Bug 88063

Summary: Radeon driver with dual monitor does not wake properly from sleep
Product: xorg Reporter: Scott Baker <scott.baker>
Component: Driver/RadeonAssignee: xf86-video-ati maintainers <xorg-driver-ati>
Status: RESOLVED MOVED QA Contact: Xorg Project Team <xorg-team>
Severity: normal    
Priority: medium    
Version: 7.5 (2009.10)   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
Xorg log
none
dmesg output
none
Updated dmesg after I fully removed fglrx none

Description Scott Baker 2015-01-05 18:50:36 UTC
I have a dual monitor configuration with Fedora 20 and now Fedora 21 setup with the following card:

01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Caicos XTX [Radeon HD 8490 / R5 235X OEM]

One monitor is using DVI, the other monitor is using a display port -> DVI connector. This setup works great for day to day usage, with no real issues. When the monitors go to sleep (when I leave for an extended period of time) the monitors don't wake up properly.

One monitor wakes up 100% fine, the other monitor wakes up but has a garbled display. My mouse will then only traverse the working monitor, as the second has dropped from the configuration. The fix is just to log out of KDE and log back in.

I'm running the following version of the radeon driver.

[959996.504] (II) LoadModule: "radeon"
[959996.504] (II) Loading /usr/lib64/xorg/modules/drivers/radeon_drv.so
[959996.505] (II) Module radeon: vendor="X.Org Foundation"
[959996.505]    compiled for 1.16.1, module version = 7.5.0
[959996.505]    Module class: X.Org Video Driver
[959996.505]    ABI class: X.Org Video Driver, version 18.0

Issue has persisted from Fedora 20 to Fedora 21. The two solutions I've found to this problem are:

1) Logout/login again
2) Run the fglrx driver instead
Comment 1 Alex Deucher 2015-01-05 21:32:54 UTC
Please attach your xorg log and dmesg output.  Also, does forcing another dpms cycle (xset dpms force off) from a terminal or doing a VT switch fix it?
Comment 2 Scott Baker 2015-01-05 21:51:02 UTC
Created attachment 111793 [details]
Xorg log
Comment 3 Scott Baker 2015-01-05 21:51:22 UTC
Created attachment 111794 [details]
dmesg output
Comment 4 Scott Baker 2015-01-06 00:00:31 UTC
I can simulate the corruption with:

sleep 5; xset dpms force off

Then working with agd5f on IRC we found the following command fixes the corruption and the second screen is usable again:

xrandr --output DVI-0 --off ; xrandr --output DVI-0 --auto --right-of DisplayPort-0
Comment 5 Alex Deucher 2015-01-06 05:09:29 UTC
I think I see the problem:

[   11.929088] fglrx: module license 'Proprietary. (C) 2002 - ATI Technologies, Starnberg, GERMANY' taints kernel.
[   11.929091] Disabling lock debugging due to kernel taint
[   11.931631] fglrx: module verification failed: signature and/or  required key missing - tainting kernel
[   11.948052] <6>[fglrx] Maximum main memory to use for locked dma buffers: 15617 MBytes.
[   11.948109] <6>[fglrx]   vendor: 1002 device: 6771 count: 1
[   11.948254] <6>[fglrx] ioport: bar 4, base 0xe000, size: 0x100
[   11.948334] <6>[fglrx] Kernel PAT support is enabled
[   11.948344] <6>[fglrx] module loaded - fglrx 14.10.2 [May  5 2014] with 1 minors

Can you make sure fglrx is completely removed before running the radeon driver?  It may be causing problems since radeon is already loaded.
Comment 6 Scott Baker 2015-01-09 21:07:57 UTC
Created attachment 112030 [details]
Updated dmesg after I fully removed fglrx
Comment 7 Scott Baker 2015-01-09 21:08:24 UTC
I fully removed (and rebooted) fglrx and I'm still getting the second screen corruption after sleep.
Comment 8 Scott Baker 2015-01-19 22:37:26 UTC
Just about every day after lunch (my monitors go to sleep after 60 minutes) my second monitor becomes corrupted. Using: 

xrandr --output DVI-0 --off ; xrandr --output DVI-0 --auto --right-of DisplayPort-0

is a semi-efficient fix and gets me back up and working. It is nonetheless an annoyance. Is there anything else I can do to provide feedback to assist in troubleshooting this issue?
Comment 9 Martin Peres 2019-11-19 07:49:27 UTC
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/xorg/driver/xf86-video-ati/issues/122.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.