Paul Gilmartin wrote:
On Fri, 2 Apr 2010 07:47:23 -0700, Edward Jaffe wrote:
Exactly! The problem is that GIMSMP is APF authorized.

I've long wondered about this.  Does this mean, in turn, that all
utilities GIMSMP invokes (IEBCOPY, Binder, Assembler, et al.)  must
likewise be authorized?  It is my understanding that an authorized
program ABENDs if it attempts to ATTACH an unauthorized subtask.

An authorized program abends with 306-C if it tries to load a module from an unauthorized library. That's all. There is no requirement that the modules it attaches be linked with AC(1).

AFAIK, that's true only because IEBCOPY requires APF authorization.
That's required because IEBCOPY uses I/O appendages. But, trust me, it's
not doing *anything* that can't also be done without I/O appendages.

Performance?

Maybe in the old days. The channel programs it uses are extremely antiquated. New channel programs, without appendages, would run circles around what they're doing now.

IMHO, the "right" fix would have been to "enhance" IEBCOPY to use
alternate I/O techniques when not running APF authorized. (BTW, that

They're 1/3 of the way there.  Ever do LOOKAT IEB1099I?  I asked IBM
in an RCF to elaborate this; IBM declined expressing an unwillingness
to document, and thereby commit to maintaining the current behavior.

This has been a "thorn" for years. They should do something about it.

--
Edward E Jaffe
Phoenix Software International, Inc
831 Parkview Drive North
El Segundo, CA 90245
310-338-0400 x318
[email protected]
http://www.phoenixsoftware.com/

----------------------------------------------------------------------
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