Thanks - that actually makes sense. You lose one bit for the difference between PC versus branch.
On Wed, 30 Jul 2014 01:41:39 -0400 Jim Mulder <d10j...@us.ibm.com> wrote: :>> The PC number is 00180305 :>> :>> Bytes 148-151 show 00080305 :>> :>> The leading 001 is missing.. :> :> In your PC number, bit 44 is one, so you need to read that :>Principles of Operation paragraph very carefully. The leading :>001 is not missing. Bit 44 has been discarded, and the leading :>001 has been shifted 1 bit to the right, so it is transformed into :>0008. To reconstruct the PC number from bytes 148-151, you need to :>first examine bits 1-12 of bytes 148-151. If any of those bits :>are one, then bit 44 must have been one in the PC number, and :>then you reconstruct the rest of the PC number by reversing the :>logic described in the Principles of Operation paragraph :>which explains what is stored in 148-151 when bit 44 of the :>PC number is one. If Bits 1-12 of bytes 148-151 are all zero, :>then the bit 44 was zero in the PC number, and you reconstruct the :>PC number by reversing the logic for when bit 44 is zero in the PC number. :> :> And if I miscounted some bits there, well, it is late at night, and :>Peter Relson can correct me, or we can find the z/OS code to do the :>PC number reconstruction that Peter must have written in z.OS 1.6 :>when he implemented ASN-and-LX Reuse (because in order for RTM2 :>to find an ARR address, we need to reconstruct the PC number :>from the linkage stack entry, so we can use it to locate the :>Entry Table Entry for the PC number, which contains the ARR address). :> :> :>> On Tue, 29 Jul 2014 17:39:44 -0400 Jim Mulder <d10j...@us.ibm.com> :>wrote: :>> :>> :>> The issued PC is 16 bit linkage index (24 bit total with the entry :>> :>number) but :>> :>> the linkage state entry at offset +x'94' (decimal 148) has bits 0-11 :> :>> :>zero. It :>> :>> is described as the "Numeric Part of PC Number". Can the entire :>value be :>> :>> returned from the linkage state entry? :>> :> :>> :> Principles of Operation says: :>> :> :>> :>Numeric Part of PC Number: In a program-call :>> :>state entry, bit positions 1-31 of bytes 148-151 contain :>> :>the numeric part of the PC number used by the :>> :>stacking PROGRAM CALL instruction that formed :>> :>the entry. When ASN-and-LX reuse is not enabled, or :>> :>when it is and bit 44 of the effective address used by :>> :>stacking PROGRAM CALL is zero, stacking PROGRAM :>> :>CALL places bits 44-63 of the effective :>> :>address, with 11 zeros appended on the left, in bit :>> :>positions 1-31 of bytes 148-151. When ASN-and-LX :>> :>reuse is enabled and bit 44 of the effective address is :>> :>one, stacking PROGRAM CALL places bits 45-63 of :>> :>the effective address, with bits 32-43 of the effective :>> :>address appended on the left, in bit positions 1-31 of :>> :>bytes 148-151. In any case, stacking PROGRAM :>> :>CALL places a zero in bit position 0 of the bytes if the :>> :>resulting addressing mode is the 24-bit or 31-bit :>> :>mode or a one in bit position 0 if the resulting :>> :>addressing mode is the 64-bit mode. :> :>Jim Mulder z/OS System Test IBM Corp. Poughkeepsie, NY :> :>---------------------------------------------------------------------- :>For IBM-MAIN subscribe / signoff / archive access instructions, :>send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- 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...@listserv.ua.edu with the message: INFO IBM-MAIN
