Bug 112400 - MST configuration breaks after resume with monitor switched to different source
Summary: MST configuration breaks after resume with monitor switched to different source
Status: RESOLVED MOVED
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: unspecified
Hardware: x86-64 (AMD64) Linux (All)
: not set not set
Assignee: Intel GFX Bugs mailing list
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard: Triaged
Keywords:
Depends on:
Blocks:
 
Reported: 2019-11-26 10:47 UTC by Simon Richter
Modified: 2019-11-29 19:52 UTC (History)
1 user (show)

See Also:
i915 platform: KBL
i915 features: display/DP MST


Attachments
Kernel log (519.76 KB, text/plain)
2019-11-26 10:49 UTC, Simon Richter
no flags Details

Description Simon Richter 2019-11-26 10:47:54 UTC
I use a MST based dual monitor setup where the first monitor in the chain has multiple inputs and allows me to switch between input sources.

When switching away from my Linux box, there is a good chance that afterwards the configuration will be invalid, and after switching back, no picture will be shown.

I usually hibernate the machines at night, and resuming with the monitor switched to the other source always triggers this condition (while switching away during normal operation only has a high probability).

Hibernating again and making sure to boot up with the monitor set to the right source does not fix the problem -- the firmware correctly shows the logo on both monitors, a text mode screen with a blinking cursor is shown during resume, then the invalid configuration is reloaded.

Attached kernel log is with a Debian standard kernel,

Linux debian 4.19.0-6-amd64 #1 SMP Debian 4.19.67-2+deb10u1 (2019-09-20) x86_64 GNU/Linux

 - Normal boot
 - X start
 - hibernate
 - resume
 - X shutdown
 - X restart
Comment 1 Simon Richter 2019-11-26 10:49:21 UTC
Created attachment 146025 [details]
Kernel log
Comment 2 Jani Saarinen 2019-11-26 13:19:55 UTC
There has been MST fixes lately on drm-tip: https://cgit.freedesktop.org/drm-tip can you test with that and report back? Not necessarily related but eg. https://bugs.freedesktop.org/show_bug.cgi?id=103643
Comment 3 Jani Saarinen 2019-11-26 13:21:35 UTC
Also send logs from drm-tip from dmesg with drm.debug=0x1e log_buf_len=4M if problem still exists.
Comment 4 Simon Richter 2019-11-27 12:03:40 UTC
Just to make sure: including 0x10 (ATOMIC)?
Comment 5 Jani Saarinen 2019-11-27 12:06:15 UTC
No, why?
Comment 6 Lakshmi 2019-11-27 12:37:13 UTC
(In reply to Simon Richter from comment #0)
> I use a MST based dual monitor setup where the first monitor in the chain
> has multiple inputs and allows me to switch between input sources.
> 
> When switching away from my Linux box, there is a good chance that
> afterwards the configuration will be invalid, and after switching back, no
> picture will be shown.
> 
> I usually hibernate the machines at night, and resuming with the monitor
> switched to the other source always triggers this condition (while switching
> away during normal operation only has a high probability).
> 
> Hibernating again and making sure to boot up with the monitor set to the
> right source does not fix the problem -- the firmware correctly shows the
> logo on both monitors, a text mode screen with a blinking cursor is shown
> during resume, then the invalid configuration is reloaded.
> 
> Attached kernel log is with a Debian standard kernel,
> 
> Linux debian 4.19.0-6-amd64 #1 SMP Debian 4.19.67-2+deb10u1 (2019-09-20)
> x86_64 GNU/Linux
> 
>  - Normal boot
>  - X start
>  - hibernate
>  - resume
>  - X shutdown
>  - X restart

Which platform is this? I couldn't find it from the kernel log.
Comment 7 Simon Richter 2019-11-27 13:00:47 UTC
@Jari, the command line you gave me has 0x1e, that's why I'm asking.

@Lakshmi, the beginning got cut off -- it's a 630 (8086:5912 rev 04).
Comment 8 Lakshmi 2019-11-28 12:48:58 UTC
(In reply to Simon Richter from comment #7)
> @Jari, the command line you gave me has 0x1e, that's why I'm asking.
> 
> @Lakshmi, the beginning got cut off -- it's a 630 (8086:5912 rev 04).

Thanks Simon, Can you reproduce this issue with drmtip kernel as Jani suggested?
Comment 9 Simon Richter 2019-11-28 16:12:24 UTC
Will try after the weekend, need to do my taxes first and build kernel packages -- then I can also test the other bug report I made.
Comment 10 Martin Peres 2019-11-29 19:52:39 UTC
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/drm/intel/issues/650.


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.