![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
https://bugzilla.novell.com/show_bug.cgi?id=339229 Summary: KDM segfault at login Product: openSUSE 10.3 Version: Final Platform: x86-64 OS/Version: openSUSE 10.3 Status: NEW Severity: Normal Priority: P5 - None Component: KDE AssignedTo: kde-maintainers@suse.de ReportedBy: dmair@novell.com QAContact: qa@suse.de Found By: Customer Created an attachment (id=182086) --> (https://bugzilla.novell.com/attachment.cgi?id=182086) kdm.log from the problem host Host system is a Lenovo Thinkpad T60p, Intel Core 2 Duo T7600 2.33GHz, 3GB RAM. Video is an ATI FireGL 5250 with 256MB of RAM. Drivers are ATI proprietary fglrx 8.39.4 (problem is reproducible with other versions of the proprietary driver). I can't remember for certain if it happens with the VESA driver but I think it does. When the system boots and the KDM login screen is visible. Attempting to login immediately results in the graphical console disappearing and the host left at a text TTY login prompt. The failure happens when I press enter having input the password. At the failure, the following appears in /var/log/messages: Nov 5 09:40:56 dmairtp kernel: kdm[4734]: segfault at 00002b3ba78349d0 rip 00002b3ba78349d0 rsp 0000000040ffe748 error 15 Nov 5 09:40:56 dmairtp kdm_greet[3403]: Can't read from core Nov 5 09:40:56 dmairtp kdm[2917]: Unknown session exit code 0 (sig 11) from manager process The rip and rsp values varies each time the failure occurs. The following is the same event in dmesg: kdm[4734]: segfault at 00002b3ba78349d0 rip 00002b3ba78349d0 rsp 0000000040ffe748 error 15 If I login as root at the text console and restart xdm and, again, attempt to login immediately kdm will segfault and I'll be back at the text TTY. However, if I wait for an indeterminate amount of time or, it appears, if I cycle the cursor through all the tab stops on the login screen then kdm will not segfault when I login. I have been able to reproduce this on three independent installs of 10.3 on the same host. One was an upgrade from 10.2, the other two were clean installs and one of those on bare metal. I attached /var/log/kdm.log. It has a stack walkback for glibc malloc() detecting a memory corruption. I can't see timestamps in the log and the address of the corruption, the rip at the point of corruption and the mapping of modules don't link well (at all) to the segfault report. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.