Victor, Chapter 25 of MVS Programming Authorized Assembler Services Guide describes (such as it is) the use of DYNALLOC as well as some suggested restrictions. There are a few places where they recommend against the use of DYNALLOC with several of those cases being the intersection of DYNALLOC and OPEN due to SYSZTIOT enqueue(s).
As others have said, the changes in allocation over the past (20?) years to use DSABs might mean that the number of current intersections has dropped. Ed seems to be blissfully unhindered by SSI OPEN and DYNALLOC and I'm quite happy to hear that. I believe I had some routines some time back that "mostly" worked just fine, too. (My routines would have been written long enough ago that they may be able to vote in the next election and I haven't used those in a long, long time.) Perhaps IBM needs to update the section entitled "Programming Considerations For Using the DYNALLOC Macro" in Chapter 25, as well as the section entitled "When to Avoid Using Dynamic Allocation" in that same chapter? Maybe that well meaning advice is dated and some of it is obsolete? Or maybe there are still some edge conditions (like maybe using SSI OPEN with DYNALLOC of a base generation data group?) that it is addressing instead of ordinary 'flat files'? I would like to suggest that IBM would be doing the user base a nice service by adding the use of SSI ALLOCATE/OPEN/CLOSE/UNALLOCATE to the "Using the Subsystem Interface" publication. The existing publication dances around the most useful, interesting and valuable part of the SSI. (Maybe they could assign a group of college co-op students to the task? Or better yet, the IBM group that didn't know about the SSI when they designed/implemented the new z/OS data encryption service! That was a lost opportunity if there ever was one.) -- Tom Schmidt Madison, WI On Wed, 3 May 2006 17:00:16 -0500, Victor Gil wrote: >I was thinking of doing dynalloc in the subtask attached by Open. Why? Any >sort of ENQ I should be aware of? I've seen products that parse the SUBSYS >dd and perform dynamic allocation. > >On Wed, 3 May 2006 16:56:14 -0500, Tom Schmidt wrote: >>You aren't planning on (or needing to be) doing dynamic allocation within >>the SSI Open/Close routines, are you? ---------------------------------------------------------------------- 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
