Bug 19833 - REGRESSION(r34754): Start profile button in profiler now profiles inspector rather than actual page
Summary: REGRESSION(r34754): Start profile button in profiler now profiles inspector r...
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: Web Inspector (Deprecated) (show other bugs)
Version: 528+ (Nightly build)
Hardware: Mac OS X 10.5
: P1 Normal
Assignee: Nobody
URL: http://nerget.com/ParticleJS/Particle...
Keywords: InRadar, Regression
Depends on:
Blocks:
 
Reported: 2008-06-30 19:31 PDT by Oliver Hunt
Modified: 2008-07-11 15:43 PDT (History)
4 users (show)

See Also:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Oliver Hunt 2008-06-30 19:31:00 PDT 
I just found that when starting a profile by clicking the button to start profiles in the inspector results in us reporting profile data for the inspector which seems wholly unhelpful.

To reproduce
* Go to http://nerget.com/ParticleJS/Particles.html
* Open the inspector
* Go to the profile pane
* Click the profile button
* Wait a wee while
* Stop the profile
* Admire the profile of the inspector.
Comment 1 Kevin McCullough 2008-07-10 14:04:48 PDT 
rdar://problem/6067178
Comment 2 Kevin McCullough 2008-07-10 14:38:04 PDT 
It looks like r34753 is the culprit.
Comment 3 Adam Barth 2008-07-10 14:43:12 PDT 
I'll take a look at this tonight.
Comment 4 Kevin McCullough 2008-07-10 15:03:33 PDT 
Although that revision does not look like it could have possibly caused the issue.  I think the inspector needs to call startUserInitiatedProfiling() instead of console.profile();
Comment 5 Kevin McCullough 2008-07-10 15:49:51 PDT 
Scratch that calling InspectorController.inspectedWindow().console should contextually be correct in that it should call the console on the inspected window.
Comment 6 Adam Barth 2008-07-10 16:19:29 PDT 
> It looks like r34753 is the culprit.

Hum...  I can't repro the bug at r34753 but I can repro it at r34754.

http://trac.webkit.org/changeset/34754

That change seems more relevant as touches lines of code that actually run in this case.
Comment 7 Kevin McCullough 2008-07-10 17:18:07 PDT 
Yeah I re-ran my tests and I also think r18754 is the real culprit.  Sorry to waste your time.  I just need to figure out how this caused the regression.
Comment 8 Kevin McCullough 2008-07-10 17:38:47 PDT 
I'm guessing that somewhere under JSInspectedObjectWrapper::wrapone of these wrappers is holding the wrong ExecState.
Comment 9 Kevin McCullough 2008-07-11 15:43:59 PDT 
Committed revision 35140.