Xubuntu 16.10: drm errors causing slow boot

[Jeff]

I have been getting drm errors since updating to 16.10 causing slow boot times. Ex:

  3.830465] jhm-Inspiron-1520 kernel: input: SynPS/2 Synaptics TouchPad as /devices/platform/i8042/serio1/input/input5

[ 13.792249] jhm-Inspiron-1520 kernel: [drm:drm_atomic_helper_commit_cleanup_done [drm_kms_helper]] *ERROR* [CRTC:26:pipe A] flip_done timed out

There is ten second gap in the above adjacent lines. I have a few of these in the journal log.
These errors also cause Xorg to stall during mode setting also, as seen in the adjacent Xorg.0.log file.

[ 55.988] (II) modeset(0): Output VGA-1 has no monitor section
[ 66.615] (II) modeset(0): Output SVIDEO-1 has no monitor section

Using the Ubuntu test kernels, the best result I have gotten is from kernel 4.5.6. At 4.6.0 the wheels start to come off the wagon. At 4.9.0 rc1 the boot stopped at the busybox initrd shell.

4.8.0 fails decompressing the hibernate image as well.

ProblemType: Bug
DistroRelease: Ubuntu 16.10
Package: linux-image-4.8.0-26-generic 4.8.0-26.28
ProcVersionSignature: Ubuntu 4.8.0-26.28-generic 4.8.0
Uname: Linux 4.8.0-26-generic i686
ApportVersion: 2.20.3-0ubuntu8
Architecture: i386
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: jhm 1262 F.... pulseaudio
CurrentDesktop: XFCE
Date: Thu Oct 27 17:02:19 2016
EcryptfsInUse: Yes
HibernationDevice: RESUME=UUID=bd7562ce-d720-4e95-995e-48f0ec1a77aa
InstallationDate: Installed on 2016-10-12 (15 days ago)
InstallationMedia: Xubuntu 16.04 LTS "Xenial Xerus" - Release i386 (20160420.1)
MachineType: Dell Inc. Inspiron 1520
ProcFB: 0 inteldrmfb
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-4.8.0-26-generic root=UUID=5dc9cf92-93d8-4e84-826d-6b4d5f6e1c0d ro plymouth:debug=1 vesafb.invalid=1
RelatedPackageVersions:
 linux-restricted-modules-4.8.0-26-generic N/A
 linux-backports-modules-4.8.0-26-generic N/A
 linux-firmware 1.161
RfKill:
 0: phy0: Wireless LAN
  Soft blocked: no
  Hard blocked: no
SourcePackage: linux
UpgradeStatus: Upgraded to yakkety on 2016-10-15 (12 days ago)
dmi.bios.date: 02/03/2008
dmi.bios.vendor: Dell Inc.
dmi.bios.version: A07
dmi.board.name: 0KY767
dmi.board.vendor: Dell Inc.
dmi.chassis.type: 8
dmi.chassis.vendor: Dell Inc.
dmi.modalias: dmi:bvnDellInc.:bvrA07:bd02/03/2008:svnDellInc.:pnInspiron1520:pvr:rvnDellInc.:rn0KY767:rvr:cvnDellInc.:ct8:cvr:
dmi.product.name: Inspiron 1520
dmi.sys.vendor: Dell Inc.
---
ApportVersion: 2.20.3-0ubuntu8
Architecture: i386
DistroRelease: Ubuntu 16.10
EcryptfsInUse: Yes
InstallationDate: Installed on 2016-10-12 (15 days ago)
InstallationMedia: Xubuntu 16.04 LTS "Xenial Xerus" - Release i386 (20160420.1)
Package: linux (not installed)
ProcEnviron:
 LANGUAGE=en_US
 TERM=xterm
 PATH=(custom, no user)
 LANG=en_US.UTF-8
 SHELL=/bin/bash
Tags: yakkety
Uname: Linux 4.6.0-040600rc1-generic i686
UnreportableReason: The running kernel is not an Ubuntu kernel
UpgradeStatus: Upgraded to yakkety on 2016-10-15 (12 days ago)
UserGroups:

_MarkForUpload: True

[Brad Figg (brad-figg)]

This change was made by a bot.

[Joseph Salisbury (jsalisbury)]

Can you give the v4.6-rc1 kernel a test? We can perform a kernel bisect if we know the last good kernel version and the first bad one.

The 4.6-rc1 kernel can be downloaded from:
http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.6-rc1-wily/

[Jeff (boardman-malibu)]

apport information

[Jeff (boardman-malibu)]

Tested 4.6.0-rc1 and drm errors appear in the journal log. I am attaching a copy of the journal using monotonic time stamps, since they show a correlation between drm errors and Xorg activity.
Question: Is there a way to get apport to use monotonic journal timestamps and..
          Is there a way to get apport to include Xorg.0.log (or any file) in a report..or
          failing that, is there a way to attach more the one file without archiving them?

[Jeff (boardman-malibu)]

Here is the corresponding Xorg.log showing a 13 second delay near the end. The modesetting timestamps correspond to the second batch of drm errors in the journal.log

[Jeff (boardman-malibu)]

I'm attaching a Xorg.log from my Xubuntu 14.04.5 LTS installation on the same machine, showing a start time of 29.904 seconds and a finish time of 32.573 , a span of only 2.669 seconds.

[Jeff (boardman-malibu)]

v4.6.0-rc1 boot test observations.. Booting without quiet,splash

Boot text at low resolution until this point: (journalctl -o short-monotonic)

    3.306180] jhm-Inspiron-1520 kernel: [drm] Initialized drm 1.1.0 20060810
    (Display goes black here)

Boot text returns at high resolution at this point: journalctl -o short-monotonic)

    5.385111] jhm-Inspiron-1520 kernel: Console: switching to colour frame buffer device 160x50
    (Text returns with colors on)
    (With later kernel versions this text return time gets later and later..25 seconds for v4.8.0

Screen goes black here: (Xorg.0.log)

    [ 32.662]
    X.Org X Server 1.18.4
    Release Date: 2016-07-19

Screen switches off for a second here: (Xorg.0.log): (It seems every version I've ever used turns off the display before going to the GUI...annoying)

    [ 49.710] (II) modeset(0): Modeline "1280x800"x0.0 75.50 1280 1344 1392 1512 800 803 809 832 -hsync -vsync (49.9 kHz eP)

After this the Xubuntu background and login appear.

[Jeff (boardman-malibu)]

While I had the time I checked 4.9.0-rc2, attaching a file showing early boot errors. The boot was confined to the initramfs only.

[Federico Leoni (effelle-gmail)]

It's not restricted to Xubuntu, but this bug affects all *buntu I've tested in the last week. Now I'm testing Ubuntu Budgie on a notebook HP 2910us and kernel 4.10.0-19-generic.

[Jeff (boardman-malibu)]

Follow Up:
Adding the option 'video=SVIDEO-1:d' (disables SVIDEO-1) resolved the problem in my case. Seems the i915 forgot how to handle this port. Something about not being 'hot-pluggable' unless kept fully powered, does not comply with the direction they are going?

[Chris Tallon (chris-loggytronic)]

video=SVIDEO-1:d fixed my laptop. Well found!

[Joseph Salisbury (jsalisbury)]

This issue appears to be an upstream bug, since you tested the latest upstream kernel. Would it be possible for you to open an upstream bug report[0]? That will allow the upstream Developers to examine the issue, and may provide a quicker resolution to the bug.

Please follow the instructions on the wiki page[0]. The first step is to email the appropriate mailing list. If no response is received, then a bug may be opened on bugzilla.kernel.org.

Once this bug is reported upstream, please add the tag: 'kernel-bug-reported-upstream'.

[0] https://wiki.ubuntu.com/Bugs/Upstream/kernel

[TJ (tj)]

This has *just* had a fix committed upstream and is hoped to be available in v5.0. Could we get it backported to the LTS kernels?

commit ed20151a7699bb2c77eba3610199789a126940c4
Author: Ville Syrjälä <email address hidden>
Date: Tue Nov 27 20:20:04 2018 +0200

    drm/vblank: Allow dynamic per-crtc max_vblank_count

commit 32db0b6501d97b09e92e70caefc74fa35aa9a8d6
Author: Ville Syrjälä <email address hidden>
Date: Tue Nov 27 22:05:50 2018 +0200

    drm/i915: Don't try to use the hardware frame counter with i965gm TV output