Bug 96866

Summary:	REGRESSION System frozen after resume
Product:	DRI	Reporter:	Bastien <bastien.bernard>
Component:	DRM/AMDgpu	Assignee:	Default DRI bug account <dri-devel>
Status:	RESOLVED FIXED	QA Contact:
Severity:	normal
Priority:	medium	CC:	alexdeucher
Version:	XOrg git
Hardware:	x86-64 (AMD64)
OS:	Linux (All)
Whiteboard:
i915 platform:		i915 features:

Description Bastien 2016-07-08 22:38:06 UTC

Problem on HP probook 450, with Radeon R7 M260/M265.
Please see:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1589379

To sum-up:
- before commit 4474b85771139f2da8f8f4f443e6fad08081e99e, resume was working OK (version 4.4.3)
- from commit 4474b85771139f2da8f8f4f443e6fad08081e99e (drm/amdgpu: remove exp hardware support from iceland), I cannot boot on the OS (version 4.4.4)
- from commit 53e609099daa023ad7771ec8351202f2a7bee1c1 (drm/amdgpu: fix topaz/tonga gmc assignment in 4.4 stable), I can boot into the OS and see the problem of system frozen on resume (this bug report)(version 4.4.5 and above).

Comment 1 Christopher M. Penalver 2016-07-09 23:34:38 UTC

1) Edited title as not a Ubuntu bug, but an upstream one as per original reporter (reproducible in mainline 4.7-rc6 and a regression).
2) Issue after suspend isn't Critical.

Comment 2 manof 2016-08-01 18:13:10 UTC

I have the same problem with Linux kernel 4.4.8 on my fresh (bought 4 days ago) HP notebook 17-y011nc, with AMD/ATI Topaz XT [Radeon R7 M260/M265].

The interesting thing is, that originally I had the 4.4.0 kernel from Ubuntu 16.04.1 LTS but it sometimes got frozen during suspending. Not everytime, just sometimes. However, it was a way too often, so it was pretty annoying.

Then I have found somewhere in forums, that this bug was resolved in newer kernels, so I switched from 4.4.0 to 4.4.8. And it surely was resolved, however after resume from suspend, it ALWAYS got frozen, so my problem in fact got much worse with the kernel upgrade 4.4.0 => 4.4.8.

Comment 3 Alex Deucher 2016-08-01 18:16:11 UTC

You can set amdgpu.runpm=0 on the kernel command line in grub or try the latest drm-next kernel code.

Comment 4 Tadej Janež 2016-09-06 08:43:27 UTC

I've encountered the same issue on Fedora 24 with kernel-4.7.2-201.fc24.

Tried booting with amdgpu.runpm=0 but that still didn't solve the problem.

Any ideas besides recompiling drm-next branch?

Comment 5 Bastien 2016-11-27 15:01:09 UTC

Good news, this seems corrected on the last upstream kernel 4.9.0-040900rc6-generic.

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.