Bug 145346
| Summary: | [GTK][Stable] Crash in WebCore::Frame::willDetachPage | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Michael Catanzaro <mcatanzaro> |
| Component: | WebKitGTK | Assignee: | Nobody <webkit-unassigned> |
| Status: | RESOLVED WONTFIX | ||
| Severity: | Normal | CC: | cgarcia, mcatanzaro, tpopela |
| Priority: | P2 | ||
| Version: | 528+ (Nightly build) | ||
| Hardware: | PC | ||
| OS: | Linux | ||
| See Also: | https://bugzilla.redhat.com/show_bug.cgi?id=1213180 | ||
Michael Catanzaro
Moving from downstream. This is probably our biggest crasher, with 2700 reports so far. It seems to be a regression introduced in 2.6.6, maybe from r181631.
Thread no. 1 (10 frames)
#0 WebCore::Frame::willDetachPage at /usr/src/debug/webkitgtk-2.6.6/Source/WebCore/page/Frame.cpp:783
#1 WebCore::FrameLoader::detachFromParent at /usr/src/debug/webkitgtk-2.6.6/Source/WebCore/loader/FrameLoader.cpp:2487
#2 WebKit::WebPage::close at /usr/src/debug/webkitgtk-2.6.6/Source/WebKit2/WebProcess/WebPage/WebPage.cpp:969
#3 handleMessage<Messages::WebPage::Close, WebKit::WebPage, void (WebKit::WebPage::*)()> at /usr/src/debug/webkitgtk-2.6.6/Source/WebKit2/Platform/IPC/HandleMessage.h:120
#4 WebKit::WebPage::didReceiveWebPageMessage at /usr/src/debug/webkitgtk-2.6.6/x86_64-redhat-linux-gnu/DerivedSources/WebKit2/WebPageMessageReceiver.cpp:592
#5 IPC::MessageReceiverMap::dispatchMessage at /usr/src/debug/webkitgtk-2.6.6/Source/WebKit2/Platform/IPC/MessageReceiverMap.cpp:87
#6 WebKit::WebProcess::didReceiveMessage at /usr/src/debug/webkitgtk-2.6.6/Source/WebKit2/WebProcess/WebProcess.cpp:599
#7 IPC::Connection::dispatchMessage at /usr/src/debug/webkitgtk-2.6.6/Source/WebKit2/Platform/IPC/Connection.cpp:828
#8 IPC::Connection::dispatchOneMessage at /usr/src/debug/webkitgtk-2.6.6/Source/WebKit2/Platform/IPC/Connection.cpp:856
#9 WTF::RunLoop::performWork at /usr/src/debug/webkitgtk-2.6.6/Source/WTF/wtf/RunLoop.cpp:119
Full backtrace downstream.
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Michael Catanzaro
Closing since this was only ever a problem with 2.6.