Bug 89198

Summary: Setting Refresh Rate on monitor causes flickering/artifacting
Product: DRI Reporter: andre35822
Component: DRM/RadeonAssignee: Default DRI bug account <dri-devel>
Status: RESOLVED DUPLICATE QA Contact:
Severity: minor    
Priority: medium    
Version: unspecified   
Hardware: Other   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
Xorg.0.log
none
dmesg log
none
dmesg after trying the first "fix"
none
xorg log with patch from comment 15
none
dmesg log with patch from comment 15 none

Description andre35822 2015-02-18 05:17:44 UTC
Setting the refresh rate of an ASUS VG248QE to 144Hz causes flickering/artifacting in various Desktop Environments, including KDE, XFCE and Cinnamon. This occurs with compositing on and off in XFCE and Cinnamon (have not tried compositing off in KDE). If I set the refresh rate to 120hz the artifiacting goes away. On windows 7 there is no artificating when set to 144hz so it is not a hardware issue. I have used the GUI tools available in XFCE and KDE to change the refresh rate, and lxrandr on Cinnamon.


I am using mesa 10.4.4 (incase that helps) and xf86-video-ati 1.7.5 I am also on Arch Linux. If you would like any log files I will be more than ready to provide you with them, ATM I have no clue what is useful.
Comment 1 Alex Deucher 2015-02-18 05:18:59 UTC
Please attach your xorg log and dmesg output.
Comment 2 andre35822 2015-02-18 05:33:58 UTC
Created attachment 113596 [details]
Xorg.0.log
Comment 3 andre35822 2015-02-18 05:35:29 UTC
Created attachment 113597 [details]
dmesg log
Comment 4 andre35822 2015-02-18 05:35:59 UTC
(In reply to Alex Deucher from comment #1)
> Please attach your xorg log and dmesg output.

Ive attached them sorry about that. I should also point my card is an AMD 7950 and is connected via Dual link DVI.
Comment 5 Alex Deucher 2015-02-18 05:44:21 UTC
This may be a duplicate of bug 87796.  Can you try the suggestions in comment 10 on that bug?  If they help can you try the patch I attached and attach the dmesg output with the patch applied?
Comment 6 andre35822 2015-02-18 13:22:02 UTC
Created attachment 113611 [details]
dmesg after trying the first "fix"

dmesg after doing echo high > /sys/class/drm/card0/device/power_dpm_force_performance_level
Comment 7 andre35822 2015-02-18 13:22:28 UTC
(In reply to Alex Deucher from comment #5)
> This may be a duplicate of bug 87796.  Can you try the suggestions in
> comment 10 on that bug?  If they help can you try the patch I attached and
> attach the dmesg output with the patch applied?
Thank you. So I tried "echo high > /sys/class/drm/card0/device/power_dpm_force_performance_level" without "" as root and without restarting I set the refresh rate to 144hz and there is no more artifacting/flickering. Setting it back to auto causes there to be artifacts/flickering. So its something that has to do with DPM? I havem my dmesg attached when setting it to high.
Comment 8 Alex Deucher 2015-02-18 14:06:49 UTC

*** This bug has been marked as a duplicate of bug 87796 ***
Comment 9 andre35822 2015-02-18 15:17:56 UTC
(In reply to Alex Deucher from comment #8)
> 
> *** This bug has been marked as a duplicate of bug 87796 ***

My only concern is if setting DPM to high is safe? Does it increase power usage/requirement, will it make my card reach high temps or anything damaging? Also is this sort of the only "fix", will anything be added in the future so it works with DPM on auto?
Comment 10 Alex Deucher 2015-02-18 15:22:05 UTC
(In reply to andre35822 from comment #9)
> (In reply to Alex Deucher from comment #8)
> > 
> > *** This bug has been marked as a duplicate of bug 87796 ***
> 
> My only concern is if setting DPM to high is safe? Does it increase power
> usage/requirement, will it make my card reach high temps or anything
> damaging? Also is this sort of the only "fix", will anything be added in the
> future so it works with DPM on auto?

There is a fix on bug 87796, but it would be helpful to get the debugging output I mentioned in that bug as well so the patch can be fine tuned.
Comment 11 andre35822 2015-02-18 19:26:26 UTC
(In reply to Alex Deucher from comment #10)
> (In reply to andre35822 from comment #9)
> > (In reply to Alex Deucher from comment #8)
> > > 
> > > *** This bug has been marked as a duplicate of bug 87796 ***
> > 
> > My only concern is if setting DPM to high is safe? Does it increase power
> > usage/requirement, will it make my card reach high temps or anything
> > damaging? Also is this sort of the only "fix", will anything be added in the
> > future so it works with DPM on auto?
> 
> There is a fix on bug 87796, but it would be helpful to get the debugging
> output I mentioned in that bug as well so the patch can be fine tuned.
I am up to help you for whatever it is that you need, I am just a bit confused. You are referring to this bug report yes? https://bugs.freedesktop.org/show_bug.cgi?id=87796 and want me to apply the patch that you put as attachment in Comment 15? Or is it Comment 19 or 20? I am also not sure how to apply the patch, it looks like something to do with git? Once I know how to apply the patch, to my understanding you then want me to enable the "problematic mode", which would be setting the screen to 144hz and then attachming a dmesg output yes?
Comment 12 Alex Deucher 2015-02-18 20:56:41 UTC
(In reply to andre35822 from comment #11)
> I am up to help you for whatever it is that you need, I am just a bit
> confused. You are referring to this bug report yes?
> https://bugs.freedesktop.org/show_bug.cgi?id=87796 and want me to apply the
> patch that you put as attachment in Comment 15? Or is it Comment 19 or 20? I
> am also not sure how to apply the patch, it looks like something to do with
> git? Once I know how to apply the patch, to my understanding you then want
> me to enable the "problematic mode", which would be setting the screen to
> 144hz and then attachming a dmesg output yes?

Yes, bug 87796.  I'm interested in the dmesg output after switching to the problematic mode from the patch in comment 15.  See comment 17 for how to apply it.

The patches in 19 and 20 will workaround the issue while still letting you use auto.
Comment 13 andre35822 2015-02-19 20:34:27 UTC
(In reply to Alex Deucher from comment #12)
> (In reply to andre35822 from comment #11)
> > I am up to help you for whatever it is that you need, I am just a bit
> > confused. You are referring to this bug report yes?
> > https://bugs.freedesktop.org/show_bug.cgi?id=87796 and want me to apply the
> > patch that you put as attachment in Comment 15? Or is it Comment 19 or 20? I
> > am also not sure how to apply the patch, it looks like something to do with
> > git? Once I know how to apply the patch, to my understanding you then want
> > me to enable the "problematic mode", which would be setting the screen to
> > 144hz and then attachming a dmesg output yes?
> 
> Yes, bug 87796.  I'm interested in the dmesg output after switching to the
> problematic mode from the patch in comment 15.  See comment 17 for how to
> apply it.
> 
> The patches in 19 and 20 will workaround the issue while still letting you
> use auto.
Just thought I would post a little update, I am in the process of getting my system to use the new kernel. Ive already compiled it with the patch and stuff, i am just having a bit of trouble creating a ramdisk (I am followeing the arch linux wiki) although I will figure it out. Will the fixes from comment 19 and 20 be implemented into the next release of video-ati or does it still need testing? I am asking this so that if I know it isn't I may as well compile another kernel with the patches from 19 and 20. Thank you btw for everything.
Comment 14 Alex Deucher 2015-02-19 20:41:22 UTC
(In reply to andre35822 from comment #13)
> Just thought I would post a little update, I am in the process of getting my
> system to use the new kernel. Ive already compiled it with the patch and
> stuff, i am just having a bit of trouble creating a ramdisk (I am followeing
> the arch linux wiki) although I will figure it out. Will the fixes from
> comment 19 and 20 be implemented into the next release of video-ati or does
> it still need testing? I am asking this so that if I know it isn't I may as
> well compile another kernel with the patches from 19 and 20. Thank you btw
> for everything.

FWIW, using make install in your kernel tree will usually handle the ramdisk update, but I'm not that familiar with arch, so if you are comfortable with that, I'd stick with that.

As for the patches from 19 and 20, I'm pretty confident they will work, but it would be nice to get some confirmation.
Comment 15 andre35822 2015-02-20 01:07:10 UTC
Created attachment 113676 [details]
xorg log with patch from comment 15
Comment 16 andre35822 2015-02-20 01:07:28 UTC
Created attachment 113677 [details]
dmesg log with patch from comment 15
Comment 17 andre35822 2015-02-20 01:09:26 UTC
(In reply to Alex Deucher from comment #14)
> (In reply to andre35822 from comment #13)
> > Just thought I would post a little update, I am in the process of getting my
> > system to use the new kernel. Ive already compiled it with the patch and
> > stuff, i am just having a bit of trouble creating a ramdisk (I am followeing
> > the arch linux wiki) although I will figure it out. Will the fixes from
> > comment 19 and 20 be implemented into the next release of video-ati or does
> > it still need testing? I am asking this so that if I know it isn't I may as
> > well compile another kernel with the patches from 19 and 20. Thank you btw
> > for everything.
> 
> FWIW, using make install in your kernel tree will usually handle the ramdisk
> update, but I'm not that familiar with arch, so if you are comfortable with
> that, I'd stick with that.
> 
> As for the patches from 19 and 20, I'm pretty confident they will work, but
> it would be nice to get some confirmation.
Alright so I tried out the kernel I compiled with the patch from comment 15 on the other bug report, and either I didn't install the patch correctly (or it didnt install correctly) or the patch is not solving the problem. I am still getting the flickering/artifacts. I have attached an xorg log and dmesg log and added comment15_patch to each of their file names so they are easily distinguishable. Should I be compiling another kernel with the patches from comment 19 and 20?
Comment 18 Alex Deucher 2015-02-20 01:12:52 UTC
(In reply to andre35822 from comment #17)
> Alright so I tried out the kernel I compiled with the patch from comment 15
> on the other bug report, and either I didn't install the patch correctly (or
> it didnt install correctly) or the patch is not solving the problem. I am
> still getting the flickering/artifacts. I have attached an xorg log and
> dmesg log and added comment15_patch to each of their file names so they are
> easily distinguishable. Should I be compiling another kernel with the
> patches from comment 19 and 20?

The patch from comment 15 doesn't fix anything.  It just adds some additional debugging output that might be useful in fine tuning the patches in comments 19 and 20.  To fix the issue, use the patches in comments 19 and 20.
Comment 19 andre35822 2015-02-20 02:15:26 UTC
(In reply to Alex Deucher from comment #18)
> (In reply to andre35822 from comment #17)
> > Alright so I tried out the kernel I compiled with the patch from comment 15
> > on the other bug report, and either I didn't install the patch correctly (or
> > it didnt install correctly) or the patch is not solving the problem. I am
> > still getting the flickering/artifacts. I have attached an xorg log and
> > dmesg log and added comment15_patch to each of their file names so they are
> > easily distinguishable. Should I be compiling another kernel with the
> > patches from comment 19 and 20?
> 
> The patch from comment 15 doesn't fix anything.  It just adds some
> additional debugging output that might be useful in fine tuning the patches
> in comments 19 and 20.  To fix the issue, use the patches in comments 19 and
> 20.
Oh okay thats good then, that means I applied the patch correctly. Shall I apply the 19 and 20 patches now or would you prefer for a later date incase you fine-tune anything more?
Comment 20 andre35822 2015-02-20 03:21:58 UTC
(In reply to andre35822 from comment #19)
> (In reply to Alex Deucher from comment #18)
> > (In reply to andre35822 from comment #17)
> > > Alright so I tried out the kernel I compiled with the patch from comment 15
> > > on the other bug report, and either I didn't install the patch correctly (or
> > > it didnt install correctly) or the patch is not solving the problem. I am
> > > still getting the flickering/artifacts. I have attached an xorg log and
> > > dmesg log and added comment15_patch to each of their file names so they are
> > > easily distinguishable. Should I be compiling another kernel with the
> > > patches from comment 19 and 20?
> > 
> > The patch from comment 15 doesn't fix anything.  It just adds some
> > additional debugging output that might be useful in fine tuning the patches
> > in comments 19 and 20.  To fix the issue, use the patches in comments 19 and
> > 20.
> Oh okay thats good then, that means I applied the patch correctly. Shall I
> apply the 19 and 20 patches now or would you prefer for a later date incase
> you fine-tune anything more?
Sorry I misunderstood/forgot what patches 19 and 20 were. I will try them and see if they solve the issue.
Comment 21 andre35822 2015-02-20 03:56:31 UTC
I am actually having trouble c(In reply to Alex Deucher from comment #18)
> (In reply to andre35822 from comment #17)
> > Alright so I tried out the kernel I compiled with the patch from comment 15
> > on the other bug report, and either I didn't install the patch correctly (or
> > it didnt install correctly) or the patch is not solving the problem. I am
> > still getting the flickering/artifacts. I have attached an xorg log and
> > dmesg log and added comment15_patch to each of their file names so they are
> > easily distinguishable. Should I be compiling another kernel with the
> > patches from comment 19 and 20?
> 
> The patch from comment 15 doesn't fix anything.  It just adds some
> additional debugging output that might be useful in fine tuning the patches
> in comments 19 and 20.  To fix the issue, use the patches in comments 19 and
> 20.
Sorry for all of the messages. I am actually having trouble compiling the kernel with patches 19 and 20 and am getting radeon errors. Here is a pastebin of the errors incase you can help me http://pastebin.com/B2rFD9R0

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.