Bug 1206733 - Regression from Gnome X Session after upgrading Fedora 21 to Fedora 22 Beta
Summary: Regression from Gnome X Session after upgrading Fedora 21 to Fedora 22 Beta
Keywords:
Status: CLOSED WORKSFORME
Alias: None
Product: Fedora
Classification: Fedora
Component: xorg-x11-server
Version: 22
Hardware: x86_64
OS: Linux
high
high
Target Milestone: ---
Assignee: X/OpenGL Maintenance List
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2015-03-27 22:21 UTC by Luya Tshimbalanga
Modified: 2015-04-03 06:38 UTC (History)
4 users (show)

Fixed In Version: 1.17.1-7.fc22
Clone Of:
Environment:
Last Closed: 2015-04-03 06:38:33 UTC
Type: Bug
Embargoed:


Attachments (Terms of Use)
Journal output (443.44 KB, text/plain)
2015-03-28 00:10 UTC, Luya Tshimbalanga
no flags Details
X log output (35.01 KB, text/plain)
2015-03-28 00:12 UTC, Luya Tshimbalanga
no flags Details
updated boot journal (461.64 KB, text/plain)
2015-03-28 17:44 UTC, Luya Tshimbalanga
no flags Details
Screenshot with laptop spec running on Wayland (39.56 KB, image/png)
2015-03-28 17:46 UTC, Luya Tshimbalanga
no flags Details
Newere X.log output (43.42 KB, text/plain)
2015-03-30 18:27 UTC, Luya Tshimbalanga
no flags Details
Journal output from Fedora 22 Beta TC6 Live media (331.34 KB, text/plain)
2015-04-01 01:00 UTC, Luya Tshimbalanga
no flags Details
Latest dmesg from F22Beta TC6 (100.13 KB, text/plain)
2015-04-01 01:01 UTC, Luya Tshimbalanga
no flags Details
Journal report from Fedora 21 (161.79 KB, text/plain)
2015-04-02 16:55 UTC, Luya Tshimbalanga
no flags Details
New X.log output after xserver update 1.17.1-7.fc22 (48.06 KB, text/plain)
2015-04-03 06:36 UTC, Luya Tshimbalanga
no flags Details

Description Luya Tshimbalanga 2015-03-27 22:21:14 UTC
Description of problem:
Booting Fedora 22 Alpha Design Suite (based on Workstation) led dropping user session and run Plymouth Quit for a while. Starting Fedora in Basic Graphic mode runs fine.

Version-Release number of selected component (if applicable):


How reproducible:


Steps to Reproduce:
1. Simply boot Fedora Live Media 22
2.
3.

Actual results:
Black screen with Plymouth message Quitting for a while

Expected results:
Should boot normally on Gnome Shell

Additional info:
Test run on Asus X550Z with AMD Quad-Core A10 7400P and Dual Graphic Radeon R6.

Comment 1 Luya Tshimbalanga 2015-03-28 00:10:34 UTC
Created attachment 1007547 [details]
Journal output

Comment 2 Luya Tshimbalanga 2015-03-28 00:12:44 UTC
Created attachment 1007548 [details]
X log output

Comment 3 Luya Tshimbalanga 2015-03-28 17:44:31 UTC
Created attachment 1007742 [details]
updated boot journal

Journal boot report without nomodeset. The system goes into Wayland login screen where I type my password. For some odd reason, the default GNOME session crashed and goes back to login screen where I switched to GNOME Wayland that worked smoothly.
This is the first time I ever see GNOME Wayland working better than the default GNOME session.

Comment 4 Luya Tshimbalanga 2015-03-28 17:46:04 UTC
Created attachment 1007743 [details]
Screenshot with laptop spec running on Wayland

Comment 5 Luya Tshimbalanga 2015-03-30 18:27:07 UTC
Created attachment 1008636 [details]
Newere X.log output

New X.log output including traceback

Comment 6 Luya Tshimbalanga 2015-04-01 01:00:09 UTC
Created attachment 1009354 [details]
Journal output from Fedora 22 Beta TC6 Live media

Booting Fedora 22 Beta TC6, same problem. Looking at the journal, here is the highlight:

Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (EE) /dev/dri/card1: failed to set DRM interface version 1.4: Permission denied

Further below at Mar 13 13:16:27 from the journal, backtrace occurred due to multiple addition of device by libinput. Not sure if the bug is to the driver or libinput because ASUS X550ZE is running fine on Wayland session. With limited skill, I am unable to find the cause of this bug.

Comment 7 Luya Tshimbalanga 2015-04-01 01:01:35 UTC
Created attachment 1009356 [details]
Latest dmesg from F22Beta TC6

Comment 8 Luya Tshimbalanga 2015-04-01 04:02:56 UTC
Looking at the journalctl report

Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (**) ETPS/2 Elantech Touchpad: Applying InputClass "evdev touchpad catchall"
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (**) ETPS/2 Elantech Touchpad: Applying InputClass "touchpad catchall"
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (**) ETPS/2 Elantech Touchpad: Applying InputClass "Default clickpad buttons"
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (**) ETPS/2 Elantech Touchpad: Applying InputClass "libinput touchpad catchall"
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (II) systemd-logind: got fd for /dev/input/event6 13:70 fd 27 paused 0
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (II) Using input driver 'libinput' for 'ETPS/2 Elantech Touchpad'
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (**) ETPS/2 Elantech Touchpad: always reports core events
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (**) Option "Device" "/dev/input/event6"
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (II) input device 'ETPS/2 Elantech Touchpad', /dev/input/event6 is tagged by udev as: Touchpad
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (II) input device 'ETPS/2 Elantech Touchpad', /dev/input/event6 is a touchpad
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (**) Option "config_info" "udev:/sys/devices/platform/i8042/serio4/input/input12/event6"
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (II) XINPUT: Adding extended input device "ETPS/2 Elantech Touchpad" (type: TOUCHPAD, id 14)
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (**) Option "AccelerationScheme" "none"
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (**) ETPS/2 Elantech Touchpad: (accel) selected scheme none/0
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (**) ETPS/2 Elantech Touchpad: (accel) acceleration factor: 2.000
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (**) ETPS/2 Elantech Touchpad: (accel) acceleration threshold: 4
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (II) input device 'ETPS/2 Elantech Touchpad', /dev/input/event6 is tagged by udev as: Touchpad
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (II) input device 'ETPS/2 Elantech Touchpad', /dev/input/event6 is a touchpad
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (II) config/udev: Adding input device ETPS/2 Elantech Touchpad (/dev/input/mouse0)
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (II) No input driver specified, ignoring this device.
Mar 31 13:16:26 localhost /usr/libexec/gdm-x-session[1650]: (II) This device may have been added with another device file.
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE)
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) Backtrace:
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) 0: /usr/libexec/Xorg (OsLookupColor+0x139) [0x599dd9]
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) 1: /lib64/libc.so.6 (__restore_rt+0x0) [0x7fb0c2284b1f]
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) 2: /lib64/libc.so.6 (__GI___strcmp_ssse3+0x16) [0x7fb0c23983d6]
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) 3: /usr/libexec/Xorg (xf86SIGIOSupported+0xa18) [0x4a36b8]
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) 4: /lib64/libdbus-1.so.3 (dbus_connection_dispatch+0x375) [0x7fb0c44a8095]
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) 5: /lib64/libdbus-1.so.3 (dbus_connection_dispatch+0x64d) [0x7fb0c44a891d]
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) 6: /usr/libexec/Xorg (config_fini+0x4c1) [0x49d7b1]
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) 7: /usr/libexec/Xorg (WakeupHandler+0x6d) [0x43ed3d]
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) 8: /usr/libexec/Xorg (WaitForSomething+0x1e7) [0x592ea7]
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) 9: /usr/libexec/Xorg (SendErrorToClient+0x111) [0x439f81]
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) 10: /usr/libexec/Xorg (remove_fs_handlers+0x41b) [0x43e26b]
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) 11: /lib64/libc.so.6 (__libc_start_main+0xf0) [0x7fb0c2270790]
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) 12: /usr/libexec/Xorg (_start+0x29) [0x428659]
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) 13: ? (?+0x29) [0x29]
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE)
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE) Segmentation fault at address 0x0
Mar 31 13:16:27 localhost /usr/libexec/gdm-x-session[1650]: (EE)

It seems to be a problem with libinput by attempting to enable the touchpad. Assigning to the right maintainer.

Comment 9 Luya Tshimbalanga 2015-04-02 16:55:33 UTC
Created attachment 1010267 [details]
Journal report from Fedora 21

Adding a journal report from Fedora 21. It now appears a regression from updating to xorg-x11-servers to 1.17 because xorg-x11-ati is stll 7.5.0. The thought about libinput causing issue was incorrect, Based on the comparison of both journal, the bug report is assigned to xorg-x11-server.

Comment 10 Luya Tshimbalanga 2015-04-03 06:36:32 UTC
Created attachment 1010501 [details]
New X.log output after xserver update 1.17.1-7.fc22

Latest update announced on http://hansdegoede.livejournal.com/15767.html addresses this bug. As explained by AdamW, it turned out a permission issue from xorg-x11-server-Xorg leading the bracktrace on Gnome Session running on, this case, AMD X550ZE, an AMD Kaveri powered laptop.

Comment 11 Luya Tshimbalanga 2015-04-03 06:38:33 UTC
Closing this bug now the functionality is restored.


Note You need to log in before you can comment on or make changes to this bug.