Bug 111659 - Kernel panic when waking up after screens go to dpms sleep
Summary: Kernel panic when waking up after screens go to dpms sleep
Status: NEW
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Radeon (show other bugs)
Version: unspecified
Hardware: x86-64 (AMD64) Linux (All)
: not set normal
Assignee: Default DRI bug account
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2019-09-11 11:39 UTC by Brad Campbell
Modified: 2019-09-11 14:26 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
Multiple instances of the Panic (10.42 KB, text/plain)
2019-09-11 11:39 UTC, Brad Campbell
no flags Details
Complete dmesg (127.86 KB, text/plain)
2019-09-11 11:39 UTC, Brad Campbell
no flags Details
Xorg log (54.94 KB, text/plain)
2019-09-11 11:39 UTC, Brad Campbell
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description Brad Campbell 2019-09-11 11:39:00 UTC
Created attachment 145332 [details]
Multiple instances of the Panic

iMac late 2011 with 2 Thunderbolt displays.

Kernel 5.2 finally got DP routing working to allow both TB displays to work, however I'm now getting lockups which appear to be triggered in   radeon_dp_needs_link_train.

I'm capturing these over netconsole as it leaves the machine paralysed.

Attachment faults.txt has 4 separate instances of the fault from 4 different boots.

This machine stays on 24/7 and this seems to occur when the displays wake up after a dpms sleep. Having said that I've also seen the fault when doing something innocuous like changing the audio volume.

Generally at least 2 of the screens wakeup, so I have displays with a lockscreen asking for a password and a mouse, but the machine is dead. 

The last example in faults.txt left the machine in a state where I could ssh in and reboot it. All the others required a hard power cycle.

I'm currently using 5.2.11. Previously I was using 4.17, but I can't roll back prior to 5.2 without losing the second TB display, and it can take hours or days to hit so bisection would be difficult.

01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Blackcomb [Radeon HD 6970M/6990M] (prog-if 00 [VGA controller])
	Subsystem: Apple Inc. Radeon HD 6970M
	Flags: bus master, fast devsel, latency 0, IRQ 79
	Memory at 90000000 (64-bit, prefetchable) [size=256M]
	Memory at a8800000 (64-bit, non-prefetchable) [size=128K]
	I/O ports at 2000 [size=256]
	Expansion ROM at a8820000 [disabled] [size=128K]
	Capabilities: [50] Power Management version 3
	Capabilities: [58] Express Legacy Endpoint, MSI 00
	Capabilities: [a0] MSI: Enable+ Count=1/1 Maskable- 64bit+
	Capabilities: [100] Vendor Specific Information: ID=0001 Rev=1 Len=010 <?>
	Capabilities: [150] Advanced Error Reporting
	Kernel driver in use: radeon
	Kernel modules: radeon
Comment 1 Brad Campbell 2019-09-11 11:39:29 UTC
Created attachment 145333 [details]
Complete dmesg
Comment 2 Brad Campbell 2019-09-11 11:39:41 UTC
Created attachment 145334 [details]
Xorg log


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.