Status: Assigned
Owner: [email protected]
Labels: Type-Bug Pri-1 OS-Mac Area-BrowserUI Size-Medium

New issue 18189 by [email protected]: Invalid write in NSRectFill() ... in  
StatusBubbleMac::Create() in ui tests on mac
http://code.google.com/p/chromium/issues/detail?id=18189

Using tools/valgrind/shard_all_tests.sh to run just the ui tests
on the mac through valgrind continuously sees the following
invalid write in roughly one of every 30 tests.
The tests this has happened in so far are

AutomationProxyTest.Cookies
AutomationProxyTest.GetActiveTabIndex
AutomationProxyTest.GetTab
AutomationProxyTest.GetTabCount
AutomationProxyTest4.NumberValueIsEchoedByDomAutomationController
AutomationProxyTest5.TestLifetimeOfDomAutomationController
DomCheckerTest.File
DownloadTest.ContentDisposition
DownloadTest.NoDownload
GoogleTest.Crash
ResourceDispatcherTest.CrossSiteNavigationNonBuffered
ResourceDispatcherTest.CrossSiteOnunloadCookie
ResourceDispatcherTest.RespectNoSniffDirective
ResourceDispatcherTest.SniffHTMLWithNoContentType
ResourceDispatcherTest.SniffNoContentTypeNoData
ResourceDispatcherTest.SyncXMLHttpRequest_Disallowed
SavePageTest.NoSave
SessionHistoryTest.BasicBackForward
SessionHistoryTest.FrameFormBackForward
SunSpiderTest.Perf
TabRestoreUITest.RestoreCrossSiteWithExistingSiteInstance
TabRestoreUITest.RestoreWindow
TabRestoreUITest.RestoreWithExistingSiteInstance
ViewSourceTest.DoesBrowserRenderInViewSource
ViewSourceTest.ViewSourceInPageMenuDisabledWhileViewingSource

I've seen it twice in
ResourceDispatcherTest.RespectNoSniffDirective
but that's probably just chance.

Here's the stack:

Invalid write of size 8
   sseCGSFill8by1 (/System/.../CoreGraphics)
   argb32_mark_constshape (/System/.../CoreGraphics)
   argb32_mark (/System/.../CoreGraphics)
   ripl_BltShape (/System/.../libRIP.A.dylib)
   ripc_Render (/System/.../libRIP.A.dylib)
   ripc_DrawRects (/System/.../libRIP.A.dylib)
   CGContextFillRects (/System/.../CoreGraphics)
   CGContextFillRect (/System/.../CoreGraphics)
   NSRectFill (/System/.../AppKit)
   -[NSView _drawRect:clip:] (/System/.../AppKit)
   -[NSView _recursiveDisplayAllDirtyWithLockFocus:visRect:]
(/System/.../AppKit)
   -[NSView
_recursiveDisplayRectIfNeededIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:topView:]
(/System/.../AppKit)
   -[NSNextStepFrame
_recursiveDisplayRectIfNeededIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:topView:]
(/System/.../AppKit)
   -[NSView
_displayRectIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:]
(/System/.../AppKit)
   -[NSView displayIfNeeded] (/System/.../AppKit)
   -[NSWindow displayIfNeeded] (/System/.../AppKit)
   -[NSWindow
_reallyDoOrderWindow:relativeTo:findKey:forCounter:force:isModal:]
(/System/.../AppKit)
   -[NSWindow orderWindow:relativeTo:] (/System/.../AppKit)
   -[NSWindow addChildWindow:ordered:] (/System/.../AppKit)
   StatusBubbleMac::Create()
(/Users/dkegel/chromium/src/xcodebuild/Debug/Chromium.app/Contents/MacOS/Chromium)

The caller of Create is

   StatusBubbleMac::SetStatus(std::basic_string<wchar_t,
std::char_traits<wchar_t>, std::allocator<wchar_t> > const&) (Chromium)
   Browser::LoadingStateChanged(TabContents*) (Chromium)
   TabContents::SetIsLoading(bool, LoadNotificationDetails*) (Chromium)

in all but one case seen so far.

Since there was a bug filed for a different crash in sseCGSFill8by1,
I'm inclined to consider this serious.


--
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings

--~--~---------~--~----~------------~-------~--~----~
Automated mail from issue updates at http://crbug.com/
Subscription options: http://groups.google.com/group/chromium-bugs
-~----------~----~----~----~------~----~------~--~---

Reply via email to