[opensuse-factory] [TW] Graphics glitches (HD530)
Hi is anyone else experiencing graphics glitches in Tumbleweed since a few snapshots back? They show up as horizontal (short) bars on fast-changing features. E.g., I can see them when resizing an xterm horizontally forth and back. On the window border that I'm extending I see jaggy features. Another way to see them is switching desktop (I use slide), often the window title bar of one or more windows on the screen I slide to shows the artifacts. Quite prominent they also show up when using optirun and the secondary nvidia card: optirun glxspheres shows a LOT of those bars in/around the big centrall ball when it pops up, or when running an unigine benchmark. I do not see them using primusrun / optirun -b primus, though. But it is not related to the nvidia card, I also see it with the card turned off, as in the other examples above. It definitely is a new issue that I did not see before (I have the laptop since August last year). Unfortunately I cannot tell when exactly it started. I think at most 2-3 weeks. HW is a Lenovo T460p (Skylake, Optimus HD530/940MX), Desktop is plasma5. TW is up-to-date (just did the gcc7 Mamut-dup...) Am I the only one? -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Peter Suetterlin wrote:
HW is a Lenovo T460p (Skylake, Optimus HD530/940MX), Desktop is plasma5. TW is up-to-date (just did the gcc7 Mamut-dup...)
Writing this (plasma5) I thought I should also test some other DE. All I have installed is IceWM, there I cannot see the artifacts. So likely a plasma/kwin issue? -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Fri, Jun 2, 2017 at 1:19 PM, Peter Suetterlin
Peter Suetterlin wrote:
HW is a Lenovo T460p (Skylake, Optimus HD530/940MX), Desktop is plasma5. TW is up-to-date (just did the gcc7 Mamut-dup...)
Writing this (plasma5) I thought I should also test some other DE. All I have installed is IceWM, there I cannot see the artifacts. So likely a plasma/kwin issue?
IceWM might not use hardware acceleration for rendering. You might want to try disabling hardware acceleration for rendering in KDE. Also try Gnome-Shell, this one surely uses hardware acceleration. Robert -- http://robert.muntea.nu/ -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 02/06/2017 12:13, Peter Suetterlin wrote:
Hi
is anyone else experiencing graphics glitches in Tumbleweed since a few snapshots back?
They show up as horizontal (short) bars on fast-changing features. E.g., I can see them when resizing an xterm horizontally forth and back. On the window border that I'm extending I see jaggy features. Another way to see them is switching desktop (I use slide), often the window title bar of one or more windows on the screen I slide to shows the artifacts.
Quite prominent they also show up when using optirun and the secondary nvidia card: optirun glxspheres shows a LOT of those bars in/around the big centrall ball when it pops up, or when running an unigine benchmark. I do not see them using primusrun / optirun -b primus, though. But it is not related to the nvidia card, I also see it with the card turned off, as in the other examples above.
It definitely is a new issue that I did not see before (I have the laptop since August last year). Unfortunately I cannot tell when exactly it started. I think at most 2-3 weeks.
HW is a Lenovo T460p (Skylake, Optimus HD530/940MX), Desktop is plasma5. TW is up-to-date (just did the gcc7 Mamut-dup...)
Am I the only one?
I've been suffering from graphical gliteches since a long time ago, more than 1 year ago, since I switched from Leap to Thumbleweed. I've been able to fix a few (not all) by lowering some aggressive hardware acceleration options. Section "Device" Identifier "Radeon" Driver "radeon" Option "AccelMethod" "EXA" Option "DRI" "3" Option "TearFree" "on" Option "ColorTiling" "off" Option "ColorTiling2D" "off" EndSection I have an old HD6850 card. 06:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Barts PRO [Radeon HD 6850] -- Andrea "Kontorotsui" Controzzi Contact: +39 392 9989834 - +39 050 644097 Skype: Kontorotsui Settore tecnico / Technical Department - www.LedMania.it ----- LedMania SRL unipersonale Via Galilei 27 56042 Lavoria (PI) P.IVA: 01941970509 -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Robert Munteanu wrote:
IceWM might not use hardware acceleration for rendering. You might want to try disabling hardware acceleration for rendering in KDE.
First thought: Um, where would I disable that? *shrug* The only reference I possibly found was Compositor in Hardware->Display and Monitor. But it is not enabled. Changed the Rendering backend nevertheless to 'XRender' and that seems to remove the issue. Playing around with the available options (XRender, OpenGL 2.0 and 3.1, initial setting been 2.0) I suddenly could not reproduce the issue anymore with any of the backends, until I noticed it only shows up if the compositor is DISabled! But then with any backend. This is weird....
Also try Gnome-Shell, this one surely uses hardware acceleration.
Don't really feel like pulling in all that gnome stuff just for a test :( -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Friday, 2 June 2017 11:13:17 BST Peter Suetterlin wrote:
Hi
is anyone else experiencing graphics glitches in Tumbleweed since a few snapshots back?
They show up as horizontal (short) bars on fast-changing features. E.g., I can see them when resizing an xterm horizontally forth and back. On the window border that I'm extending I see jaggy features. Another way to see them is switching desktop (I use slide), often the window title bar of one or more windows on the screen I slide to shows the artifacts.
Quite prominent they also show up when using optirun and the secondary nvidia card: optirun glxspheres shows a LOT of those bars in/around the big centrall ball when it pops up, or when running an unigine benchmark. I do not see them using primusrun / optirun -b primus, though. But it is not related to the nvidia card, I also see it with the card turned off, as in the other examples above.
It definitely is a new issue that I did not see before (I have the laptop since August last year). Unfortunately I cannot tell when exactly it started. I think at most 2-3 weeks.
HW is a Lenovo T460p (Skylake, Optimus HD530/940MX), Desktop is plasma5. TW is up-to-date (just did the gcc7 Mamut-dup...)
Am I the only one?
I am running an old Toshiba (Satellite A500-1GL) with Nvidia 330m chipset. No such problems here. It is about seven years old. Regards. -- Sudhir Anand -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
02.06.2017 13:13, Peter Suetterlin пишет:
Hi
is anyone else experiencing graphics glitches in Tumbleweed since a few snapshots back?
They show up as horizontal (short) bars on fast-changing features. E.g., I can see them when resizing an xterm horizontally forth and back. On the window border that I'm extending I see jaggy features. Another way to see them is switching desktop (I use slide), often the window title bar of one or more windows on the screen I slide to shows the artifacts.
Are you using intel or modesetting Xorg driver? -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Andrei Borzenkov wrote:
02.06.2017 13:13, Peter Suetterlin пишет:
Hi
is anyone else experiencing graphics glitches in Tumbleweed since a few snapshots back?
They show up as horizontal (short) bars on fast-changing features. E.g., I can see them when resizing an xterm horizontally forth and back. On the window border that I'm extending I see jaggy features. Another way to see them is switching desktop (I use slide), often the window title bar of one or more windows on the screen I slide to shows the artifacts.
Are you using intel or modesetting Xorg driver?
No idea :o I'm running on default settings and never checked what 'default' is(*).... Checking the log I see [ 6.371] (==) Matched modesetting as autoconfigured driver 0 [ 6.371] (==) Matched intel as autoconfigured driver 1 ... [ 6.371] (II) LoadModule: "modesetting" ... [ 6.372] (II) LoadModule: "intel" ... [ 6.537] (II) UnloadModule: "intel" So it's the modesetting one. (*)googling a bit it seems this is a recently ongoing issue, switching to modesetting. Any word if for TW such a change happened recently? -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Peter Suetterlin composed on 2017-06-05 12:39 (UTC+0100):
Andrei Borzenkov wrote:
Are you using intel or modesetting Xorg driver?
No idea :o I'm running on default settings and never checked what 'default' is(*)....
Checking the log I see [ 6.371] (==) Matched modesetting as autoconfigured driver 0 [ 6.371] (==) Matched intel as autoconfigured driver 1 ... [ 6.371] (II) LoadModule: "modesetting" ... [ 6.372] (II) LoadModule: "intel" ... [ 6.537] (II) UnloadModule: "intel"
So it's the modesetting one.
(*)googling a bit it seems this is a recently ongoing issue, switching to modesetting. Any word if for TW such a change happened recently?
Check if the pci ID of your gfxchip matches any listed in /etc/X11/xorg_pci_ids/modesetting.ids. -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Felix Miata wrote:
Peter Suetterlin composed on 2017-06-05 12:39 (UTC+0100):
So it's the modesetting one.
(*)googling a bit it seems this is a recently ongoing issue, switching to modesetting. Any word if for TW such a change happened recently?
Check if the pci ID of your gfxchip matches any listed in /etc/X11/xorg_pci_ids/modesetting.ids.
Obviously it is (8086191B), else it wouldn't use the modesetting one :) The question rather was has that always been the case or/and has the ranking of modesetting vs. intel changed lately? Probably not, rpm -q --changelog xorg-x11-server lists * Sun May 08 2016 eich@suse.com - modesetting.ids: Add file for PCI IDs of ASICs which the modesetting rather than the native driver should be used for. This includes all Intel Gen9+ hardware (boo#978954). as last relevant change... -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
05.06.2017 14:39, Peter Suetterlin пишет:
Andrei Borzenkov wrote:
02.06.2017 13:13, Peter Suetterlin пишет:
Hi
is anyone else experiencing graphics glitches in Tumbleweed since a few snapshots back?
They show up as horizontal (short) bars on fast-changing features. E.g., I can see them when resizing an xterm horizontally forth and back. On the window border that I'm extending I see jaggy features. Another way to see them is switching desktop (I use slide), often the window title bar of one or more windows on the screen I slide to shows the artifacts.
Are you using intel or modesetting Xorg driver?
No idea :o I'm running on default settings and never checked what 'default' is(*)....
Checking the log I see [ 6.371] (==) Matched modesetting as autoconfigured driver 0 [ 6.371] (==) Matched intel as autoconfigured driver 1 ... [ 6.371] (II) LoadModule: "modesetting" ... [ 6.372] (II) LoadModule: "intel" ... [ 6.537] (II) UnloadModule: "intel"
So it's the modesetting one.
Personally I have rather negative experience with modesetting driver on Intel HD 5500; I switched back to intel. You may try whether it makes any difference. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Peter Suetterlin composed on 2017-06-05 14:20 (UTC+0100):
Obviously it is (8086191B), else it wouldn't use the modesetting one :)
The question rather was has that always been the case or/and has the ranking of modesetting vs. intel changed lately? Probably not, rpm -q --changelog xorg-x11-server lists
* Sun May 08 2016 eich@suse.com - modesetting.ids: Add file for PCI IDs of ASICs which the modesetting rather than the native driver should be used for. This includes all Intel Gen9+ hardware (boo#978954).
as last relevant change...
That same changelog has a March 17 entry pointing to https://bugzilla.opensuse.org/show_bug.cgi?id=1021803 that looks suspiciously like a possible cause of which your OP complained: "graphics glitches...show up as horizontal (short) bars on fast-changing features". -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Felix Miata wrote:
Peter Suetterlin composed on 2017-06-05 14:20 (UTC+0100):
rpm -q --changelog xorg-x11-server lists
* Sun May 08 2016 eich@suse.com - modesetting.ids: Add file for PCI IDs of ASICs which the modesetting rather than the native driver should be used for. This includes all Intel Gen9+ hardware (boo#978954).
as last relevant change...
That same changelog has a March 17 entry pointing to https://bugzilla.opensuse.org/show_bug.cgi?id=1021803 that looks suspiciously like a possible cause of which your OP complained:
"graphics glitches...show up as horizontal (short) bars on fast-changing features".
Whee, I wouldn't have spotted that from the changelog itself :o But reading through the quite extended thread I'm not sure if it is the same thing - there it really was rather single pixels missing, if I read this correct (but TBH I did not understand too much of that discussion at all...) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Andrei Borzenkov wrote:
Personally I have rather negative experience with modesetting driver on Intel HD 5500; I switched back to intel. You may try whether it makes any difference.
So far I couldn't complain TBH. Nevertheless, I commented out the entry for my HD530 in modesetting.ids and now it uses the intel one. No artifacts with that driver. Guess I'll stay with it, too. :) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Tue, Jun 6, 2017 at 12:57 PM, Peter Suetterlin
Andrei Borzenkov wrote:
Personally I have rather negative experience with modesetting driver on Intel HD 5500; I switched back to intel. You may try whether it makes any difference.
So far I couldn't complain TBH. Nevertheless, I commented out the entry for my HD530 in modesetting.ids and now it uses the intel one.
No artifacts with that driver. Guess I'll stay with it, too. :)
This really calls for bug report. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Andrei Borzenkov wrote:
On Tue, Jun 6, 2017 at 12:57 PM, Peter Suetterlin
wrote: Andrei Borzenkov wrote:
Personally I have rather negative experience with modesetting driver on Intel HD 5500; I switched back to intel. You may try whether it makes any difference.
So far I couldn't complain TBH. Nevertheless, I commented out the entry for my HD530 in modesetting.ids and now it uses the intel one.
No artifacts with that driver. Guess I'll stay with it, too. :)
This really calls for bug report.
Sorry, yes, should have posted the link immediately: https://bugzilla.opensuse.org/show_bug.cgi?id=1042873 -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
participants (6)
-
Andrea Controzzi - LedMania.it
-
Andrei Borzenkov
-
Felix Miata
-
Peter Suetterlin
-
Robert Munteanu
-
Sudhir Anand