Unity displaying with black overlay with fglrx driver

[Jeff Lane ]

Testing Oneiric Alpha 3 on an Insprion 1122 with an ATI video card. I had issues with screen corruption when connecting an external display using the OSS driver, so I opened Jockey and tried using the Proprietary driver instead.

This was even worse, however, because I was almost unable to use the system while the proprietary driver was loaded. I had to drop to a shell and use jockey-text to remove it before I could use the system again.

So, the proprietary driver that Jockey is suggesting for ATI cards in Oneiric Alpha 3 breask the system. The driver needs to be fixed, or blacklisted from Jockey so that users don't end up with an unusable Oneiric system.
---
Architecture: i386
CompizPlugins: [core,bailer,detection,composite,opengl,decor,mousepoll,vpswitch,regex,animation,snap,expo,move,compiztoolbox,place,grid,imgpng,gnomecompat,wall,ezoom,workarounds,staticswitcher,resize,fade,unitymtgrabhandles,scale,session,unityshell]
DistUpgraded: Fresh install
DistroCodename: oneiric
DistroRelease: Ubuntu 11.10
DistroVariant: ubuntu
DkmsStatus: fglrx, 8.861, 3.0.0-8-generic, i686: installed
GraphicsCard:
 ATI Technologies Inc Device [1002:9803] (prog-if 00 [VGA controller])
   Subsystem: Dell Device [1028:04df]
InstallationMedia: Ubuntu 11.10 "Oneiric Ocelot" - Alpha i386 (20110803.1)
JockeyStatus: xorg:fglrx - ATI/AMD proprietary FGLRX graphics driver (Proprietary, Enabled, In use)
LightdmGreeterLog:

MachineType: Dell Inc. Inspiron 1122
NonfreeKernelModules: fglrx
Package: fglrx-installer (not installed)
ProcEnviron:
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-3.0.0-8-generic root=UUID=0fc2125d-a95d-4528-9164-206c91790f5c ro quiet splash initcall_debug vt.handoff=7
ProcVersionSignature: Ubuntu 3.0.0-8.10-generic 3.0.1
Tags: oneiric running-unity ubuntu single-occurrence reproducible
Uname: Linux 3.0.0-8-generic i686
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare
dmi.bios.date: 12/02/2010
dmi.bios.vendor: Dell Inc.
dmi.bios.version: X09
dmi.board.name: 0XDYKV
dmi.board.vendor: Dell Inc.
dmi.chassis.type: 8
dmi.chassis.vendor: Dell Inc.
dmi.chassis.version: X09
dmi.modalias: dmi:bvnDellInc.:bvrX09:bd12/02/2010:svnDellInc.:pnInspiron1122:pvrX09:rvnDellInc.:rn0XDYKV:rvr:cvnDellInc.:ct8:cvrX09:
dmi.product.name: Inspiron 1122
dmi.product.version: X09
dmi.sys.vendor: Dell Inc.
version.compiz: compiz 1:0.9.5.0-0ubuntu2
version.fglrx-installer: fglrx-installer N/A
version.libdrm2: libdrm2 2.4.26-1ubuntu1
version.libgl1-mesa-dri: libgl1-mesa-dri 7.11-0ubuntu2
version.libgl1-mesa-dri-experimental: libgl1-mesa-dri-experimental N/A
version.libgl1-mesa-glx: libgl1-mesa-glx 7.11-0ubuntu2
version.xserver-xorg: xserver-xorg 1:7.6+7ubuntu6
version.xserver-xorg-input-evdev: xserver-xorg-input-evdev 1:2.6.0-1ubuntu13
version.xserver-xorg-video-ati: xserver-xorg-video-ati 1:6.14.2-1ubuntu2
version.xserver-xorg-video-intel: xserver-xorg-video-intel 2:2.15.901-1ubuntu2
version.xserver-xorg-video-nouveau: xserver-xorg-video-nouveau 1:0.0.16+git20110411+8378443-1

[Jeff Lane  (bladernr)]

Here is a brief video showing what the proprietary driver Jockey installs does to the system.

[Bryce Harrington (bryce)]

Hi Jeff, sorry to hear you had such troubles. Unfortunately this bug report as written is non-actionable. Here's a few tips for improving your bug report:

1. Attach log files. Without log files developers cannot diagnose the problem. Generally, running 'apport-collect 823588' should attach the required files.

2. When writing a bug report description, try to think of it more like a high school lab or a police report - state the facts clearly and describe the situation completely. Don't describe things as "random" or "unusable" or "broken" - those terms are so broad they actually tell us very little, instead try to characterize the behavior as if you were studying the problem scientifically.

3. Avoid assuming that a bug you see on your system is seen by everyone on all systems; while some X bugs are widespread, the majority tend to be somewhat hardware-specific or even specific to some configuration that you've done. Indeed, there are relatively few bug reports about -fglrx open right now. Instead, try to think about what might make *your* system different than other people's - it can give strong clues as to what went wrong.

4. You've mentioned Alpha 3 specifically, which makes me wonder if you had successfully used fglrx on oneiric previously. If you have, and know a bug is a regression from an earlier point in time, it is very valuable to mention this; often in those situations we can look at what got upgraded when as a clue to finding the bug.

Hope this all helps, and good luck improving your bug report.

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

apport information

[Jeff Lane  (bladernr)]

On 08/10/2011 01:59 AM, Bryce Harrington wrote:
> 1. Attach log files. Without log files developers cannot diagnose the
> problem. Generally, running 'apport-collect 823588' should attach the
> required files.

Done. Sorry for neglecting to do that yesterday. I'm in the lab and
have a whole list of systems in test, and in the last two days alone
I've filed 30 bugs or so :( things slip through the cracks... but I've
remedied that now.

> 2. When writing a bug report description, try to think of it more like
> a high school lab or a police report - state the facts clearly and
> describe the situation completely. Don't describe things as "random" or
> "unusable" or "broken" - those terms are so broad they actually tell us
> very little, instead try to characterize the behavior as if you were
> studying the problem scientifically.

Unfortunately, I do not have anything more descriptive for you. Thus, I
attached a video showing EXACTLY what was happening. I was assuming
that if a picture is worth a thousand words, a video should be "War and
Peace Unabridged"... It's just that simple. Install driver. Boot.
System is unusable because it's a black screen, then not a black screen,
then a black screen, then not a black screen... :)

> 3. Avoid assuming that a bug you see on your system is seen by everyone
> on all systems; while some X bugs are widespread, the majority tend to
> be somewhat hardware-specific or even specific to some configuration
> that you've done. Indeed, there are relatively few bug reports about
> -fglrx open right now. Instead, try to think about what might make
> *your* system different than other people's - it can give strong clues
> as to what went wrong.

I assume nothing, which is why I stated specifically in the bug that I
was operating on an Inspiron 1122 with an ATI video card. That's pretty
narrow, I think. But hopefully, the logs attached from #1 above will
help further narrow it down.

> 4. You've mentioned Alpha 3 specifically, which makes me wonder if you
> had successfully used fglrx on oneiric previously. If you have, and
> know a bug is a regression from an earlier point in time, it is very
> valuable to mention this; often in those situations we can look at what
> got upgraded when as a clue to finding the bug.

Negative... I flew up to Lexington specifically to do a help do a
certification test run on the systems stored there in the Cert Lab.
I've not touched this system since I came up here to do the 11.04 cert
run back in April. In fact, as far as I can tell by the system entry on
the cert site, that was the last time anyone tested on this machine.

> Hope this all helps, and good luck improving your bug report.

It does, and I do apologize for the lack of logs and such, as I said, I
have a whole list of systems to get through and the first two systems
lone generated 30 bugs and took two full days of my time to get through.

--
Jeff Lane - Hardware Certification Engineer and Test Tools Developer
Ubuntu Ham: W4KDH
Freenode IRC: bladernr or bladernr_
gpg: 1024D/3A14B2DD 8C88 B076 0DD7 B404 1417 C466 4ABD 3635 3A14 B2DD

[Ara Pulido (ara)]

Setting to medium and assigning to the desktop team

[Chris Halse Rogers (raof)]

Wow. That's a very funky video. Setting priority to high; that's clearly unusable.

There doesn't seem to be a spree of similar issues in Launchpad, though, so I think we should discover the extent of the problem before turning off the driver in jockey, which would inhibit our ability to work out where it's broken.

The logs don't seem to show any unusual activity; I'll see if Alberto has anything to say about this.

[Alberto Milone (albertomilone)]

I think I've already seen that problem on a system with switchable graphics when using the discrete card.

I also remember that if I switched to Unity 2D from the login manager I was not able to reproduce the issue.

This is probably a problem in fglrx triggered by nux on some graphics chipsets.

[Bryce Harrington (bryce)]

I was able to reproduce what I think is this same bug on my test box - see bug #825605. It's just an ordinary desktop with an ati card; no switchable graphics. I didn't try Unity 2D, but the FOSS -ati driver works fine on it.

Notably none of the unity toolbar stuff shows up, which makes me also think it's something conflicting with nux/unity.

[Bryce Harrington (bryce)]

One thing I found was that if I typed into the xterm or moved the mouse, the blackness would display only down to where the cursor was.

https://bugs.launchpad.net/ubuntu/+source/fglrx-installer/+bug/825605/+attachment/2273487/+files/100_2276.JPG

https://bugs.launchpad.net/ubuntu/+source/fglrx-installer/+bug/825605/+attachment/2273488/+files/100_2278.JPG

[Bryce Harrington (bryce)]

The PCI ID for the card in this report is 1002:9803; that PCI ID isn't listed anywhere so I'm guessing it's a very new card. The card I was testing was a RV730XT [Radeon HD 4670] [1002:9490], which is also relatively recent.

[Bryce Harrington (bryce)]

An AMD engineer has reported apparently they've also seen the bug on their end, as a regression between alpha-2 and alpha-3.

[Ara Pulido (ara)]

After chatting on IRC a bit with Alberto, it seems that it is a regression with AMD and Unity.
Jay is talking with AMD about it, so assigning to him to follow up.

[LostinSpacetime (lostinspacetime)]

I have exactly the same problem with a HP 6735b. ATI HD 3200 graphics.