Re: VM: redefine virtual storage from an EXEC

[Alan Altmark]

On Mon, 18 Jul 2022 17:33:26 +, Seymour J Metz  wrote:

>How do CP and CMS handle LINERD  LOGICAL=YES?

LOGICAL=YES applies only to the CMS VSCREEN (fullscreen CMS).   The CMS SET 
LINEND command command affects whether the console handler uses LOGICAL=YES 
(SET LINEND ON) or LOGICAL=NO (SET LINEND OFF).   Applications are free to 
LINERD with either option.

This is what I meant about fullscreen applications setting their own rules.  CP 
doesn't scan the 3270 data stream for any specific content when the guest is 
using DIAGNOSE 0x58 or using native 3270 I/O.  It's up to the app or OS. 

Alan Altmark
Senior Managing z/VM Consultant
IBM Technology Services

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Seymour J Metz]

How do CP and CMS handle LINERD  LOGICAL=YES?


From: IBM Mainframe Discussion List  on behalf of 
Alan Altmark 
Sent: Monday, July 18, 2022 11:12 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: VM: redefine virtual storage from an EXEC

On Thu, 14 Jul 2022 21:57:33 +, Seymour J Metz  wrote:

>The relevant statement is "Which component recognizes the '#' or the '15'x?  
>CP or VM (or XEDIT)?".
> XEDIT can run in either line mode or full screen; in neither case is the CMS 
> command line relevant.

When #CP is used from the line mode command line, the target is CP, not the 
guest.   With a couple of exceptions, # characters are detected by the 3270 
device driver and are converted to 0x15.   Each line is then executed.  The 
guest is never involved.   So the DEFINE STORAGE and IPL succeed.

Without the #CP, the data is split on # characters and each line is queued to 
the guest virtual console with an attention interrupt.  CMS reads that data in 
puts each line in the CMS' internal console input buffer.   On its way back to 
"idle" state, the CMS command processor reads the first buffered line and runs 
it.  The DEFINE STORAGE command is passed to CP (SET IMPCP ON), causing the 
guest to be reset.  The still-queued "IPL" command is never executed.

(Note that # is really the current LINEND character according to the CP 
TERMINAL setting.)

Within XEDIT, Fullscreen CMS, and other fullscreen applications, #CP is a 
programming construct implemented by the application.   Fullscreen CMS sees the 
#CP and hands the entire line to CP via DIAG 8 after converting # to 0x15.  
XEDIT, on the other hand, has no special behavior for #CP.  It  simply breaks 
lines at the linend character and passes them to CMS if they aren't recognized 
(per SET IMPCPCMS).   So what works in fullscreen CMS won't work the same way 
in XEDIT.

Alan Altmark
z/VM Consultant
IBM Technology Services

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Alan Altmark]

On Thu, 14 Jul 2022 21:57:33 +, Seymour J Metz  wrote:

>The relevant statement is "Which component recognizes the '#' or the '15'x?  
>CP or VM (or XEDIT)?". 
> XEDIT can run in either line mode or full screen; in neither case is the CMS 
> command line relevant.

When #CP is used from the line mode command line, the target is CP, not the 
guest.   With a couple of exceptions, # characters are detected by the 3270 
device driver and are converted to 0x15.   Each line is then executed.  The 
guest is never involved.   So the DEFINE STORAGE and IPL succeed.

Without the #CP, the data is split on # characters and each line is queued to 
the guest virtual console with an attention interrupt.  CMS reads that data in 
puts each line in the CMS' internal console input buffer.   On its way back to 
"idle" state, the CMS command processor reads the first buffered line and runs 
it.  The DEFINE STORAGE command is passed to CP (SET IMPCP ON), causing the 
guest to be reset.  The still-queued "IPL" command is never executed.

(Note that # is really the current LINEND character according to the CP 
TERMINAL setting.)

Within XEDIT, Fullscreen CMS, and other fullscreen applications, #CP is a 
programming construct implemented by the application.   Fullscreen CMS sees the 
#CP and hands the entire line to CP via DIAG 8 after converting # to 0x15.  
XEDIT, on the other hand, has no special behavior for #CP.  It  simply breaks 
lines at the linend character and passes them to CMS if they aren't recognized 
(per SET IMPCPCMS).   So what works in fullscreen CMS won't work the same way 
in XEDIT.

Alan Altmark
z/VM Consultant
IBM Technology Services

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Seymour J Metz]

Maybe you don't have control over what LINERD options XEDIT uses, but XEDIT 
does. 

I wonder whether the MODIFY subcommand uses the same options as reading CMS 
commands from the command line. It would seem more useful if it didn't.

From: IBM Mainframe Discussion List  on behalf of 
Phil Smith III 
Sent: Sunday, July 17, 2022 10:10 PM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: VM: redefine virtual storage from an EXEC

Seymour J Metz wrote, re XEDIT reading commands:
>Not so; the behavior depends on the LINERD options.



Sorry, no-XEDIT uses what it uses. You don't have any control over LINERD 
options when using XEDIT. You were asking "Which component
recognizes the '#' or the '15'x?  CP or VM (or XEDIT)?", remember.


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Paul Gilmartin]

On Sun, 17 Jul 2022 22:10:39 -0400, Phil Smith III wrote:

>Seymour J Metz wrote, re XEDIT reading commands:
>>Not so; the behavior depends on the LINERD options.
>
>Sorry, no-XEDIT uses what it uses. You don't have any control over LINERD 
>options when using XEDIT. You were asking "Which component
>recognizes the '#' or the '15'x?  CP or VM (or XEDIT)?", remember.
> 
Is this relevant?:

SET LINENd  Defines a line end character.

Sometimes I use "¾".  It's mostly harmless.

-- 
gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Phil Smith III]

Seymour J Metz wrote, re XEDIT reading commands:
>Not so; the behavior depends on the LINERD options.

 

Sorry, no-XEDIT uses what it uses. You don't have any control over LINERD 
options when using XEDIT. You were asking "Which component
recognizes the '#' or the '15'x?  CP or VM (or XEDIT)?", remember.


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Seymour J Metz]

Not so; the behavior depends on the LINERD options.


From: IBM Mainframe Discussion List  on behalf of 
Phil Smith III 
Sent: Friday, July 15, 2022 5:51 PM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: VM: redefine virtual storage from an EXEC

Seymour J Metz wrote:

> The relevant statement is "Which component recognizes the '#' or the
'15'x?  CP or VM (or XEDIT)?". XEDIT can run in either line mode or full
screen; in neither case is the CMS command line relevant.



Your distinction, while reasonable, turns out to be a distinction without a
difference: in linemode, it's the same command line, just being read by
XEDIT. CMS has two modes: command line and full-screen. Anything that isn't
full-screen is command line. (Yes, full-screen CMS blurred this, but it's
dead, right?)


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Tony Harminc]

On Sat, 16 Jul 2022 at 17:29, Phil Smith III  wrote:
>
> Dave: I stand (well, sit) corrected! I would have sworn I'd read the 
> full-screen CMS was being removed a while ago.

https://www.mxg.com/thebuttonman/html/button880.htm

Tony H.

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Phil Smith III]

Dave: I stand (well, sit) corrected! I would have sworn I'd read the 
full-screen CMS was being removed a while ago.

 

.phsiii ("Senility-not just a river in Africa")


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Dave Jones]

Phil Smith III wrote:

"Your distinction, while reasonable, turns out to be a distinction without a
difference: in linemode, it's the same command line, just being read by
XEDIT. CMS has two modes: command line and full-screen. Anything that isn't
full-screen is command line. (Yes, full-screen CMS blurred this, but it's
dead, right?)"

Baloney, Phil. I use full-screen CMS all the time here. :-) I think its great.
Take care.
DJ

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Phil Smith III]

Seymour J Metz wrote:

> The relevant statement is "Which component recognizes the '#' or the
'15'x?  CP or VM (or XEDIT)?". XEDIT can run in either line mode or full
screen; in neither case is the CMS command line relevant.

 

Your distinction, while reasonable, turns out to be a distinction without a
difference: in linemode, it's the same command line, just being read by
XEDIT. CMS has two modes: command line and full-screen. Anything that isn't
full-screen is command line. (Yes, full-screen CMS blurred this, but it's
dead, right?)


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Seymour J Metz]

The relevant statement is "Which component recognizes the '#' or the '15'x?  CP 
or VM (or XEDIT)?". XEDIT can run in either line mode or full screen; in 
neither case is the CMS command line relevant.


--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3


From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of 
Phil Smith III [li...@akphs.com]
Sent: Thursday, July 14, 2022 3:39 PM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: VM: redefine virtual storage from an EXEC

Seymour J Metz wrote:

>Isn't that true only for line mode?


Yes, as I said, "if you type DEF STOR 16M#IPL on the CMS command line".
Fullscreen environments are not the CMS command line.


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Paul Gilmartin]

On Thu, 14 Jul 2022 15:39:37 -0400, Phil Smith III wrote:

>Seymour J Metz wrote:
>
>>Isn't that true only for line mode?
>
>
>Yes, as I said, "if you type DEF STOR 16M#IPL on the CMS command line".
>Fullscreen environments are not the CMS command line.
> 
And Fullscreen CMS?

When a command is parsed by SVC 202 (DMSCSF?) what indication is there
that the command was issued by a program rather than on the command line?
And is this information passed on to DIAG 8?

My dislike of csh intensified when II learned that "#" (again!) introduces a 
line
comment if it appears in a script but is a command operand when it is typed
on the command line.

And bash has a special treatment of "!" on the command line.  I turn it off
in ~/$ENV.  The "!" behavior is not POSIX, yet it persists when I start bash
with the POSIX option.

Grrr...

-- 
gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Phil Smith III]

Seymour J Metz wrote:

>Isn't that true only for line mode?


Yes, as I said, "if you type DEF STOR 16M#IPL on the CMS command line".
Fullscreen environments are not the CMS command line.


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Seymour J Metz]

Isn't that true only for line mode?


From: IBM Mainframe Discussion List  on behalf of 
Phil Smith III 
Sent: Thursday, July 14, 2022 12:27 PM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: VM: redefine virtual storage from an EXEC

Paul Gilmartin wrote:

>Which component recognizes the '#' or the '15'x?  CP or VM (or XEDIT)?



CP. But if you type DEF STOR 16M#IPL on the CMS command line, then CP splits
it into two lines that CMS sees, so the IPL gets lost. If you do CP DEF STOR
16M#IPL, same thing. If you do #CP DEF STOR 16M#IPL, it works because the
#CP goes directly to CP and CMS never sees the command at all.



>I've generally turned '#' off; it's a pitfall when I enter a command that
contains

>that character.  This astonishes and dismays experts who try to assist me.

Right, perverts :) like you are troublemakers.



>But it's a valid filename character and occurs in some files distributed by
IBM.

Sure. Just as \ is a valid character in a *ix filename (as is * for that
matter).



My usual way to do this from an EXEC is DIAG 8-try this, it won't reIPL but
will prove the point (and then it also works with DEF STOR 16M#IPL):

call diag 8, 'M * HI1' '15'X 'M * HI2'


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Paul Gilmartin]

On Thu, 14 Jul 2022 12:27:49 -0400, Phil Smith III wrote:

>Paul Gilmartin wrote:
>
>>Which component recognizes the '#' or the '15'x?  CP or VM (or XEDIT)?
>
>CP. But if you type DEF STOR 16M#IPL on the CMS command line, then CP splits
>it into two lines that CMS sees, so the IPL gets lost. If you do CP DEF STOR
>
IBM couldn't have done it worse.  Obviously.  Because if they could have, they 
would have.

>16M#IPL, same thing. If you do #CP DEF STOR 16M#IPL, it works because the
>#CP goes directly to CP and CMS never sees the command at all.
> 
I use PA1 to do about the same thing.

>>I've generally turned '#' off; it's a pitfall when I enter a command that
>contains
>
>>that character.  This astonishes and dismays experts who try to assist me.
>
>Right, perverts :) like you are troublemakers.
> 
Long story.  I tried to view a file *distributed* by*IBM* with XEDIT.  My 
PROFILE XEDIT had such as:
ADDRESS CMS
... do some CMS stuff ...
ADDRESS XEDIT
'LOAD' FN FT FM  /* Splits at the "#" in FN.  */

I couldn't turn "#" off because of the implied LOAD.
>
To their credit, IBM took an APAR and made the XEDIT initial setting of
LINEND that from CP TERM.

>>But it's a valid filename character and occurs in some files distributed by 
>>IBM.
>
>Sure. Just as \ is a valid character in a *ix filename (as is * for that 
>matter).
>
but:  598 $ date >"foo \\ *
> bar"; printf '%s\n' foo*
foo \ *
bar
... they  leave me a way around it.  UNIX designers were thoughtful; IBM's 
werem't.

Could I protect it with "#?  I doubt it.

>My usual way to do this from an EXEC is DIAG 8-try this, it won't reIPL but
>will prove the point (and then it also works with DEF STOR 16M#IPL):
>
>call diag 8, 'M * HI1' '15'X 'M * HI2'
> 
If you're not careful, you're left at CP READ with your VM reset.

-- 
gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Phil Smith III]

Paul Gilmartin wrote:

>Which component recognizes the '#' or the '15'x?  CP or VM (or XEDIT)?

 

CP. But if you type DEF STOR 16M#IPL on the CMS command line, then CP splits
it into two lines that CMS sees, so the IPL gets lost. If you do CP DEF STOR
16M#IPL, same thing. If you do #CP DEF STOR 16M#IPL, it works because the
#CP goes directly to CP and CMS never sees the command at all.

 

>I've generally turned '#' off; it's a pitfall when I enter a command that
contains

>that character.  This astonishes and dismays experts who try to assist me.

Right, perverts :) like you are troublemakers.

 

>But it's a valid filename character and occurs in some files distributed by
IBM.

Sure. Just as \ is a valid character in a *ix filename (as is * for that
matter).

 

My usual way to do this from an EXEC is DIAG 8-try this, it won't reIPL but
will prove the point (and then it also works with DEF STOR 16M#IPL):

call diag 8, 'M * HI1' '15'X 'M * HI2'


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Paul Gilmartin]

On Thu, 14 Jul 2022 09:26:45 -0400, Tony Harminc wrote:
>...
>The old-style method would be to put a X'15' byte between the commands
>(in place of your second #). I'm too out of touch with modern VM to
>know if it still works.
>
Which component recognizes the '#' or the '15'x?  CP or VM (or XEDIT)?

I've generally turned '#' off; it's a pitfall when I enter a command that 
contains
that character.  This astonishes and dismays experts who try to assist me.

But it's a valid filename character and occurs in some files distributed by 
IBM..

-- 
gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[David Spiegel]

Hi Binyamin AMV"SH,
Place a x'15' between  commands

Regards,
David

On 2022-07-14 09:16, Binyamin Dissen wrote:

I have not been successfully making an exec re-IPL CM'S after redefining
storage.

If I directly issue

#cp def stor 32m#cp ipl cms parm autocr

it changes the storage size and re-ipls cms.

How would I be able to do this from rexx?

--
Binyamin Dissen 
https://nam12.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.dissensoftware.com%2Fdata=05%7C01%7C%7Ccda0842621c24055338e08da659b2621%7C84df9e7fe9f640afb435%7C1%7C0%7C637934014333517791%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7Csdata=tPzm8GXjdvd9ZyUUXZgD%2BhilhEygdYxu6%2Fa5OqYnU0g%3Dreserved=0

Director, Dissen Software, Bar & Grill - Israel

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Binyamin Dissen]

On Thu, 14 Jul 2022 09:26:45 -0400 Tony Harminc  wrote:

:>On Thu, 14 Jul 2022 at 09:17, Binyamin Dissen
:> wrote:

:>> I have not been successfully making an exec re-IPL CM'S after redefining
:>> storage.
>
:>> If I directly issue

:>>#cp def stor 32m#cp ipl cms parm autocr

:>> it changes the storage size and re-ipls cms.

:>> How would I be able to do this from rexx?

:>The old-style method would be to put a X'15' byte between the commands
:>(in place of your second #). I'm too out of touch with modern VM to
:>know if it still works.

Thanks. Worked like a champ.

--
Binyamin Dissen 
http://www.dissensoftware.com

Director, Dissen Software, Bar & Grill - Israel

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: VM: redefine virtual storage from an EXEC

[Tony Harminc]

On Thu, 14 Jul 2022 at 09:17, Binyamin Dissen
 wrote:
>
> I have not been successfully making an exec re-IPL CM'S after redefining
> storage.
>
> If I directly issue
>
>#cp def stor 32m#cp ipl cms parm autocr
>
> it changes the storage size and re-ipls cms.
>
> How would I be able to do this from rexx?

The old-style method would be to put a X'15' byte between the commands
(in place of your second #). I'm too out of touch with modern VM to
know if it still works.

Tony H.

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

VM: redefine virtual storage from an EXEC

[Binyamin Dissen]

I have not been successfully making an exec re-IPL CM'S after redefining
storage.

If I directly issue

   #cp def stor 32m#cp ipl cms parm autocr   

it changes the storage size and re-ipls cms.

How would I be able to do this from rexx?

--
Binyamin Dissen 
http://www.dissensoftware.com

Director, Dissen Software, Bar & Grill - Israel

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN