On Wed, 23 Nov 2005 09:09:00 -0800, Edward E. Jaffe wrote: >Bill Neiman wrote: > >>All broadcast SSI requests, which include WTO, command, and task >>termination processing, run the entire SSCT chain. Having unused >>subsystems defined slows down this process. Only directed SSI requests - >>those sent to a particular named subsystem - use a hash table lookup. > >That's true. But, the additional path length is negligible. Only active >subsystems that capture SVC 35 will actually get control. A "dummy" >subsystem without an SSVT (no INIT routine created one) adds only three >additional instructions. One with an SSVT adds just a few more. No big >deal. We're not talking about thousands of entries here.
No, but we are talking about millions of calls (especially from task termination processing on some software - your software probably isn't task challenged, Ed). Count the WTOs in a syslog for an estimate of the WTO cost per subsystem. Tasks are harder to get a handle on, but there is always SVC sampling (RYO) for ATTACH/DETACH if you are curious. -- Tom Schmidt Madison, WI ---------------------------------------------------------------------- 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
