Bug 164920 - [GTK] Web process crash destroying PlatformDisplayX11 in exit handler
Summary: [GTK] Web process crash destroying PlatformDisplayX11 in exit handler
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: WebKitGTK (show other bugs)
Version: WebKit Nightly Build
Hardware: PC Linux
: P2 Normal
Assignee: Nobody
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2016-11-18 05:42 PST by Michael Catanzaro
Modified: 2016-11-30 05:47 PST (History)
3 users (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Michael Catanzaro 2016-11-18 05:42:26 PST
Some search provider (I guess epiphany) web process crash destroying PlatformDisplayX11 in exit handler:

Thread 1 (Thread 0x7fa240ba3440 (LWP 1808)):
#0  0x0000000000000000 in ?? ()
No symbol table info available.
#1  0x00007fa23f9c2319 in WebCore::PlatformDisplayX11::~PlatformDisplayX11 (this=0x7fa229992240, __in_chrg=<optimized out>) at /usr/src/debug/webkitgtk-2.13.90/Source/WebCore/platform/graphics/x11/PlatformDisplayX11.cpp:58
No locals.
#2  0x00007fa23278a420 in __run_exit_handlers (status=status@entry=0, listp=0x7fa232b105b8 <__exit_funcs>, run_list_atexit=run_list_atexit@entry=true, run_dtors=run_dtors@entry=true) at exit.c:83
        atfct = <optimized out>
        onfct = <optimized out>
        cxafct = <optimized out>
        f = <optimized out>
#3  0x00007fa23278a47a in __GI_exit (status=status@entry=0) at exit.c:105
No locals.
#4  0x00007fa23e8e9ed3 in IPC::Connection::didFailToSendSyncMessage (this=this@entry=0x7fa2299e8000) at /usr/src/debug/webkitgtk-2.13.90/Source/WebKit2/Platform/IPC/Connection.cpp:814
No locals.
#5  0x00007fa23e8ecc11 in IPC::Connection::sendSyncMessage (this=this@entry=0x7fa2299e8000, syncRequestID=28, encoder=std::unique_ptr<IPC::Encoder> containing 0x7fa1b8f73a00, timeout=..., timeout@entry=..., sendSyncOptions=sendSyncOptions@entry=...) at /usr/src/debug/webkitgtk-2.13.90/Source/WebKit2/Platform/IPC/Connection.cpp:474
        locker = {m_lockable = 0x7fa2299e80c8}
        protect = <optimized out>
        reply = std::unique_ptr<IPC::Decoder> containing 0x7fffec1d4000
#6  0x00007fa23ea142dd in IPC::Connection::sendSync<Messages::WebProcessProxy::ShouldTerminate>(Messages::WebProcessProxy::ShouldTerminate&&, Messages::WebProcessProxy::ShouldTerminate::Reply&&, unsigned long, std::chrono::duration<long, std::ratio<1l, 1000l> >, WTF::OptionSet<IPC::SendSyncOption>) (sendSyncOptions=..., timeout=..., destinationID=0, reply=<optimized out>, message=<optimized out>, this=0x7fa2299e8000) at /usr/src/debug/webkitgtk-2.13.90/Source/WebKit2/Platform/IPC/Connection.h:353
        syncRequestID = 28
        encoder = std::unique_ptr<IPC::Encoder> containing 0x0
        replyDecoder = std::unique_ptr<IPC::Decoder> containing 0x7fa23e1c3ee8 <bmalloc::PerProcess<bmalloc::Heap>::s_mutex>
#7  WebKit::WebProcess::shouldTerminate (this=<optimized out>) at /usr/src/debug/webkitgtk-2.13.90/Source/WebKit2/WebProcess/WebProcess.cpp:616
        shouldTerminate = false
#8  0x00007fa23e8f7f4a in WebKit::ChildProcess::terminationTimerFired (this=0x55746952c570) at /usr/src/debug/webkitgtk-2.13.90/Source/WebKit2/Shared/ChildProcess.cpp:160
No locals.
#9  0x00007fa23eabba67 in WebKit::WebPage::close (this=this@entry=0x7fa2299ce000) at /usr/src/debug/webkitgtk-2.13.90/Source/WebKit2/WebProcess/WebPage/WebPage.cpp:1113
        isRunningModal = false
#10 0x00007fa23ebfeb5b in IPC::callMemberFunctionImpl<WebKit::WebPage, void (WebKit::WebPage::*)(), std::tuple<>>(WebKit::WebPage*, void (WebKit::WebPage::*)(), std::tuple<>&&, std::integer_sequence<unsigned long>) (args=<optimized out>, function=<optimized out>, object=0x7fa2299ce000) at /usr/src/debug/webkitgtk-2.13.90/Source/WebKit2/Platform/IPC/HandleMessage.h:13
No locals.

Longer backtrace downstream.
Comment 1 Michael Catanzaro 2016-11-18 05:44:23 PST
Two user reports:

"Go to GNOME Activites, type "problem reporter" (crashes about two symbols before the end)"

"I ran into this crash when closing yelp after searching for something in a help document."

The actual crash count is useless (it's >22,000 but it's obviously wrong).
Comment 2 Carlos Garcia Campos 2016-11-30 05:47:09 PST
This is 2.13.90, I think this crash was fixed later in 2.14 branch by merging r205852. Feel free to reopen if you see this crash again with 2.14 or trunk.