Created attachment 6713 [details]
Edited version of Xorg.0.log

I have edited Xorg.0.log to show the intial setup and a few of the BUG lines including the final segfault.

Corrected misspelling in Title.

Summary: X crashing and also throwing huge number of errors on Dexx X13 9343 machine. => X crashing and also throwing huge number of errors on Dell X13 9343 machine.

OK, I finally managed to get a number of Xorg.0.log files from when X crashes. 
I have put one up as an attachment. In all cases /etc/X11/X segfaults and then dies.

This is still a critical fault, which needs to be fixed.

Summary: X crashing and also throwing huge number of errors on Dell X13 9343 machine. => X segfaulting and also throwing huge number of errors on Dell X13 9343 machine.
Source RPM: X11 (?) => x11-server-xorg

Created attachment 6776 [details]
Xorg.o.log.old file from just after an X crash.

Here is the Xorg.0.log.old from just after X crashes with a segfault.
I am not sure if the original attachment is actually from after such a crash, but am sure of this one. I have another 6 examples all of which show the segfault similar to this one. 
x11-server-xorg-1.16.4-2.mga5

Attachment 6713 is obsolete: 0 => 1

(changed version to 5 from cauldron.)

Version: Cauldron => 5

Are there interesting kernel messages too?

Thierry and Thomas, is it likely to be an XOrg bug or rather a kernel one?

CC: (none) => tmb
Summary: X segfaulting and also throwing huge number of errors on Dell X13 9343 machine. => X segfaulting on Dell X13 9343 machine (BUG: synaptics.c:3020 in UpdateTouchState())

It's also seen on RH, Suse & Ubuntu (for quite some time):
https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-input-synaptics/+bug/1172094
https://bugzilla.novell.com/show_bug.cgi?id=818883
https://bugzilla.redhat.com/show_bug.cgi?id=1194990

Can you try rebuilding latest x11-driver-input-synaptics SRPM from cauldron for mga5 and see if it's better?
http://distrib-coffee.ipsl.jussieu.fr/pub/linux/Mageia//distrib/cauldron/SRPMS/core/release/x11-driver-input-synaptics-1.8.2-1.mga6.src.rpm

Upstream changes are:
One significant change here: an artificial delay is now used between
the button events of a doubletap. Some applications previously dropped the
events because they had the same timestamp, with the delay this should now
work fine.

Gabriele Mazzotta (2):
      Update machine state diagram
      Add a delay between the second button down-up event of double taps

Peter Hutterer (3):
      eventcomm: prevent possibly division by zero
      eventcomm: ignore fake and broken MT devices


If it still shows the bug, can you rebuild it without the patches and try it?

Keywords: (none) => NEEDINFO
Source RPM: x11-server-xorg => kernel, x11-driver-input-synaptics

Will try. The machine itself is 500 miles away now, so I have to get my colleague
to test the package I have compile. Is this driver loaded at boot, or just by X?
(this is also why it is hard to see if the kernel also has bug logs at the same time).


Reading the changelog, those patches seem to be the difference between 1.8.1 and 1.8.2 and thus without the patches it would seem to be identical to the Mageia 5 version which does not work. Or am I misunderstanding.

Our 1.8.1 pacakge has patches that are no upstream.
So there's 2 things you can try:
- rebuilding 1.8.1 w/o those patches
- updating to 1.8.2
(hoping that one of those would work)

Have recompiled 1.8.2 for mga5, and installed it. So far so good-- no synaptics bugs after about 3 hours, mind you not very heavy use.

Will keep a watch on it for a couple of days.

Nope.Still Problems. So fqr no segfault, but 23 MB today of error messages like


(EE) [dix] DLL0665:01 06CB:76AD UNKNOWN: unable to find touch point 0
(EE) [dix] DLL0665:01 06CB:76AD UNKNOWN: unable to find touch point 0
(EE) [dix] DLL0665:01 06CB:76AD UNKNOWN: unable to find touch point 0
(EE) BUG: triggered 'if (priv->num_active_touches > priv->num_slots)'
(EE) BUG: synaptics.c:3017 in UpdateTouchState()
(EE) 
(EE) Backtrace:
(EE) 0: /etc/X11/X (xorg_backtrace+0x49) [0x58b709]
(EE) 1: /usr/lib64/xorg/modules/input/synaptics_drv.so (0x7f0a7d400000+0x520e) [0x7f0a7d40520e]
(EE) 2: /usr/lib64/xorg/modules/input/synaptics_drv.so (0x7f0a7d400000+0x74a2) [0x7f0a7d4074a2]
(EE) 3: /etc/X11/X (0x400000+0x73e68) [0x473e68]
(EE) 4: /etc/X11/X (0x400000+0x9dc53) [0x49dc53]
(EE) 5: /lib64/libc.so.6 (0x7f0a83b53000+0x336b0) [0x7f0a83b866b0]
(EE) 6: /lib64/libc.so.6 (__select+0x33) [0x7f0a83c40c63]
(EE) 7: /etc/X11/X (WaitForSomething+0x19c) [0x588b2c]
(EE) 8: /etc/X11/X (0x400000+0x36aa1) [0x436aa1]
(EE) 9: /etc/X11/X (0x400000+0x3acb6) [0x43acb6]
(EE) 10: /lib64/libc.so.6 (__libc_start_main+0xf0) [0x7f0a83b72fd0]
(EE) 11: /etc/X11/X (0x400000+0x2618e) [0x42618e]

They look slightly different than the old ones- different addresses in synaptics_drv.so, but that is probably just the recompiline etc.

Will try the one without the 2 mageia patches.

Note that the problem seems to occur after a hibernation.




Here

It was 22000 occurances in 10 hrs.

Another data point. I also compiled 1.8.2 without the two Mageia patches, and that
version was tried. The button clicks/taps on the trackpad ceased to work. The trackpad scrolling worked, but tapping did not.

Thus 1.8.2 with the 4 patches (2 Fedora and 2 Mageia) gives huge numbers of error messages after hibernation. Before hibernation, those error messages do not seem to be there.
1.8.2 with only the two Fedora patches loses the tapping/button functionality.
Have not tried to see what would happen if one hibernated and woke up that version.

I can confirm that the version from cauldron doesn't fix the issue, no matter whether the additional patches are disabled or not. But there is a patch available that fixes the problem for me

https://bugs.freedesktop.org/show_bug.cgi?id=88903

It doesn't apply cleanly because of other changes in the meantime, that's why I'll attach a rediffed version that  applies cleanly to mageia's 1.8.1 and 1.8.2 packages.

I'm running a 1.8.2 version with the patch for a couple of days now (along with the "disable psmouse as much as possible" configuration from bug#15660#c4 without the flooding in the logs, and more important: no x-segfault after a while.

Keywords: NEEDINFO => PATCH
URL: (none) => https://bugs.freedesktop.org/show_bug.cgi?id=88903

Created attachment 6929 [details]
patch to fix the problem - taken from fdo#88903 and just rediffed to apply

@ w unruh,

Thank you for having taken the needed time to report this issue!

@ Christian

Thanks for the patch :-)

@ All,

Did this bug get fixed? If so, please change its status to RESOLVED - FIXED

If it didn't, then we regret that we weren't able to fix it in Mageia 5. Mageia 5 has officially reached its End of Life on December 31st, 2017 https://blog.mageia.org/en/2017/11/07/mageia-5-eol-postponed/
It only continued to get important security updates since then, because we are waiting for a big Plasma5 update in Mageia 6, that'll fix many of the Mageia 5 => 6 upgrade issues.

If you haven't seen that this bug got fixed, then please check whether this bug still exists in Mageia 6. If it does, then please change the Version (near the top, at the left) to "6". If you know it exists in Cauldron, then change Version to Cauldron. If you see it in both Cauldron and Mageia 6, then please set Version to Cauldron and add MGA6TOO on the Whiteboard.

Thanks,
Marja

CC: (none) => marja11

FYI: I don't own the XPS13 anymore - (and new one uses libinput) - so cannot help whether it is actually fixed (except checking whether the abovementioned patch is included in the package...)

(In reply to Christian Lohmaier from comment #17)
> FYI: I don't own the XPS13 anymore - (and new one uses libinput) - so cannot
> help whether it is actually fixed (except checking whether the
> abovementioned patch is included in the package...)

No problem :-)

No one else replied, so we don't seem to need to keep this report open for Mageia 6.

Closing as OLD, since Mga5 is no longer maintained.

Status: NEW => RESOLVED
Resolution: (none) => OLD