Bug 102499 - i915 i5-5200U i5-5287U disconnects second external display
Summary: i915 i5-5200U i5-5287U disconnects second external display
Status: CLOSED WORKSFORME
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: XOrg git
Hardware: Other All
: medium normal
Assignee: Intel GFX Bugs mailing list
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2017-08-31 20:42 UTC by Harry
Modified: 2018-04-25 06:59 UTC (History)
1 user (show)

See Also:
i915 platform: BDW
i915 features: power/Other


Attachments

Description Harry 2017-08-31 20:42:12 UTC
GNOME 3.22 Wayland Ubuntu 17.04

Running with internal 2560 + 1 external 4k display work fine
Arranging 2560+4K externals side-by-side and internal 2560 below:
second external monitor disconnects within 10 s
(cannot have wider than 8k pixels)

disconnects means that one or all displays go black for 1 s then behaves like they were just plugged in

sometime it works for a while
as of late: never works

Used to be lot worse, all monitors frequently disconnecting when:
updating Wayland window
moving mouse cursor across display boundary
having Wayland window straddling a display boundary

Could become eternally cyclic when disconnecting a display placed a Wayland (Terminal) window across a display boundary, causing yet another reset

MacBook Pro 13" 2015

Using 4.12.8
Have to use Kernel 4.11.6+ due to #99295

Used to work prior to 17.04/4.10+


Log Output
Aug 31 09:55:03 c89 kernel: [686894.015206] thunderbolt 0000:07:00.0: resetting error on 0:b.
Aug 31 09:55:03 c89 kernel: [686894.015255] thunderbolt 0000:07:00.0: 0:b: hotplug: scanning
Aug 31 09:55:03 c89 kernel: [686894.015259] thunderbolt 0000:07:00.0: 0:b: hotplug: no switch found
Aug 31 09:55:04 c89 kernel: [686894.514572] thunderbolt 0000:07:00.0: resetting error on 0:b.
Aug 31 09:55:04 c89 kernel: [686894.514625] thunderbolt 0000:07:00.0: 0:b: got unplug event for disconnected port, ignoring
Comment 1 Harry 2017-08-31 20:51:54 UTC
After I finished filing this bug, now it works gain.
I filed it because I could not get it to work.
Let's see for how long it will work this time?

No. after finishing typing this, second external display disconnected again, after maybe 20 s.
Settings Displays has it in Off
Comment 2 Harry 2017-08-31 20:56:13 UTC
Seems to now be logging this:
Aug 31 13:55:32 c89 kernel: [701321.523059] [drm] Reducing the compressed framebuffer size. This may lead to less power savings than a non-reduced-size. Try to increase stolen memory size if available in BIOS.
Comment 3 Harry 2017-08-31 21:19:15 UTC
cat /sys/module/i915/parameters/enable_psr
0
Comment 4 Harry 2017-08-31 21:23:47 UTC
cat /sys/module/i915/parameters/enable_fbc 
1
Comment 5 Harry 2017-08-31 21:47:23 UTC
0 s: Enable the second external display
produces this log:
Aug 31 14:40:06 c89 kernel: [703995.721847] [drm] Reducing the compressed framebuffer size. This may lead to less power savings than a non-reduced-size. Try to increase stolen memory size if available in BIOS. 

13 s: you click Keep Changes
produces 4 of these after 4 s
Aug 31 14:40:11 c89 gnome-terminal-[19749]: Allocating size to GtkBox 0x49fbb5a10 without calling gtk_widget_get_preferred_width/height(). How does the code know the size to allocate?   

24 s: all monitors go to black

32 s: thunderbolt logs

Mucking with Wayland windows makes it go down quickly
Doing nothing in a good state goes to reset, too, after 51 s or longer
Comment 6 Harry 2017-08-31 21:53:49 UTC
Likely never worked for GNOME 3.22 4.10.19+
probably last worked 4.8.0.46

Was not discovered sooner because bug 99295 forced frequent restarts
Comment 7 Harry 2017-08-31 22:21:02 UTC
tried to use two external displays and internal off, does not work either.

When things don't work out for GNOME it comes up with any random display configuration, just never two external displays

The GNOME Display settings is bad, too, you must make one change at a time instead of enabling and arranging all displays the desired way and then clicking apply once like on the macky mac

macOS has these kinds of troubles, too, but if you wait 10 s between each cable plugin, things work as they are supposed to.

Impact is that users are limited to one external display. Used to work
Comment 8 Jani Saarinen 2018-03-29 07:11:21 UTC
First of all. Sorry about spam.
This is mass update for our bugs. 

Sorry if you feel this annoying but with this trying to understand if bug still valid or not.
If bug investigation still in progress, please ignore this and I apologize!

If you think this is not anymore valid, please comment to the bug that can be closed.
If you haven't tested with our latest pre-upstream tree(drm-tip), can you do that also to see if issue is valid there still and if you cannot see issue there, please comment to the bug.
Comment 9 Jani Saarinen 2018-04-25 06:59:47 UTC
Closing, please re-open is issue still exists.


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.