Bug 98324 - [DC] amd-staging-4.7: problems with unblanking displays when monitors are switched off
Summary: [DC] amd-staging-4.7: problems with unblanking displays when monitors are swi...
Status: RESOLVED MOVED
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/AMDgpu (show other bugs)
Version: unspecified
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Default DRI bug account
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2016-10-19 17:23 UTC by Darren Salt
Modified: 2019-11-19 08:10 UTC (History)
3 users (show)

See Also:
i915 platform:
i915 features:


Attachments
Kernel log extract, showing the unblanking failure. (14.52 KB, text/plain)
2016-10-19 17:23 UTC, Darren Salt
no flags Details
Kernel log showing normal blank & unblank. (2.50 KB, text/plain)
2017-08-22 14:56 UTC, Darren Salt
no flags Details
Kernel log showing blanking & as-if-reconnected unblanking. (12.25 KB, text/plain)
2017-08-22 14:58 UTC, Darren Salt
no flags Details
Apparent EDID errors (2.63 KB, text/plain)
2017-10-01 13:11 UTC, Darren Salt
no flags Details
Good EDID output (4.17-rc3 + DC) (13.18 KB, text/plain)
2018-05-03 20:51 UTC, Darren Salt
no flags Details
Bad EDID output (4.17-rc3 + DC) (9.15 KB, text/plain)
2018-05-03 20:52 UTC, Darren Salt
no flags Details

Description Darren Salt 2016-10-19 17:23:32 UTC
Created attachment 127405 [details]
Kernel log extract, showing the unblanking failure.

Setup: RX 470 with two monitors attached – one via DVI-and one via HDMI. (Both are AOC monitors.)

* Ensure that outputs are blanked (via xset, time-out or similar).
* Switch off the monitors.
* Unblank the outputs.
* Switch on the monitors.

Invariably, in this situation, the DVI-connected monitor is unblanked and displaying what it should be, but the HDMI-connected monitor is receiving no signal (or at least nothing to display) and is therefore displaying nothing.

The HDMI-connected monitor can be restored by using xrandr to change mode or by blanking and unblanking – so long as it's switched on while doing so.
Comment 1 dwagner 2017-08-20 23:23:08 UTC
JFYI: A similar, but more dramatic symptom that may be related I reported in https://bugs.freedesktop.org/show_bug.cgi?id=102323
Comment 2 Darren Salt 2017-08-22 14:54:52 UTC
Currently running amd-staging-drm-next 0313e8bfbbdd. The behaviour is now improved, though is still regressed vs. mainline.

Unblanking works. If I allow the problem monitor to display “no signal”, it becomes effectively disconnected, and becomes reconnected (as if physically freshly connected) on unblanking.
Comment 3 Darren Salt 2017-08-22 14:56:50 UTC
Created attachment 133671 [details]
Kernel log showing normal blank & unblank.
Comment 4 Darren Salt 2017-08-22 14:58:31 UTC
Created attachment 133673 [details]
Kernel log showing blanking & as-if-reconnected unblanking.
Comment 5 dwagner 2017-08-22 22:34:43 UTC
JFYI: A patch attached to report https://bugs.freedesktop.org/show_bug.cgi?id=102323 provided some improved behaviour regarding switched-off HDMI displays.
Comment 6 Darren Salt 2017-08-23 01:12:34 UTC
That patch makes no difference to this problem.
Comment 7 Darren Salt 2017-09-14 15:32:20 UTC
Seems to be working properly with current amd-staging-drm-next (43dd6fde5df4).
Comment 8 Darren Salt 2017-09-28 13:02:09 UTC
Broken (same symptoms, though the kernel log looks a little different) with drm-next-4.15-dc (fbf0c14dd75) (merged with 4.13.4).
Comment 9 Darren Salt 2017-10-01 13:11:07 UTC
Created attachment 134594 [details]
Apparent EDID errors

HDMI monitor set to 1920×1080, but on unblanking gets set to 1152×864.

This is the second time that I've seen this. Both times it's happened on the first unblanking after boot; both times, using drm-next-4.15-dc.
Comment 10 Darren Salt 2018-05-03 20:51:53 UTC
Created attachment 139325 [details]
Good EDID output (4.17-rc3 + DC)
Comment 11 Darren Salt 2018-05-03 20:52:59 UTC
Created attachment 139326 [details]
Bad EDID output (4.17-rc3 + DC)

Once this happens (after VESA blanking long enough for the problem monitor to fully go to sleep), it persists until reboot.
Comment 12 Edward Kigwana 2018-05-05 00:32:05 UTC
I remember this having this issue and found that switching to the console (CTRL-ALT-F1) and then back to X once or twice woke the monitor.
Comment 13 Darren Salt 2018-06-16 21:40:04 UTC
EDID problems are gone in drm-next-4.18. However, the disconnection issue remains.
Comment 14 Martin Peres 2019-11-19 08:10:56 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/amd/issues/108.


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.