Created attachment 130969 [details]
dmesg with drm.debug=0x1e log_buf_len=1M with xorg

Created attachment 130970 [details]
xrandr --verbose

Created attachment 130971 [details]
dpkg libreoffice

My laptop is a DELL XPS 13 9345

A new hang but this time with XWayland (on the same laptop)

[77401.948200] [drm] GPU HANG: ecode 9:0:0x85dffffb, in Xwayland [17361], reason: Hang on render ring, action: reset
[77401.948203] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
[77401.948204] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
[77401.948205] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
[77401.948206] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
[77401.948208] [drm] GPU crash dump saved to /sys/class/drm/card0/error
[77401.948281] drm/i915: Resetting chip after gpu hang
[77401.948346] [drm] RC6 on
[77401.964901] [drm] GuC firmware load skipped
[77413.947343] drm/i915: Resetting chip after gpu hang
[77413.947410] [drm] RC6 on
[77413.963504] [drm] GuC firmware load skipped

Created attachment 131037 [details]
GPU hang error with xwayland

Created attachment 131038 [details]
GPU hang dmesg with xwayland

On the xwayland case, I was in working with LibreOffice Writer and Calc with a graph (copy/paste the graph from Calc to Writer)

Oh! a new GPU Hang with Xwayland just after a reboot of my laptop, a update of mesa (from Debian 9, 13.0.6-1 --> 13.0.6-1+b1), and with only LibreOffice Writer open.

(in this case, I've the debug of drm enable, see attachment)


# dpkg -l|grep mesa
ii  libegl1-mesa:amd64                    13.0.6-1+b1                          amd64        free implementation of the EGL API -- runtime
ii  libegl1-mesa:i386                     13.0.6-1+b1                          i386         free implementation of the EGL API -- runtime
ii  libgl1-mesa-dri:amd64                 13.0.6-1+b1                          amd64        free implementation of the OpenGL API -- DRI modules
ii  libgl1-mesa-glx:amd64                 13.0.6-1+b1                          amd64        free implementation of the OpenGL API -- GLX runtime
ii  libglapi-mesa:amd64                   13.0.6-1+b1                          amd64        free implementation of the GL API -- shared library
ii  libgles1-mesa:amd64                   13.0.6-1+b1                          amd64        free implementation of the OpenGL|ES 1.x API -- runtime
ii  libgles2-mesa:amd64                   13.0.6-1+b1                          amd64        free implementation of the OpenGL|ES 2.x API -- runtime
ii  libglu1-mesa:amd64                    9.0.0-2.1                            amd64        Mesa OpenGL utility library (GLU)
ii  libwayland-egl1-mesa:amd64            13.0.6-1+b1                          amd64        implementation of the Wayland EGL platform -- runtime
ii  libwayland-egl1-mesa:i386             13.0.6-1+b1                          i386         implementation of the Wayland EGL platform -- runtime
ii  mesa-common-dev:amd64                 13.0.6-1+b1                          amd64        Developer documentation for Mesa
ii  mesa-utils                            8.3.0-3                              amd64        Miscellaneous Mesa GL utilities
ii  mesa-vdpau-drivers:amd64              13.0.6-1+b1                          amd64        Mesa VDPAU video acceleration drivers

Created attachment 131039 [details]
/sys/class/drm/card0/error with GPU Hang with Xwayland just after a reboot

Created attachment 131040 [details]
dmesg with drm.debug=0x1e log_buf_len=1M with xwayland just after a reboot

Another hang gpu on my laptop but in this case with RC6 off into grub:
GRUB_CMDLINE_LINUX="i915.enable_rc6=0 drm.debug=0x1e log_buf_len=1M"


Always when I works with libreoffice (on  GNOME/Xwayland)

See attachments *20170426-1231*

Created attachment 131051 [details]
/sys/class/drm/card0/error with GPU Hang with Xwayland with RC6 off - error-20170426-1231

Created attachment 131052 [details]
dmesg with drm.debug=0x1e log_buf_len=1M with xwayland with RC6 off - dmesg-20170426-1231

And another gpu hang with kernel 4.10.7-1~exp1 from Debian Experimental repo and RC6 on.
Always with Libreoffice

Created attachment 131059 [details]
/sys/class/drm/card0/error with GPU Hang with Xwayland and 4.10.7-1~exp1

Created attachment 131060 [details]
dmesg with drm.debug=0x1e log_buf_len=1M with xwayland and kernel 4.10.7-1~exp1

Another GPU Hang with these parameters in grub:

# cat /proc/cmdline
BOOT_IMAGE=/boot/vmlinuz-4.9.0-2-amd64 root=UUID=1a7380f6-83de-4a8a-970f-7f61d9ace0ef ro i915.disable_power_well=0 i915.enable_psr=0 i915.enable_rc6=0 drm.debug=0x1e log_buf_len=1M quiet

Seems this issue is not related to RC6...?

May  3 10:13:46 ender org.gnome.Shell.desktop[6748]: libinput error: kernel bug: Touch jump detected and discarded.
May  3 10:13:46 ender org.gnome.Shell.desktop[6748]: See https://wayland.freedesktop.org/libinput/doc/1.6.3/touchpad_jumping_cursor.html for details
May  3 10:13:53 ender kernel: [402838.974220] [drm] GPU HANG: ecode 9:0:0x85dffffb, in Xwayland [6753], reason: Hang on render ring, action: reset
May  3 10:13:53 ender kernel: [402838.974222] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
May  3 10:13:53 ender kernel: [402838.974223] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
May  3 10:13:53 ender kernel: [402838.974223] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
May  3 10:13:53 ender kernel: [402838.974224] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
May  3 10:13:53 ender kernel: [402838.974226] [drm] GPU crash dump saved to /sys/class/drm/card0/error
May  3 10:13:53 ender kernel: [402838.974319] drm/i915: Resetting chip after gpu hang
May  3 10:13:53 ender kernel: [402838.974419] [drm] RC6 off
May  3 10:13:53 ender kernel: [402838.989766] [drm] GuC firmware load skipped
May  3 10:14:05 ender kernel: [402851.005775] drm/i915: Resetting chip after gpu hang
May  3 10:14:05 ender kernel: [402851.005968] [drm] RC6 off
May  3 10:14:05 ender kernel: [402851.018071] [drm] GuC firmware load skipped
May  3 10:14:05 ender org.gnome.Shell.desktop[6748]: intel_do_flush_locked failed: Input/output error
May  3 10:14:05 ender org.gnome.Nautilus[3565]: X IO Error
May  3 10:14:05 ender firefox-esr.desktop[5055]: firefox-esr: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.
May  3 10:14:05 ender thunderbird.desktop[2521]: thunderbird: Fatal IO error 11 (Ressource temporairement non disponible) on X server :0.
May  3 10:14:05 ender shutter.desktop[10713]: shutter: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.
May  3 10:14:05 ender org.a11y.atspi.Registry[6765]: XIO:  fatal IO error 11 (Resource temporarily unavailable) on X server ":0"
May  3 10:14:05 ender org.a11y.atspi.Registry[6765]:       after 308755 requests (308755 known processed) with 0 events remaining.
May  3 10:14:05 ender gnome-session[6654]: gnome-session-binary[6654]: WARNING: App 'gnome-settings-daemon.desktop' exited with code 1
May  3 10:14:05 ender org.gnome.Shell.desktop[6748]: xcb_connection_has_error() returned true
May  3 10:14:05 ender gnome-settings-[3454]: failed to connect to device: Failed to connect to missing device /org/freedesktop/ColorManager/devices/xrandr_Sharp_Corporation_bru
no_1000
May  3 10:14:05 ender gnome-settings-[20938]: Unable to initialize GTK+

Created attachment 131182 [details]
/sys/class/drm/card0/error with GPU Hang with Xwayland and 4.9.18 from Debian 9

Created attachment 131183 [details]
dmesg with drm.debug=0x1e log_buf_len=1M with xwayland and kernel 4.9.18 from Debian 9

I can confirm this bug on Arch Linux, kernel version 4.11.9-1-ARCH.

When working with LibreOffice, the system freezes spontaneously. I observed this for three times now, twice in LibreOffice Impress, once (today) in LibreOffice Calc. Did nothing special which could have triggered it - no 3D/animation stuff or similar.

The driver I use is the kernel one (not xf86-video-intel). Version of Mesa is 17.1.4-1.

Further information:

$ uname -a
Linux ixion 4.11.9-1-ARCH #1 SMP PREEMPT Wed Jul 5 18:23:08 CEST 2017 x86_64 GNU/Linux

$ cat /proc/cmdline             
initrd=\intel-ucode.img initrd=\initramfs-linux.img cryptdevice=/dev/disk/by-partuuid/0eb66195-acc1-449f-8ee5-7a6e4986b54b:cryptolvm:allow-discards root=/dev/mapper/CryptoLVM-root resume=/dev/mapper/CryptoLVM-swap rootflags=noatime,compress=lzo,ssd,subvol=_root fsck.mode=skip rw

$ lspci
00:00.0 Host bridge: Intel Corporation Skylake Host Bridge/DRAM Registers (rev 08)
00:02.0 VGA compatible controller: Intel Corporation HD Graphics 520 (rev 07)
00:08.0 System peripheral: Intel Corporation Skylake Gaussian Mixture Model
00:14.0 USB controller: Intel Corporation Sunrise Point-LP USB 3.0 xHCI Controller (rev 21)
00:14.2 Signal processing controller: Intel Corporation Sunrise Point-LP Thermal subsystem (rev 21)
00:16.0 Communication controller: Intel Corporation Sunrise Point-LP CSME HECI #1 (rev 21)
00:17.0 SATA controller: Intel Corporation Sunrise Point-LP SATA Controller [AHCI mode] (rev 21)
00:1c.0 PCI bridge: Intel Corporation Device 9d10 (rev f1)
00:1c.2 PCI bridge: Intel Corporation Device 9d12 (rev f1)
00:1f.0 ISA bridge: Intel Corporation Sunrise Point-LP LPC Controller (rev 21)
00:1f.2 Memory controller: Intel Corporation Sunrise Point-LP PMC (rev 21)
00:1f.3 Audio device: Intel Corporation Sunrise Point-LP HD Audio (rev 21)
00:1f.4 SMBus: Intel Corporation Sunrise Point-LP SMBus (rev 21)
00:1f.6 Ethernet controller: Intel Corporation Ethernet Connection I219-LM (rev 21)
02:00.0 Unassigned class [ff00]: Realtek Semiconductor Co., Ltd. RTS522A PCI Express Card Reader (rev 01)
04:00.0 Network controller: Intel Corporation Wireless 8260 (rev 3a)

$ journalctl
[...]
Jul 21 09:39:09 ixion kernel: [drm] GPU HANG: ecode 9:0:0x86dffffd, in Xorg [724], reason: Hang on render ring, action: reset
Jul 21 09:39:09 ixion kernel: [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
Jul 21 09:39:09 ixion kernel: [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
Jul 21 09:39:09 ixion kernel: [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
Jul 21 09:39:09 ixion kernel: [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.
Jul 21 09:39:09 ixion kernel: [drm] GPU crash dump saved to /sys/class/drm/card0/error
Jul 21 09:39:09 ixion kernel: drm/i915: Resetting chip after gpu hang
Jul 21 09:39:09 ixion kernel: [drm] RC6 on
Jul 21 09:39:09 ixion kernel: [drm] GuC firmware load skipped
Jul 21 09:39:21 ixion kernel: drm/i915: Resetting chip after gpu hang
Jul 21 09:39:21 ixion kernel: [drm] RC6 on
Jul 21 09:39:21 ixion kernel: [drm] GuC firmware load skipped
Jul 21 09:39:29 ixion kernel: drm/i915: Resetting chip after gpu hang
Jul 21 09:39:29 ixion kernel: [drm] RC6 on
Jul 21 09:39:29 ixion kernel: [drm] GuC firmware load skippedport 67
Jul 21 09:39:37 ixion kernel: drm/i915: Resetting chip after gpu hang
Jul 21 09:39:37 ixion kernel: [drm] RC6 on
Jul 21 09:39:37 ixion kernel: [drm] GuC firmware load skippedport 67
Jul 21 09:39:51 ixion kernel: drm/i915: Resetting chip after gpu hang
Jul 21 09:39:51 ixion kernel: [drm] RC6 on
Jul 21 09:39:51 ixion kernel: [drm] GuC firmware load skipped
Jul 21 09:39:51 ixion okular[5832]: The X11 connection broke (error 1). Did the X11 server die?
[...]

Created attachment 132811 [details]
/sys/class/drm/card0/error from GPU hang on 2017-07-21 (Xorg)

Created attachment 132812 [details]
xrandr --verbose output from GPU hang on 2017-07-21

Same problem with OpenSUSE Tumbleweed: I got 2 hangs and X.org resets then by a watchdog. I've seen this problem before, usually for me it happened when I'm working with presentation in Impress. I have an example of such presentation where issue might be reproduced with high probability - to reproduce it all you have to do is just start editing it and after some time it will be hanged.

Problem isn't dependent on Libreoffice as I've seen it with different LO version, also I asked my friend to check if he's in trouble with editing presentation in Impress - he said that he has no problem. The only difference between my config and his one is that he's not using i915.

I can say this is a very old problem which I seen with earlier versions of 4.x kernels.

I'm using MS GS40 6QE laptop, that's a Skylake machine. I'm using i915 module from the kernel.

mynote:/home/ivan # uname -a
Linux mynote.levshin.site 4.15.4-1-default #1 SMP PREEMPT Sat Feb 17 09:02:00 UTC 2018 (7169cf1) x86_64 x86_64 x86_64 GNU/Linux

mynote:/home/ivan # rpm -qa|grep -i mesa
Mesa-dri-nouveau-18.0.0-187.1.x86_64
Mesa-libGL-devel-18.0.0-187.1.x86_64
Mesa-libva-18.0.0-187.1.x86_64
Mesa-libEGL-devel-18.0.0-187.1.x86_64
Mesa-libGL1-18.0.0-187.1.x86_64
Mesa-gallium-18.0.0-187.1.x86_64
Mesa-dri-18.0.0-187.1.x86_64
Mesa-18.0.0-187.1.x86_64
Mesa-libEGL1-18.0.0-187.1.x86_64
Mesa-libglapi0-18.0.0-187.1.x86_64

mynote:/home/ivan # rpm -qa | grep -i xorg
xorg-x11-server-wayland-1.19.6-3.1.x86_64
xorg-x11-essentials-7.6_1-16.4.noarch
xorg-x11-libX11-ccache-7.6-20.5.noarch
xorg-x11-fonts-core-7.6-32.3.noarch
xorg-x11-fonts-7.6-32.3.noarch
xorg-x11-server-1.19.6-3.1.x86_64
xorg-x11-driver-input-7.6_1-14.3.noarch
xorg-x11-driver-video-7.6_1-17.1.x86_64

mynote:/home/ivan # rpm -qa | grep -i libreoffice
libreoffice-6.0.1.1-1.3.x86_64
libreoffice-gtk3-6.0.1.1-1.3.x86_64
libreoffice-l10n-ru-6.0.1.1-1.3.noarch
libreoffice-kde4-6.0.1.1-1.3.x86_64
libreoffice-l10n-en-6.0.1.1-1.3.noarch
libreoffice-draw-6.0.1.1-1.3.x86_64
libreoffice-writer-6.0.1.1-1.3.x86_64
libreoffice-base-drivers-mysql-6.0.1.1-1.3.x86_64
libreoffice-writer-extensions-6.0.1.1-1.3.x86_64
libreoffice-calc-6.0.1.1-1.3.x86_64
libreoffice-share-linker-1-4.3.noarch
libreoffice-branding-upstream-6.0.1.1-1.3.noarch
libreoffice-mailmerge-6.0.1.1-1.3.x86_64
libreoffice-pyuno-6.0.1.1-1.3.x86_64
libreoffice-impress-6.0.1.1-1.3.x86_64
libreoffice-filters-optional-6.0.1.1-1.3.x86_64
libreoffice-gnome-6.0.1.1-1.3.x86_64
libreoffice-base-6.0.1.1-1.3.x86_64
libreofficekit-6.0.1.1-1.3.x86_64
libreoffice-icon-themes-6.0.1.1-1.3.noarch
libreoffice-math-6.0.1.1-1.3.x86_64
libreoffice-calc-extensions-6.0.1.1-1.3.x86_64

Created attachment 137588 [details]
dmesg output with an error

Created attachment 137589 [details]
Result of cat /sys/class/drm/card0/error

Created attachment 137590 [details]
Output of xrandr --verbose

Hello everyone, 

Milamber, Felix, this issue may be related to bug 99720, could you please try a newer mesa release, 17.3.5 or latest mesa 18rc version, and at least kernel 4.13 to get more relevant information in the error state? https://www.mesa3d.org

Ivan, your issue seems to be different, could you please open a new bug with the same information and share the sample presentation to try to reproduce? 
Thank you.

Hello Elizabeth,

As suggested in bug 104689 I'm running a current version of mesa from Git since 12 February (f4e85ba93f, replaced 17.3.3 with which I had up to several hangs per day). Using Arch Linux, stock kernel 4.15. No GPU hangs so far. Will report if I'm again experiencing any.

Felix

As a side note why I'm referring to bug 104689: I experienced both "0x86dffffd" and "0x85dffffb" hangs, approximately equally often.

-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/mesa/mesa/issues/1590.