开云体育

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Kevin Leonard
Sent: Monday, January 4, 2021 5:24 PM
To: [email protected]
Subject: [H390-OSVS1] Announcement: VS1 6.0 DLIBs are available

This announcement was supposed to have gotten posted before the files were, but it didn't work out that way. So...

Exciting news: We've located VS1 Release 6.0 DLIBs, and they function as designed to build a working VS1 system.

It's not an official IBM distribution, but is rather a tape I've built with all the DLIBs on it.

We also have an IBM VS1 6.0 starter system.

It would be really nice if I could give you a prebuilt system. There will be one eventually, but I wanted to make the DLIBs available so you can start playing with generating your own system. I've got a collection of files including the jobs I've been using to ensure that we can create a system, and that collection is posted along with the DLIBs. Ordinarily, what you'd see would be the finished jobs to create the turnkey system. What you'll see here is more of a work in progress; everything worked for me, but you need to verify everything to make sure it will work for you.

To generate a new VS1 system, you have two options. You can use an existing system, perhaps the MVS Turnkey system, as the generating system. Or you can use the VS1 starter system as your generating system. The OS system generation process imposes a peculiar requirement: distribution libraries must have "SYS1" names (there's no way to specify a high-level qualifier for DLIBs), and they must be cataloged (you can't specify a DLIB volser). If you're using an existing OS/VS system as your generating system, you can create a VSAM user catalog that has the DLIBs cataloged in it, and put that catalog as a JOBCAT in your stage 2 jobs. Alternatively, you can use the VS1 starter system as your generating system, and catalog the VS1 DLIBs in its system catalog. More on generation options later.

I've provided and tested generation jobs for each option. Job files with "starter" in the file name set up the environment for using the VS1 starter system and its system catalog. Job files with "nonstarter" will build the DLIB user catalog and insert a DD statement for it in stage 2 jobs. You can do it either way. If you go the "nonstarter" route, your generating system must support VSAM and have a VSAM master catalog (which is guaranteed to be true for every MVS system).

The files are in the "Files" section of the H390-OSVS1 group, in the "VS1 Rel. 6.0" directory. They are:

o zdlf60_vs1-6.0-unloaded-dlibs_2020-10-03.zip
o zdlf60_vs1-6.0-unloaded-dlibs_tapemap.txt

A zipped labeled AWS tape (ZDLF60) of VS1 6.0 DLIBs unloaded with IEBCOPY, and a map of the ZDLF60 tape.

o vs1r60_starter_3330.zip
o vs1r60_starter_3330_tapemap.txt

A zipped unlabeled AWS tape of the IBM VS1 6.0 starter system, and a map of the starter system tape. This is a pretty basic system that doesn't have VSAM or SMF or most other options. The first file on the tape, which you won't need, has IBCDASDI followed by IBCDMPRS. The second file on the tape is an IEHDASDR dump of a 3330 VS1 6.0 sysres volume, which you can restore using IEHDASDR.

o vs160gen_2020-10-06.zip

A zip file containing the jobs that I've been using to test building a system from these DLIBs. The system I've been working on is the one I will eventually use to generate a prebuilt system, hence the name "generating-system". I'll post a more detailed description of the files in this zip file in a bit.

There has been some discussion of making available OS/360 compilers for VS1 use. I've built a tape containing all the OS compilers that IBM says were supported under SVS, along with their subroutine libraries. It's in the "Files" section of the H390-OSVS1 group, in the "Support-Programs" directory:

o zsup2a_sys2link_2020-08-17.aws
o zsup2a_tapemap.txt

An labeled AWS tape (ZSUP2A) of a SYS2.LINKLIB containing OS COBOL, FORTRAN and PL/I compilers and their subroutine libraries, and a tape map of the ZSUP2A tape.

There has also been some discussion of how it might be a good idea to put SMP4 on the VS1 system. That's a future objective. The original tapes did not have the SMP3 ACDS on them, which limits the options for creating an SMP4 ACDS. I've been working on a Perl script to build UCLIN to create an ACDS from IEHLIST LISTPDS listings, and that may produce results in the future.

There are a number of VS1 manuals at Bitsavers. A page with pointers to Bitsavers VS1 manuals is at:



Some user modifications are also under development, including Y2K tolerance. More on them later.

We have found some but not all the source tapes for VS1 6.0. More later on them as well.

--
Kevin

On Monday, January 4, 2021, 04:24:28 PM MST, Kevin Leonard <groups-list@...> wrote:

This announcement was supposed to have gotten posted before the files were, but it didn't work out that way.? So...

Exciting news:? We've located VS1 Release 6.0 DLIBs, and they function as designed to build a working VS1 system.

It's not an official IBM distribution, but is rather a tape I've built with all the DLIBs on it.

We also have an IBM VS1 6.0 starter system.

It would be really nice if I could give you a prebuilt system.? There will be one eventually, but I wanted to make the DLIBs available so you can start playing with generating your own system.? I've got a collection of files including the jobs I've been using to ensure that we can create a system, and that collection is posted along with the DLIBs.? Ordinarily, what you'd see would be the finished jobs to create the turnkey system.? What you'll see here is more of a work in progress; everything worked for me, but you need to verify everything to make sure it will work for you.

To generate a new VS1 system, you have two options.? You can use an existing system, perhaps the MVS Turnkey system, as the generating system.? Or you can use the VS1 starter system as your generating system.? The OS system generation process imposes a peculiar requirement:? distribution libraries must have "SYS1" names (there's no way to specify a high-level qualifier for DLIBs), and they must be cataloged (you can't specify a DLIB volser).? If you're using an existing OS/VS system as your generating system, you can create a VSAM user catalog that has the DLIBs cataloged in it, and put that catalog as a JOBCAT in your stage 2 jobs.? Alternatively, you can use the VS1 starter system as your generating system, and catalog the VS1 DLIBs in its system catalog.? More on generation options later.

I've provided and tested generation jobs for each option.? Job files with "starter" in the file name set up the environment for using the VS1 starter system and its system catalog.? Job files with "nonstarter" will build the DLIB user catalog and insert a DD statement for it in stage 2 jobs.? You can do it either way.? If you go the "nonstarter" route, your generating system must support VSAM and have a VSAM master catalog (which is guaranteed to be true for every MVS system).

The files are in the "Files" section of the H390-OSVS1 group, in the "VS1 Rel. 6.0" directory.? They are:

o? zdlf60_vs1-6.0-unloaded-dlibs_2020-10-03.zip
o? zdlf60_vs1-6.0-unloaded-dlibs_tapemap.txt

A zipped labeled AWS tape (ZDLF60) of VS1 6.0 DLIBs unloaded with IEBCOPY, and a map of the ZDLF60 tape.

o? vs1r60_starter_3330.zip
o? vs1r60_starter_3330_tapemap.txt

A zipped unlabeled AWS tape of the IBM VS1 6.0 starter system, and a map of the starter system tape.? This is a pretty basic system that doesn't have VSAM or SMF or most other options.? The first file on the tape, which you won't need, has IBCDASDI followed by IBCDMPRS.? The second file on the tape is an IEHDASDR dump of a 3330 VS1 6.0 sysres volume, which you can restore using IEHDASDR.

o? vs160gen_2020-10-06.zip

A zip file containing the jobs that I've been using to test building a system from these DLIBs.? The system I've been working on is the one I will eventually use to generate a prebuilt system, hence the name "generating-system".? I'll post a more detailed description of the files in this zip file in a bit.

There has been some discussion of making available OS/360 compilers for VS1 use.? I've built a tape containing all the OS compilers that IBM says were supported under SVS, along with their subroutine libraries.? It's in the "Files" section of the H390-OSVS1 group, in the "Support-Programs" directory:

o? zsup2a_sys2link_2020-08-17.aws
o? zsup2a_tapemap.txt

An labeled AWS tape (ZSUP2A) of a SYS2.LINKLIB containing OS COBOL, FORTRAN and PL/I compilers and their subroutine libraries, and a tape map of the ZSUP2A tape.

There has also been some discussion of how it might be a good idea to put SMP4 on the VS1 system.? That's a future objective.? The original tapes did not have the SMP3 ACDS on them, which limits the options for creating an SMP4 ACDS.? I've been working on a Perl script to build UCLIN to create an ACDS from IEHLIST LISTPDS listings, and that may produce results in the future.

There are a number of VS1 manuals at Bitsavers.? A page with pointers to Bitsavers VS1 manuals is at:

?

Some user modifications are also under development, including Y2K tolerance.? More on them later.

We have found some but not all the source tapes for VS1 6.0.? More later on them as well.

--
Kevin

On Sunday, January 31, 2021, 05:45:55 PM MST, Mark Waterbury <mark.s.waterbury@...> wrote:

Gary,

See this manual:
?

See page 87 (page 91 in Adobe Acrobat PDF viewers).

Did you update the Program Properties Table IEFSDPPT per that documentation?

Hope that helps,

Mark