[Bug 109955] amdgpu [RX Vega 64] system freeze while gaming

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=109955

--- Comment #4 from Mauro Gaspari <ilvipero at gmx.com> ---
Issue still happens despite kernel updates and mesa updates on openSUSE
Tumbleweed. Same happens on Kubuntu with oibaf ppa, and on Arch.

It seems this bug affects many people on linux using AMDGPUS, and found some
interesting workarounds. Had a look at kernel options, applied to grub, and so
far it has been 2 weeks of extensive testing, and I did not have a single
system freeze or hang.

-> BEGIN KENEL PARAMETERS <-
This is what I am using now. Please note that some of those settings are to
enable debugging and should not left there forever. I will remove those once
I am confident with the stability of the system.

AMDGPU
amdgpu.dc=1 amdgpu.vm_update_mode=0 amdgpu.dpm=-1
amdgpu.ppfeaturemask=0xffffffff amdgpu.vm_fault_stop=2 amdgpu.vm_debug=1
amdgpu.gpu_recovery=0


- Kernel parameters explained from:
https://www.kernel.org/doc/html/latest/gpu/amdgpu.html

--- dc (int)
Disable/Enable Display Core driver for debugging (1 = enable, 0 = disable).
The default is -1 (automatic for each asic).


--- dpm (int)
Override for dynamic power management setting (1 = enable, 0 = disable). The
default is -1 (auto).

--- vm_update_mode (int)
Override VM update mode. VM updated by using CPU (0 = never, 1 = Graphics
only, 2 = Compute only, 3 = Both). The default is -1 (Only in large BAR(LB)
systems Compute VM tables will be updated by CPU, otherwise 0, never).

--- ppfeaturemask (uint)
Override power features enabled. See enum PP_FEATURE_MASK in
drivers/gpu/drm/amd/include/amd_shared.h. The default is the current set of
stable power features.

--- vm_fault_stop (int)
Stop on VM fault for debugging (0 = never, 1 = print first, 2 = always). The
default is 0 (No stop).

--- vm_debug (int)
Debug VM handling (0 = disabled, 1 = enabled). The default is 0 (Disabled).

-gpu_recovery (int)
Set to enable GPU recovery mechanism (1 = enable, 0 = disable). The default
is -1 (auto, disabled except SRIOV).

-> END KERNEL PARAMETERS <-

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/dri-devel/attachments/20190411/52cc2233/attachment-0001.html>