Bug 15189 - xmodmap segfaults server when Wacom InputDevices are registered
Summary: xmodmap segfaults server when Wacom InputDevices are registered
Status: RESOLVED FIXED
Alias: None
Product: xorg
Classification: Unclassified
Component: Server/General (show other bugs)
Version: 7.3 (2007.09)
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Xorg Project Team
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2008-03-24 06:08 UTC by martin f. krafft
Modified: 2009-09-02 21:56 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
configuration (5.33 KB, text/plain)
2008-03-24 06:08 UTC, martin f. krafft
no flags Details
log output (80.71 KB, text/plain)
2008-03-24 06:13 UTC, martin f. krafft
no flags Details
The log (35.20 KB, application/octet-stream)
2008-08-23 00:55 UTC, Andraž \'ruskie\' Levstik
no flags Details

Description martin f. krafft 2008-03-24 06:08:09 UTC
Created attachment 15427 [details]
configuration

My Xserver (Debian sid, 1:7.3+10), running with a self-compiled ATI driver (6.6 branch, a5570f3), segfaults if I run xmodmap on it, but only if the three (or any of the three) InputDevice lines in the configuration are uncommented. If they're commented out, it all works.

Here is the backtrace, reproducible at will:

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x2b3aafa442c0 (LWP 25679)]
XkbSendNotification (kbd=0x8a2650, pChanges=0x7ffffcf1ab00, 
    cause=<value optimized out>) at ../../xkb/xkbEvents.c:770
770	../../xkb/xkbEvents.c: No such file or directory.
	in ../../xkb/xkbEvents.c
(gdb) bt
#0  XkbSendNotification (kbd=0x8a2650, pChanges=0x7ffffcf1ab00, 
    cause=<value optimized out>) at ../../xkb/xkbEvents.c:770
#1  0x00000000005401df in XkbApplyMappingChange (kbd=0x8a2650, request=32 ' ', 
    firstKey=<value optimized out>, num=248 'ø', client=0x9a9320)
    at ../../xkb/xkbUtils.c:580
#2  0x0000000000526ec1 in SendDeviceMappingNotify (client=0x42, 
    request=0 '\0', firstKeyCode=0 '\0', count=1 '\001', dev=0x8a2650)
    at ../../Xi/exevents.c:966
#3  0x000000000044641e in ProcSetModifierMapping (client=0x9a9320)
    at ../../dix/devices.c:1382
#4  0x000000000044e300 in Dispatch () at ../../dix/dispatch.c:502
#5  0x0000000000436add in main (argc=1, argv=0x7ffffcf1b208, 
    envp=<value optimized out>) at ../../dix/main.c:452
Comment 1 martin f. krafft 2008-03-24 06:13:34 UTC
Created attachment 15428 [details]
log output

This log file includes me telling gdb to 'continue' after the segfault, resulting the the SIGABRT.
Comment 2 Andraž \'ruskie\' Levstik 2008-08-23 00:55:27 UTC
Created attachment 18474 [details]
The log

I have exactly the same issue with an aiptek tablet
Comment 3 Andraž \'ruskie\' Levstik 2008-08-23 02:01:39 UTC
Well not running xmodmap certainly helps I'm guessing xmodmap in this case would be like using a hammer where a fine screwdriver would suffice. Or in other words
the tablets tend to have some keys on them and I'm guessing the driver for them
doesn't take into account that someone could use xmodmap with them...
Comment 4 Peter Hutterer 2009-09-02 21:55:27 UTC
I can't remember the exact commit but I fixed this a number of months ago. This seems to work fine now.


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.