Kernel bug in amdgpu system freezing

[Janghou]

Running any flatpak package like Foliate of Visual Studio Code freezes system in Ubuntu 19.10.

Worked fine in 19.04

Description: Ubuntu 19.10
Release: 19.10

Only see it when using flatpak. For the rest system is quite stable.

Lots of errors in syslog:

mdgpu 0000:09:00.0: [gfxhub] retry page fault (src_id:0 ring:0 vmid:3 pasid:32769, for process Xorg pid 2137 thread Xorg:cs0 pid 2138)
amdgpu 0000:09:00.0: in page starting at address 0x000000011a500000 from 27
amdgpu 0000:09:00.0: VM_L2_PROTECTION_FAULT_STATUS:0x00301031

Also errors in Gnone-shell:

gnome-shell[2440]: [GFX1]: Shader compilation failure, cfg: features: 0 multiplier: 1 op: CompositionOp::OP_OVER
gnome-shell[2440]: message repeated 5 times: [ [GFX1]: Shader compilation failure, cfg: features: 0 multiplier: 1 op: CompositionOp::OP_OVER]
gnome-shell[2440]: Object .Gjs_WindowSwitcherPopup (0x557f7238f500), has been already deallocated — impossible to access it. This might be caused by the object having been destroyed from C code using something such as destroy(), dispose(), or remove() vfuncs

ProblemType: Bug
DistroRelease: Ubuntu 19.10
Package: linux-image-5.3.0-23-generic 5.3.0-23.25
ProcVersionSignature: Ubuntu 5.3.0-23.25-generic 5.3.7
Uname: Linux 5.3.0-23-generic x86_64
ApportVersion: 2.20.11-0ubuntu8.2
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC1: ben 2111 F.... pulseaudio
 /dev/snd/controlC0: ben 2111 F.... pulseaudio
CurrentDesktop: ubuntu:GNOME
Date: Sat Nov 16 20:08:36 2019
InstallationDate: Installed on 2019-07-19 (120 days ago)
InstallationMedia: Ubuntu 19.04 "Disco Dingo" - Release amd64 (20190416)
MachineType: System manufacturer System Product Name
ProcFB: 0 amdgpudrmfb
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-5.3.0-23-generic root=UUID=7fb304cc-a08f-4ccf-945c-f9f1556d4342 ro quiet splash vt.handoff=7
RelatedPackageVersions:
 linux-restricted-modules-5.3.0-23-generic N/A
 linux-backports-modules-5.3.0-23-generic N/A
 linux-firmware 1.183.2
RfKill:

SourcePackage: linux
UpgradeStatus: Upgraded to eoan on 2019-10-28 (19 days ago)
dmi.bios.date: 02/28/2019
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: 0809
dmi.board.asset.tag: Default string
dmi.board.name: PRIME B450M-A
dmi.board.vendor: ASUSTeK COMPUTER INC.
dmi.board.version: Rev X.0x
dmi.chassis.asset.tag: Default string
dmi.chassis.type: 3
dmi.chassis.vendor: Default string
dmi.chassis.version: Default string
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvr0809:bd02/28/2019:svnSystemmanufacturer:pnSystemProductName:pvrSystemVersion:rvnASUSTeKCOMPUTERINC.:rnPRIMEB450M-A:rvrRevX.0x:cvnDefaultstring:ct3:cvrDefaultstring:
dmi.product.family: To be filled by O.E.M.
dmi.product.name: System Product Name
dmi.product.sku: SKU
dmi.product.version: System Version
dmi.sys.vendor: System manufacturer

[Ubuntu Kernel Bot (ubuntu-kernel-bot)]

This change was made by a bot.

[Kai-Heng Feng (kaihengfeng)]

Please test latest mainline kernel:
https://kernel.ubuntu.com/~kernel-ppa/mainline/v5.4-rc8/

[Janghou (janghou)]

Unfortunately no improvement.

I would say even worse, because I could not get a VT, so it needed a hard shutdown, after opening a something in `foliate`, Only thing working the mouse.

With old kernel at least (after a time) I could open up a VT, to close and reboot.

uname -a
Linux desk 5.4.0-050400rc8-generic #201911171930 SMP Mon Nov 18 00:33:15 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux

89.533687] [drm:amdgpu_dm_commit_planes.constprop.0 [amdgpu]] *ERROR* Waiting for fences timed out!
kernel: [ 94.397626] gmc_v9_0_process_interrupt: 111 callbacks suppressed
kernel: [ 94.397633] amdgpu 0000:09:00.0: [gfxhub0] no-retry page fault (src_id:0 ring:24 vmid:3 pasid:32769, for process Xorg pid 2272 thread Xorg:cs0 pid 2275)
kernel: [ 94.397638] amdgpu 0000:09:00.0: in page starting at address 0x0000000117d00000 from client 27
kernel: [ 94.397641] amdgpu 0000:09:00.0: VM_L2_PROTECTION_FAULT_STATUS:0x00301031
kernel: [ 94.397643] amdgpu 0000:09:00.0: MORE_FAULTS: 0x1
kernel: [ 94.397646] amdgpu 0000:09:00.0: WALKER_ERROR: 0x0
kernel: [ 94.397648] amdgpu 0000:09:00.0: PERMISSION_FAULTS: 0x3
kernel: [ 94.397650] amdgpu 0000:09:00.0: MAPPING_ERROR: 0x0
kernel: [ 94.397651] amdgpu 0000:09:00.0: RW: 0x0
kernel: [ 94.397662] amdgpu 0000:09:00.0: [gfxhub0] no-retry page fault (src_id:0 ring:24 vmid:3 pasid:32769, for process Xorg pid 2272 thread Xorg:cs0 pid 2275)
kernel: [ 94.397664] amdgpu 0000:09:00.0: in page starting at address 0x0000000117d00000 from client 27
kernel: [ 94.397666] amdgpu 0000:09:00.0: VM_L2_PROTECTION_FAULT_STATUS:0x00301031
kernel: [ 94.397668] amdgpu 0000:09:00.0: MORE_FAULTS: 0x1
kernel: [ 94.397670] amdgpu 0000:09:00.0: WALKER_ERROR: 0x0
kernel: [ 94.397672] amdgpu 0000:09:00.0: PERMISSION_FAULTS: 0x3
kernel: [ 94.397674] amdgpu 0000:09:00.0: MAPPING_ERROR: 0x0
kernel: [ 94.397676] amdgpu 0000:09:00.0: RW: 0x0

[Janghou (janghou)]

Related issues:
https://gitlab.freedesktop.org/mesa/mesa/issues/1899
https://discussion.fedoraproject.org/t/some-flatpaks-kill-the-machine-on-f31sb-beta-testing/7673/4

So is it a Mesa, and not a kernel bug?

AFAICS it can be solved/workaround by Flatpak packages moving to a newer Mesa version.

Still a bug if only machines with AMD cards are hit.

[Nowaker (nowaker)]

I've encountered VM_L2_PROTECTION_FAULT_STATUS two times today. No Flatpack involved. Using the latest Mesa on Arch Linux 20200924.