|Summary:||System freezes after resuming from suspend to ram the second time x86 i7|
|Product:||Drivers||Reporter:||John Smith (john-s-84)|
|Severity:||blocking||CC:||kugel, rui.zhang, yu.c.chen|
Description John Smith 2018-05-03 16:02:11 UTC
Created attachment 275751 [details] journalctl_log This happends after the second time of resuming from suspend. The first time I press the power button everything seems to be ok. Please see the dmesg log (suspend_error_log_after_1._resume). When I press a button the second time suspend, the system freezes with screen on. The keyboard/mouse is not working anymore. But I am quiet sure the suspendig mem state is never reached, because the backlight of my keys are still recognising the key pressing. After that, I can only hard reset the system. Moreover it is not booting anymore. I need to plug-in the power-supply to reboot. There is no difference between pressing the power-button or calling $echo mem > /sys/power/state Please let me know, which additional logs do you need. I do not know how to catch error logs before completly freezing.
Comment 1 Chen Yu 2018-05-07 02:47:19 UTC
There are two issues after resumed 1st: 1. Mai 03 17:46:28 bang kernel: BUG: unable to handle kernel paging request at ffffa6a481eacfec Mai 03 17:46:28 bang kernel: IP: smu7_populate_single_firmware_entry.isra.4+0x55/0xd0 [amdgpu] which is a amdgpu driver bug, please blacklist this driver and try again. 2. xhci issue usb 1-7:1.0: rebind failed: -517 The amdgpu is the most suspicious one broke your second suspend.
Comment 2 John Smith 2018-05-07 15:18:56 UTC
Thank you Chen. Blacklisting amdgpu seems to be the cause. So what is next, do I need to open a new issue for the amdgpu driver?
Comment 3 Chen Yu 2018-05-08 01:26:32 UTC
You can send the panic information to firstname.lastname@example.org and CC David (ChunMing) Zhou <David1.Zhou@amd.com> Alex Deucher <email@example.com> Christian König <firstname.lastname@example.org>
Comment 4 John Smith 2018-05-08 17:55:11 UTC
Comment 5 John Smith 2018-05-11 13:49:30 UTC
Comment 6 Thomas Martitz 2018-05-15 05:25:09 UTC
Apparently the happened also for iceland gpus in the past, but was fixed (see https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=885919). I don't know which commit fixed it for iceland, though. FWIW, I am also affected. After I found a fix for https://bugzilla.kernel.org/show_bug.cgi?id=199693 I trapped into this issue.