On Thu, 7 Dec 2017 16:53:22 -0800, Ed Jaffe wrote:
>
>Here is an actual fragment from our code:
>
>       LLGT  R4,TSOPSCB              Get PSCB address
>       DOEXIT LTGR,R4,R4,Z           Exit if no PSCB
>_4    USING PSCB,R4                <Synchronize PSCB
>       DOEXIT CLI,_4.PSCBUSER,LE,C' ' Exit if no User ID
>       LLGT  R15,FLCCVT              Get CVT address
>_15   USING CVTMAP,R15             <Synchronize CVTMAP
>       IF TM,_15.CVTOSLV6,           If z/OS 2.3 or higher
>                       CVTZOS_V2R3,O ..
>         MVC   TSOUSER,_4.PSCBUID8     Save the User ID
>       ELSE  ,                       Else back-level release
>         MVC   TSOUSER,_4.PSCBUSER     Save the User ID
>         MVI   TSOUSER+7,C' '          (same)
>       ENDIF ,                       EndIf
>       DROP  _15                    <Drop CVTMAP
>       DROP  _4                     <Drop PSCB
> 
Will this still cause assembly errors on an older level of z/OS?  (Probably 
doesn't
affect you, but AIF might help.)

I looked at definition of ISPF stats at 2.1.  Apparently back then IBM was
anticipating 8-character User IDs; in extended stats it's padded with a blank.
But should there be a tool to convert stats to extended format so code isn't
forever burdened with a test-and-branch?

What other control block definitions are affected?

-- gil

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

Reply via email to