------------------------<snip>----------------------
So why doesn't the very straightforward solution of a DEBUG parm and
calls to debug logic in the code work? You can have multiple debug flags
(routine entry exit, record read, etc.).
--------------------------<unsnip>-------------------
In many, probably most, cases, that will work just fine; certainly for
ARCHIVER or my RACF report goodies. I was hoping to find/develop a more
"elegant" mechanism that would be less dependant on imbedded code,
mainly because of size and re-entranterability considerations. I had
even considered the use of S-Cons to keep parm lists short! My aim
was/is a generalized mechanism to dump registers and selected storage
areas at specific points in the code, like parm flags, storage buffers,
selected control blocks etc. A single SPIE/ESPIE/STAE/ESTAE parm list
and a single exit routine might satisfy all these needs, whereas a CALL
to another routine would (PROBABLY) destroy registers 14 and 15, and
possibly 0 and 1. At one time, I even considered using SDUMP's and a
special processor for AMDPRDMP!
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
