On Sun, 18 Dec 2011 01:24:40 -0500 Jim Mulder <d10j...@us.ibm.com> wrote:
:>> :>My understanding, which might be incorrect or incomplete, is that WAIT :>sets :>> :>the wait bits, stores the RB address in the ECBs, sets the wait :>> count in the :>> :>RB and puts the task in a wait state. At that point, isn't WAIT :>finished? :>> :>POST sets the post bit, clears the wait bit, decrements the wait :>> count if it :>> :>is greater than zero, then if it is zero, makes the task dispatchable. :>> :>The ECBs are not reset. :>> That was my point. If the ECBLIST has 5 ECBs and the wait count is 1, :>posting :>> one of the ECBs causes the task to be dispatched without resetting the :>wait :>> bit in the other ECBs. :> Quoting from the MVS/XA SP2.1.0 version of IEAVEPST (5 years :>before it became OCO in MVS/ESA SP3.1.0, so some of you :>probably have this on microfiche that you have squirreled away :>somewhere): :>* WHEN THE WAIT COUNT BECOMES ZERO, THIS CHECKS IF THERE WAS A LIST :>* OF ECB'S BEING WAITED ON BIGGER THAN THE WAIT COUNT. IF SO, THE :>* LIST ADDRESS IS FOUND (GETTING REG 1 FROM WHERE THE RB'S REGISTERS :>* WERE SAVED). ALL OF THE ECB'S WAIT FLAGS (EXCEPT FOR THE ECB BEING :>* POSTED) ARE RESET TO ZERO. :> SPACE 1 :>*/* D (NO,TCBREADY,YES,) RB SEARCH BIT IS ON*/ :> TM RBSTAB2,RBECBWT ECB'S SEARCH FLAG SET :> BZ TCBREADY NO, BRANCH :>*/*LOOP3: P FIND TCB- GET SAVED ECBLIST ADDRESS*/ :> L R10ECBP,TCBGRS1 ASSUME REGS IN TCB :> CL R5RB,TCBRBP ECB'S RB TOP OF QUEUE :> BE RESETWT YES, BRANCH :> L R3WORK,TCBRBP GET ADDRESS OF TOP RB :>* NOTE -- HIGH BYTE OF R3 MUST REMAIN ZERO THROUGH LOOP :>LOOP3 L R10ECBP,RBGRS1-RBSECT(,R3WORK) LOAD RB REG 1 :> ICM R3WORK,M0111,RBLINKB-RBSECT(R3WORK) GET NEXT RB ADDR :>* (HIGH BYTE ALREADY CLEARED) :> CLR R3WORK,R5RB FOUND WAITING RB :> BNE LOOP3 NO, BRANCH :> I could not find anything in the manuals which says this, but :>it has worked this way for at least 30 years. Very interesting. So I have useless logic after ECBLIST waits, resetting the non-posted ECBs to zero. The status will either be posted or zero. -- Binyamin Dissen <bdis...@dissensoftware.com> http://www.dissensoftware.com Director, Dissen Software, Bar & Grill - Israel Should you use the mailblocks package and expect a response from me, you should preauthorize the dissensoftware.com domain. I very rarely bother responding to challenge/response systems, especially those from irresponsible companies. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
