Re: Re: Info on the mvs source volumes

________________________________
From: halfmeg <[email protected]>
To: [email protected]
Sent: Sat, October 23, 2010 8:29:16 AM
Subject: [turnkey-mvs] Re: Info on the mvs source volumes

  


> Harold Bell wrote:

> I am thinking about modifying the the MVS3.8j source and have been
> unable to get a handle on what needs to be done to compile MVS.

Is there a specific area you are thinking about?

Installation of 3.8j is done from the distribution library tape which contains 
object or load modules not much source other than macros.  AFAIK there is no 
assembly or link procedure distributed or published which starts with the source 
files as input.

The DLIB tape is at a software level where numerous PTFs were applied by IBM 
before distribution.  The source that we have doesn't necessarily match what is 
running.  The addition of PTFs to the system, most of which were distributed in 
object format without source adds additional mismatches between what is running 
and the source we have.  As if that weren't enough, not all the source was 
included in the distributions various people received and some of what was 
received was truncated in the middle of the file.

To summarize:

We don't have source for all the modules running - some source missing 
altogether.

We don't have source which matches running modules - PTFs applied but no source 
provided.

We don't have source because partial source is all we have - end of source 
member simply missing on the source tape.

> Googling doesn't seem to move me forward so I was wondering if there
> is perhaps an early 1980's book that delves into the OS building
> process or internals I might try to find on ebay or where ever.  Any
> web resources that anyone knows of would be appreciated.  Thanks.
>  Buddy

http://www.bitsavers.org/ has numerous manuals on IBM hardware and software.  
They have several mirrors available.

Phil




I was considering fairly radical action to improve performance.  If the entire 
source were available I was thinking of cross compiling to gcc and redoing the 
i/o structure for linux.  Wild idea.  After the info I have gotten on the state 
of the source I am now leaning towards a maybe wackier idea and modifing 
Hercules to be able to switch between emulated 370 instructions and segments of 
native linux gcc.  If I could get it to work I could recode some of the source 
that is available and leave the 370 stuff without source alone, piece by piece.
Buddy