Phil Smith wrote:
I'm trying to use APA to grok some performance issues, and I see this in the
output:
Address Size Location Percent of CPU Time * 10.00% >0.4%
*....1....2....3....4....5....6....7....8
1AACF228 64 .P000236+041640 10.88 *****
1AACF268 64 .P000236+041680 10.31 *****
1AACF2A8 64 .P000236+0416C0 2.89 *
1AACF1E8 64 .P000236+041600 2.43 *
There's no entry point ".P000236" in our code. There's no mention of that in
the doc, either, although it does have an example with .P000014 shown (but not explained).
Anyone know where this value comes from?
Just to provide some info to chew on:
1) Dignus doesn't (by default) generate labels of that form. You
could add options to PLINK to cause that to happen.
2) The routine in question appears to only be 64 bytes long?
Pretty small - any chance of looking at a dump or something
and seeing what is at that address?
3) The Dignus stack-management routines could be involved here,
they are allocated dynamically, so they don't have symbols
in the load module. The .P000236 may simply be the nearest
symbol the APA could find... (although, we don' thave 3 such
routines, and none of them is 64 bytes - I don't think.)
Anyway - more things to look at/think about.
- Dave Rivers -
--
[email protected] Work: (919) 676-0847
Get your mainframe programming tools at http://www.dignus.com
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
