Nonexistent VGA monitor detected on AMD A8-3850 / Asus F1A75-V PRO

[Steven Wilson]

CPU/GPU: AMD A8-3850 w/onboard GPU ("Radeon 6550D" according to AMD marketing; "Device 9640" according to lspci)
Mainboard: Asus F1A75-V PRO
Monitor (TV): Sony KDL46EX701

Ubuntu version: 11.10 Beta 2 amd64
xserver-xorg-video-ati version: 1:6.14.99~git20110811.g93fc084-0ubuntu1

My system is connected to a 1920x1080 LCD TV via an HDMI cable. No other video cable is connected. Despite this, the driver says:

[ 19.096] (II) RADEON(0): Output VGA-0 connected
[ 19.096] (II) RADEON(0): Output HDMI-0 connected
[ 19.096] (II) RADEON(0): Output DisplayPort-0 disconnected
[ 19.096] (II) RADEON(0): Using fuzzy aspect match for initial modes
[ 19.096] (II) RADEON(0): Output VGA-0 using initial mode 1024x768
[ 19.096] (II) RADEON(0): Output HDMI-0 using initial mode 1024x768

By default, this results in a 1024x768 desktop being mirrored across the TV and the "Unknown" VGA monitor. I can change this in "Displays" and get a sane configuration, but that only takes effect after logging in; the login screen goes back to 1024x768.

The kernel correctly detects that only the HDMI output is connected, and I get the expected full-screen 1920x1080 display if I switch to a terminal with Ctrl+Alt+F1:

[ 7.144058] [drm] Radeon display connector VGA-1: No monitor connected or invalid EDID
[ 7.247568] [drm] Radeon display connector HDMI-A-1: Found valid EDID
[ 7.724031] [drm] Radeon display connector DP-1: No monitor connected or invalid EDID

Launchpad bug #151790 / Freedesktop bug #18719 describes a similar problem, but I'm filing a new bug since that one is almost 3 years old and its fix seems to be tied to a specific hardware quirk in an older hardware family.

[madbiologist (me-again)]

If this is still occurring on the final release of Oneiric, please run apport-collect 864155

[madbiologist (me-again)]

Also, there is a patch in the upstream 3.1-rc10 kernel which may fix this issue:

author Alex Deucher Tue, 4 Oct 2011 16:23:24 +0000 (12:23 -0400)
committer Dave Airlie Wed, 5 Oct 2011 09:31:40 +0000 (10:31 +0100)
commit 6777a4f6898a53974ef7fe7ce09ec41fae0f32db
tree e5680a33625e3a69d78c11e1fb56849e2c573beb
parent 4f332844cc87c5f99c5300f788abbe8a8c731390

drm/radeon/kms: fix dp_detect handling for DP bridge chips

The HPD pin is not reliable for detecting whether a monitor
is connected or not. Skip HPD and just use DDC or load
detection.

Fixes phantom VGA connected bugs.

[Michel: fixes phantom VGA bugs on his llano system.]

Signed-off-by: Alex Deucher
Reviewed-by: Michel Dänzer
Tested-by: Michel Dänzer
Cc: <email address hidden>
Signed-off-by: Dave Airlie

A PPA of this kernel is available at http://kernel.ubuntu.com/~kernel-ppa/mainline/
It has also been cc'd to the stable kernel tree and should therefore eventually make it's way to oneiric-updates as an SRU.

[Steven Wilson (excyber)]

It still happens with an updated system after uninstalling fglrx, but I can't test beyond that right now because the uninstall of fglrx somehow broke Unity so that I can't launch programs (it looks like I only have Nautilus after logging in), and I don't have time right now to do a full reinstall.

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[karlrt (karlrt)]

just installed the most recent mainline http://kernel.ubuntu.com/~kernel-ppa/mainline/daily/2011-10-24-oneiric/ and it is fixed there!

I have exactly the same hardware as the original poster, the radeon hd6550D, no problem with dvi output, but with hdmi for the lcd tv.

[madbiologist (me-again)]

@Steven - the breakage of Unity caused by uninstalling fglrx is bug #855943. There is a fix in oneiric-proposed which has been verified and will therefore be coming soon via oneiric-updates. If you want to get the fix in the meantime, see https://wiki.ubuntu.com/Testing/EnableProposed for info on how to enable and use proposed.

[Steven Wilson (excyber)]

This no longer occurs with linux-image-3.1.0-030100-generic_3.1.0-030100.201110241006_amd64.deb from kernel-ppa.

Sorry for the delay.

[Bryce Harrington (bryce)]

Hey excyber,

Hi, thanks for reporting this issue during the development period of
Ubuntu.

I notice there's not been further comments to the bug report since the
release came out, would you mind updating us on the status of it in the
release?

Are you still able to reproduce the issue? If not, do you think the bug
report can be closed, or do you think we should continue tracking it?

[Steven Wilson (excyber)]

Bryce, there have been several comments updating on the post-release status of this bug. In case you're reading via email and lost those comments, the summary is: yes, it's still reproducible on oneiric release, and it was fixed in the 3.1-rc kernel series (likely commit 6777a4f6898a53974ef7fe7ce09ec41fae0f32db per madbiologist at comment #2). Does this qualify as "Fix Committed"?

[alexandros (efalexspa)]

This bug affect me,please fix this plz!!!!

[Julian Wiedmann (jwiedmann)]

The latest kernel update (3.0.0-14.23) brings several radeon/kms fixes. Could you see if that helps?

[Steven Wilson (excyber)]

Still reproducible on an updated system, for which "uname -r" gives 3.0.0-14. I'm not sure whether this is "3.0.0-14.23".

[madbiologist (me-again)]

uname - a should reveal that.

[madbiologist (me-again)]

oops, I meant uname -a

[Steven Wilson (excyber)]

FWIW, I can't reproduce this with Precise Alpha 2 (not surprisingly).