Bug 98522 - No autorelease pool in place, causing buildup of autoreleased objects.
Summary: No autorelease pool in place, causing buildup of autoreleased objects.
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: WebCore Misc. (show other bugs)
Version: 528+ (Nightly build)
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Jer Noble
URL:
Keywords: InRadar
Depends on:
Blocks:
 
Reported: 2012-10-05 08:32 PDT by Jer Noble
Modified: 2012-10-05 11:34 PDT (History)
3 users (show)

See Also:

Attachments
Patch (1.82 KB, patch)
2012-10-05 08:36 PDT, Jer Noble 		ap: review+
Details | Formatted Diff | Diff
View All 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 Jer Noble 2012-10-05 08:32:53 PDT 
No autorelease pool in place, causing buildup of autoreleased objects.
Comment 1 Jer Noble 2012-10-05 08:36:42 PDT 
Created attachment 167332 [details]
Patch
Comment 2 Alexey Proskuryakov 2012-10-05 10:42:34 PDT 
Comment on attachment 167332 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=167332&action=review

> Source/WebCore/ChangeLog:4
> +        https://bugs.webkit.org/show_bug.cgi?id=98522

If there is a Radar bug tracking this, please post its number to Bugzilla and to ChangeLog.

> Source/WebCore/platform/cf/RunLoopTimerCF.cpp:49
> +    AutodrainedPool pool;

This won't break Windows build, will it? With a cursory look, I couldn't find a no-op implementation of this class that Windows could be using.
Comment 3 Alexey Proskuryakov 2012-10-05 10:45:23 PDT 
My mistake - I was looking at a wrong AutodrainedPool.h.
Comment 4 Jer Noble 2012-10-05 11:14:58 PDT 
(In reply to comment #2)
> (From update of attachment 167332 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=167332&action=review
> 
> > Source/WebCore/ChangeLog:4
> > +        https://bugs.webkit.org/show_bug.cgi?id=98522
> 
> If there is a Radar bug tracking this, please post its number to Bugzilla and to ChangeLog.

Sure thing.

> > Source/WebCore/platform/cf/RunLoopTimerCF.cpp:49
> > +    AutodrainedPool pool;
> 
> This won't break Windows build, will it? With a cursory look, I couldn't find a no-op implementation of this class that Windows could be using.

Yep, as I think you discovered in comment #3, there's a no-op implementation for non-ObjC CF-ports.
Comment 5 Jer Noble 2012-10-05 11:15:49 PDT 
<rdar://problem/11647950>
Comment 6 Jer Noble 2012-10-05 11:34:05 PDT 
Committed r130534: <http://trac.webkit.org/changeset/130534>