Bug 111980

Summary: Rebbot and shutdown doesn't work on specific hardware
Product: DRI Reporter: lei.pero
Component: DRM/RadeonAssignee: Default DRI bug account <dri-devel>
Status: RESOLVED FIXED QA Contact:
Severity: not set    
Priority: not set CC: sebastiankapela
Version: XOrg git   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
See Also: https://bugzilla.kernel.org/show_bug.cgi?id=205147
Whiteboard:
i915 platform: i915 features:

Description lei.pero 2019-10-11 16:05:12 UTC 
When system shutdown/reboot is initiated, everything works as it should, journal doesn't show any error and it reaches target reboot (for example), display shuts down properly, but PC stays powered on (requiring hard reset) indefinitely.

This happened after kernel upgrade, after bisectiing, commit that was to blame was bellow, after removing just that specific commit, it would work as expected:

894c414129a8d9ef1b2de443015e4dde6085f64f is the first bad commit
commit 894c414129a8d9ef1b2de443015e4dde6085f64f
Author: KyleMahlkuch <kmahlkuc@linux.vnet.ibm.com>
Date:   Wed Jul 31 17:10:14 2019 -0500

    drm/radeon: Fix EEH during kexec
    
    [ Upstream commit 6f7fe9a93e6c09bf988c5059403f5f88e17e21e6 ]
    
    During kexec some adapters hit an EEH since they are not properly
    shut down in the radeon_pci_shutdown() function. Adding
    radeon_suspend_kms() fixes this issue.
    
    Signed-off-by: KyleMahlkuch <kmahlkuc@linux.vnet.ibm.com>
    Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
    Signed-off-by: Sasha Levin <sashal@kernel.org>

 drivers/gpu/drm/radeon/radeon_drv.c | 8 ++++++++
 1 file changed, 8 insertions(+)

Issue is described here, seems like others have similar issue that might be connected with this as well:
https://bbs.archlinux.org/viewtopic.php?id=249787
Comment 1 Tim Cuthbertson 2019-10-11 18:05:28 UTC 
My system does this, too, since kernel 5.3.5-arch1-1-ARCH on Arch Linux.
Comment 2 lei.pero 2019-10-11 18:54:57 UTC 
(In reply to Tim Cuthbertson from comment #1)
> My system does this, too, since kernel 5.3.5-arch1-1-ARCH on Arch Linux.

Can you test it without this commit? Same for me 5.3.5-arch1-1-ARCH introduced the bug.

I've used
git revert 894c414129a8d9ef1b2de443015e4dde6085f64f
to exclude that commit in stable branch (after bisecting) for 5.4-rc2 and it worked fine without it.
Comment 3 Tim Cuthbertson 2019-10-15 14:11:47 UTC 
This morning, my kernel was upgraded to version 5.3.6-arch1-1-ARCH dated Oct 11. The problem persists.
Comment 4 Tim Cuthbertson 2019-10-18 13:20:15 UTC 
This morning, my kernel was upgraded to version 5.3.7-arch1-1-ARCH dated Oct 18. The problem persists.
Comment 6 lei.pero 2019-10-18 16:21:23 UTC 
(In reply to Alex Deucher from comment #5)
> Reverted:
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/
> ?id=8d13c187c42e110625d60094668a8f778c092879

Great, I can close this bug report now?
Comment 7 Alex Deucher 2019-10-18 16:42:47 UTC 
Yes.

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.