Hide Forgot
Description of problem: Display blinks randomly. Version-Release number of selected component (if applicable): kernel 4.6.0.rc3.git0.1.fc23 How reproducible: Occurs randomly Actual results: Display blinks (or brightness goes to 0%, not possible to differ). Occurs very randomly, I haven't found any patter to reproduce this behavior yet. Problems started with kernel 4.6 rc1 (4.5 is OK). Expected results: Display should not randomly blink. Additional info: I am running kernel from my COPR repository: https://copr.fedorainfracloud.org/coprs/frantisekz/NodebugFedora23Kernel/ . Other components are from Fedora 23 updates-testing (updated recently). GPU: Intel HD 4400
(In reply to František Zatloukal from comment #0) > Description of problem: > Display blinks randomly. > > Version-Release number of selected component (if applicable): > kernel 4.6.0.rc3.git0.1.fc23 > > How reproducible: > Occurs randomly > > Actual results: > Display blinks (or brightness goes to 0%, not possible to differ). Occurs > very randomly, I haven't found any patter to reproduce this behavior yet. > Problems started with kernel 4.6 rc1 (4.5 is OK). > > Expected results: > Display should not randomly blink. > > Additional info: > I am running kernel from my COPR repository: > https://copr.fedorainfracloud.org/coprs/frantisekz/NodebugFedora23Kernel/ . > Other components are from Fedora 23 updates-testing (updated recently). > GPU: Intel HD 4400 The kernel doesn't use ncurses. There's really no reason for this COPR to exist afaik. Can you please test the official F25 build?
(In reply to Josh Boyer from comment #1 > The kernel doesn't use ncurses. There's really no reason for this COPR to > exist afaik. I remember I was creating that repository after i wasn't able to upgrade kernel because of some ncurses dependency cannot been satisfied. Anyway, some people are using that repo right now, so I don't think it's good idea to drop it. > Can you please test the official F25 build? Yes, I've tested this one: http://koji.fedoraproject.org/koji/buildinfo?buildID=753397 and result is same.
(In reply to František Zatloukal from comment #2) > (In reply to Josh Boyer from comment #1 > > The kernel doesn't use ncurses. There's really no reason for this COPR to > > exist afaik. > I remember I was creating that repository after i wasn't able to upgrade > kernel because of some ncurses dependency cannot been satisfied. Anyway, > some people are using that repo right now, so I don't think it's good idea > to drop it. The more people that use it, the more people that will be left with an unsupported kernel. We typically just close those as wontfix. If you would like to continue to provide it, please at least change the disttag to reflect the fact that it was built against a different Fedora release. > > > Can you please test the official F25 build? > Yes, I've tested this one: > http://koji.fedoraproject.org/koji/buildinfo?buildID=753397 and result is > same. Please attach the output of dmesg after the issue has happened.
(In reply to Josh Boyer from comment #3) > (In reply to František Zatloukal from comment #2) > > (In reply to Josh Boyer from comment #1 > > > The kernel doesn't use ncurses. There's really no reason for this COPR to > > > exist afaik. > > I remember I was creating that repository after i wasn't able to upgrade > > kernel because of some ncurses dependency cannot been satisfied. Anyway, > > some people are using that repo right now, so I don't think it's good idea > > to drop it. > > The more people that use it, the more people that will be left with an > unsupported kernel. We typically just close those as wontfix. If you would > like to continue to provide it, please at least change the disttag to > reflect the fact that it was built against a different Fedora release. Thanks for the tip, but I think copr already do this for me (dnf info kernel returns fc23) or am I wrong? (In reply to Josh Boyer from comment #3) > Please attach the output of dmesg after the issue has happened. There is nothing suspicious after blinking occurred, the last line is: [14861.828160] usb 2-6: reset full-speed USB device number 4 using xhci_hcd
(In reply to František Zatloukal from comment #4) > (In reply to Josh Boyer from comment #3) > > (In reply to František Zatloukal from comment #2) > > > (In reply to Josh Boyer from comment #1 > > > > The kernel doesn't use ncurses. There's really no reason for this COPR to > > > > exist afaik. > > > I remember I was creating that repository after i wasn't able to upgrade > > > kernel because of some ncurses dependency cannot been satisfied. Anyway, > > > some people are using that repo right now, so I don't think it's good idea > > > to drop it. > > > > The more people that use it, the more people that will be left with an > > unsupported kernel. We typically just close those as wontfix. If you would > > like to continue to provide it, please at least change the disttag to > > reflect the fact that it was built against a different Fedora release. > Thanks for the tip, but I think copr already do this for me (dnf info kernel > returns fc23) or am I wrong? No, you are correct. I thought it did that too, but was confused by the webpage because I was looking at the SRPM you used which just points directly to koji. > (In reply to Josh Boyer from comment #3) > > Please attach the output of dmesg after the issue has happened. > > There is nothing suspicious after blinking occurred, the last line is: > [14861.828160] usb 2-6: reset full-speed USB device number 4 using xhci_hcd Could you please just attach it anyway?
Created attachment 1146559 [details] dmesg after blink
(In reply to Josh Boyer from comment #5) > Could you please just attach it anyway? I've attached it :)
Bug appears even after upgrade to Fedora 24 (with same version of kernel).
It's probably problem with brightness - external monitor is not dimming at all.
Issue still present in 4.6.0.rc6.
Created attachment 1182304 [details] Dmesg file of the current boot I've Intel Iris Pro 5200 on Intel Haswell Core i7-4770HQ with Fedora 24, and I notice this issue every once in a while (happens very random, but not if I'm watching a video). I didn't notice the brightness thing, but it sure blink for a moment in random times. This issue happens under both X.org and Wayland. I'm using GNOME 3 if it's important to know. This issue appeared with Kernel-4.6.x, but If I booted the old Kernel-4.5.x, it disappears completely. I've also tried the option i915.enable_rc6=0 in the linux cmdline (/etc/default/grup) but nothing happened. Kernels: kernel-4.6.4-301.fc24.x86_64 (affected, but less than the previous one) kernel-4.6.3-300.fc24.x86_64 (affected with a lot of random blinking) kernel-4.5.7-300.fc24.x86_64 (not affected) Intel Driver: xorg-x11-drv-intel-2.99.918-1.20160709.fc24.x86_64 (Installed from bollocks intelblood Copr, but the issue existed before + kernel-4.5.x does work with it without a problem). Attached with this message, the dmesg output for the current boot.
Created attachment 1182305 [details] Xorg.0.log Attached the X.org log file for the current session, in which I've seen the blinking. Maybe there's some information that's useful for troubleshooting the issue. If you need any other log files, please tell me.
For more information about my system, check this fpaste: https://paste.fedoraproject.org/393099/46906584/
This bug appears to have been reported against 'rawhide' during the Fedora 25 development cycle. Changing version to '25'.
Using Fedora 24 on a i7-4702MQ having HD Graphics 4600, I can reproduce the flickering. The flickering does not occur while I am moving the mouse cursor but happens immediately when I stop moving. Passing the i915.enable_psr=0 option to the kernel stops the flickering. https://paste.fedoraproject.org/396329/24927146/
Same problem here. Using: Intel(R) Core(TM) i7-4712HQ CPU @ 2.30GHz 00:02.0 VGA compatible controller: Intel Corporation 4th Gen Core Processor Integrated Graphics Controller (rev 06) The flickering problem really became serious after the upgrade to Fedora 25. The laptop's internal display is fine. External displays cannot stay sync'd or something. The external display is a Samsung, EDID data 'SAM', 'U28E590', 'HTPGC00137' and it is running (or trying to) at "mode 3840x2160, preferred, current" Another interesting data point is that if I run glxgears the external monitor stops flickering. I assume this iss because the GPU is kept busy and does not idle.
Oh, and this problem is happening to me with Fedora 25's kernel 4.8.0-0.rc4.git0.1.fc25.x86_64
Note a possible workaround for this is adding "i915.enable_psr=0" to your kernel commandline.
Yes that does fix it.
(In reply to Jonathan Briggs from comment #19) > Yes that does fix it. Ok good to know, can you please file a bug for this upstream, as documented here: https://01.org/linuxgraphics/documentation/how-report-bugs And make sure to note that i915.enable_psr=0 fixes it. We've been seeing quite a few "flickering" bugs related to psr recently. Hopefully upstream will be able to find some common denominator and come up with a proper fix.
It appears this bug will be fixed in kernel 4.8. See https://lkml.org/lkml/2016/8/1/652
I am running the same issue but in my case the monitor gets black like if it was shut down and after a sec or two it comes back. If anyone needs something from my side let me know
The bug is still present in kernel version 4.8.1-1.fc25.x86_64. This is the version used in the Fedora 25 Wayland Test Day Image. (https://fedoraproject.org/wiki/Test_Day:2016-10-13_Wayland) My upstream bug report: https://bugs.freedesktop.org/show_bug.cgi?id=97875
I am experiencing this issue in Fedora 25 with 4.8.6-300.fc25.x86_64 on a laptop with Haswell and an AMD HD8750M discrete gpu. I do not experience this issue in Windows 10, Ubuntu 16.10 on Unity, Tumbleweed/KDE, nor did experience it in F20-F23, but do experience it in Tumbleweed/GNOME using both Wayland and X11. In KDE I had a possibly related issue with also-seemingly random intermittent artifacting that was resolved by changing the compositor from the default OpenGL2.0 to either OpenGL3.1 or KWin. There is nothing in dmesg or journal that coincides with the blinks, and upgrading to a X11/Mesa repo which tracks git on tumbleweed (on the advice of #amdgpu) did not fix the issue there. lspci and lsmod below: https://paste.fedoraproject.org/469881/31093147/ http://paste.fedoraproject.org/469883/82312101/
(In reply to perosredo from comment #24) > I am experiencing this issue in Fedora 25 with 4.8.6-300.fc25.x86_64 on a > laptop with Haswell and an AMD HD8750M discrete gpu. > > I do not experience this issue in Windows 10, Ubuntu 16.10 on Unity, > Tumbleweed/KDE, nor did experience it in F20-F23, but do experience it in > Tumbleweed/GNOME using both Wayland and X11. > > In KDE I had a possibly related issue with also-seemingly random > intermittent artifacting that was resolved by changing the compositor from > the default OpenGL2.0 to either OpenGL3.1 or KWin. > > There is nothing in dmesg or journal that coincides with the blinks, and > upgrading to a X11/Mesa repo which tracks git on tumbleweed (on the advice > of #amdgpu) did not fix the issue there. lspci and lsmod below: > > > https://paste.fedoraproject.org/469881/31093147/ > http://paste.fedoraproject.org/469883/82312101/ Note a possible workaround for this is adding "i915.enable_psr=0" to your kernel commandline, habe you tried this ?
(In reply to Hans de Goede from comment #25) > Note a possible workaround for this is adding "i915.enable_psr=0" to your > kernel commandline, habe you tried this ? The workaround does fix the blinking in F25 on both X and Wayland, albeit I have no idea what the command does, but does seem to introduce a single almost-persistent horizontal scanline only on X. Also might not explain why the bug does not occur in Ubuntu/Unity or Tumbleweed/KDE on contemporary kernels.
(In reply to perosredo from comment #26) > Also might not explain why the bug does not occur in Ubuntu/Unity or > Tumbleweed/KDE on contemporary kernels. For Panel Self Refresh to kick in to start saving power, the GPU has to stop updating the screen. If the compositor keeps writing updates then that can't happen. That's why the flickering only happens with some software combinations. PSR is worth the temporary annoyance of these bugs. It saves much more power than idling SATA, gigabit Ethernet or the like. That eDP update link to the display is normally pushing gigabits per second.
I´ve removed xorg-x11-drv-intel and I am no longer encountering this issue. You can try this too, system will fall back to using generic 2D driver. This is going to be default configuration (if I am not mistaken) in Fedora 26. @ajax, Should I close this bug?
PSR has been disabled by default for Haswell/Broadwell in kernel 4.10 (see https://bugs.freedesktop.org/show_bug.cgi?id=97875). I'm still using Fedora 24 (on 4.9.7) and have not been able to test it.
Upgraded to 4.9.13-100.fc24.x86_64 and see no more flickering. # cat /sys/module/i915/parameters/enable_psr 0 PSR is disabled by default for my Haswell processor.
This message is a reminder that Fedora 25 is nearing its end of life. Approximately 4 (four) weeks from now Fedora will stop maintaining and issuing updates for Fedora 25. It is Fedora's policy to close all bug reports from releases that are no longer maintained. At that time this bug will be closed as EOL if it remains open with a Fedora 'version' of '25'. Package Maintainer: If you wish for this bug to remain open because you plan to fix it in a currently maintained version, simply change the 'version' to a later Fedora version. Thank you for reporting this issue and we are sorry that we were not able to fix it before Fedora 25 is end of life. If you would still like to see this bug fixed and are able to reproduce it against a later version of Fedora, you are encouraged change the 'version' to a later Fedora version prior this bug is closed as described in the policy above. Although we aim to fix as many bugs as possible during every release's lifetime, sometimes those efforts are overtaken by events. Often a more recent Fedora release includes newer upstream software that fixes bugs or makes them obsolete.
Fixed for me on Fedora 26.
Yeah, I removed the `enable_psr` a long time ago, and it never happened to me again. I think it's fixed. I'm using F27 now.
Fedora 25 changed to end-of-life (EOL) status on 2017-12-12. Fedora 25 is no longer maintained, which means that it will not receive any further security or bug fix updates. As a result we are closing this bug. If you can reproduce this bug against a currently maintained version of Fedora please feel free to reopen this bug against that version. If you are unable to reopen this bug, please file a new report against the current release. If you experience problems, please add a comment to this bug. Thank you for reporting this bug and we are sorry it could not be fixed.