On Sat, 1 Dec 2018 20:46:19 +0000, Robert Prins wrote: >On 2018-11-30 12:33, John Eells wrote: >> >> IND$FILE, as part of z/OS, remains supported. As far as I know, support for >> it >> has never lapsed (certainly not since its inclusion into OS/390). > >If it's supported, then would you be willing to make a small change to it, to >use SDB, rather than whatever hardcoded blocksize it's using now. FWIW, I >zapped >it, and I believe there there is a list of locations to zap for various (older) >versions around. > One for each version, or several for each? If the latter, I hope it's an EQU where a one-line change suffices.
Sounds like an RFE. Or has IBM a blanket policy of replacing hardcoded blocksize with SDB? IIRC, IBM changed IEBGENER to rely on SDB, then needed to backtrack because customers depended on the prior behavior. And IBM's changing Rexx EXECIO to rely on SDB broke some of my EXECs because SDB chose an invalid blocksize in some cases where the Rexx runtime had chosen a valid one. SR suggested an existing PTF, intended for JES, that solved my problem. But then cautioned, "That PTF causes Rexx to behave contrary to specification, and the adverse behavior might return if a customer reported a problem. You should always specify BLKSIZE." -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
