Ubuntu turns monitor permanently off during boot

[Jason Gunthorpe]

This is a regression, at least trusty (and prior) worked fine on this hardware.

Booting the Xenial installer image off USB also worked fine.

The system boots fine, however the monitor is in DPMS power-off and does not ever turn back on. Switching VTs doesn't help, I have found no way to get it switch on.

Removing '$vt_handoff' from grub resolves the problem and the system seems to work fine.

ProblemType: Bug
DistroRelease: Ubuntu 16.04
Package: linux-image-4.4.0-31-generic 4.4.0-31.50
ProcVersionSignature: Ubuntu 4.4.0-31.50-generic 4.4.13
Uname: Linux 4.4.0-31-generic x86_64
ApportVersion: 2.20.1-0ubuntu2.1
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC1: jgg 2224 F.... pulseaudio
 /dev/snd/controlC0: jgg 2224 F.... pulseaudio
CurrentDesktop: Unity
Date: Sun Jul 17 14:04:30 2016
HibernationDevice: RESUME=UUID=b254872c-e84b-427c-bebe-343c031cbdcd
InstallationDate: Installed on 2016-04-22 (86 days ago)
InstallationMedia: Ubuntu 16.04 LTS "Xenial Xerus" - Release amd64 (20160420.1)
IwConfig:
 lo no wireless extensions.

 enp6s0 no wireless extensions.
MachineType: System manufacturer System Product Name
ProcFB: 0 inteldrmfb
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-4.4.0-31-generic.efi.signed root=/dev/mapper/main2-xenial64 ro quiet splash
RelatedPackageVersions:
 linux-restricted-modules-4.4.0-31-generic N/A
 linux-backports-modules-4.4.0-31-generic N/A
 linux-firmware 1.157.2
RfKill:

SourcePackage: linux
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 11/25/2010
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: 0805
dmi.board.asset.tag: To be filled by O.E.M.
dmi.board.name: P8H67-M EVO
dmi.board.vendor: ASUSTeK Computer INC.
dmi.board.version: Rev 1.xx
dmi.chassis.asset.tag: Asset-1234567890
dmi.chassis.type: 3
dmi.chassis.vendor: Chassis Manufacture
dmi.chassis.version: Chassis Version
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvr0805:bd11/25/2010:svnSystemmanufacturer:pnSystemProductName:pvrSystemVersion:rvnASUSTeKComputerINC.:rnP8H67-MEVO:rvrRev1.xx:cvnChassisManufacture:ct3:cvrChassisVersion:
dmi.product.name: System Product Name
dmi.product.version: System Version
dmi.sys.vendor: System manufacturer

[Brad Figg (brad-figg)]

This change was made by a bot.

[Joseph Salisbury (jsalisbury)]

Does booting back into the Trusty kernel on the Xenial installed system make the bug go away?

[Jason Gunthorpe (jgunthorpe)]

On Mon, Jul 18, 2016 at 2:48 PM, Joseph Salisbury <
<email address hidden>> wrote:

> Does booting back into the Trusty kernel on the Xenial installed system
> make the bug go away?
>

Yes. I've done multiple reboot experiments now and have learned the bug
only happens about 80% of the time, so I did all of these will multiple
reboots each:

3.16.0-77 GOOD
3.19.0-65 GOOD
4.2.0-42 BAD
4.4.0-31 BAD

All generic versions, 4.4 was using the .efi.signed variant.

I've attached the dmesg for 3.19-65

BTW, none of the kernels manage a seamless graphical boot like Windows
does. All successes sit at black screens for a long time then transition to
the splash for maybe 0.5s then into the graphical login (after another
black screen flash). Come to think of it, I don't think I've ever seen
Ubuntu do this properly on any release on any hardware I've tried (at least
with EFI boot). Even my HP's at work sit at a black screen for more time
than the splash. So I don't understand why Canonical bothers with the
vt_handoff stuff, when it seems *worse* on than off. At least without
vt_handoff you get something that looks like a running computer instead of
black screen.. That said, I think it works better with BIOS boot ??

Windows 7 & 10 both do this perfectly in EFI mode, the monitor is never
black.

Jason

[Joseph Salisbury (jsalisbury)]

Would it be possible to test the upstream stable 4.0 and 4.1 kernels? They can be downloaded from:

4.0: http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.0-vivid/
4.1: http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.1-wily/

We can perform a kernel bisect once we identify the last good and first bad kernel versions.

[Jason Gunthorpe (jgunthorpe)]

4.1.0-040100-generic works
4.4.14-040414-generic works

Didn't test others.

The boot sequence is different with the main line kernels. There is a black
screen period with a couple of messages (1-2s), then the console font load,
then it briefly switches to the splash, black screen flash back to the
prior screen then the login prompt. There is no DPMS flash (the monitor
displays a dialog for 1s when it is DPMS powered off)

Whereas when, say, 4.2-ubuntu rarely randomly works, or with 3.19-ubuntu,
it just stays at a black screen, briefly flashes a DPMS power off
notification, and then goes to the login screen without showing the couple
of messages.

Sounds like an Ubuntu kernel bug?

Obnoxiously all the console messages are not captured in the journal, that
seems like another bug..

On Tue, Jul 19, 2016 at 1:21 PM, Joseph Salisbury <
<email address hidden>> wrote:

> Would it be possible to test the upstream stable 4.0 and 4.1 kernels?
> They can be downloaded from:
>
> 4.0: http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.0-vivid/
> 4.1: http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.1-wily/
>
> We can perform a kernel bisect once we identify the last good and first
> bad kernel versions.
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1603810
>
> Title:
> Ubuntu turns monitor permanently off during boot
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1603810/+subscriptions
>

[Jason Gunthorpe (jgunthorpe)]

As a clarification, the mainline boot behaviour is identical to the Ubuntu
kernel with $vt_handoff removed. I double checked that the option is
present in grub for all the tests. I guess this isn't surprising since that
option isn't part of mainline.

Jason

On Wed, Jul 20, 2016 at 8:30 PM, Jason Gunthorpe <email address hidden>
wrote:

> 4.1.0-040100-generic works
> 4.4.14-040414-generic works
>
> Didn't test others.
>
> The boot sequence is different with the main line kernels. There is a
> black screen period with a couple of messages (1-2s), then the console font
> load, then it briefly switches to the splash, black screen flash back to
> the prior screen then the login prompt. There is no DPMS flash (the monitor
> displays a dialog for 1s when it is DPMS powered off)
>
> Whereas when, say, 4.2-ubuntu rarely randomly works, or with 3.19-ubuntu,
> it just stays at a black screen, briefly flashes a DPMS power off
> notification, and then goes to the login screen without showing the couple
> of messages.
>
> Sounds like an Ubuntu kernel bug?
>
> Obnoxiously all the console messages are not captured in the journal, that
> seems like another bug..
>
>
>
> On Tue, Jul 19, 2016 at 1:21 PM, Joseph Salisbury <
> <email address hidden>> wrote:
>
>> Would it be possible to test the upstream stable 4.0 and 4.1 kernels?
>> They can be downloaded from:
>>
>> 4.0: http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.0-vivid/
>> 4.1: http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.1-wily/
>>
>> We can perform a kernel bisect once we identify the last good and first
>> bad kernel versions.
>>
>> --
>> You received this bug notification because you are subscribed to the bug
>> report.
>> https://bugs.launchpad.net/bugs/1603810
>>
>> Title:
>> Ubuntu turns monitor permanently off during boot
>>
>> To manage notifications about this bug go to:
>>
>> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1603810/+subscriptions
>>
>
>

[Joseph Salisbury (jsalisbury)]

Can you see if the bug still happens when booting a Trusty kernel on the Xenial system? This can be down by downloading the kernel from:

https://launchpad.net/~canonical-kernel-team/+archive/ubuntu/ppa/+build/10184623

You will need to install both the linux-image and linux-image-extra .deb packages.

If the bug still happens, it may indicate a bug in GRUB.

[Jason Gunthorpe (jgunthorpe)]

Wasn't testing with 3.16 and 3.19 ubuntu patched kernels enough to disprove
the grub theory?

Jason

On Mon, Jul 25, 2016 at 1:44 PM, Joseph Salisbury <
<email address hidden>> wrote:

> Can you see if the bug still happens when booting a Trusty kernel on the
> Xenial system? This can be down by downloading the kernel from:
>
> https://launchpad.net/~canonical-kernel-
> team/+archive/ubuntu/ppa/+build/10184623
>
> You will need to install both the linux-image and linux-image-extra .deb
> packages.
>
> If the bug still happens, it may indicate a bug in GRUB.
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1603810
>
> Title:
> Ubuntu turns monitor permanently off during boot
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1603810/+subscriptions
>

[Joseph Salisbury (jsalisbury)]

Ahh, yes it was. Comment 8 is a duplicate request, sorry about that.

[Joseph Salisbury (jsalisbury)]

This does appear to be an Ubuntu specific kernel. I'll build an Ubuntu kernel based on 4.0 and post a link to it for testing.

[Jason Gunthorpe (jgunthorpe)]

Were you still wanting to test this new kernel?

On Mon, Jul 25, 2016 at 2:33 PM, Joseph Salisbury <
<email address hidden>> wrote:

> This does appear to be an Ubuntu specific kernel. I'll build an Ubuntu
> kernel based on 4.0 and post a link to it for testing.
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1603810
>
> Title:
> Ubuntu turns monitor permanently off during boot
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/
> 1603810/+subscriptions
>

[Joseph Salisbury (jsalisbury)]

Sorry for the delay. The 4.0 based kernel can be downloaded from:
https://launchpad.net/ubuntu/+source/linux/4.0.0-4.7/+build/7636518

You will need to install both the linux-image and linux-image-extra .deb packages.

[Launchpad Janitor (janitor)]

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