Created attachment 37394 [details] distortions on external display I have a notebook with switchable graphics (discrete ATI Mobility Radeon HD 5650 and integrated Intel Arrandale). The DisplayPort only works with the discrete adapter but this limitation also exists in Windows, so I guess the DisplayPort is hard-wired to the discrete card. If I use the radeon driver, it detects whenever I plug something into the DisplayPort. E.g. I have a DisplayPort-to-VGA adapter (for projectors) and test it with my LCD monitor that luckily has a VGA input, too. xrandr detects the connection and suggests the correct resolution and frequency (1680x1050 @60Hz). But the image is totally distorted. It looks like a old tube monitor that runs a too high frequency, you can see this on the attached screenshot. What you cannot see, is that the distortions move quite fast and are not static. I switched back to 1024x768@60Hz but the problem persists and the optional 75.1Hz do not make any difference, too. The same settings work fine with Windows, so I suspect this issue to be caused by the radeon driver. Is there any way to see what's going wrong and maybe fix this? - Kernel 2.6.34 - XOrg 7.5 - xf86-driver-ati 6.13.0
I forgot to mention that even with the distorions, you can see that the output shows a clone of the desktop how it is supposed to be. I can see this by the colors on the external display. Compare the attached screenshot to http://desk08.customize.org/wallpaper/54, which is my wallpaper.
Please attach your xorg log and dmesg output.
Created attachment 37407 [details] dmesg output
Created attachment 37408 [details] [review] Xorg logfile
I attached both dmesg output and Xorg logfile. Since my system comes with hybrid graphics, I have blacklisted the radeon driver in order to see the console (and hot have radeon and i915 conflicting each other). Then I booted this way: 1. plug external monitor (to make sure it's not a hotplug issue) 2. turn system on, boot to runlevel 3 3. load radeon module 4. use vga_switcheroo to switch to ATI card 5. turn Intel card off 6. switch to runlevel 5 and start X
Any news on this?
Is there any way to boot with the ATI card as the primary? There should be a bios option.
(In reply to comment #7) > Is there any way to boot with the ATI card as the primary? There should be a > bios option. Unfortunally, the Envy 14 does not provide such a BIOS option. I already filed a complainment about this fact to HP but I do not really expect something to happen soon. :(
Any idea or anything I could test?
I have a similar issue with my radeon card. I can give some additional information because I only have the radeon card on my machine and am able to boot with the displayport as the primary output. My corruption occurs in two situations: 1. When resuming from suspend 2. When hotplugging the monitor after booting only using the laptop display. Note, I have the same problem using the VGA adapter, so I don't think this is a display port specific issue. My laptop is an HP Elitebook 8540w with an HD Mobility Radeon 5800 series card. I'll attach my dmesg and Xorg logfile as well.
Created attachment 38221 [details] dmesg output for jgeisler
Created attachment 38222 [details] Xorg log for jgeisler
Do things work any better on Dave's drm-fixes branch: http://git.kernel.org/?p=linux/kernel/git/airlied/drm-2.6.git;a=shortlog;h=refs/heads/drm-fixes There are a number of evergreen related display fixes there. Also, Jonathan, can you attach a copy of your vbios? As per your log: [ 1.975933] [drm] Connector 4: [ 1.976127] [drm] VGA [ 1.976318] [drm] DDC: no ddc bus - possible BIOS bug - please report to xorg-driver-ati@lists.x.org [ 1.976627] [drm] Encoders: [ 1.976818] [drm] CRT1: INTERNAL_KLDSCP_DAC1 it looks like there is a bios bug there that may be part of the problem in your case. To access your vbios (as root): cd /sys/bus/pci/devices/<pci bus id> echo 1 > rom cat rom > /tmp/vbios.rom echo 0 > rom
Created attachment 38228 [details] vbios for jgeisler
I've attached the vbios as requested and will report back on the updated kernel version soon.
Created attachment 38229 [details] [review] fix ddc line for VGA This patch fixes the ddc line for VGA.
To use Dave's drm-fixes branch you have to check it out. The master branch on his repo is very old. So after you clone the repo, you need to: git checkout -b drm-fixes origin/drm-fixes
I've run the drm-fixes kernel and the symptoms remain the same. Additionally, I can confirm that the patch from Alex removes the error message. I've tried to run the kernel with and without the patch, but there is no visible difference in the behavior of the output.
Created attachment 38408 [details] drm-fixes kernel logfile for jgeisler
I've added an attachment that includes the kernel log file when drm.debug=1 is enabled on boot with the drm-fixes kernel.
Hi, I have the same problem on my Envy 14, but thanks to a tip from the helpful folks in #radeon on FreeNode I found that if I turn off my laptop panel with xrandr then my external display immediately becomes ungarbled. Specifically, the command: xrandr --output LVDS-1 --off (and then, because it starts in mirror mode with a non-native resolution of my monitor) xrandr --output DisplayPort-0 --auto I'm using Arch and am selecting the ATI card via vgaswitcheroo first. Here's my uname-a Linux bender 2.6.35-ARCH #1 SMP PREEMPT Fri Aug 27 17:14:28 CEST 2010 x86_64 Intel(R) Core(TM) i5 CPU M 450 @ 2.40GHz GenuineIntel GNU/Linux Package versions: local/xorg-server 1.8.1.902-1 local/xf86-video-ati 6.13.1-1 (xorg-video-drivers) local/ati-dri 7.8.2-1 Hope this helps anyone who may find this trade-off suitable until a proper fix is found.
I just tried to switch the internal LVDS off. Instantly, the external display shows a correct picture. At least a way to make use of a projector. But running the internal and an external display at the same time still needs to be fixed.
I can also confirm that doing the workaround allows me to use the external device exclusive to my laptop screen--but not both at the same time. This is a big improvement, but not the complete solution.
Does this set of kernel patches help (against recent 2.6.36): http://people.freedesktop.org/~agd5f/pll_fixes/
(In reply to comment #25) > Does this set of kernel patches help (against recent 2.6.36): > http://people.freedesktop.org/~agd5f/pll_fixes/ No. I see the same behavior with these patches.
Can you try a newer kernel? 3.0 or 3.1? If not, please try this patch as well: http://lists.freedesktop.org/archives/dri-devel/2011-October/014993.html
I am now able to switch adapters without a problem. I forget which kernel this started working, but I am currently running 2.6.39-gentoo-r3 without issues. As far as I am concerned this bug has been fixed and I forgot to note that improvement on the bug.
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.