[H390-MVS] A question about loadparm.

[Gerhard Postpischil [e-mail] [H390-MVS]]

And I'm wondering how difficult would be a usermod for MVS3.8j
to support the loadparm 7 character semantic, the IMSI character.

Way back when we used a SHARE modification that allowed a single
character response to automate the IPL process; you might consider that
as an option.

You might check what Hercules does with the LOADPARM text; if it stores
it in memory, that should be relatively easy to handle (I like the
alternate catalog option for system crash recovery). If it does
something nasty that requires an interface with the service console you
might be out of luck, as that's proprietary.

I guess for MVS3.8j, at least for me, would be enough to support
the IMSI characters S/T.

The standard way to specify an alternate nucleus is to put a hardware
stop at location 80, and modify storage with the 8th digit of the
nucleus name. A similar kludge could be used to enter arbitrary
information during IPL.

If I understand the IMSI description, getting changes to work would also
require modifications to the console task?


Gerhard Postpischil
Bradford, VT

---
This email has been checked for viruses by AVG.
https://www.avg.com

[Laddie Hanus [e-mail] [H390-MVS]]

Loadparm is accessed by the service call instruction. It is in XA and up but I recall it available in 370 on newer processors (those that supported XA and ESA in addition to 370. Those values are only used during IPL and nip (nucleus initialization).

After master scheduler the console definitions from system generation are used. I don’t remember when console definitions were moved to parmlib my guess is XA.

About the only documentation i have seen on service call is in the Hercules source files service.c and service.h the instruction opcode is B220


Laddie Hanus

Sent from whatever device I am using.

https://www.ibm.com/support/knowledgecenter/en/SSLTBW_2.2.0/com.ibm.zos.v2r2.ieag100/a2field.htm
I guess for Z/OS 2.2.
And I'm wondering how difficult would be a usermod for MVS3.8j
to support the loadparm 7 character semantic, the IMSI character.
I guess for MVS3.8j, at least for me, would be enough to support
the IMSI characters S/T.
I hope this is a topic which may be discussed on a public list,
as it is for MVS3.8j actually, not for Z/OS.
Peppe.

[Giuseppe Vitillaro [e-mail] [H390-MVS]]

Post by Gerhard Postpischil ***@charter.net [H390-MVS]

And I'm wondering how difficult would be a usermod for MVS3.8j
to support the loadparm 7 character semantic, the IMSI character.

Way back when we used a SHARE modification that allowed a single
character response to automate the IPL process; you might consider that
as an option.
You might check what Hercules does with the LOADPARM text; if it stores
it in memory, that should be relatively easy to handle (I like the
alternate catalog option for system crash recovery). If it does
something nasty that requires an interface with the service console you
might be out of luck, as that's proprietary.

I guess for MVS3.8j, at least for me, would be enough to support
the IMSI characters S/T.

The standard way to specify an alternate nucleus is to put a hardware
stop at location 80, and modify storage with the 8th digit of the
nucleus name. A similar kludge could be used to enter arbitrary
information during IPL.
If I understand the IMSI description, getting changes to work would also
require modifications to the console task?

I'm not begging for a full IMSI implementation, Gerhard.

What I looking for is actually only the "S" feature:

IMSI Display Inf.MSG Prompt Master Catalog Prompt System Parameter
A Yes Yes Yes
C No Yes No
D Yes Yes No
M Yes No No
P No Yes Yes
S No No Yes
T Yes No Yes

of this table, for MVS3.8j, which shdouldn't requires any console task
modification, isn't it?

What I've in mind is a "ZERO" default which allows MVS3.8j to IPL WITHOUT
asking any System Parameter and the "S" character which ask MVS3.8j to
behave in the very same way it actually do now, at IPL time:

IEA101A SPECIFY SYSTEM PARAMETERS FOR RELEASE 03.8 .VS2

with a loadparm support for an autmatic choice of the "sysp=xx" parameter,
in the very same way Z/OS looks to loadparm for the IPLPARM/PARMLIB LOADxx
PDS member.

Something like

LOADPARM ppx

where "pp" is the sysp value and "x" the IMSI character.

I understand is more an estetic feature that a really functional
usermod. The hercules HAO automatic operator can easily handle
an automatic IPL, but, still, for example under VM/370, it may
actually be a useful feature.

Once understood where the loadparm string is stored, it should
be just a question of a simple "if-then-else" branch
somewhere, in the MVS3.8j IPL code, isn't it?

The other features are actually more complex functionalities,
which may be even more useful, but, I guess, would requires
more massive changes in many sources of the MVS code.

Peppe.

[Gerhard Postpischil [e-mail] [H390-MVS]]

Post by Giuseppe Vitillaro ***@vitillaro.org [H390-MVS]
IMSI Display Inf.MSG Prompt Master Catalog Prompt System Parameter
A Yes Yes Yes
C No Yes No
D Yes Yes No
M Yes No No
P No Yes Yes
S No No Yes
T Yes No Yes
of this table, for MVS3.8j, which shdouldn't requires any console task
modification, isn't it?

I'm not certain what the "display informational messages" entails; if it
refers to NIP related information, then it could be handled. But if it
refers to console options, you would need to pass the request to console
initialization or find some way to override the options.

Post by Giuseppe Vitillaro ***@vitillaro.org [H390-MVS]
Once understood where the loadparm string is stored, it should
be just a question of a simple "if-then-else" branch
somewhere, in the MVS3.8j IPL code, isn't it?

If the other response is correct, you should be able to issue a B220
instruction, and that might require a minor change to Hercules for the
370 instruction set. And you could define the 8 character string as
anything you want.

Gerhard Postpischil
Bradford, VT

---
This email has been checked for viruses by AVG.
https://www.avg.com

[Gerhard Postpischil [e-mail] [H390-MVS]]

Understood Gerhard. That change for B220 instruction 370 support
seems easy, I think I know what has to be changed in spinhawk.
1) how the B220 instruction works, is documented somewhere?

I don't know of any documentation, other than Hercules code, but I would
try DC X'B220',SL2(DB) - give it an aligned double word, and see what
happens.

2) how to change the MVS nucleus, do we have the sources?

Some of the components are on the (tk optional) disks SRC000-SRC002:

MVSSRC.SYM101.F12
IEAVNP13 000201
IEAVNP14 001401
MVSSRC.SYM101.F17
IEAVNPS5 000201
MVSSRC.SYM101.F18
IEAVNP08 006701
MVSSRC.SYM101.F19
IEAVNPA5 002D01
IEAVNP05 003D01
MVSSRC.SYM101.F20
IEAVNPA1 007201
IEAVNP01 009001
MVSSRC.SYM102.F03
IEAVNPA6 015601
MVSSRC.SYM102.F09
IEAVNPB5 00AB01
MVSSRC.SYM102.F16
IEAVNP10 000201
MVSSRC.SYM102.F19
IEAVNP02 000201
MVSSRC.SYM103.F01
IEAVNP07 000201
MVSSRC.SYM103.F02
IEAVNPB6 012803
IEAVNP09 012901
MVSSRC.SYM103.F04
IEAVNIPM 000201
IEAVNIPX 002E01
IEAVNIP0 004801
IEAVNPM2 008E01
IEAVNPM3 00AA01
IEAVNP03 00C301
MVSSRC.SYM103.F20
IEAVNP17 011801
MVSSRC.SYM601.F09
IEAVNP1A 003901
IEAVNP1B 005E01
IEAVNP11 006201
IEAVNP12 007601
MVSSRC.SYM602.F07
IEAVNP16 000201
MVSSRC.SYM701.F05
IEAVNP19 00FE01

Also Paul has a dozen or so data sets with the common prefix of
D.DEV.ALL on his pack, containing disassembled code. Also JCL.

Gerhard Postpischil
Bradford, VT

---
This email has been checked for viruses by AVG.
https://www.avg.com

[Giuseppe Vitillaro [e-mail] [H390-MVS]]

Post by Gerhard Postpischil ***@charter.net [H390-MVS]

Understood Gerhard. That change for B220 instruction 370 support
seems easy, I think I know what has to be changed in spinhawk.
1) how the B220 instruction works, is documented somewhere?

I don't know of any documentation, other than Hercules code, but I would
try DC X'B220',SL2(DB) - give it an aligned double word, and see what
happens.

Damn, if even "you" don't know how it works ...
well ... nobody even tried to use B220 for the 370
arch up to now ... ;-)

Thanks Gerhard, Peppe.

[Harold Grovesteen [e-mail] [H390-MVS]]

I would not be comfortable adding B220 to the S/370 instruction table
without an analysis of which aspects of the instruction should be
disabled under S/370 and the information returned when executed under
370 mode.

If the desire is to simply access the loadparm information, I would
recommend simply adding a Hercules specific DIAGNOSE code that returns
the loadparm data in a register pair or the two registers as specified
by the DIAGNOSE instruction.  These have been codes starting at
X'F00'. 

If such a DIAGNOSE code already exists in VM, although I did not find
one, then that would be the preferred choice.

Much simpler and less intrusive into Hercules itself.  Of course if I
were doing it, it would be in Hyperion and not spinhawk.

Harold Grovesteen

On Tue, 2018-09-18 at 11:52 +0200, Giuseppe Vitillaro

https://www.ibm.com/support/knowledgecenter/en/SSLTBW_2.2.0/com.ibm.z
os.v2r2.ieag100/a2field.htm
I guess for Z/OS 2.2.
And I'm wondering how difficult would be a usermod for MVS3.8j
to support the loadparm 7 character semantic, the IMSI character.
I guess for MVS3.8j, at least for me, would be enough to support
the IMSI characters S/T.
I hope this is a topic which may be discussed on a public list,
as it is for MVS3.8j actually, not for Z/OS.
Peppe.
 

[Mike Stramba [e-mail] [H390-MVS]]

http://hercules-390.yahoogroups.narkive.com/rTC7neOC/loadparm-bare-hercules-and-mvs

laddiehanus
8 years ago

Service call example

"This from a standalone program that runs DAT off"

L R2,SCPINFO Load in READ_SCP_INFO command word
L R4,SCPINAD Address of SCCB is at loc X'2000'
LHI R5,4096 Length of SCCB (4K max)
STH R5,0(R4) Save length in SCCB
XC 2(6,R4),6(R4) Clear rest of header
MVI 2(R4),X'80' Indicate Syncronous request
MVI 5(R4),X'80' Indicate variable request
DC X'B2200024'
..
..
..
SCPINFO DC X'00020001' READ_SCP_INFO Command Word
SCPINAD DC A(8192) SCCB located at X'2000'


memory display after the servc has executed

r 2000
R:00002000:K:06=10008000 00800010 04000104 00010000 ................
R:00002010:K:06=00200080 000002BF F0F3F4C5 F0F0D4C9 ........034E00MI
R:00002020:K:06=00000000 00000100 00000000 00000000 ................
R:00002030:K:06=C4040840 00000000 00000000 00000000 D.. ............

The load parm is at 2018

Refer to Hercules code service.c service.h

z/OS 1.11 data areas vol 5 look for SCCB

Laddie Hanus

Post by Harold Grovesteen ***@tx.rr.com [H390-MVS]
I would not be comfortable adding B220 to the S/370 instruction table
without an analysis of which aspects of the instruction should be
disabled under S/370 and the information returned when executed under
370 mode.
If the desire is to simply access the loadparm information, I would
recommend simply adding a Hercules specific DIAGNOSE code that returns
the loadparm data in a register pair or the two registers as specified
by the DIAGNOSE instruction.  These have been codes starting at
X'F00'.
If such a DIAGNOSE code already exists in VM, although I did not find
one, then that would be the preferred choice.
Much simpler and less intrusive into Hercules itself.  Of course if I
were doing it, it would be in Hyperion and not spinhawk.
Harold Grovesteen
On Tue, 2018-09-18 at 11:52 +0200, Giuseppe Vitillaro

https://www.ibm.com/support/knowledgecenter/en/SSLTBW_2.2.0/com.ibm.z
os.v2r2.ieag100/a2field.htm
I guess for Z/OS 2.2.
And I'm wondering how difficult would be a usermod for MVS3.8j
to support the loadparm 7 character semantic, the IMSI character.
I guess for MVS3.8j, at least for me, would be enough to support
the IMSI characters S/T.
I hope this is a topic which may be discussed on a public list,
as it is for MVS3.8j actually, not for Z/OS.
Peppe.

------------------------------------
------------------------------------
------------------------------------
Yahoo Groups Links

[João Reginato [e-mail] [H390-MVS]]

Should it be
XC 2(6,R4),2(R4) Clear rest of header


Not
XC 2(6,R4),6(R4) Clear rest of header



De: H390-***@yahoogroups.com [mailto:H390-***@yahoogroups.com]
Enviada em: sexta-feira, 21 de setembro de 2018 17:00
Para: H390-***@yahoogroups.com
Assunto: Re: [H390-MVS] A question about loadparm.


<http://hercules-390.yahoogroups.narkive.com/rTC7neOC/loadparm-bare-hercules-and-mvs> http://hercules-390.yahoogroups.narkive.com/rTC7neOC/loadparm-bare-hercules-and-mvs

laddiehanus
8 years ago

Service call example

"This from a standalone program that runs DAT off"

L R2,SCPINFO Load in READ_SCP_INFO command word
L R4,SCPINAD Address of SCCB is at loc X'2000'
LHI R5,4096 Length of SCCB (4K max)
STH R5,0(R4) Save length in SCCB
XC 2(6,R4),6(R4) Clear rest of header
MVI 2(R4),X'80' Indicate Syncronous request
MVI 5(R4),X'80' Indicate variable request
DC X'B2200024'
...
...
...
SCPINFO DC X'00020001' READ_SCP_INFO Command Word
SCPINAD DC A(8192) SCCB located at X'2000'

memory display after the servc has executed

r 2000
R:00002000:K:06=10008000 00800010 04000104 00010000 ................
R:00002010:K:06=00200080 000002BF F0F3F4C5 F0F0D4C9 ........034E00MI
R:00002020:K:06=00000000 00000100 00000000 00000000 ................
R:00002030:K:06=C4040840 00000000 00000000 00000000 D.. ............

The load parm is at 2018

Refer to Hercules code service.c service.h

z/OS 1.11 data areas vol 5 look for SCCB

Laddie Hanus

Post by Harold Grovesteen ***@tx.rr.com [H390-MVS]
I would not be comfortable adding B220 to the S/370 instruction table
without an analysis of which aspects of the instruction should be
disabled under S/370 and the information returned when executed under
370 mode.
If the desire is to simply access the loadparm information, I would
recommend simply adding a Hercules specific DIAGNOSE code that returns
the loadparm data in a register pair or the two registers as specified
by the DIAGNOSE instruction. These have been codes starting at
X'F00'.
If such a DIAGNOSE code already exists in VM, although I did not find
one, then that would be the preferred choice.
Much simpler and less intrusive into Hercules itself. Of course if I
were doing it, it would be in Hyperion and not spinhawk.
Harold Grovesteen
On Tue, 2018-09-18 at 11:52 +0200, Giuseppe Vitillaro

<https://www.ibm.com/support/knowledgecenter/en/SSLTBW_2.2.0/com.ibm.z> https://www.ibm.com/support/knowledgecenter/en/SSLTBW_2.2.0/com.ibm.z
os.v2r2.ieag100/a2field.htm
I guess for Z/OS 2.2.
And I'm wondering how difficult would be a usermod for MVS3.8j
to support the loadparm 7 character semantic, the IMSI character.
I guess for MVS3.8j, at least for me, would be enough to support
the IMSI characters S/T.
I hope this is a topic which may be discussed on a public list,
as it is for MVS3.8j actually, not for Z/OS.
Peppe.

------------------------------------
------------------------------------
------------------------------------
Yahoo Groups Links

[Giuseppe Vitillaro [e-mail] [H390-MVS]]

Post by Mike Stramba ***@gmail.com [H390-MVS]
http://hercules-390.yahoogroups.narkive.com/rTC7neOC/loadparm-bare-hercules-and-mvs
laddiehanus
8 years ago
Service call example
"This from a standalone program that runs DAT off"
L R2,SCPINFO Load in READ_SCP_INFO command word
L R4,SCPINAD Address of SCCB is at loc X'2000'
LHI R5,4096 Length of SCCB (4K max)
STH R5,0(R4) Save length in SCCB
XC 2(6,R4),6(R4) Clear rest of header
MVI 2(R4),X'80' Indicate Syncronous request
MVI 5(R4),X'80' Indicate variable request
DC X'B2200024'
..
..
..
SCPINFO DC X'00020001' READ_SCP_INFO Command Word
SCPINAD DC A(8192) SCCB located at X'2000'
memory display after the servc has executed
r 2000
R:00002000:K:06=10008000 00800010 04000104 00010000 ................
R:00002010:K:06=00200080 000002BF F0F3F4C5 F0F0D4C9 ........034E00MI
R:00002020:K:06=00000000 00000100 00000000 00000000 ................
R:00002030:K:06=C4040840 00000000 00000000 00000000 D.. ............
The load parm is at 2018
Refer to Hercules code service.c service.h
z/OS 1.11 data areas vol 5 look for SCCB
Laddie Hanus

Nice example! I guess the NUCLEUS, at the stage
where loadparm has to be checked may be considered
a "standalone program" with DAT off?

Peppe.

[[e-mail] [H390-MVS]]

Just what are you trying to accomplish? Or did you see something in a manual that you thought would be nice to have. What you want out of Loadparm exactly?



Do you know how MVS lPL and NIP work? From your question I don't think so. BTW starting with MVS/XA IPL was totally rewritten.

IPL - Initial Program Loader. Its main function is to load the nucleus into real storage.

NIP - Nucleus Initialization Program. It does all the initialization to get MVS to the point that ASID 1 (the Master Scheduler can run). At that point all the facilities for normal running are available. The master scheduler gets the consoles up and running and get JES2 (or JES3) running. Then normal processing takes over.



The issues with making any changes are:
No one has found a copy of MVS System Initialization Logic.

Most of MVS was written in a language called PL/S what we have as source code Is actually output of that complier. IBM did not generally make it available and recalled the few copies that made it out of IBM, Any way we dont have the PL/S input, The output of the compiler has the following issues:

1. Lot of the comments were taken out, especially the ones at the start of the module where change logs and logic doc were,
2. All macros were expanded especially the macros that were data areas. Offsets and lengths of data areas are essentially hard coded. For example the macro that describes page table entries. It is in 370 a 2 byte field. In XA is a 4 byte field. So you change it to 4 bytes and move the bit fields as needed. If you want to reassemble the source code that manipulates those bits the macro you changed will be ignored.

3. most of the code is actually at the MVS 3.7 level and does not match the running system as there are 5+ years of PTFs along with the changes from MVS3.8A to MVS3.8J. Getting the code to state where it can be changed is a fairly laborious process involving disassemblers and finding the differences between different versions of the source.



So lets look at LOADPARM. Granted the content is operating system specific. In z/OS MVS the first 4 bytes are the unitaddress of the IODF volume and the next 2 bytes of the LOADXX is the suffix of the LOADXX parmlib member that has some initial parms. Neither applies in S/370 refer to the XA or later documentation for details.



Byte 7 is where you want the S and T parameters supported. Are informational messages such a problem that you want to suppress them. I see very few if any and the screen clears quickly and the console task takes over message processing which does not use this. A subsystem to process messages would be the correct way to go, or maybe the wto exit IEECVXIT can handle this. I havent used it in years as MPF provides for message handling in recent years. I am pretty sure that it is in OS/390 and newer may go back to XA. Refer to system initialization and tuning for a newer system for the MPFLSTxx parameter. If you want to bypass the IEA101A prompt maybe a Hercules Automatic Operator rule would be good and a lot less work.



Byte 8 is the alternate Nucleus prompt. If you need to ipl an alternate Nucleus there is a procedure to do that.


Also there is no such thing as a standalone program during IPL/NIP It has to be explicitly called. But the SCCB area in the b220 instruction has to be a real address. Dat is turned on in NIP byte early in the process. Then there is the issue of saving the data you want somewhere where it is not overridden.



Laddie Hanus

[Giuseppe Vitillaro [e-mail] [H390-MVS]]

Post by ***@yahoo.com [H390-MVS]
Just what are you trying to accomplish? Or did you see something in a manual that you thought would be nice to have. What you want out of Loadparm exactly?

I wrote what I wish, the "requirement", few posts above,
it looks quite a simple requirement.

You are correct: I haven't the "know how" for the changes
needed, otherwise I wouldn't have asked, I would have
just changed the way MVS3.8j nucleus looks for "sysp=xx",
really my single wish.

Any way, beside practical purpouses, it is interesting
to understand how loadparm works.

Curiosity almost always pay, on the long run ;-)

Peppe.

[[e-mail] [H390-MVS]]

---In H390-***@yahoogroups.com, <***@...> wrote :

The issues with making any changes are: No one has found a copy of MVS System Initialization Logic.


Laddie Hanus
Are you referring to an S/370 IPL logic book or any version ? For the latter, I have the hardcopy version of the ESA System Initialization Logic manual. It's about four inches thick and very informative. I have a lot of XA and ESA system manuals that I saved over the years. I always made sure I never lost my HSM logic manual, I considered it to be worth its weight in gold.

[[e-mail] [H390-MVS]]

I went and got the book to make sure it's what you were talking about. Mine is LY28-1847-1
MVS/ESA Component Diagnosis and Logic: System Initialization


I have a bunch of these CD & L manuals for ESA, and a few for XA. I didn't realize at first how valuable they would be in the future. I sure wish I had saved some of the microfiche.


---In H390-***@yahoogroups.com, <***@...> wrote :




---In H390-***@yahoogroups.com, <***@...> wrote :

The issues with making any changes are: No one has found a copy of MVS System Initialization Logic.


Laddie Hanus
Are you referring to an S/370 IPL logic book or any version ? For the latter, I have the hardcopy version of the ESA System Initialization Logic manual. It's about four inches thick and very informative. I have a lot of XA and ESA system manuals that I saved over the years. I always made sure I never lost my HSM logic manual, I considered it to be worth its weight in gold.

['M. Khalid Khan' [e-mail] [H390-MVS]]

Would you be able to scan them and make them generally available?
Khalid

Post by ***@sympatico.ca [H390-MVS]
I went and got the book to make sure it's what you were talking about. Mine is LY28-1847-1
MVS/ESA Component Diagnosis and Logic: System Initialization
I have a bunch of these CD & L manuals for ESA, and a few for XA. I
didn't realize at first how valuable they would be in the future. I sure
wish I had >saved some of the microfiche.
No one has found a copy of MVS System Initialization Logic.
Laddie Hanus
Are you referring to an S/370 IPL logic book or any version ? For the
latter, I have the hardcopy version of the ESA System Initialization
Logic >manual. It's about four inches thick and very informative. I have
a lot of XA and ESA system manuals that I saved over the years. I always
made sure I >never lost my HSM logic manual, I considered it to be worth
its weight in gold.

[Laddie Hanus [e-mail] [H390-MVS]]

I have read that manual. The issue is that IPL was completely rewritten for MVS/XA and it does a lot more than our IEAIPL00 in MVS/370.

It’s worth hanging on to but a number of Object Code only code is missing it’s not even in the index. The one I was looking for at the time was dual address space and pc call setup logic.

Laddie

Sent from whatever device I am using.

Post by ***@sympatico.ca [H390-MVS]
I went and got the book to make sure it's what you were talking about. Mine is LY28-1847-1
MVS/ESA Component Diagnosis and Logic: System Initialization
I have a bunch of these CD & L manuals for ESA, and a few for XA. I didn't realize at first how valuable they would be in the future. I sure wish I had saved some of the microfiche.
No one has found a copy of MVS System Initialization Logic.
Laddie Hanus
Are you referring to an S/370 IPL logic book or any version ? For the latter, I have the hardcopy version of the ESA System Initialization Logic manual. It's about four inches thick and very informative. I have a lot of XA and ESA system manuals that I saved over the years. I always made sure I never lost my HSM logic manual, I considered it to be worth its weight in gold.

[[e-mail] [H390-MVS]]

Well I can have a look in some of the other logics, see if there's anything there. Like I said I have most of them for ESA, and a few XA ones. Unfortunately I only have the source code to the ESA IEAIPL00 itself, I don't have any of the IRIMS or NIP.


---In H390-***@yahoogroups.com, <***@...> wrote :

I have read that manual. The issue is that IPL was completely rewritten for MVS/XA and it does a lot more than our IEAIPL00 in MVS/370.

It’s worth hanging on to but a number of Object Code only code is missing it’s not even in the index. The one I was looking for at the time was dual address space and pc call setup logic.


Laddie

Sent from whatever device I am using.

On Sep 24, 2018, at 7:55 PM, ***@... mailto:***@... [H390-MVS] <H390-***@yahoogroups.com mailto:H390-***@yahoogroups.com> wrote:



I went and got the book to make sure it's what you were talking about. Mine is LY28-1847-1
MVS/ESA Component Diagnosis and Logic: System Initialization


I have a bunch of these CD & L manuals for ESA, and a few for XA. I didn't realize at first how valuable they would be in the future. I sure wish I had saved some of the microfiche.


---In H390-***@yahoogroups.com mailto:H390-***@yahoogroups.com, <***@...> wrote :




---In H390-***@yahoogroups.com mailto:H390-***@yahoogroups.com, <***@...> wrote :

The issues with making any changes are: No one has found a copy of MVS System Initialization Logic.


Laddie Hanus
Are you referring to an S/370 IPL logic book or any version ? For the latter, I have the hardcopy version of the ESA System Initialization Logic manual. It's about four inches thick and very informative. I have a lot of XA and ESA system manuals that I saved over the years. I always made sure I never lost my HSM logic manual, I considered it to be worth its weight in gold.

[[e-mail] [H390-MVS]]

So I actually have the C&D for ESA PC/AUTH but it doesn't say anything about initialization; whole thing is only about 20 pages long. Not even a mention of IEAVXMAS. It makes me crazy when I think of all the micro-fiche we threw out then. Who knew it would be so valuable 40 years later.


---In h390-***@yahoogroups.com, <***@...> wrote :

I have read that manual. The issue is that IPL was completely rewritten for MVS/XA and it does a lot more than our IEAIPL00 in MVS/370.

It’s worth hanging on to but a number of Object Code only code is missing it’s not even in the index. The one I was looking for at the time was dual address space and pc call setup logic.


Laddie

['\'Fish\' (David B. Trout)' [e-mail] [H390-MVS]]

FYI: those interested might want to bookmark the following page:

http://www.cbttape.org/~jjaeger/opcodes.html

http://www.cbttape.org/~jjaeger/opcodes.html#servc


HTH

['\'Fish\' (David B. Trout)' [e-mail] [H390-MVS]]

FYI: those interested might also be interested in the following manual (especially pages 505-510):


"GA32-0939-02 zOS V2R1 - MVS Data Areas - Volume 5 (LDA - SJRSP)"

http://publibfp.dhe.ibm.com/epubs/pdf/iea3d502.pdf



(It looks like IBM finally fixed their web server!)