Bug 34948 - tryMakeString should fail on error in length calculation
Summary: tryMakeString should fail on error in length calculation
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: JavaScriptCore (show other bugs)
Version: 528+ (Nightly build)
Hardware: PC OS X 10.5
: P2 Normal
Assignee: Gavin Barraclough
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-02-15 10:51 PST by Gavin Barraclough
Modified: 2010-02-15 14:03 PST (History)
0 users

See Also:

Attachments
The patch (6.55 KB, patch)
2010-02-15 10:52 PST, Gavin Barraclough 		ggaren: 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 Gavin Barraclough 2010-02-15 10:51:33 PST 
The sum of the length of substrings could overflow.
Comment 1 Gavin Barraclough 2010-02-15 10:52:57 PST 
Created attachment 48759 [details]
The patch

No perf impact.
Comment 2 Geoffrey Garen 2010-02-15 12:40:47 PST 
Comment on attachment 48759 [details]
The patch

r=me
Comment 3 Gavin Barraclough 2010-02-15 13:00:06 PST 
Sending        JavaScriptCore/ChangeLog
Sending        JavaScriptCore/runtime/UString.h
Transmitting file data ..
Committed revision 54788.
Comment 4 Gavin Barraclough 2010-02-15 14:03:48 PST 
Errk! - thanks darin/aroben,  made the output bool a reference in r54795.

Darin - I will separately check the performance of using |=.  My instinct is often to go with the branch, since on x86 the setcc instruction used to be multiple cycle, and a well predicted branch can be free.

In this case there may be little performance difference, since I don't think makeString is heavily used or performance critical.  As such perhaps I should switch to |=, since it's a line shorter.  :o)