Bug 173831 – [GTK] WebProcess from WebKitGtk+ 2.17.4 SIGSEVs in WTF::RunLoop::TimerBase::updateReadyTime() at Source/WTF/wtf/glib/RunLoopGLib.cpp:192

• Component: WebKitGTK
• Status: NEW

[Andres Gomez Garcia]

Created attachment 313834 [details]
BT from gdb for the WebProcess

Epiphany 3.24.1 and WebKit 2.17.4.

The compilation was done with CMake args:

'-DDEBUG_FISSION=OFF -DPORT=GTK -DCMAKE_BUILD_TYPE=Release -DENABLE_MINIBROWSER=ON -DCMAKE_C_FLAGS_RELEASE="-O0 -g -DNDEBUG -DG_DISABLE_CAST_CHECKS" -DCMAKE_CXX_FLAGS_RELEASE="-O0 -g -DNDEBUG -DG_DISABLE_CAST_CHECKS"'

After visiting several pages, eventually, the WebProcess hits a SIGSEV. Epiphany is completely closed.

This bug is not reproducible in a predictable way.

[Andres Gomez Garcia]

Created attachment 313835 [details]
BT from gdb for the WebProcess - 2

During epiphany's crashing/closing, this SIGSEV also happened.

[Yusuke Suzuki]

(In reply to Andres Gomez Garcia from comment #1)
> Created attachment 313835 [details]
> BT from gdb for the WebProcess - 2
> 
> During epiphany's crashing/closing, this SIGSEV also happened.

What makes this crash? g_source_set_ready_time's source is dangling?

[Andres Gomez Garcia]

(In reply to Yusuke Suzuki from comment #2)
> (In reply to Andres Gomez Garcia from comment #1)
> > Created attachment 313835 [details]
> > BT from gdb for the WebProcess - 2
> > 
> > During epiphany's crashing/closing, this SIGSEV also happened.
> 
> What makes this crash? g_source_set_ready_time's source is dangling?

These 2 BTs are the ones showing the crash. The actual reason why they happened in the first place, I don't know :(

[Andres Gomez Garcia]

Created attachment 314719 [details]
BT from gdb for the WebProcess - 3

Another similar BT that I got recently.

I cannot explain how it happened neither reproduce.

[Andres Gomez Garcia]

*** Bug 176102 has been marked as a duplicate of this bug. ***