Bug 118611

Summary: Using the 4.6 Kernel causes one Radeon screen to intermittently power cycle.
Product: Drivers Reporter: K. Paden (kpaden1)
Component: Video(DRI - non Intel)Assignee: drivers_video-dri
Status: RESOLVED CODE_FIX    
Severity: normal CC: lenb
Priority: P1    
Hardware: x86-64   
OS: Linux   
Kernel Version: linux 4.6 Subsystem:
Regression: No Bisected commit-id:
Attachments: Xorg.0.log
DMESG.txt

Description K. Paden 2016-05-20 13:55:33 UTC 
I am using an AMD/ATI Cape Verde PRO [Radeon HD 7750/8740 / R7 250E with 4 screens. 

I upgraded to 4.6 get some of the BTRFS performance enhancements. 

When I upgraded to the 4.6 kernel one of the screens started randomly power-cycling.  It would turn-off and then back on. The answer was to switch back to an earlier kernel.  

-Khalid
kpaden1@engineer.com
Comment 1 Len Brown 2016-05-23 23:02:51 UTC 
I would expect regression is likely to be related to the radeon driver.
Perhaps you can git-bisect which kernel version caused it to break?
Comment 2 Michel Dänzer 2016-05-24 02:35:58 UTC 
Also, please attach the dmesg output and Xorg log file (assuming you're using X), preferably captured after the problem occurred.
Comment 3 K. Paden 2016-05-25 00:08:05 UTC 
Created attachment 217331 [details]
Xorg.0.log

Here you are,

-Khalid

On 05/23/2016 09:35 PM, bugzilla-daemon@bugzilla.kernel.org wrote:
> https://bugzilla.kernel.org/show_bug.cgi?id=118611
>
> --- Comment #2 from Michel Dänzer <michel@daenzer.net> ---
> Also, please attach the dmesg output and Xorg log file (assuming you're using
> X), preferably captured after the problem occurred.
>
Comment 4 K. Paden 2016-05-25 00:08:06 UTC 
Created attachment 217341 [details]
DMESG.txt
Comment 5 K. Paden 2016-05-27 04:24:06 UTC 
Sorry, I am not familiar with git-bisect. It is kernel 4.6.
Comment 6 Michel Dänzer 2016-05-27 06:34:57 UTC 
Enter "git bisect howto" in your favourite web search engine.
Comment 7 K. Paden 2016-08-05 11:57:00 UTC 
The answer was to go into the display control screen, and lower the resolution of the screen that was powering off. I noticed in Kernel 4.5.3, this was a problem. But it is ok now with the one screen at a lower resolution.