RE: Re: GCIC Fullscreen applications - Update

Hi Mr. BFN
 
 
IND$FILE is completely unrelated to the editor,
Right about the full screen and IND$FILE
 
 
Can't use the real IND$FILE because that uses DIAG 58 - And the GCIC uses
SIO and interrupts
 
Needed the FullScreen and then an apolication. 
So now vanilla vm/370 has both full screen and IND$FILE.
 
The GCIC application has a "BROWSER" an "EDITOR" and an "FLIST" as major
funtions.
All of the above in FULL SCREEN
 
I just added another Major Map called XFER and from that one you can do
IND$FILE.
Unrelated to the editor except that they both work under GCIC
 
 
 
You can collate, collaborate, but most likely, elaborate. :-)

Elaborate - collaborate is a bit difficult - but if anyone has specific
questions - Most of the answers on how it works are fresh in my mind or on
little bits of paper that the wife will eventually throw out when  I allow
her to get close to my desk.
 
Thre was No help from the IBM world - "THIS DOCUMENT IS  IBM CONFIDENTIAL".

 
But  I wrote a good trace routine and then looking at the X3270.C code was
enough to understand what was going on.
 
The "C" solved some of the trickier problems with the logic - Check sums
etc.
And the trace showed the  exact protocol and the responses from the
mainframe side.
 
At least I can understand my own code so it shoud not be a problem to
collate, collaborate, but most likely, elaborate
 
-----
About the EC MODE and BC MODE
 
I think that can be  very loosely translated to - 360 and 370 -  Under CMS
certain functions work slightly different in EC MODE than IN BC MODE.
Probably one of the CMS gurus can tell us what. Certainly for Guest
operating systems there is a big difference.
 
As a CMS  end user I think that for the most part - it is transparent. 
I know the TIMER behaves differently . (or can behave differently)
I have never had any problem in BC Mode except for some EXEC's that were
dependent on Timer interrupts ,
 
However  the difference between the two is rather significant in the way IO
interrpts are handled. PSWE format is very different
 
IN BC mode the interrupt is returned in the IOPSW along with device causing
the interrupt  - In Ecmode the interrupt and other information is returned
in low core -(I think bytes 128-129) 
Since I handle all the IO interrupts in te program --- (the multitasking
part) I would have to change that to wrok in EC Mode.. Quite a bit of LOGIC
to change . 
Probably the channels that you are trapping interrupts will have to be set
in the Control registers as well instead of the PSW.
All very doable.
Not sure if its worth it.
 
The only user in my system that is defined to come up EC mode is MAINT.
 

 
 
Roc
 



  _____  

From: [email protected] [mailto:[email protected]] On Behalf Of
kerravon86
Sent: Friday, September 03, 2010 15:13 
To: [email protected]
Subject: [H390-VM] Re: GCIC Fullscreen applications - Update


  

--- In [email protected] <mailto:H390-VM%40yahoogroups.com> , "Rocky"
<rocsystems@...> wrote:
>
> 2) Nice to get it to work in EC mode as well as BC Mode.

Is there a point in doing that while CMS is BC mode?

> I have some notes about the specs for IND$FILE that 
> I made while I was
> analysing it.
> If anybody is intereted - I can collaberate. 

You can collate, collaborate, but most likely, elaborate. :-)

IND$FILE is completely unrelated to the editor,
right?

And IND$FILE didn't previously exist for VM/370,
because it needs a fullscreen 3270, right?

Thanks. Paul.