RESOLVED FIXED 83881
[GTK] fast/dom/gc-11.html flaky crash 
https://bugs.webkit.org/show_bug.cgi?id=83881
Summary [GTK] fast/dom/gc-11.html flaky crash
Philippe Normand
Reported 2012-04-13 03:57:58 PDT
Happens on the Debug bot. Trace: #0 0x00007f7653fd61ce in WTF::RefPtr<WebCore::Document>::get (this=0x560) at ../../Source/WTF/wtf/RefPtr.h:58 58 T* get() const { return m_ptr; } #0 0x00007f7653fd61ce in WTF::RefPtr<WebCore::Document>::get (this=0x560) at ../../Source/WTF/wtf/RefPtr.h:58 #1 0x00007f7653fd5af2 in WebCore::Frame::document (this=0x0) at ../../Source/WebCore/page/Frame.h:291 #2 0x00007f7654848284 in WebCore::DocumentLoader::subresource (this=0x128f06c0, url=...) at ../../Source/WebCore/loader/DocumentLoader.cpp:547 #3 0x00007f7653fee89c in WebKit::FrameLoaderClient::dispatchDidFinishLoading (this=0x12b5cdf0, loader=0x128f06c0, identifier=11489) at ../../Source/WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:995 #4 0x00007f76548a41f6 in WebCore::ResourceLoadNotifier::dispatchDidFinishLoading (this=0x12aed750, loader=0x128f06c0, identifier=11489, finishTime=0) at ../../Source/WebCore/loader/ResourceLoadNotifier.cpp:142 #5 0x00007f76548a3d3f in WebCore::ResourceLoadNotifier::didFinishLoad (this=0x12aed750, loader=0x128b78e0, finishTime=0) at ../../Source/WebCore/loader/ResourceLoadNotifier.cpp:89 #6 0x00007f76548a2900 in WebCore::ResourceLoader::didFinishLoadingOnePart (this=0x128b78e0, finishTime=0) at ../../Source/WebCore/loader/ResourceLoader.cpp:327 #7 0x00007f76548a2823 in WebCore::ResourceLoader::didFinishLoading (this=0x128b78e0, finishTime=0) at ../../Source/WebCore/loader/ResourceLoader.cpp:313 #8 0x00007f7654896500 in WebCore::MainResourceLoader::didFinishLoading (this=0x128b78e0, finishTime=0) at ../../Source/WebCore/loader/MainResourceLoader.cpp:481 #9 0x00007f76548a304d in WebCore::ResourceLoader::didFinishLoading (this=0x128b78e0, finishTime=0) at ../../Source/WebCore/loader/ResourceLoader.cpp:452 #10 0x00007f7654a47350 in WebCore::readCallback (source=0x9047700, asyncResult=0x90b6ea0, data=0x1249c930) at ../../Source/WebCore/platform/network/soup/ResourceHandleSoup.cpp:686 #11 0x00007f7651c75b6f in async_ready_callback_wrapper () from /home/slave/webkitgtk/gtk-linux-64-debug/build/WebKitBuild/Dependencies/Root/lib64/libgio-2.0.so.0 #12 0x00007f7651c8ca83 in g_simple_async_result_complete () from /home/slave/webkitgtk/gtk-linux-64-debug/build/WebKitBuild/Dependencies/Root/lib64/libgio-2.0.so.0 #13 0x00007f7651c8cc4f in complete_in_idle_cb_for_thread () from /home/slave/webkitgtk/gtk-linux-64-debug/build/WebKitBuild/Dependencies/Root/lib64/libgio-2.0.so.0 #14 0x00007f7651b183f3 in g_idle_dispatch () from /home/slave/webkitgtk/gtk-linux-64-debug/build/WebKitBuild/Dependencies/Root/lib64/libglib-2.0.so.0 #15 0x00007f7651b15c8a in g_main_dispatch () from /home/slave/webkitgtk/gtk-linux-64-debug/build/WebKitBuild/Dependencies/Root/lib64/libglib-2.0.so.0 #16 0x00007f7651b16950 in g_main_context_dispatch () from /home/slave/webkitgtk/gtk-linux-64-debug/build/WebKitBuild/Dependencies/Root/lib64/libglib-2.0.so.0 #17 0x00007f7651b16b3a in g_main_context_iterate () from /home/slave/webkitgtk/gtk-linux-64-debug/build/WebKitBuild/Dependencies/Root/lib64/libglib-2.0.so.0 #18 0x00007f7651b16f70 in g_main_loop_run () from /home/slave/webkitgtk/gtk-linux-64-debug/build/WebKitBuild/Dependencies/Root/lib64/libglib-2.0.so.0 #19 0x00007f76529ede99 in gtk_main () from /home/slave/webkitgtk/gtk-linux-64-debug/build/WebKitBuild/Dependencies/Root/lib64/libgtk-3.so.0 #20 0x000000000045b77c in runTest (testPathOrURL=...) at ../../Tools/DumpRenderTree/gtk/DumpRenderTree.cpp:705 #21 0x000000000045adf4 in runTestingServerLoop () at ../../Tools/DumpRenderTree/gtk/DumpRenderTree.cpp:492 #22 0x000000000045de10 in main (argc=2, argv=0x7fffd385aef8) at ../../Tools/DumpRenderTree/gtk/DumpRenderTree.cpp:1387 Maybe a missing m_frame NULL check in DocumentLoader::subresource()?
Attachments
Add attachment proposed patch, testcase, etc.
Philippe Normand
Comment 1 2012-04-13 04:00:17 PDT
What do you think about adding a test on m_frame in DocumentLoader::subresource() Adam?
Philippe Normand
Comment 2 2012-04-13 04:25:49 PDT
Updated test_expectations in r114112
Adam Barth
Comment 3 2012-04-13 09:30:41 PDT
> What do you think about adding a test on m_frame in DocumentLoader::subresource() Adam? That's a good question for Nate. Nate, thoughts?
Zan Dobersek
Comment 4 2012-07-15 10:47:09 PDT
This crash was (most likely) caused by the fast/dom/gc-10.html test being run before this test. That test is now skipped (bug #86771) so the crash in gc-11.html doesn't occur anymore, ergo test expectation was removed in http://trac.webkit.org/changeset/118474. Closing the bug.
Note You need to log in before you can comment on or make changes to this bug.