Bug 112781 - radeon: f95429e "only init fbdev if we have connectors" breaks suspend on Clevo P170EM laptop
Summary: radeon: f95429e "only init fbdev if we have connectors" breaks suspend on Cle...
Status: RESOLVED CODE_FIX
Alias: None
Product: Drivers
Classification: Unclassified
Component: Video(DRI - non Intel) (show other bugs)
Hardware: All Linux
: P1 normal
Assignee: drivers_video-dri
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2016-02-21 13:17 UTC by Christoph Haag
Modified: 2016-03-27 08:23 UTC (History)
3 users (show)

See Also:
Kernel Version: 4.5-rc2
Tree: Mainline
Regression: No


Attachments
dmesg (78.68 KB, text/plain)
2016-02-21 13:17 UTC, Christoph Haag
Details
fix (3.31 KB, patch)
2016-03-14 14:36 UTC, Alex Deucher
Details | Diff

Description Christoph Haag 2016-02-21 13:17:01 UTC
Created attachment 204201 [details]
dmesg

On a Clevo P170EM with these GPUs:

00:02.0 VGA compatible controller: Intel Corporation 3rd Gen Core processor Graphics Controller (rev 09)
01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Wimbledon XT [Radeon HD 7970M] (rev ff)

With rc2 suspend stopped working. It starts to suspend, turns off the display, etc., but then it fails to actually enter the suspend state. The power stays on and after a while the radeon GPU fan starts spinning up. There's nothing in the logs relating to the problem I could find.

I bisected to f95429eccc570dc45d589c327bfcfddcdc3e8228 and confirmed it with latest linux git (4.5-rc5) with this commit reverted - suspend works. Attached is a dmesg with this setup including one successful suspend and wake up.
Comment 1 Christoph Haag 2016-03-12 09:27:01 UTC
Uhm... bump?

With the latest linux git, I tried

+       radeon_fbdev_init(rdev);
        if (!list_empty(&rdev->ddev->mode_config.connector_list)) {
-               radeon_fbdev_init(rdev);
                drm_kms_helper_poll_init(rdev->ddev);
        }

and suspend works.
Comment 2 Alex Deucher 2016-03-14 14:36:28 UTC
Created attachment 209091 [details]
fix

The attached patch should fix it.
Comment 3 Christoph Haag 2016-03-14 16:10:25 UTC
4.5 with the patch seems to suspend without problems, thanks.
Comment 4 Christoph Haag 2016-03-27 08:23:25 UTC
e5f243bd2edd95c6cc1d90c1878f821068e83fba is in 4.6-rc1.

Thanks.

Note You need to log in before you can comment on or make changes to this bug.