On Wed, 2 Apr 2014 11:58:35 +0300, Binyamin Dissen wrote:
>Sorry for the typo.
>
>Look at your BPXWDYN error messages - look at the S99MSG stem
>
There were no error messages.
>:>>You need to specify the PERM[C] attribute, otherwise the files are
>temporarily
>:>>concatenated and can split.
>:>>
Here's a test case and its output. I'd welcome any analysis:
/* **************************************** */
/* Rexx */ signal on novalue /*
Doc: determine DDNAME leakage.
*/
trace C
call BPXWDYN 'alloc dummy msg(2)'
call BPXWDYN 'alloc dd(SYSLIB) shr dsn(SYS1.MACLIB) msg(2)'
call BPXWDYN 'alloc rtddn(CAT1) shr dsn(SYS1.MODGEN) msg(2)'
say; say 'Allocated:'
call DDinfo
call BPXWDYN 'concat ddlist(SYSLIB,'CAT1') msg(2)'
say; say 'Catenated:'
call DDinfo
call BPXWDYN 'free dd(SYSLIB) msg(2)'
say; say 'Freed; note that' CAT1 'is not freed:'
call DDinfo
call BPXWDYN 'alloc dd(SYSLIB) shr dsn(SYS1.MACLIB) msg(2)'
call BPXWDYN 'concat ddlist(SYSLIB,'CAT1') permc msg(2)'
say; say 'Catenated, with PERMC; note that' CAT1 'is not catenated:'
call DDinfo
return( 0 )
DDinfo:
do I = 1 until last<>0
Call bpxwdyn 'info inrelno('I')' ,
'inrtddn(ddn) inrtdsn(dsn) inrtlst(last)'
say I ddn copies( '(catenated)', ddn='') dsn; end I
return( 0 )
/* **************************************** */
user@HOST: catdd
Allocated:
1 SYS00001 NULLFILE
2 SYSLIB SYS1.MACLIB
3 SYS00002 SYS1.MODGEN
Catenated:
1 SYS00001 NULLFILE
2 SYSLIB SYS1.MACLIB
3 (catenated) SYS1.MODGEN
Freed; note that SYS00002 is not freed:
1 SYS00001 NULLFILE
2 SYS00002 SYS1.MODGEN
Catenated, with PERMC; note that SYS00002 is not catenated:
1 SYS00001 NULLFILE
2 SYS00002 SYS1.MODGEN
3 SYSLIB SYS1.MACLIB
user@HOST:
/* **************************************** */
Thanks,
gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
