Bugzilla – Bug 851044
system halt on resume from hibernation: segfault in libdirectfb
Last modified: 2014-11-03 14:10:13 UTC
User-Agent: Mozilla/5.0 (X11; Linux i686; rv:25.0) Gecko/20100101 Firefox/25.0 Whenever I hibernate my pb dots e2 and then try to resume it stops after reading in 100% of all pages apparently when it tries to set the video mode. /var/log/messages shows a segafult in libdirectfb. It is possible to escape with the SysRq keys. Reproducible: Always
Created attachment 567979 [details] /var/log/messages
Created attachment 567981 [details] pm-suspend.log Is there a way to disable the console framebuffer? It seems to have some bugs as it does also display the boot image in a wrong position. Maybe I could set vga=0 or anything thelike or will that stay without effect unless I disable KMS completely?
Why this has become a bug of systemd-maintainers@suse.de ? ... from attachments /usr/lib/pm-utils/pm-functions: Zeile 134: 24494 Speicherzugriffsfehler s2disk --config $S2DISK_CONF
Adding the maintainers of pm-utils and DirectFB to Carbon Copy list as well as re-assigning to the bugowner of pm-utils @ Elmar : A core dump could help here even if the kernel tells about kernel: [11361.253888] s2disk[24494]: segfault at 45454551 ip b729f8e7 sp bf9d4200 error 4 in libdirectfb-1.6.so.0.3.0[b7230000+133000]
At least there is no *.core in /usr and /var or directly in / or /root or not even in any other dir except the wine mono runtime.
(In reply to comment #0) > User-Agent: Mozilla/5.0 (X11; Linux i686; rv:25.0) Gecko/20100101 > Firefox/25.0 > > Whenever I hibernate my pb dots e2 and then try to resume it stops after > reading in 100% of all pages apparently when it tries to set the video mode. > /var/log/messages shows a segafult in libdirectfb. It is possible to escape > with the SysRq keys. > > Reproducible: Always Is the affected system running a 32 bit kernel ? what is the output of uname -a ?
Linux linux-4wml 3.11.6-4-desktop #1 SMP PREEMPT Wed Oct 30 18:04:56 UTC 2013 (e6d4a27) i686 i686 i386 GNU/Linux
(In reply to comment #7) > Linux linux-4wml 3.11.6-4-desktop #1 SMP PREEMPT Wed Oct 30 18:04:56 UTC 2013 > (e6d4a27) i686 i686 i386 GNU/Linux All these "failure at resume" bugs in 32 bit x86 are caused by the same set of kernel bugs. https://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git/commit/?h=linux-3.11.y&id=83da8ac7ee52d56e9bf5626502cca6e8e169580c https://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git/commit/?h=linux-3.11.y&id=83da8ac7ee52d56e9bf5626502cca6e8e169580c other symptoms http://lists.opensuse.org/opensuse/2013-12/msg00248.html In short, stuff using epoll or select are affected.. fix is only on 3.11.8 or later.
If this is a known bug then why not provide an update from 3.11.6 to 3.11.8?
(In reply to comment #9) > If this is a known bug then why not provide an update from 3.11.6 to 3.11.8? AFAIK, There is an update to 3.12.x in the works.
*** Bug 861488 has been marked as a duplicate of this bug. ***
Increase priority as I've been hit now several times by the same kernel bug
*** Bug 862678 has been marked as a duplicate of this bug. ***
Updated kernels have hit the update repos by now.. 3.11.10 should have this bugs fixed..
*** Bug 839170 has been marked as a duplicate of this bug. ***
Closing as we've long since updated past 3.11.8.