Dump reading is a lost art: For newbies it was one of the principal ways to 
learn how to debug assembler code. As you became more experienced is was more 
of a "to do"  if you encountered problems.
Never like using those Abend Aid "do-me squaggies" .. Just ridiculously 
cumbersome.

C/++  on a mainframe is rare .. If you have had the luxury of using both you 
are  one of the lucky few.. as wading through the manual C/C++  set is a time 
consuming effort. You'd also have to be able to decipher  the macro coding 
since most macro (RACF comes to mind ) will generate both C and Assembler code. 
 Worst of all You'd have to know USS and Unix (XPG4 anyone ..?? ) inside out, 
no small feat either.

Your candidate profile  should include those who an code marco's ( ye olde 
SETA, SETB, SETC), Dump Reading , GTF Tracing,,, and C/C++ which does not mean 
cursory skill on z/OS, as SSL programmer  and TCIP Programmer guides have 
specific coding styles. There are no crutches for mainframe C/C++ programmers 
.. sorry guy's you either know it or you don't.  Object programming does not 
lend itself well to TCPIP programming, unless you are using Java on z/OS but 
that's another story.

Good Luck;

T'dell Sparks: Retired  aug/2013
Not interested in working anymore.. I had enough

-----Original Message-----
From: IBM Mainframe Assembler List [mailto:[email protected]] On 
Behalf Of John Gilmore
Sent: Friday, July 26, 2013 9:48 AM
To: [email protected]
Subject: Re: 3 job openings for mainframe Assembler/C programmers, dump readers

I agree with Ed Jaffe that 'Attitude has a lot to do with it.'

I think, however, that attitudes are is often reflections of skills and 
experiences.
Those who 1) have learned to read dumps and 2) have had the habit of reading 
them reinforced by successes in extracting valuable diagnostic information from 
them are very likely to use them.  Others are likely to try to avoid using them.

Dump reading is often, of course, premature.  I have seen programmers studying 
dumps in situations in which brief examination of their source programs by 
someone else quickly disclosed their errors.

EJ's second point, that some programmers write more reliable routines than 
others, is another important one, but I will leave its discussion for another 
day.

John Gilmore, Ashland, MA 01721 - USA

Reply via email to