IBM Mainframe Discussion List <IBM-MAIN@BAMA.UA.EDU> wrote on 11/27/2007 11:07:20 PM:
> Jim, > > Sorry for my off-topic question. > > What does it mean when we say 'interrupt handling completes'? > > From what you said about SRB, I guess it means FLIH have done its work and > scheduled SLIH. And that's why normally SRB doesn't need a save area: FLIH > will directly give back control to SRB so there is no need to move > reg/PSW/etc from PSA and LCCA under such a situation. (FLIH is running > disabled. ) > > In the past I assumed that interrupt handling completes when SLIH has > finished execution. But this understanding cannot explain many things. > > On Nov 28, 2007 10:39 AM, Jim Mulder <[EMAIL PROTECTED]> wrote: > > > IBM Mainframe Discussion List <IBM-MAIN@BAMA.UA.EDU> wrote on 11/27/2007 > > 04:35:36 PM: > > > > And until SUSPEND/RESUME for > > SRBs in SP 4.1.0, there was no external interface for waiting in > > an SRB - they only time they waited was due to lock suspends and > > page faults. So SRB creation does not allocate areas for > > saving registers/PSW/etc. They are allocated when the SRB > > has to wait. > > A FLIH calls (not schedules) a SLIH. The SLIH returns to the FLIH. The FLIH returns to the interrupted work unit, or goes to the dispatcher, or goes to RTM to abend the interrupted work unit. That is what I would usually refer to as 'interrupt handling completes'. Jim Mulder z/OS System Test IBM Corp. Poughkeepsie, NY ---------------------------------------------------------------------- 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
