[Webkit-unassigned] [Bug 55516] Possible data race on JSC::Yarr::Interpreter::interpret

Wed Mar 2 10:13:47 PST 2011

https://bugs.webkit.org/show_bug.cgi?id=55516

Timur Iskhodzhanov <timurrrr at google.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |timurrrr at google.com

--- Comment #4 from Timur Iskhodzhanov <timurrrr at google.com>  2011-03-02 10:13:47 PST ---
Hi,

I'm one of the authors of this ThreadSanitizer tool.
It's not aimed at detecting OOB accesses (we have Valgrind Memcheck for that, right?)
instead it detects data races.
Here you can see some info on the data in the report
http://code.google.com/p/data-race-test/wiki/UnderstandingThreadSanitizerReports 

In this particular case I think the problem is that the WebKit thread and two Chrome threads are accessing the same memory  without any synchronization.
I bet it's a Chromium-specific problem or it could be caused by YARR trying to write OOB or to some wrong memory given by the memory allocator
(around third_party/WebKit/Source/JavaScriptCore/yarr/YarrInterpreter.cpp:111 )

Do you have any ideas how the latter may happen?

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.