[VX800] Ubuntu 10.04 alpha 3 boot corrupts screen (A110 netbook)

[Maik Merten]

When booting the screen-content will suddenly disintegrate (on X startup?). No graphical login is possible. After startup it's possible to change to a text console (screen won't recover, though) and press ctrl-alt-del for a machine shutdown/restart.

In-depth hardware description: http://a110wiki.de/wiki/One_A110

Videos of the bootup process:

Standard options: http://people.xiph.org/~maikmerten/a110-ubuntu/standardboot.ogv
With removed "quiet splash": http://people.xiph.org/~maikmerten/a110-ubuntu/boot-noquiet-nosplash.ogv

Graphical installation from USB stick works, with at most a few corrupted pixels at the top of the display.

Unsure what package - could be kernel (is there kernel-modesetting for VIA hardware?), framebuffer or the X driver.

Expected result: Graphical login screen.
What happend instead: Corrupted screen.

[Maik Merten (maikmerten)]

Ubuntu 10.04 Beta 1 is affected in the same way.

[Bartosz Kosiorek (gang65)]

Attach your /var/log/Xorg.0.log file.
If you can from after reproducing this issue (or /var/log/Xorg.0.log.old after reboot).

[Maik Merten (maikmerten)]

Attached Xorg.0.log

[Florian Tobias Schandinat (florianschandinat)]

This happens as VGA is not the output connected to your hardware. You'll likely see the correct screen if you plug in a monitor to the VGA port of the netbook.
viafb will work if you give it the options (as explained in viafb.txt)
viafb_active_dev=DVI viafb_mode=800x480
I don't know how to fix the problem for X as I'm only using the framebuffer driver (with -fbdev if X is needed).

[Maik Merten (maikmerten)]

Thank you, I can confirm X outputs to VGA. The picture there is fine, although with a corrupted mouse pointer (rectangle filled with noise).

So apparently this is an issue with output-device selection.

[Aurélien COUDERC (coucouf)]

I tried a Lucid installation with monday's daily iso and almost the same device (One A115) and didn't experience your issue.
However I was able to reproduce it with adding video=viafb to the kernel boot line. It seems that the viafb somehow breaks the openchrome X driver…

You could try the following kernel boot options and report the results:
video=vesa vga=0x312

- video=vesa should avoid the viafb framebuffer driver being loaded, and openchrome selecting the wrong display for X;
- vga=0x312 will avoid the broken VTs/Plymouth if your device behaves like mine.

If you can confirm this works for you, the bug should be affected either to xserver-xorg-video-openchrome which is the default X driver for this device, or more likely to the linux-image-xxx package for wrongly assuming that the viafb module would work with a VIA chip. :-)

[Florian Tobias Schandinat (florianschandinat)]

I'm wondering why video=viafb makes a difference for you as normally this gets autoloaded or doesn't react at all.
Does it work correctly if you add something like this instead?
video=viafb:viafb_active_dev=DVI,viafb_mode=800x480
Yeah, it might sound odd that its called 'DVI' but the whole output device setup is inside viafb and as far as I know in all free drivers a chaos at least.
Maybe openchrome assumes that the current setting is correct which is probably not the worst choice but should be avoided if possible. That said its probably a bug in viafb and in openchrome with viafb making the additional assumption that the output device is on VGA/CRT/primary output which is often correct but sometimes wrong. On the long term the whole output device setup mess has to be cleaned up but whether this is possible and how long it will take depends heavily on VIA and won't be finished under a year, I fear.

[Maik Merten (maikmerten)]

Sadly none of the the kernel options above fix the problem. Thank you for providing me with something to toy around, though :)

I still wonder why booting the installation CD from USB gives a working X vanishing with installation.

[Jeremy Foshee (jeremyfoshee)]

Hi Maik,

This bug was reported a while ago and there hasn't been any activity in it recently. We were wondering if this is still an issue? Can you try with the latest development release of Ubuntu? ISO CD images are available from http://cdimage.ubuntu.com/releases/ .

If it remains an issue, could you run the following command from a Terminal (Applications->Accessories->Terminal). It will automatically gather and attach updated debug information to this report.

apport-collect -p linux 538430

Also, if you could test the latest upstream kernel available that would be great. It will allow additional upstream developers to examine the issue. Refer to https://wiki.ubuntu.com/KernelMainlineBuilds . Once you've tested the upstream kernel, please remove the 'needs-upstream-testing' tag. This can be done by clicking on the yellow pencil icon next to the tag located at the bottom of the bug description and deleting the 'needs-upstream-testing' text. Please let us know your results.

Thanks in advance.

    [This is an automated message. Apologies if it has reached you inappropriately; please just reply to this message indicating so.]

[Maik Merten (maikmerten)]

The current Maverick ISO won't boot for me.

FATAL: Error inserting ramzswap (lib/modules/2.6.34-5-generic/kernel/drivers/staging/ramzswap/ramzswap.ko): Unknown symbol in module, or unknown parameter (see dmesg)
FATAL: Error in inserting vesafb (/lib/modules/2.6.34-5-generic/initrd/vesafb.ko): No such device
ALERT! does not exist. Dropping to a shell!

I guess testing will have to wait until things look a bit better ;)

[Maik Merten (maikmerten)]

Maverick alpha 2 installs and boots correctly into X, initializing the correct screen. Outside of X and during boot the screen is still mangled, but this would be a different issue, I'd guess.

[Alberto Jovito (thedemon007)]

Test Precise Pangolin or latest ubuntu.