Bug 58004 - Memory Leak in CoreIPC::decodeResourceError
Summary: Memory Leak in CoreIPC::decodeResourceError
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: WebKit2 (show other bugs)
Version: 528+ (Nightly build)
Hardware: Mac OS X 10.5
: P2 Normal
Assignee: Michael Saboff
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2011-04-06 18:35 PDT by Michael Saboff
Modified: 2011-04-07 09:08 PDT (History)
0 users

See Also:

Attachments
Patch to wrap NSError allocation with autorelease (1.45 KB, patch)
2011-04-06 18:41 PDT, Michael Saboff 		mjs: 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 Michael Saboff 2011-04-06 18:35:50 PDT 
The local NSError object created in decodeResourceError() has a retain count of 2 when the function exits.  One is needed for the ResourceError reference argument, the other should be released directly or via an autorelease.
Comment 1 Michael Saboff 2011-04-06 18:41:01 PDT 
Created attachment 88555 [details]
Patch to wrap NSError allocation with autorelease
Comment 2 Maciej Stachowiak 2011-04-06 18:46:54 PDT 
Comment on attachment 88555 [details]
Patch to wrap NSError allocation with autorelease

r=me

(But why autorelease instead of release? Explicit release is usually better.)
Comment 3 Darin Adler 2011-04-06 20:37:00 PDT 
Comment on attachment 88555 [details]
Patch to wrap NSError allocation with autorelease

View in context: https://bugs.webkit.org/attachment.cgi?id=88555&action=review

> Source/WebKit2/Shared/mac/WebCoreArgumentCodersMac.mm:188
> +    NSError *nsError = [[[NSError alloc] initWithDomain:nsString(domain) code:code userInfo:userInfo] autorelease];

A plain old release is much faster than autorelease, so it would be better to do it that way.
Comment 4 Michael Saboff 2011-04-07 09:08:11 PDT 
Committed r83177: <http://trac.webkit.org/changeset/83177>