[rkward] [Bug 387394] Carsh of rkward

[Thomas Friedrichsmeier]

https://bugs.kde.org/show_bug.cgi?id=387394

--- Comment #3 from Thomas Friedrichsmeier <thomas.friedrichsmeier at ruhr-uni-bochum.de> ---
> If it is a webkit problem any pointers were to look into?

Well, not sure how to track this down. But from the two backtraces: Thread 1 is
the active thread, i.e. the one to look at. Here are some relevant lines from
that:

[just a few more step to the crash, now, looks like an invalid free()]
#9  0x00007fc59460e599 in _int_free () at /usr/lib/libc.so.6
[everything in between happens inside jre/lib/libdeploy.so]
#18 0x00007fc5126a67ea in  () at /usr/lib/jvm/java-9-jre/jre/lib/libdeploy.so
#19 0x00007fc59d63a51a in call_init.part () at /lib64/ld-linux-x86-64.so.2
[...]
#27 0x00007fc5918b2f22 in dlopen () at /usr/lib/libdl.so.2
[...]
#29 0x00007fc59529f429 in QLibrary::load() () at /usr/lib/libQt5Core.so.5
[...]
#35 0x00007fc59b17e19c in WebCore::Page::pluginData() const () at
/usr/lib/libQt5WebKit.so.5

So, in essence, what's happening is that QtWebKit is trying to load some
plugin. That plugin loads /usr/lib/jvm/java-9-jre/jre/lib/libdeploy.so
(possibly as one out of several libraries). Then while initializing that, there
is an invalid free.

What we can conclude up to here is that it's some java-based plugin. Which
exactly? You could try attaching strace to the running session (before you
trigger the crash) as

  strace -p PID_OF_RKWARD (not rkward.rbackend)

That might generate some further hints.

-- 
You are receiving this mail because:
You are the assignee for the bug.