Select Fullscreen) and/or RxMB (Rexx MessageBox) to use VSCREENs and WINDOWs. (I can send them to you if you like).? The native use of VSCREENs and WINDOWs isn't that easy.
?
--
Kris Buelens
|
Re: DMSCIT171T Permanent console error; re-IPL CMS after TERM CONMODE 3270
Thank you,
your answer helped me understand how it works.
|
Re: DMSCIT171T Permanent console error; re-IPL CMS after TERM CONMODE 3270
Thank you for answer.
CMS require 3215 conmode, but programs for 3270 work, it confused me.
|
Re: DMSCIT171T Permanent console error; re-IPL CMS after TERM CONMODE 3270
TERM CONMODE 3270 is useful to run guests.? The CMS abend isn't nice.? So to avoid that
- All manual:
SYSTEM RESET? ? to kill CMS
TERM CONMODE 3270
IPL xxxx
- At the console on 1 command
#CP TERM CONMODE 3270#IPL xxxx
- In a REXX exec
nl='15'x
address command 'CP TERM CONMODE 3270' || nl || 'IPL xxxx'
Often a DEF STOR is done as well before starting the guest: insert that before the IPL command
--
Kris Buelens
|
Re: DMSCIT171T Permanent console error; re-IPL CMS after TERM CONMODE 3270
On Sat, Nov 23, 2024 at 02:02 PM, Evgenii Shikhovtsev wrote:
Why after TERM CONMODE 3270 i get DMSCIT171T ?
?
?
Because virtual machines running CMS require CONMODE 3215 (but you don't say if you run the command from the CMS Ready prompt).
?
Cheers,
?
Rene FERLAND, Montreal
|
DMSCIT171T Permanent console error; re-IPL CMS after TERM CONMODE 3270
z/VM 4.4 CMS level 20 service 301
Why after TERM CONMODE 3270 i get DMSCIT171T ?
|
Re: VM/ESA 370 (was Re: VM/370 new thrilling experience)
At that time, I was myself in charge of the applications for a
French Research Lab. and my only problem was to ensure that they
will go through the millennium gate safely... A VM-PC helped me a
lot : That was the very first time I was allowed to log on
OPERATOR or even MAINT... Today I'm allowed to but have not the
required knowledge to tweak the system : the only assembly
language code I ever committed was the implementation of a
QUICKSORT? and an IUCV interface for VS-PASCAL programs for
internal use. (Truth to be told, I'm not sure I would still be
able to do so today... But reading about your experiences, and
your achievements around VM-CE make me dream again abount the
nicest years of my professional life).
?Thank you so much for this.
Jean-Pierre Cabani¨¦
Le 16/11/2024 ¨¤ 03:25, Mark A. Stevens
via groups.io a ¨¦crit?:
On Mon, Nov 11, 2024 at 02:59 AM, Dave Wade wrote:
It would be when the short lived VM/XA appeared. Its
been a while but doesn't VM/ESA 370 feature still have DMK CP
Modules.
Yes, it did. I can remember that much.
?
?... Mark S.
--
J¦Ðr
|
Re: VM/ESA 370 (was Re: VM/370 new thrilling experience)
On Mon, Nov 11, 2024 at 02:59 AM, Dave Wade wrote:
It would be when the short lived VM/XA appeared. Its been a while but doesn't VM/ESA 370 feature still have DMK CP Modules.
Yes, it did. I can remember that much.
?
?... Mark S.
|
Re: VM/ESA 370 (was Re: VM/370 new thrilling experience)
In the late 90s I "ran" a shop where we had VM/SP 6 running on a 4381(P02?) as it had an AP, so that was configured in CP. It was the late 90s, I was itching to get out of that shop as the environment was dying due to numerous reasons.
?
The concern about running VM after Y2K was the ONLY thing I was allowed to be concerned about. I didn't know anything about XA, and was unwilling to try to bring it up second level and certainly didn't know enough to get it running first level, during my service time slot.
?
VM/ESA 370 Feature would allow that to happen, so I got it up and running second level and switched it to primary when I was ready. That kept the shop up and running over Y2K and I lost track of what they were doing, as I had enough on my hands with my new job, plus my old boss was literally out for my skin (nothing job related other than I visited our director in the hospital after she had a stroke, to say good bye. He heard about it and was pissed because he was VERY protective of her.)
?
So for those sites, who may have had less capable Sys Progs, or were stuck with S/370 architecture. VM/ESA 370 Feature gave us some breathing room for Y2K.
?
?... Mark S.
|
Re: VM/ESA 370 (was Re: VM/370 new thrilling experience)
On Tue, 12 Nov 2024 at 11:36, Drew Derbyshire via <swhobbit= [email protected]> wrote: [...]
On Mon, Nov 11, 2024 at 12:41 AM, William Denton wrote:
I'm not even sure VM/XA CP even had any VM/SP or VM/HPO code. The progression was:
?
CPTOOL (???) ¡ª>
VM/XA Migration Aid (1982?) ¡ª>
VM/XA System Facility (1985?) ¡ª>
VM/XA SP (1988) ¡ª>
VM/ESA ¡ª>
z/VM
?
CPTOOL was internal use only to help get MVS/XA up. I don't know from who, when, or where it came, but time wise it may have predated VM/HPO.
?
I doubt the CPTOOL changes came from the official VM/370 team; the CPTOOL team may have just grabbed the virgin VM/370 source and ran with it.
I was told that the VM/SP HPO folks did make serious progress to run on XA hardware, but the VM/XA official releases all came from the CPTOOL tree.?
Some of this is discussed in Melinda's History of VM paper.? [soon after 1976] "...several of the most knowledgeable CP people who did go to Poughkeepsie, including Dick Newson and Per Jonas, were put into a separate group whose purpose was to build a tool that could be used for the development of MVS/XA. Pete Tallman, of VMA fame, also joined what came to be known as the ¡°VM Tool¡± project as soon as it moved to Poughkeepsie. Starting with VM/370 Release 3, PLC 06, the VM Tool group began building a fast, stripped-down CP that would create XA virtual machines on a real 370 so that the MVS developers could test MVS/XA." and later
"Meanwhile, hidden away in a corner of Poughkeepsie, the VM Tool group, which had grown to a dozen people by 1980, continued to work on building an XA CP system and SIE microcode to support MVS/XA development and testing. By 1977, they had gotten together a system that would support XA guests (in 24-bit addressing mode) on a real S/370 with SIE emulated in software. They first IPLed an XA processor with SIE in microcode in October, 1979. By August, 1980, they had an XA VM system using SIE in production. Although their official purpose was to support MVS development, the group was composed of passionate VM loyalists who very much wanted to create an XA VM for VM customers."
Tony H.
|
Re: VM/ESA 370 (was Re: VM/370 new thrilling experience)
NOTE: Personally, I was fat, dumb, & happy on S/370 hardware until late 1987, which was well into the VM/XA SF era. What follows are what I recall from casual conversations with people shortly after that.? So I wasn't there at the beginning, my memories are 35 years old, and I have no notes.??
?
On Mon, Nov 11, 2024 at 12:41 AM, William Denton wrote:
I'm not even sure VM/XA CP even had any VM/SP or VM/HPO code. The progression was:
?
CPTOOL (???) ¡ª>
VM/XA Migration Aid (1982?) ¡ª>
VM/XA System Facility (1985?) ¡ª>
VM/XA SP (1988) ¡ª>
VM/ESA ¡ª>
z/VM
?
CPTOOL was internal use only to help get MVS/XA up. I don't know from who, when, or where it came, but time wise it may have predated VM/HPO.
?
I doubt the CPTOOL changes came from the official VM/370 team; the CPTOOL team may have just grabbed the virgin VM/370 source and ran with it.
I was told that the VM/SP HPO folks did make serious progress to run on XA hardware, but the VM/XA official releases all came from the CPTOOL tree.??
?
-ahd-
?
p.s. An example of the IBM VM/XA folks going their own way: After DIRECT on VM/SP acquired profiles (with one copy of each profile accessed by all users), DIRECTXA on VM/XA got it too, but its?implementation had the profiles expanded in line. This resulted in a large shop having 25,000 identical copies of the profile in the directory.? They were not happy.??
--
Drew Derbyshire
Software Hobbit (SRE Emeritus)
Kendra Electronic Wonderworks
Kenmore, WA
|
Re: VM/ESA 370 (was Re: VM/370 new thrilling experience)
On 12/11/2024 08:30, William Denton via
groups.io wrote:
One of the main things I really remember about the
DMK-->HCP transition was that IBM abandoned the whole notion
of source code maintenance. HCP started to have OCO modules!!!
Up until that point, even lots of the code was poorly written by
today's standards, at least we had the source code
and we could do stuff with it. Source code based VM in all its
forms was always a collaborative journey with IBM and its
customers.
Whilst this is true, and you may find some of my rants about it in
the vmshare archives at , in practice
for much of CP the source code wasn't of use. Lots of the CP
function was built into the hardware microcode as part of the SIE
instruction. So even if we still had it, it wouldn't have been
terribly useful.
?
Lots of the VM functional evolution happened because of
customer driven change. The EDGAR Editor grew out of the PE
Editor and all gave rise to XEDIT. PROFS was originally
developed by Amaco, BNR, and others and picked up by IBM as one
of the most successful mainframe productivity tools ever. Little
things like TERM MORE and HOLD are direct copies of customer
mods. The list goes on
?
For a glimpse back at how part of this, I would encourage
y'all to find a copy of the LSRAD Report. The hardcopy is out
there someplace. There was also an accompanying 35mm slide deck
that went with the public pitch... don't know if that still
exists. (My shop, Home Savings of America, was one of the five
featured customers in that presentation.)
?
cheers,
William
_._,_._,_
Dave
G4UGM
|
Re: VM/ESA 370 (was Re: VM/370 new thrilling experience)
One of the main things I really remember about the DMK-->HCP transition was that IBM abandoned the whole notion of source code maintenance. HCP started to have OCO modules!!! Up until that point, even lots of the code was poorly written by today's standards, at least we had the source code and we could do stuff with it. Source code based VM in all its forms was always a collaborative journey with IBM and its customers.
?
Lots of the VM functional evolution happened because of customer driven change. The EDGAR Editor grew out of the PE Editor and all gave rise to XEDIT. PROFS was originally developed by Amaco, BNR, and others and picked up by IBM as one of the most successful mainframe productivity tools ever. Little things like TERM MORE and HOLD are direct copies of customer mods. The list goes on
?
For a glimpse back at how part of this, I would encourage y'all to find a copy of the LSRAD Report. The hardcopy is out there someplace. There was also an accompanying 35mm slide deck that went with the public pitch... don't know if that still exists. (My shop, Home Savings of America, was one of the five featured customers in that presentation.)
?
cheers,
William
|
Re: VM/ESA 370 (was Re: VM/370 new thrilling experience)
Hi, Tony,
?
What I recall is, IBM PokLabs was trying hard to "kill off" the whole VM product line around that time.
?
What I heard is this:
Fortunately, Tom Watson Jr. was smart enough to realize what was going on, politically, inside IBM, and so he asked the PokLabs MVS-XA developers if they could create MVS/XA without using the internal version of VM/XA, and of course, they reluctantly admitted that they could not get it done in a timely manner without VM/XA.? ?With that, Tom made an executive decision and told them to stop trying to kill VM, and figure out how to make a VM/XA product to assist IBM's customers wanting to migrate to MVS/XA.
?
That is what the original VM/XA Migration Aid was all about.? No enhancements to CMS at all; it was just the barest minimum to run MVS/XA under VM, and you could run "side-by-side" with MVS/370 in one VM and MVS/XA in a second VM.? ?This was well before the arrival of LPARs.
?
Eventually, it evolved into VM/XA SP, and ultimately, VM/ESA.
?
Another reason for the "HCP" designation (I heard it stood for "Hypervisor Control Program") was that for "XA", IBM had invented the 370-XA mode SIE instruction, that basically had put much of "CP" into the microcode.? So, the new HCP version of "CP" in VM/XA and above was very different from the older versions of VM/370 and VM/SP or HPO that came before on the original 370 hardware (before the arrival of 370-XA mode and the SIE instruction.)
?
Mark S. Waterbury
|
Re: VM/ESA 370 (was Re: VM/370 new thrilling experience)
On 11/11/2024 08:41, William Denton via wrote:
[...]
> I am trying to remember... when did DMK become HCP? I am thinking it
> was before ESA and z but I just don't remember... I sort of tuned out
> when IBM decided to no longer support VM and CMS at the source level.
It would be when the short lived VM/XA appeared.
The first appearance outside IBM of a CP that ran on an XA machine was the VM/XA Migration Aid, sometimes called XA/MA or XAMA, announced in 1981 (the same year as HPO for S/370).?
XA/MA was for running MVS/XA and MVS/370 guests during migration, but had only the barest support for anything CMS related. (Kind of like CMS on zVM today...?) A full-function VM/XA took several more years, and a bimodal CMS a few more (1988).
? Its been a while but doesn't VM/ESA 370 feature still have DMK CP Modules.
So 370 mode == DMK,
XA/ESA/Z == HCP
I'm pretty sure that's right.
Tony H.
|
Re: VM/ESA 370 (was Re: VM/370 new thrilling experience)
VM/HPO as all DMK.
?
HCP was for VM/XA and its follow-up, VM/ESA (the real one), z/VM.
--
Kris Buelens
|
Ren¨¦ Ferland
Jean-Pierre Cabani¨¦
Mark A. Stevens