In <[EMAIL PROTECTED]>,
on 12/21/2006
   at 03:15 PM, "McKown, John" <[EMAIL PROTECTED]> said:

>And just to confuse things futher, I cannot see any significant
>difference between a type 2, type 3, or type 4 SVC.

What are the most important three factors in real estate? The
difference is location. Type 2 is in the nucleus, and thus external
symbols in the nucleus will be resolved at IPL time. Types 3 and for
are in the LPA list and external references can only be resolved at
link/bind time.

>In the past, a type 3 SVC was dynamically loaded from SVCLIB into a
> transient area (I think that's what it was called). 

Yes. Likewise type 4.

>This area was fixed in size, so a type 3 could not be larger than
>that (4K?).

1 kiB in OS/360, 2 in OS/VS1.

>The only difference between a type 3 and a type 4 was that a type 4
>could do an XCTL to another "segment" of itself (an independant
>module in SVCLIB).

That was a documentation difference; the Supervisor actually supported
XCTL from types 2, 3 and 4.

>IIRC, the nucleus is still non-pageable (true?).

Aren't there both pageable and nonpageable csects in the nucleus these
days?

>As far as I can see, in today's z/OS environment, everything could
>be architected as a type 1, 3, or 6 SVC with no real impact on
>functionality.

No. You need type 2 to resolve references to other nucleus symbols.

>If I'm blowing smoke, please educate me on the advantages of a type 2
>(other than no paging, I guess) or a type 4.

Over what? The advantage of type 2 over 3 or 4 is as above. The
advantage over type 1 is that you can issue other SVCs.

-- 
     Shmuel (Seymour J.) Metz, SysProg and JOAT
     ISO position; see <http://patriot.net/~shmuel/resume/brief.html> 
We don't care. We don't have to care, we're Congress.
(S877: The Shut up and Eat Your spam act of 2003)

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

Reply via email to