[Bug 1156254] New: External monitor flicker: [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe A or B
http://bugzilla.suse.com/show_bug.cgi?id=1156254 Bug ID: 1156254 Summary: External monitor flicker: [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe A or B Classification: openSUSE Product: openSUSE Tumbleweed Version: Current Hardware: x86-64 OS: Other Status: NEW Severity: Normal Priority: P5 - None Component: Kernel Assignee: kernel-maintainers@forge.provo.novell.com Reporter: dawei.pang@suse.com QA Contact: qa-bugs@suse.de Found By: --- Blocker: --- Working with recent Tumbleweed build (for example: 5.3.8-1-default), the external monitor connected to laptop by HDMI port randomly entered black then back. Find the related information as following: [ 665.983269] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=39753 end=39754) time 397 us, min 1073, max 1079, scanline start 1069, end 1096 [20229.291090] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=406108 end=406109) time 389 us, min 1073, max 1079, scanline start 1065, end 1091 ... [22495.478664] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=542080 end=542081) time 402 us, min 1073, max 1079, scanline start 1069, end 1096 [22495.628672] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=542089 end=542090) time 392 us, min 1073, max 1079, scanline start 1069, end 1096 My laptop is Dell Inc. Precision 5530, the discrete graphics card (01:00.0 NVIDIA Corporation GP107GLM) is disabled by bbswitch. Search some bugs filled already: https://bugs.freedesktop.org/show_bug.cgi?id=106107 https://bugs.freedesktop.org/show_bug.cgi?id=105537 -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1156254
Dawei Pang
http://bugzilla.suse.com/show_bug.cgi?id=1156254
http://bugzilla.suse.com/show_bug.cgi?id=1156254#c1
Takashi Iwai
http://bugzilla.suse.com/show_bug.cgi?id=1156254
http://bugzilla.suse.com/show_bug.cgi?id=1156254#c2
Dawei Pang
Is this a regression from some older versions?
Yeah, I think this is a regression, because the issue did not happen before (about half or one month ago). -Dawei -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1156254
http://bugzilla.suse.com/show_bug.cgi?id=1156254#c3
Takashi Iwai
http://bugzilla.suse.com/show_bug.cgi?id=1156254
http://bugzilla.suse.com/show_bug.cgi?id=1156254#c4
--- Comment #4 from Dawei Pang
Supposing it's a kernel regression, could you try to check which kernel version started showing the problem? You can find a few old TW packages in history repo, http://download.opensuse.org/history/
Just downgrade the kernel from that repo, and find out which one broke, a la bisection.
I'd recommend you to increase the number of max installable kernels before testing that, though. Edit /etc/zypp/zypp.conf and give more entries in the multiversion.kernels line, e.g.: multiversion.kernels = latest,latest-1,latest-2,latest-3,running
Hello Takashi, Since this issue is randomly reproduced, my testing may need some time, I will let know once finding the broken one. Thanks, Dawei -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1156254
http://bugzilla.suse.com/show_bug.cgi?id=1156254#c5
Dawei Pang
http://bugzilla.suse.com/show_bug.cgi?id=1156254
http://bugzilla.suse.com/show_bug.cgi?id=1156254#c6
--- Comment #6 from Takashi Iwai
http://bugzilla.suse.com/show_bug.cgi?id=1156254
http://bugzilla.suse.com/show_bug.cgi?id=1156254#c7
--- Comment #7 from Dawei Pang
So it's still happening with a month-old kernel. You may try 5.2.x kernel or 5.1.x kernel from my kernel repos, OBS home:tiwai:kernel:5.2 and OBS home:tiwai:kernel:5.1, too, just to confirm that it's really a kernel regression.
Thanks for your information! Have tried kernel 5.2.14-1.gaf75f09-default with extracted i915 firmware comes from 5.3.x kernel, the error information is reported on it. [ 593.757502] fuse: init (API version 7.31) [ 1789.376127] [drm:intel_cpu_fifo_underrun_irq_handler [i915]] *ERROR* CPU pipe A FIFO underrun [ 3759.840917] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=225373 end=225374) time 381 us, min 1073, max 1079, scanline start 1071, end 1097 [ 6706.548502] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=402175 end=402176) time 399 us, min 1073, max 1079, scanline start 1065, end 1092 ~> uname -a Linux 5.2.14-1.gaf75f09-default #1 SMP Tue Sep 10 10:23:57 UTC 2019 (af75f09) x86_64 x86_64 x86_64 GNU/Linux I will try 5.1 kernel and feedback result later. -Dawei -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1156254
http://bugzilla.suse.com/show_bug.cgi?id=1156254#c8
--- Comment #8 from Dawei Pang
http://bugzilla.suse.com/show_bug.cgi?id=1156254
http://bugzilla.suse.com/show_bug.cgi?id=1156254#c9
Takashi Iwai
http://bugzilla.suse.com/show_bug.cgi?id=1156254
http://bugzilla.suse.com/show_bug.cgi?id=1156254#c10
--- Comment #10 from Dawei Pang
The fact that 5.1.x still shows the problem implies that the issue isn't about the kernel code change. A firmware update might be the culprit, indeed.
The kernel-firmware package is basically kernel-version agnostic, and you can upgrade / downgrade as you like (unless the latest kernel driver requires a newer firmware). That is, download some old kernel-firmware.rpm from the TW history repo, and install it via zypper instlal --oldpackage option. It'll rebuild the initrd's, too.
Let me know if this is really a kernel firmware problem.
Hello Takashi, On my laptop, i915 driver loads firmware i915/kbl_dmc_ver1_04.bin, the last modified date of this file is 2017-11-01 in linux-firmware git, hope SUSE does not touch it recently. So this issue may be not related to kernel firmware. Thanks, Dawei -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1156254
http://bugzilla.suse.com/show_bug.cgi?id=1156254#c11
--- Comment #11 from Takashi Iwai
http://bugzilla.suse.com/show_bug.cgi?id=1156254
http://bugzilla.suse.com/show_bug.cgi?id=1156254#c12
--- Comment #12 from Dawei Pang
Hm, then possibly it's other changes, e.g. now runtime PM is enabled or such. Could you try to boot with the following options? i915.enable_dc=0 915.disable_power_well=0
Hello, I added kernel boot parameter "i915.enable_dc=0 915.disable_power_well=0" as your mentioned, but i915 still reports the similar information as following: [16697.976015] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe A (start=213482 end=213483) time 392 us, min 1073, max 1079, scanline start 1069, end 1096 Thanks, Dawei -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.suse.com/show_bug.cgi?id=1156254
http://bugzilla.suse.com/show_bug.cgi?id=1156254#c13
Dawei Pang
http://bugzilla.suse.com/show_bug.cgi?id=1156254
http://bugzilla.suse.com/show_bug.cgi?id=1156254#c14
Patrik Jakobsson
http://bugzilla.suse.com/show_bug.cgi?id=1156254
http://bugzilla.suse.com/show_bug.cgi?id=1156254#c15
--- Comment #15 from Patrik Jakobsson
https://bugzilla.suse.com/show_bug.cgi?id=1156254
https://bugzilla.suse.com/show_bug.cgi?id=1156254#c16
Miroslav Beneš
participants (2)
-
bugzilla_noreply@novell.com
-
bugzilla_noreply@suse.com