This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.

Bug 73421 - Strange results measured on Mac OS X Intel
Summary: Strange results measured on Mac OS X Intel
Status: CLOSED WONTFIX
Alias: None
Product: profiler
Classification: Unclassified
Component: Base (show other bugs)
Version: 5.x
Hardware: PC Mac OS X
: P3 blocker (vote)
Assignee: issues@profiler
URL:
Keywords:
Depends on:
Blocks: 73536
  Show dependency tree
 
Reported: 2006-03-09 13:15 UTC by ehucka
Modified: 2007-02-20 18:17 UTC (History)
0 users

See Also:
Issue Type: DEFECT
Exception Reporter:


Attachments
screenshot of snapshot (19.38 KB, image/png)
2006-03-09 13:16 UTC, ehucka
Details

Note You need to log in before you can comment on or make changes to this bug.
Description ehucka 2006-03-09 13:15:58 UTC
Mac OS X running on intel

I profiled an web application (cpu profiling, entire application) and I've got
into a state when most of measured values are shown as '0.0 ms'. I have not
execute calibration between the right measurement and the wrong.
Comment 1 ehucka 2006-03-09 13:16:50 UTC
Created attachment 29187 [details]
screenshot of snapshot
Comment 2 Jiri Sedlacek 2006-03-09 13:21:48 UTC
You should perform calibration once more and rerun the profiling again. If some 
results seemed correct and other after were incorrect without any calibration 
between, it could be caused by some other application previously running and 
significantly loading the system when calibrating and running the first 
(correct) profiling.
Comment 3 ehucka 2006-03-09 13:26:13 UTC
I will try it but the snapshot is not count correctly I think. There is a method
with total time 0.0 but its self time is measured right - 500 ms.
Comment 4 ehucka 2006-03-09 13:49:07 UTC
I have executed calibration twice and than start profiling again but the results
are the same. All times except self times are zero.
Comment 5 Tomas Hurka 2006-03-09 14:43:59 UTC
You are not right that all times except self times are zero. Look at your screenshot again.
Comment 6 ehucka 2006-03-09 16:15:46 UTC
aha, I found out the first submit to the web application is measured right - all
methods have some non-zero times but second and others submits are measured with
the wrong values.
Comment 7 ehucka 2006-03-09 16:45:09 UTC
Is it possible to check it on some other mac on intel? If it is reproducible it
should be set as p2 I think.
Comment 8 Tomas Hurka 2006-03-31 12:03:38 UTC
This seems to be JDK bug. I fixed it by installing latest (J2SE 5.0 Release 4 Developer Preview 7)  JDK 1.5 
on Mac OS X/Intel.
Comment 9 ehucka 2006-10-25 14:30:58 UTC
verified
Comment 10 Alexander Kouznetsov 2007-02-20 10:18:04 UTC
Closing old issues
Comment 11 Alexander Kouznetsov 2007-02-20 18:17:57 UTC
Reverting to original Target Milestone value changed by mistake. Sorry for
inconvenience.