Right. And XPLINK implies PDSE. Nothing is simple. I think I only have one assembler routine that is used after initialization, so I could re-write that and live with a performance hit on all of the others.
FWIW, no, I never wrote the words "register 13." I want a traceback. For all I care it could be based on R13 or breadcrumbs or footprints in the sand. No, switching to AMODE 64 and calling __le_traceback() is I would think a recipe for agony. I have a solution, somewhat outside of the box that has been pursued of late in this thread, sent to me off-list, that sounds very straightforward and promising. I am going to give that a shot. Assuming the author does not object I will follow up here, assuming I can make it work. Charles -----Original Message----- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Tom Marchant Sent: Friday, January 23, 2015 8:26 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: In C++ any way to get traceback info in a catch()? On Thu, 22 Jan 2015 20:11:12 -0800, Charles Mills wrote: >Mmm. I'm going to make the 64-bit leap one of these days. I should set >the option and compile everything and see what errors I get. That would >be a first clue as to how big a leap. Be aware that C/C++ running AMODE 64 is XPLINK. Program linkage is very different, and switching between XPLINK and standard linkage is expensive. IIRC you mentioned in an earlier post that you were trying to trace your way back via R13. That isn't the way it is done with XPLINK. And for the comment that someone (Bernd?) made about switching to AMODE 64 before calling TRACEBCK, that wouldn't work at all, for the same reason. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
