Intel GMA965 DRM hanging on yakkety

[Fred Vkigg]

Upgrades to Yakkety in feature freeze with kernel 4.8.0-21-generic.

System regularly hangs for 10 second intervals which is very noticeable in boot process. Cause seems to be the Intel GMA965 DRM support causing regular kernel warnings:
 WARNING: CPU: 0 PID: 1183 at /build/linux-JqzU2_/linux-4.8.0/drivers/gpu/drm/drm_irq.c:1224 drm_wait_one_vblank+0x1b6/0x1c0 [drm]

The problem seems to start very early in the boot process, possible already in GRUB, definitely when the kernel image is first started.

Workaround:
Boot into Ubuntu recovery mode and then resume into graphical system. In this case the problem does not occur.

Disabling GRUB graphics with GRUB_TERMINAL=console does not help

ProblemType: Bug
DistroRelease: Ubuntu 16.10
Package: linux-image-4.8.0-21-generic 4.8.0-21.23
ProcVersionSignature: Ubuntu 4.8.0-21.23-generic 4.8.0
Uname: Linux 4.8.0-21-generic x86_64
ApportVersion: 2.20.3-0ubuntu7
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: michael 2377 F.... pulseaudio
Date: Sun Oct 9 11:10:10 2016
HibernationDevice: RESUME=UUID=34b8481d-d92b-4611-820b-c0c2b607f6ab
InstallationDate: Installed on 2012-08-21 (1509 days ago)
InstallationMedia: Ubuntu 12.04 LTS "Precise Pangolin" - Release amd64 (20120425)
MachineType: LENOVO 8943DNG
PccardctlIdent:
 Socket 0:
   no product info available
PccardctlStatus:
 Socket 0:
   no card
ProcEnviron:
 LANGUAGE=en_US
 TERM=xterm-256color
 PATH=(custom, no user)
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcFB: 0 inteldrmfb
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-4.8.0-21-generic root=UUID=2835f778-bc9c-463b-8ee1-b337e0efe6d7 ro quiet splash vt.handoff=7
PulseList:
 Error: command ['pacmd', 'list'] failed with exit code 1: Home directory not accessible: Permission denied
 No PulseAudio daemon running, or not running as session daemon.
RelatedPackageVersions:
 linux-restricted-modules-4.8.0-21-generic N/A
 linux-backports-modules-4.8.0-21-generic N/A
 linux-firmware 1.161
SourcePackage: linux
UpgradeStatus: Upgraded to yakkety on 2016-10-08 (0 days ago)
dmi.bios.date: 10/22/2007
dmi.bios.vendor: LENOVO
dmi.bios.version: 7QET27WW (1.09 )
dmi.board.name: 8943DNG
dmi.board.vendor: LENOVO
dmi.board.version: Not Available
dmi.chassis.asset.tag: No Asset Information
dmi.chassis.type: 10
dmi.chassis.vendor: LENOVO
dmi.chassis.version: Not Available
dmi.modalias: dmi:bvnLENOVO:bvr7QET27WW(1.09):bd10/22/2007:svnLENOVO:pn8943DNG:pvrThinkPadR61:rvnLENOVO:rn8943DNG:rvrNotAvailable:cvnLENOVO:ct10:cvrNotAvailable:
dmi.product.name: 8943DNG
dmi.product.version: ThinkPad R61
dmi.sys.vendor: LENOVO

[Fred Vkigg (fvkigg)]

Attached dmesg with complete kernel errors with stack traces

[Fred Vkigg (fvkigg)]

Attach Xorg.log for completeness although this does not show any problems

[Brad Figg (brad-figg)]

This change was made by a bot.

[Joseph Salisbury (jsalisbury)]

Did this issue start happening after an update/upgrade? Was there a kernel version where you were not having this particular problem? This will help determine if the problem you are seeing is the result of a regression, and when this regression was introduced. If this is a regression, we can perform a kernel bisect to identify the commit that introduced the problem.

[Fred Vkigg (fvkigg)]

As the problem report shows the system was upgraded, over many releases originally from Ubuntu 12.04 LTS "Precise Pangolin" - Release amd64
The lastest 16.04 kernel did not have the problem. I guess that gives us two bisection points.

[Jeff (boardman-malibu)]

I can confirm this is also a problem on my Dell Inspiron 1520. It seems to only happen using the 4.8.0 kernel. Resume from hibernation fails while decompressing the hibernation image. The boot time is much longer than when I use the 4.4.0 kernel on my Xubuntu 16.10 installation. It seems the drm/i915 module has undergone some big changes between 4.4.0 and 4.8.0, and now is asking for non-existant firmware during update-initramfs operations.

[Fred Vkigg (fvkigg)]

I think the non-existant firmware is not relevant for the older Intel Chips such as GMA965. This firmware is only for the latest chips.

[Jeff (boardman-malibu)]

To Fred,

Looking at you Xlog I seen the same problem as in mine

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

Almost 10 seconds gone there and:

[ 46.876] (II) modeset(0): EDID for output VGA-1
[ 57.393] (II) modeset(0): EDID for output SVIDEO-1

another 11 there. Lastly:

[ 126.003] (II) modeset(0): Modeline "1280x1024"x0.0 108.00 1280 1328 1440 1688 1024 1025 1028 1066 +hsync +vsync (64.0 kHz e)

Over two minutes hammering away on your video card.
All the time is eaten up modesetting and polling outputs, when the DRM module was supposed to have done that.

I took up the suggestion somebody mentioned of using the Ubuntu test kernels and I got up to 4.5.6 with boots looking good.

I also started to learn git and downloaded the test kernel repository. I see that there is a lot going on with drm/i915 drivers, in an attempt to move higher level functions away from Xorg ,and a lot of commits (changes) devoted to the same area and problems (fix one thing,break two more).

Maybe a one size fits all driver (9 generations), is too much.

Your first time stamp is 36.282 which is when the system is booted and going into user-mode. Mine is similar using 4.5.6. You can use 'systemd-analyze' to see your boot times.

The call traces in your kernel log don't seem to eat much time at all, but obviously something is afoot.

BTW I tried the 4.9.0-rc1 kernel and the thing stop at the initramfs shell, with alot of "error -22"
messages. I think maybe they're heading in the wrong direction.

[Launchpad Janitor (janitor)]

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