It's been on all the z machines. I don't remember which was the
earliest machine. But I'm sure somebody does and will tell us.

On Wed, May 22, 2013 at 7:57 AM, Robin Atwood <[email protected]> wrote:
> John -
> That's very interesting to know, thanks a lot. Is that a comparatively
> recent register? It would have been useful in the past!
>
> Cheers
> -Robin
>
> -----Original Message-----
> From: IBM Mainframe Discussion List [mailto:[email protected]] On
> Behalf Of John McKown
> Sent: 22 May 2013 20:38
> To: [email protected]
> Subject: Re: Mysterious Abend 0C1
>
> I didn't see mentioned, but if you ever think that the problem is due to a
> "wild branch" of any sort, remember that the BEA Register contains the
> address last successfully "branched from". It is in the SYSUDUMP under the
> PRB of the program which was in execution. So I'd look at the instruction at
> that address and see if it could cause execution to proceed to where the
> S0C1 was reported. If so, you probably have your culprit.
>
> On Wed, May 22, 2013 at 7:28 AM, Tom Marchant <[email protected]>
> wrote:
>> On Wed, 22 May 2013 07:24:01 -0500, Tom Marchant
> <[email protected]> wrote:
>>
>>>On Wed, 22 May 2013 19:20:29 +0800, Robin Atwood wrote:
>>>
>>>>Here is a puzzle. The program has taken an 0C1 half-way through an
>>>>instruction.
>>>
>>>If the PSW that you listed is the correct PSW, the code you have shown
>>>is not the correct code.  You could not have received an operation
>>>exception for trying to execute a X'5A'
>>>opcode.  In fact, unless you are on a _very_ old processor, the only
>>>halfword that shows an invalid opcode is at 00008008.
>>>
>>>>It cannot have got there via the normal instruction sequence, so it
>>>>must have been branched to. However, none of the registers have a
>>>>value less than or equal to the PSW.
>>>
>>>You show registers 1, 2, 3, 5 and 6 as being zero.  And have you
>>>forgotten register 0?  What about relative branches?
>>
>> Please disregard most of the preceding paragraph.  Register zero and
>> the registers containing zero are not important.  Relative branches
>> could still have been used.
>>
>>>
>>>>This is compiled C code so no PC/PR
>>>>instructions are involved, AFAIK. How could the PSW get to be where it
> is?
>>>>Absolute address X'A9E' is all zeros so we didn't go there.
>>>
>>>ITYM virtual address A9E, which would be the same as real address A9E.
>>>Absolute address A9E is something altogether different and is not
>>>relevant.
>>>
>>>--
>>>Tom Marchant
>>>
>>>>
>>>>
>>>>
>>>>
>>>>Using  ASXB519  TCB: 007CC950  Abend Code: 0C1000  ILC: 00  Int: 04
>>>>
>>>>PSW: 000180E100008000      Csect EP:
>>>>
>>>>
>>>>
>>>>      R0            R1       R2       R3       R4       R5       R6
> R7
>>>>
>>>>7FFFF000      00000000 00000000 00000000 00FD6D40 00000000 00000000
> 000A0000
>>>>
>>>>      R8            R9      R10      R11      R12      R13      R14
> R15
>>>>
>>>>000140E1      000A0000 000150E1 000A0000 000160E1 000A0000 000170E1
> 000A0000
>>>>
>>>>
>>>>
>>>>00007FF6 00000016 92D5 7055            MVI    85(7),X'D5'
>>>>
>>>>00007FFA 0000001A 91FF 7056            TM     86(7),X'FF'
>>>>
>>>>00007FFE 0000001E 4770 5A9E            BC     7,2718(0,5)
>>>>
>>>>00008002 00000022 92D5 7056            MVI    86(7),X'D5'
>>>>
>>>>00008006 00000026 41A0 61C8            LA     10,456(0,6)
>>>
>>>----------------------------------------------------------------------
>>>For IBM-MAIN subscribe / signoff / archive access instructions, send
>>>email to [email protected] with the message: INFO IBM-MAIN
>>
>> ----------------------------------------------------------------------
>> For IBM-MAIN subscribe / signoff / archive access instructions, send
>> email to [email protected] with the message: INFO IBM-MAIN
>
>
>
> --
> This is a test of the Emergency Broadcast System. If this had been an actual
> emergency, do you really think we'd stick around to tell you?
>
> Maranatha! <><
> John McKown
>
> ----------------------------------------------------------------------
> For IBM-MAIN subscribe / signoff / archive access instructions, send email
> to [email protected] with the message: INFO IBM-MAIN
>
> ----------------------------------------------------------------------
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to [email protected] with the message: INFO IBM-MAIN



-- 
This is a test of the Emergency Broadcast System. If this had been an
actual emergency, do you really think we'd stick around to tell you?

Maranatha! <><
John McKown

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to