[Bug 102972] Flickering since 4.11

* [Bug 102972] Flickering since 4.11
@ 2017-09-25 12:20 bugzilla-daemon
  2017-09-25 12:20 ` bugzilla-daemon
                   ` (10 more replies)
  0 siblings, 11 replies; 12+ messages in thread
From: bugzilla-daemon @ 2017-09-25 12:20 UTC (permalink / raw)
  To: dri-devel

[-- Attachment #1.1: Type: text/plain, Size: 2181 bytes --]

https://bugs.freedesktop.org/show_bug.cgi?id=102972

            Bug ID: 102972
           Summary: Flickering since 4.11
           Product: DRI
           Version: XOrg git
          Hardware: All
                OS: Linux (All)
            Status: NEW
          Severity: normal
          Priority: medium
         Component: DRM/other
          Assignee: dri-devel@lists.freedesktop.org
          Reporter: lionel-fdo@lefolgoc.net

Created attachment 134458
  --> https://bugs.freedesktop.org/attachment.cgi?id=134458&action=edit
bisect log (drm)

Hi,

(I've only Intel hardware available but since git bisect shows the culprit in
drm and not drm/i915, I've filed it against DRM/other, feel free to reassign)

Since 4.11, I've noticed major flickering on some displays I have (reproducible
100%). By flickering I mean the screen is mostly black and every few seconds I
can see a flash with the screen content and it goes black again.
It only occurs with the preferred mode, i.e. when I force a modeline to another
mode in xorg.conf the flickering does not occur.

I've tried all versions up to latest 4.14.0-rc2 and it is still reproducible,
4.10.x were fine. The display driver used (intel or modesetting) does not make
a difference.

I tried on all hardware I have here (NUC gen4 d5420, NUC5i5 and NUC6i5SYB) and
the behavior is the same. I tried miniDP and HDMI, and it does not change
anything.

The captured logs below have been run on a NUC6i5SYB connected with miniDP,
running CentOS 7.3, with the kernel 4.11.8 (from mainline, no ct7 or rhel
patches). The git bisect was also run on this machine.

Git bisect identified fcc8a22cc9053a8d1bbb94833ec103cd5961feef as first bad
commit.
With the log files below, 4.11.8-1 is "original" 4.11.8 kernel (bad), and
4.11.8-2 includes the commit reverted (good).

>From the git bisect log and the fact it can be worked around with another
modeline, I'm not even sure this is a regression in the kernel, as this might
be that the screen is advertising something it cannot support properly...

Thanks.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[-- Attachment #1.2: Type: text/html, Size: 3584 bytes --]

[-- Attachment #2: Type: text/plain, Size: 160 bytes --]

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

^ permalink raw reply	[flat|nested] 12+ messages in thread