Bug 110862 - Dual-monitors invalid state after turning on
Summary: Dual-monitors invalid state after turning on
Status: RESOLVED NOTOURBUG
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/AMDgpu (show other bugs)
Version: XOrg git
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Default DRI bug account
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2019-06-08 19:19 UTC by denisgolovan
Modified: 2019-06-12 09:36 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments
xorg.log after bug occured (183.34 KB, text/plain)
2019-06-11 18:04 UTC, denisgolovan
no flags Details
dmesg after bug occured (255.49 KB, text/plain)
2019-06-11 18:04 UTC, denisgolovan
no flags Details
xrandr (1.65 KB, text/plain)
2019-06-11 18:05 UTC, denisgolovan
no flags Details
xorg log after successful X restart (89.29 KB, text/plain)
2019-06-11 18:06 UTC, denisgolovan
no flags Details
dmesg after successfull X restart (255.42 KB, text/plain)
2019-06-11 18:07 UTC, denisgolovan
no flags Details

Description denisgolovan 2019-06-08 19:19:23 UTC
Hi

I experience annoying bug when I turn on my dual-monitor PC after some time is turned off state.
Basically it results in following:

+-----------+
|   OK      |
|           |
+-----------+------------------+
|           |                  |
|           |                  |
| GARBAGE   |  GARBAGE         |
|           |                  |
|           +------------------+
|           |
|           |
+-----------+

X Windows render a small part of screen ("OK" area), everything else is garbage, quite often flicking. Changing VT does not help.

Tested under Linux x64 5.1.6 + xorg-server-1.20.5 + amdgpu driver from git.
Hardware: XFX RX580 8GB
Comment 1 Michel Dänzer 2019-06-11 09:02:39 UTC
Please attach the corresponding Xorg log file and output of xrandr and dmesg.
Comment 2 Michel Dänzer 2019-06-11 09:05:35 UTC
Probably best if you attach all of those from a bad and good boot each.
Comment 3 denisgolovan 2019-06-11 18:04:04 UTC
Created attachment 144506 [details]
xorg.log after bug occured
Comment 4 denisgolovan 2019-06-11 18:04:34 UTC
Created attachment 144507 [details]
dmesg after bug occured
Comment 5 denisgolovan 2019-06-11 18:05:31 UTC
Created attachment 144508 [details]
xrandr
Comment 6 denisgolovan 2019-06-11 18:06:53 UTC
Created attachment 144509 [details]
xorg log after successful X restart
Comment 7 denisgolovan 2019-06-11 18:07:21 UTC
Created attachment 144511 [details]
dmesg after successfull X restart
Comment 8 Michel Dänzer 2019-06-12 09:36:28 UTC
The xrandr output is probably different in the bad case as well, isn't it?

From the bad case Xorg log file:

 [421376.800] (II) AMDGPU(0): Allocate new frame buffer 2160x738

This means Xorg calls into the driver to resize the desktop to 2160x738 (which explains why that area is OK, whereas the remaining area has undefined contents). This is most likely in turn due to a client (probably some component of the desktop environment you're using) requesting so via the RandR extension. So it's either a bug in that or in the X server, not in the driver.


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.