i915 multiple display halt system when moving windows

Bug #1647084 reported by Richard on 2016-12-03
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
mutter (Ubuntu)
High
Unassigned

Bug Description

When moving certain windows across displays, the system may suddenly halt and has to be powered off

SYMPTOM
Either the window being dragged suddenly jumps to the top-left position of the external display to the right (ie. display 2, the first external display) and the system halts, or the system halts while the window is being dragged across the internal display

Affected applications:
Virtual Machine Manager, Vuirtual Machine window
LibreOffice Writer

Other problem
If a Libre Office Writer window is opened on the internal display, ie by launching LibreOffice Writer, the system halts.

IMPACT
Critical: in-app data loss, fsck data loss, forced system reboot

GETAROUND
Do not drag halting apps between displays
To move a LibreOffice Writer window, close it, and launch LibreOffice from a terminal window on the destination display. On command libreoffice --writer, LibreOffice opens on the display where
the terminal window is located.
For Virtual machines, close the window, move virtual machine manager to the destination display and open the virtual machine window there.

The halt is similar to how the system halts when the kernel command line parameter i915.enable_rc6=0 is missing.

ProblemType: Bug
DistroRelease: Ubuntu 16.10
Package: linux-image-4.8.0-28-generic 4.8.0-28.30
ProcVersionSignature: Ubuntu 4.8.0-28.30-generic 4.8.6
Uname: Linux 4.8.0-28-generic x86_64
ApportVersion: 2.20.3-0ubuntu8
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC1: foxyboy 4662 F.... pulseaudio
 /dev/snd/controlC0: foxyboy 4662 F.... pulseaudio
CurrentDesktop: GNOME
Date: Sat Dec 3 11:12:07 2016
HibernationDevice: RESUME=/dev/mapper/C89-SWAP
MachineType: Apple Inc. MacBookPro12,1
ProcFB: 0 inteldrmfb
ProcKernelCmdLine: BOOT_IMAGE=/vmlinuz-4.8.0-28-generic root=/dev/mapper/C89-RFS ro i915.enable_rc6=0
RelatedPackageVersions:
 linux-restricted-modules-4.8.0-28-generic N/A
 linux-backports-modules-4.8.0-28-generic N/A
 linux-firmware 1.161
SourcePackage: linux
UpgradeStatus: Upgraded to yakkety on 2016-11-16 (17 days ago)
dmi.bios.date: 10/26/2015
dmi.bios.vendor: Apple Inc.
dmi.bios.version: MBP121.88Z.0167.B15.1510261437
dmi.board.name: Mac-E43C1C25D4880AD6
dmi.board.vendor: Apple Inc.
dmi.board.version: MacBookPro12,1
dmi.chassis.type: 9
dmi.chassis.vendor: Apple Inc.
dmi.chassis.version: Mac-E43C1C25D4880AD6
dmi.modalias: dmi:bvnAppleInc.:bvrMBP121.88Z.0167.B15.1510261437:bd10/26/2015:svnAppleInc.:pnMacBookPro12,1:pvr1.0:rvnAppleInc.:rnMac-E43C1C25D4880AD6:rvrMacBookPro12,1:cvnAppleInc.:ct9:cvrMac-E43C1C25D4880AD6:
dmi.product.name: MacBookPro12,1
dmi.product.version: 1.0
dmi.sys.vendor: Apple Inc.

Richard (ismail-a) wrote :
Richard (ismail-a) wrote :

uname --nodename --kernel-release --kernel-version && date --utc --rfc-3339=date && lsb_release --description --codename --short && dpkg --status libwayland-server0 | egrep "(^V|^Pa)" && dpkg --status gnome-session | egrep "(^V|^Pa)"
c89 4.8.0-28-generic #30-Ubuntu SMP Fri Nov 11 14:03:52 UTC 2016
2016-12-03
Ubuntu 16.10
yakkety
Package: libwayland-server0
Version: 1.11.0-2
Package: gnome-session
Version: 3.20.2-1ubuntu7

This change was made by a bot.

Changed in linux (Ubuntu):
status: New → Confirmed
Joseph Salisbury (jsalisbury) wrote :

Did this issue start happening after an update/upgrade? Was there a prior kernel version where you were not having this particular problem?

Would it be possible for you to test the latest upstream kernel? Refer to https://wiki.ubuntu.com/KernelMainlineBuilds . Please test the latest v4.9 kernel[0].

If this bug is fixed in the mainline kernel, please add the following tag 'kernel-fixed-upstream'.

If the mainline kernel does not fix this bug, please add the tag: 'kernel-bug-exists-upstream'.

Once testing of the upstream kernel is complete, please mark this bug as "Confirmed".

Thanks in advance.

[0] http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.9-rc8

Changed in linux (Ubuntu):
importance: Undecided → High
status: Confirmed → Incomplete
tags: added: kernel-da-key
Richard (ismail-a) wrote :

It’s 16.10/GNOME/Wayland, did not happen in 16.04
I was previously on 16.10 KDE/X where it does not happen either.
When Wayland is used, certain actions halts i915 and the cpu.

When GNOME 3.20 was in staging, LibreOffice was so bad it could not be used. Now it only halts when a Libreoffice window is dragged across the hidpi display. If that isn't done, the system does not halt.

It is practically certain to be an i915 dri problem

It may be related to GtkStyleContext and having multiple displays both low dpi and hidpi

These two applications, when in gtk3, do something that halts in i915

One could fix the bug in i915 or make LibreOffice not do whatever it is that causes the halt

Richard (ismail-a) wrote :

This is fixed in mutter 3.23.3

affects: linux (Ubuntu) → mutter (Ubuntu)
tags: removed: kernel-da-key
Richard (ismail-a) wrote :

Even with updated mutter version 3.22.2-2ubuntu1~ubuntu16.10.1
that was just built at
http://ppa.launchpad.net/gnome3-team/gnome3-staging/ubuntu

this bug is present in GNOME 3.22/Wayland

No fix yet

Launchpad Janitor (janitor) wrote :

[Expired for mutter (Ubuntu) because there has been no activity for 60 days.]

Changed in mutter (Ubuntu):
status: Incomplete → Expired
Richard (ismail-a) on 2017-02-20
Changed in mutter (Ubuntu):
status: Expired → New
status: New → Confirmed
Richard (ismail-a) wrote :

The freeze seems to apply to all xwayland windows being dragged across internal HiDPI display

It is unlikely to be kernel related, because system stability is not affected, just the GNOME session analogous with the myriad of mutter freezes

All LibreOffice dialogs just went to 25% display, so mutter are not testing for HiDPI. Another reason to not use GNOME 3.22/Wayland

Richard (ismail-a) wrote :

Correction to the initial report: this was separate from the i915 problems that halts the system. Although it looks the same, in this case the system is still running, so gdm can be restarted using ssh from another computer.

The i915 problem turned out to be a required kernel parameter i915.enable_rc6=0 without which mouse cursor changes will randomly halt the system.

mutter has lots of flickering and freeze risk in the magic area 40 px into an external display bordering the hidpi one. bugs.

I went back to X that now actually works for multiple-display. I’ll try Wayland again for 3.24

To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers