Bug 24337

Summary:	Xorg server crashes in TryClientEvents()
Product:	xorg	Reporter:	Jay Cotton <jay.cotton>
Component:	Server/General	Assignee:	Peter Hutterer <peter.hutterer>
Status:	RESOLVED FIXED	QA Contact:	Xorg Project Team <xorg-team>
Severity:	normal
Priority:	medium	CC:	peter.hutterer, przanoni
Version:	unspecified
Hardware:	Other
OS:	All
Whiteboard:
i915 platform:		i915 features:

Description Jay Cotton 2009-10-05 16:17:08 UTC

This is intermittent, occurs very seldom.


0: /usr/X11/bin/amd64/Xorg'xorg_backtrace+0x33 [0x4b7363]
1: /usr/X11/bin/amd64/Xorg'xf86SigHandler+0x86 [0x4e2a86]
2: /lib/amd64/libc.so.1'__sighndlr+0x6 [0xfffffd7fff2be066]
3: /lib/amd64/libc.so.1'call_user_handler+0x2a7 [0xfffffd7fff2b0f0f]
4: /usr/X11/bin/amd64/Xorg'TryClientEvents+0xad [0x49192d]
5: /usr/X11/bin/amd64/Xorg'DeliverEventsToWindow+0x1a2 [0x491bc2]
6: /usr/X11/bin/amd64/Xorg'DeliverDeviceEvents+0x24a [0x49265a]
7: /usr/X11/bin/amd64/Xorg'ProcessOtherEvent+0x408 [0x59fa28]
8: /usr/X11/bin/amd64/Xorg'ProcessKeyboardEvent+0xc4 [0x5cb224]
9: /usr/X11/bin/amd64/Xorg'mieqProcessInputEvents+0x3c3 [0x53ace3]
10: /usr/X11/bin/amd64/Xorg'ProcessInputEvents+0xd [0x4e24ed]
11: /usr/X11/bin/amd64/Xorg'Dispatch+0x108 [0x4819c8]
12: /usr/X11/bin/amd64/Xorg'main+0x6dc [0x49f87c]
13: /usr/X11/bin/amd64/Xorg'0x71fcc [0x471fcc]

Comment 1 Jay Cotton 2009-10-05 16:18:30 UTC

This patch was added to catch the segv.  pEvents is the first pointer reference
in TryClientEvents and the offset is small.

--- events.c.orig       Mon Oct  5 12:02:59 2009
+++ events.c    Mon Oct  5 12:02:30 2009
@@ -1889,6 +1889,9 @@
     {
        if (grab && !SameClient(grab, client))
            return -1; /* don't send, but notify caller */
+
+       if(!pEvents)  /* guard for segv */
+               return -1;
        type = pEvents->u.u.type;
        if (type == MotionNotify)
        {

Comment 2 Peter Hutterer 2009-10-06 04:30:08 UTC

please try to find a testcase where this occurs. Just fixing up and "ignoring" the crash isn't enough, we need to know if this has wider repercussions.

From the backtrace I can see that it's caused by a key event, that's about as much info as I can tell. Does it happen with the same application each time?

Comment 3 Peter Hutterer 2010-05-04 23:07:23 UTC

any updates on this? Do you still see this issue Jay?

Comment 4 Peter Hutterer 2013-01-11 05:32:49 UTC

Claiming this is fixed since I haven't seen this in a while. Please reopen if it's still here

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.