I agree with Syrmour's method. When I am stumped, I'll start an email to my boss (the best assembler coder I know) asking what I've done wrong. While putting as much information into the email as possible, so he doesn't think I'm taking the easy way out, 99 times out of 100 I'll find my issue. Usually, it's something fairly minor that I've simply overlooked as "obviously correct" or "obviously not the area of the problem." When I point this out to my boss, he usually says "desk check" your code. But I live by the motto that was posted here some time in the past - Months of coding and debugging beats hours of desk checking any day. (or something very close to that LOL). Ramsey
On Fri, Feb 26, 2021 at 11:09 AM Seymour J Metz <[email protected]> wrote: > With regard to Tom's second method, often *I* spot the error when I'm > asking for help and explaining the code. Somehow it seems to sometimes cure > a mental blind spot. > > > -- > Shmuel (Seymour J.) Metz > http://mason.gmu.edu/~smetz3 > > ________________________________________ > From: IBM Mainframe Discussion List [[email protected]] on behalf > of Charles Mills [[email protected]] > Sent: Friday, February 26, 2021 11:20 AM > To: [email protected] > Subject: Re: Assembler - Authorized program debug > > I really second Tom's latter method. Try walking through the code with > someone else -- explain to them how it works instruction by instruction. I > have good luck with that method using my wife as a sounding board -- even > though she doesn't know L from ST. > > I think many respondents are answering the wrong question. The OP's > question is not "how do I debug or prevent an S047?" He understands the > S047. His question is "how do I debug this code without triggering an > unrelated but well-deserved S047?" > > Charles > > > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] On > Behalf Of Tom Brennan > Sent: Thursday, February 25, 2021 11:46 PM > To: [email protected] > Subject: Re: Assembler - Authorized program debug > > Is the TSO TESTAUTH command still around? I have to admit I can't > remember ever trying it. My debugging method of such code typically > consisted of multiple temporary WTO's to let me know where the program > was at before it failed, and also display fields or registers I was > interested in. Usually within a few iterations of that method, I'd > figure out my problem. > > Another method: After looking at your code for hours and hours, have > someone else peek over your shoulder and invariably they will see the > problem in seconds. > > ---------------------------------------------------------------------- > 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 > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
