[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Re: Crash info with no stack trace
From:       Thomas_Lübking <thomas.luebking () gmail ! com>
Date:       2012-01-26 16:03:13
Message-ID: op.v8o77nae9bmiid () localhost ! localdomain
[Download RAW message or body]

Am 26.01.2012, 10:33 Uhr, schrieb Lindsay Mathieson  
<lindsay.mathieson@gmail.com>:

> Is this more helpful? I attached after startup but the crash didn't  
> happen for
> quite some time - maybe half an hour. While I was switching between apps
> everything froze up for a few seconds and gdb showed the segfault

It's at least similar to the pending thread crashes in bug #281955 (both)  
and pretty much bug #281964

Since the other crashes lead to segfaults in malloc(), read(), select(),  
pthread_cond_wait(), and yours into a statement in QMutex (given you're  
using Qt 4.8) which is completely crash safe (or if not 4.7 in a dangeling  
d pointer?), i'd say here's clearly some stack corruption which seems to  
point libsoprano or thread handling, but could also be due to an invalid  
(outdated and ABI incompatible) plugin library or ABI incompatible soprano  
version... *shrug*

-> how did you obtain KDE 4.8 and did you get updates for libsoprano with  
that as well?


Cheers,
Thomas

PS: i'm not aware that one would require root permissions to load the  
debug symbols from the other library, sounds like their permissions are  
not set correctly.

>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<
[prev in list] [next in list] [prev in thread] [next in thread]