Navi Reset Kernel Patch

that is currently the expected behavior with Navi10. you’re doing nothing wrong.

1 Like

Have you tried the v2 Navi reset patch?

Nope, but I will now :slight_smile:

What does it build against? I guess I can figure out if it does not apply cleanly, but I will be trying it against the Fedora 31 kernel 5.4.6ish.

Thanks!

It was developed against a git master at the time, however it should cleanly apply to pretty much every 5 series kernel as 99.9% of it is it’s just additions, not alterations.

1 Like

It’s not too bad you clone git://git.proxmox.com/git/pve-kernel.git drop the patch in the patches/kernel folder and install any dependencies and asks you for after running make then you should be able to install the deb files and be good to go.

2 Likes

Is it possible to perform this reset without applying a kernel patch? For example, by running a script as root to do the same fiddling with powerplay tables? I ask because I’m using Fedora Silverblue, and I’m not sure how to run a custom kernel, or if it’s even possible - and if it is possible, it goes against the idea of Silverblue.

The current workaround that I am using is to use devcon within my Windows VM to disable the GPU upon Windows shutdown (which leaves the GPU still displaying a black output), as well as the trick of removing devices in Linux, suspending to RAM, and then rescanning. For whatever reason, these two workarounds don’t seem to work on their own, but only when I use both at the same time.

This works, but it’s quite a hassle having to suspend my entire computer every time I shut down the VM. I’d love a simpler solution.

No.

That’s disappointing, but thanks for the answer. I hope it’ll get merged into the kernel soon then.

afaik it wont even be submitted to mainline until the patch is stable.
and the patch cannot be made stable until AMD decides to help @gnif with this effort.

Proxmox 6.1 kernel 5.3.13-1 + Navi patch #2: pve-kernel-5.3.13-1-pve_5.3.13-1_amd64.deb

Patch #2

Installation:
sudo dpkg -i pve-kernel-5.3.13-1-pve_5.3.13-1_amd64.deb pve-headers-5.3.13-1-pve_5.3.13-1_amd64.deb

Post installation:

sudo apt-mark hold pve-kernel-5.3.13-1-pve
sudo apt-mark hold pve-headers-5.3.13-1-pve

Proxmox PCI passthrough wiki

First, thank you @ApplePi for the respond, I did just that. The installation was successful but it does does not work, at least for my setup.

Thank @xracer. I’ll thy that.

Hi everyone,

First of all, a big thank you @gnif ! Just applied the Navi v2 patch on my Fedora 31 system, works like a charm ! :smiley:

These are my specs :

  • Fedora 31
  • Kernel 5.4.7-200.fc31.x86_64
  • Windows 10 VM using VFIO for MSI RX 5700 XT passthrough

Rebooted several times, no more reset bug.

Many, many thanks !

1 Like

Hi,

I tried to pass through a RX 5500 XT with the patched proxmox kernel, no change the system freezes immediately.

@gnif, would it work if i just add { PCI_VENDOR_ID_ATI, 0x7340, reset_amd_navi10 } to your patch?

System: Proxmox 6.1 - 5.3.13-1-pve, Supermicro X10SRi-F, Xeon 1650v3

########
balloon: 0
bios: ovmf
bootdisk: virtio1
cores: 4
cpu: host,hidden=1
efidisk0: ssd2:vm-102-disk-0,size=1M
hostpci0: 07:00,pcie=1,x-vga=on,romfile=RX5500XT.rom
hostpci1: 07:00.1,pcie=1
machine: pc-q35-3.1
memory: 4012
name: Libreelec
net0: virtio=BA:5D:18:FC:8C:D1,bridge=vmbr0,firewall=1
numa: 0
ostype: l26
scsihw: virtio-scsi-pci
smbios1: uuid=37973275-6bbb-435b-8c02-e7405a8a4b4f
sockets: 1
virtio1: ssd2:vm-102-disk-1,size=25548M
vmgenid: d919af08-f527-4a00-b40e-0c6de12941eb
########

07:00.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Device [1002:7340] (rev c5) (prog-if 00 [VGA controller])
Subsystem: Sapphire Technology Limited Device [1da2:e421]
Flags: fast devsel, IRQ 11, NUMA node 0
Memory at 383fe0000000 (64-bit, prefetchable) [disabled] [size=256M]
Memory at 383ff0000000 (64-bit, prefetchable) [disabled] [size=2M]
I/O ports at d000 [disabled] [size=256]
Memory at fab00000 (32-bit, non-prefetchable) [disabled] [size=512K]
Expansion ROM at fab80000 [disabled] [size=128K]
Capabilities: [48] Vendor Specific Information: Len=08 <?>
Capabilities: [50] Power Management version 3
Capabilities: [64] Express Legacy Endpoint, MSI 00
Capabilities: [a0] MSI: Enable- Count=1/1 Maskable- 64bit+
Capabilities: [100] Vendor Specific Information: ID=0001 Rev=1 Len=010 <?>
Capabilities: [150] Advanced Error Reporting
Capabilities: [200] #15
Capabilities: [240] Power Budgeting <?>
Capabilities: [270] #19
Capabilities: [2a0] Access Control Services
Capabilities: [2b0] Address Translation Service (ATS)
Capabilities: [2c0] Page Request Interface (PRI)
Capabilities: [2d0] Process Address Space ID (PASID)
Capabilities: [320] Latency Tolerance Reporting
Capabilities: [400] #25
Capabilities: [410] #26
Capabilities: [440] #27
Kernel driver in use: vfio-pci

07:00.1 Audio device [0403]: Advanced Micro Devices, Inc. [AMD/ATI] Navi 10 HDMI Audio [1002:ab38]
Subsystem: Advanced Micro Devices, Inc. [AMD/ATI] Navi 10 HDMI Audio [1002:ab38]
Flags: bus master, fast devsel, latency 0, IRQ 10, NUMA node 0
Memory at faba0000 (32-bit, non-prefetchable) [size=16K]
Capabilities: [48] Vendor Specific Information: Len=08 <?>
Capabilities: [50] Power Management version 3
Capabilities: [64] Express Legacy Endpoint, MSI 00
Capabilities: [a0] MSI: Enable- Count=1/1 Maskable- 64bit+
Capabilities: [100] Vendor Specific Information: ID=0001 Rev=1 Len=010 <?>
Capabilities: [150] Advanced Error Reporting
Capabilities: [2a0] Access Control Services
Kernel driver in use: vfio-pci
Kernel modules: snd_hda_intel

That’s Navi 14, not 10, it might work but it’s doubtful

Love your work gnif.

I just set up a vfio system with the Ryzen R9 3950X, a MSI MEG X570 Unify and a RX 5700.

The VM will not boot a second time, although I applied the patch. While the VM starts up the system gets unresponsive.
I attached the dmesg output.

[  317.400741] vfio-pci 0000:30:00.0: Navi10: performing BACO reset
[  318.397274] scsi 8:0:0:0: Direct-Access     StoreJet Transcend        2AR1 PQ: 0 ANSI: 2 CCS
[  318.397726] sd 8:0:0:0: [sdc] 1953525167 512-byte logical blocks: (1.00 TB/932 GiB)
[  318.398161] sd 8:0:0:0: [sdc] Write Protect is off
[  318.398163] sd 8:0:0:0: [sdc] Mode Sense: 28 00 00 00
[  318.398548] sd 8:0:0:0: [sdc] No Caching mode page found
[  318.398549] sd 8:0:0:0: [sdc] Assuming drive cache: write through
[  318.441859]  sdc: sdc1
[  318.443288] sd 8:0:0:0: [sdc] Attached SCSI disk
[  318.689612] audit: type=2500 audit(1578946472.537:174): pid=2722 uid=0 auid=4294967295 ses=4294967295 msg='virt=kvm op=stop reason=shutdown vm="win10" uuid=73933314-6734-48cf-b7bb-19cfbc564803 vm-pid=-1 exe="/usr/bin/libvirtd" hostname=? addr=? terminal=? res=success'
[  341.307813] audit: type=2502 audit(1578946495.157:175): pid=2722 uid=0 auid=4294967295 ses=4294967295 msg='virt=kvm vm="win10" uuid=73933314-6734-48cf-b7bb-19cfbc564803 vm-ctx=+1000:+992 img-ctx=+1000:+992 model=dac exe="/usr/bin/libvirtd" hostname=? addr=? terminal=? res=success'
[  341.320338] virbr0: port 2(vnet0) entered blocking state
[  341.320340] virbr0: port 2(vnet0) entered disabled state
[  341.320387] device vnet0 entered promiscuous mode
[  341.320407] audit: type=1700 audit(1578946495.171:176): dev=vnet0 prom=256 old_prom=0 auid=4294967295 uid=0 gid=0 ses=4294967295
[  341.320530] virbr0: port 2(vnet0) entered blocking state
[  341.320531] virbr0: port 2(vnet0) entered listening state
[  341.320581] audit: type=2501 audit(1578946495.171:177): pid=2722 uid=0 auid=4294967295 ses=4294967295 msg='virt=kvm resrc=net reason=open vm="win10" uuid=73933314-6734-48cf-b7bb-19cfbc564803 net=52:54:00:bd:85:bf path="/dev/net/tun" rdev=0A:C8 exe="/usr/bin/libvirtd" hostname=? addr=? terminal=? res=success'
[  341.320779] virbr1: port 2(vnet1) entered blocking state
[  341.320780] virbr1: port 2(vnet1) entered disabled state
[  341.320822] device vnet1 entered promiscuous mode
[  341.320833] audit: type=1700 audit(1578946495.171:178): dev=vnet1 prom=256 old_prom=0 auid=4294967295 uid=0 gid=0 ses=4294967295
[  341.320918] virbr1: port 2(vnet1) entered blocking state
[  341.320920] virbr1: port 2(vnet1) entered listening state
[  341.320956] audit: type=2501 audit(1578946495.171:179): pid=2722 uid=0 auid=4294967295 ses=4294967295 msg='virt=kvm resrc=net reason=open vm="win10" uuid=73933314-6734-48cf-b7bb-19cfbc564803 net=52:54:00:7a:c0:ff path="/dev/net/tun" rdev=0A:C8 exe="/usr/bin/libvirtd" hostname=? addr=? terminal=? res=success'
[  341.367370] audit: type=2501 audit(1578946495.217:180): pid=2722 uid=0 auid=4294967295 ses=4294967295 msg='virt=kvm resrc=cgroup reason=deny vm="win10" uuid=73933314-6734-48cf-b7bb-19cfbc564803 cgroup="/sys/fs/cgroup/devices/machine.slice/machine-qemu\x2d2\x2dwin10.scope/" class=all exe="/usr/bin/libvirtd" hostname=? addr=? terminal=? res=success'
[  341.367402] audit: type=2501 audit(1578946495.217:181): pid=2722 uid=0 auid=4294967295 ses=4294967295 msg='virt=kvm resrc=cgroup reason=allow vm="win10" uuid=73933314-6734-48cf-b7bb-19cfbc564803 cgroup="/sys/fs/cgroup/devices/machine.slice/machine-qemu\x2d2\x2dwin10.scope/" class=major category=pty maj=88 acl=rw exe="/usr/bin/libvirtd" hostname=? addr=? terminal=? res=success'
[  341.367430] audit: type=2501 audit(1578946495.217:182): pid=2722 uid=0 auid=4294967295 ses=4294967295 msg='virt=kvm resrc=cgroup reason=allow vm="win10" uuid=73933314-6734-48cf-b7bb-19cfbc564803 cgroup="/sys/fs/cgroup/devices/machine.slice/machine-qemu\x2d2\x2dwin10.scope/" class=path path="/dev/kvm" rdev=0A:E8 acl=rw exe="/usr/bin/libvirtd" hostname=? addr=? terminal=? res=success'
[  341.367461] audit: type=2501 audit(1578946495.217:183): pid=2722 uid=0 auid=4294967295 ses=4294967295 msg='virt=kvm resrc=cgroup reason=allow vm="win10" uuid=73933314-6734-48cf-b7bb-19cfbc564803 cgroup="/sys/fs/cgroup/devices/machine.slice/machine-qemu\x2d2\x2dwin10.scope/" class=path path="/dev/input/by-id/usb-04b4_6014-event-kbd" rdev=0D:43 acl=rw exe="/usr/bin/libvirtd" hostname=? addr=? terminal=? res=success'
[  341.367491] audit: type=2501 audit(1578946495.217:184): pid=2722 uid=0 auid=4294967295 ses=4294967295 msg='virt=kvm resrc=cgroup reason=allow vm="win10" uuid=73933314-6734-48cf-b7bb-19cfbc564803 cgroup="/sys/fs/cgroup/devices/machine.slice/machine-qemu\x2d2\x2dwin10.scope/" class=path path="/dev/input/by-id/usb-Logitech_Gaming_Mouse_G502_1461355F3630-event-mouse" rdev=0D:4A acl=rw exe="/usr/bin/libvirtd" hostname=? addr=? terminal=? res=success'
[  341.859949] vfio-pci 0000:30:00.0: enabling device (0400 -> 0403)
[  341.958851] vfio-pci 0000:30:00.0: vfio_ecap_init: hiding ecap 0x19@0x270
[  341.958867] vfio-pci 0000:30:00.0: vfio_ecap_init: hiding ecap 0x1b@0x2d0
[  341.958872] vfio-pci 0000:30:00.0: vfio_ecap_init: hiding ecap 0x25@0x400
[  341.958874] vfio-pci 0000:30:00.0: vfio_ecap_init: hiding ecap 0x26@0x410
[  341.958876] vfio-pci 0000:30:00.0: vfio_ecap_init: hiding ecap 0x27@0x440
[  342.201836] AMD-Vi: Completion-Wait loop timed out
[  342.347216] AMD-Vi: Completion-Wait loop timed out
[  342.471658] AMD-Vi: Completion-Wait loop timed out
[  342.596308] AMD-Vi: Completion-Wait loop timed out
[  342.721135] AMD-Vi: Completion-Wait loop timed out
[  342.873640] AMD-Vi: Completion-Wait loop timed out
[  342.997766] AMD-Vi: Completion-Wait loop timed out
[  343.061324] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a0740]
[  343.276029] AMD-Vi: Completion-Wait loop timed out
[  343.400472] AMD-Vi: Completion-Wait loop timed out
[  343.486604] virbr1: port 2(vnet1) entered learning state
[  343.486625] virbr0: port 2(vnet0) entered learning state
[  343.525221] AMD-Vi: Completion-Wait loop timed out
[  343.649109] AMD-Vi: Completion-Wait loop timed out
[  343.772995] AMD-Vi: Completion-Wait loop timed out
[  343.931591] AMD-Vi: Completion-Wait loop timed out
[  344.055360] AMD-Vi: Completion-Wait loop timed out
[  344.150040] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a0af0]
[  345.499023] AMD-Vi: Completion-Wait loop timed out
[  345.619937] virbr0: port 2(vnet0) entered forwarding state
[  345.619944] virbr0: topology change detected, propagating
[  345.619976] virbr1: port 2(vnet1) entered forwarding state
[  345.619979] virbr1: topology change detected, propagating
[  345.623825] AMD-Vi: Completion-Wait loop timed out
[  345.748787] AMD-Vi: Completion-Wait loop timed out
[  345.873606] AMD-Vi: Completion-Wait loop timed out
[  346.018204] AMD-Vi: Completion-Wait loop timed out
[  346.143007] AMD-Vi: Completion-Wait loop timed out
[  346.268047] AMD-Vi: Completion-Wait loop timed out
[  346.358340] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1500]
[  346.512615] AMD-Vi: Completion-Wait loop timed out
[  346.637629] AMD-Vi: Completion-Wait loop timed out
[  346.780588] AMD-Vi: Completion-Wait loop timed out
[  346.930387] AMD-Vi: Completion-Wait loop timed out
[  347.073050] AMD-Vi: Completion-Wait loop timed out
[  347.197101] AMD-Vi: Completion-Wait loop timed out
[  347.340056] AMD-Vi: Completion-Wait loop timed out
[  347.358869] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a15d0]
[  347.483171] AMD-Vi: Completion-Wait loop timed out
[  347.607807] AMD-Vi: Completion-Wait loop timed out
[  347.732377] AMD-Vi: Completion-Wait loop timed out
[  347.856857] AMD-Vi: Completion-Wait loop timed out
[  347.981395] AMD-Vi: Completion-Wait loop timed out
[  348.105872] AMD-Vi: Completion-Wait loop timed out
[  348.230417] AMD-Vi: Completion-Wait loop timed out
[  348.354886] AMD-Vi: Completion-Wait loop timed out
[  348.358337] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1670]
[  348.483393] AMD-Vi: Completion-Wait loop timed out
[  348.608109] AMD-Vi: Completion-Wait loop timed out
[  348.732886] AMD-Vi: Completion-Wait loop timed out
[  348.857744] AMD-Vi: Completion-Wait loop timed out
[  349.005508] AMD-Vi: Completion-Wait loop timed out
[  349.130313] AMD-Vi: Completion-Wait loop timed out
[  349.255200] AMD-Vi: Completion-Wait loop timed out
[  349.358920] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1710]
[  349.482982] AMD-Vi: Completion-Wait loop timed out
[  349.607515] AMD-Vi: Completion-Wait loop timed out
[  349.732221] AMD-Vi: Completion-Wait loop timed out
[  349.856731] AMD-Vi: Completion-Wait loop timed out
[  349.981316] AMD-Vi: Completion-Wait loop timed out
[  350.105987] AMD-Vi: Completion-Wait loop timed out
[  350.229802] AMD-Vi: Completion-Wait loop timed out
[  350.353651] AMD-Vi: Completion-Wait loop timed out
[  350.358337] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1760]
[  350.482185] AMD-Vi: Completion-Wait loop timed out
[  350.605986] AMD-Vi: Completion-Wait loop timed out
[  350.764572] AMD-Vi: Completion-Wait loop timed out
[  350.888425] AMD-Vi: Completion-Wait loop timed out
[  351.012141] AMD-Vi: Completion-Wait loop timed out
[  351.136249] AMD-Vi: Completion-Wait loop timed out
[  351.280705] AMD-Vi: Completion-Wait loop timed out
[  351.359127] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a17b0]
[  351.482392] AMD-Vi: Completion-Wait loop timed out
[  351.606639] AMD-Vi: Completion-Wait loop timed out
[  351.731100] AMD-Vi: Completion-Wait loop timed out
[  351.855838] AMD-Vi: Completion-Wait loop timed out
[  351.980324] AMD-Vi: Completion-Wait loop timed out
[  352.104898] AMD-Vi: Completion-Wait loop timed out
[  352.229504] AMD-Vi: Completion-Wait loop timed out
[  352.354074] AMD-Vi: Completion-Wait loop timed out
[  352.358335] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1820]
[  352.483074] AMD-Vi: Completion-Wait loop timed out
[  352.607757] AMD-Vi: Completion-Wait loop timed out
[  352.752399] AMD-Vi: Completion-Wait loop timed out
[  352.878127] AMD-Vi: Completion-Wait loop timed out
[  353.002142] AMD-Vi: Completion-Wait loop timed out
[  353.142617] AMD-Vi: Completion-Wait loop timed out
[  353.273500] AMD-Vi: Completion-Wait loop timed out
[  353.358333] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1870]
[  353.483662] AMD-Vi: Completion-Wait loop timed out
[  353.630418] AMD-Vi: Completion-Wait loop timed out
[  353.755629] AMD-Vi: Completion-Wait loop timed out
[  353.892699] AMD-Vi: Completion-Wait loop timed out
[  354.016161] AMD-Vi: Completion-Wait loop timed out
[  354.153170] AMD-Vi: Completion-Wait loop timed out
[  354.276682] AMD-Vi: Completion-Wait loop timed out
[  354.358333] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a18c0]
[  354.483312] AMD-Vi: Completion-Wait loop timed out
[  354.608204] AMD-Vi: Completion-Wait loop timed out
[  354.733309] AMD-Vi: Completion-Wait loop timed out
[  354.858464] AMD-Vi: Completion-Wait loop timed out
[  354.983211] AMD-Vi: Completion-Wait loop timed out
[  355.107848] AMD-Vi: Completion-Wait loop timed out
[  355.249844] AMD-Vi: Completion-Wait loop timed out
[  355.358340] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1910]
[  355.495538] AMD-Vi: Completion-Wait loop timed out
[  355.632700] AMD-Vi: Completion-Wait loop timed out
[  355.769789] AMD-Vi: Completion-Wait loop timed out
[  355.906799] AMD-Vi: Completion-Wait loop timed out
[  356.044929] AMD-Vi: Completion-Wait loop timed out
[  356.186707] AMD-Vi: Completion-Wait loop timed out
[  356.323716] AMD-Vi: Completion-Wait loop timed out
[  356.358341] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1960]
[  356.495537] AMD-Vi: Completion-Wait loop timed out
[  356.620380] AMD-Vi: Completion-Wait loop timed out
[  356.757389] AMD-Vi: Completion-Wait loop timed out
[  356.895190] AMD-Vi: Completion-Wait loop timed out
[  357.032439] AMD-Vi: Completion-Wait loop timed out
[  357.169561] AMD-Vi: Completion-Wait loop timed out
[  357.311350] AMD-Vi: Completion-Wait loop timed out
[  357.358338] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a19b0]
[  357.497388] AMD-Vi: Completion-Wait loop timed out
[  357.659186] AMD-Vi: Completion-Wait loop timed out
[  357.821528] AMD-Vi: Completion-Wait loop timed out
[  357.973920] AMD-Vi: Completion-Wait loop timed out
[  358.098176] AMD-Vi: Completion-Wait loop timed out
[  358.222245] AMD-Vi: Completion-Wait loop timed out
[  358.346225] AMD-Vi: Completion-Wait loop timed out
[  358.358341] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1a30]
[  358.482363] AMD-Vi: Completion-Wait loop timed out
[  358.606321] AMD-Vi: Completion-Wait loop timed out
[  358.730201] AMD-Vi: Completion-Wait loop timed out
[  358.854465] AMD-Vi: Completion-Wait loop timed out
[  358.978566] AMD-Vi: Completion-Wait loop timed out
[  359.102551] AMD-Vi: Completion-Wait loop timed out
[  359.226450] AMD-Vi: Completion-Wait loop timed out
[  359.358336] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1ac0]
[  359.482359] AMD-Vi: Completion-Wait loop timed out
[  359.606356] AMD-Vi: Completion-Wait loop timed out
[  359.730270] AMD-Vi: Completion-Wait loop timed out
[  359.854538] AMD-Vi: Completion-Wait loop timed out
[  359.978682] AMD-Vi: Completion-Wait loop timed out
[  360.102710] AMD-Vi: Completion-Wait loop timed out
[  360.226673] AMD-Vi: Completion-Wait loop timed out
[  360.358338] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1b50]
[  360.507010] AMD-Vi: Completion-Wait loop timed out
[  360.656815] AMD-Vi: Completion-Wait loop timed out
[  360.836905] AMD-Vi: Completion-Wait loop timed out
[  360.968019] AMD-Vi: Completion-Wait loop timed out
[  361.117703] AMD-Vi: Completion-Wait loop timed out
[  361.254740] AMD-Vi: Completion-Wait loop timed out
[  361.358335] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1be0]
[  361.508690] AMD-Vi: Completion-Wait loop timed out
[  361.658788] AMD-Vi: Completion-Wait loop timed out
[  361.795826] AMD-Vi: Completion-Wait loop timed out
[  361.944460] AMD-Vi: Completion-Wait loop timed out
[  362.068388] AMD-Vi: Completion-Wait loop timed out
[  362.205397] AMD-Vi: Completion-Wait loop timed out
[  362.355165] AMD-Vi: Completion-Wait loop timed out
[  362.358338] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1c60]
[  362.494990] AMD-Vi: Completion-Wait loop timed out
[  362.619440] AMD-Vi: Completion-Wait loop timed out
[  362.744219] AMD-Vi: Completion-Wait loop timed out
[  362.868293] AMD-Vi: Completion-Wait loop timed out
[  363.018168] AMD-Vi: Completion-Wait loop timed out
[  363.142114] AMD-Vi: Completion-Wait loop timed out
[  363.266054] AMD-Vi: Completion-Wait loop timed out
[  363.358314] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1ce0]
[  363.493677] AMD-Vi: Completion-Wait loop timed out
[  363.631113] AMD-Vi: Completion-Wait loop timed out
[  363.770812] AMD-Vi: Completion-Wait loop timed out
[  363.931164] AMD-Vi: Completion-Wait loop timed out
[  364.081107] AMD-Vi: Completion-Wait loop timed out
[  364.205866] AMD-Vi: Completion-Wait loop timed out
[  364.348265] AMD-Vi: Completion-Wait loop timed out
[  364.358314] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1d50]
[  364.483285] AMD-Vi: Completion-Wait loop timed out
[  364.613703] AMD-Vi: Completion-Wait loop timed out
[  364.738273] AMD-Vi: Completion-Wait loop timed out
[  364.882757] AMD-Vi: Completion-Wait loop timed out
[  365.007281] AMD-Vi: Completion-Wait loop timed out
[  365.157193] AMD-Vi: Completion-Wait loop timed out
[  365.281942] AMD-Vi: Completion-Wait loop timed out
[  365.358331] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1dd0]
[  365.482913] AMD-Vi: Completion-Wait loop timed out
[  365.607487] AMD-Vi: Completion-Wait loop timed out
[  365.732483] AMD-Vi: Completion-Wait loop timed out
[  365.857281] AMD-Vi: Completion-Wait loop timed out
[  365.982214] AMD-Vi: Completion-Wait loop timed out
[  366.106913] AMD-Vi: Completion-Wait loop timed out
[  366.231869] AMD-Vi: Completion-Wait loop timed out
[  366.356606] AMD-Vi: Completion-Wait loop timed out
[  366.358327] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1e30]
[  366.483104] AMD-Vi: Completion-Wait loop timed out
[  366.607817] AMD-Vi: Completion-Wait loop timed out
[  366.732776] AMD-Vi: Completion-Wait loop timed out
[  366.857512] AMD-Vi: Completion-Wait loop timed out
[  366.982506] AMD-Vi: Completion-Wait loop timed out
[  367.107243] AMD-Vi: Completion-Wait loop timed out
[  367.232163] AMD-Vi: Completion-Wait loop timed out
[  367.358328] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1e90]
[  367.483201] AMD-Vi: Completion-Wait loop timed out
[  367.607881] AMD-Vi: Completion-Wait loop timed out
[  367.732830] AMD-Vi: Completion-Wait loop timed out
[  367.857499] AMD-Vi: Completion-Wait loop timed out
[  368.000264] AMD-Vi: Completion-Wait loop timed out
[  368.125020] AMD-Vi: Completion-Wait loop timed out
[  368.265676] AMD-Vi: Completion-Wait loop timed out
[  368.358312] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1f10]
[  368.529954] AMD-Vi: Completion-Wait loop timed out
[  368.674390] AMD-Vi: Completion-Wait loop timed out
[  368.846456] AMD-Vi: Completion-Wait loop timed out
[  368.971081] AMD-Vi: Completion-Wait loop timed out
[  369.142686] AMD-Vi: Completion-Wait loop timed out
[  369.267198] AMD-Vi: Completion-Wait loop timed out
[  369.358335] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a1fc0]
[  369.522742] AMD-Vi: Completion-Wait loop timed out
[  369.659549] AMD-Vi: Completion-Wait loop timed out
[  369.802654] AMD-Vi: Completion-Wait loop timed out
[  369.973414] AMD-Vi: Completion-Wait loop timed out
[  370.146062] AMD-Vi: Completion-Wait loop timed out
[  370.318416] AMD-Vi: Completion-Wait loop timed out
[  370.358332] iommu ivhd0: AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=30:00.0 address=0x81b5a0070]

funny how different the results are.
The combination Asus X570-e gaming/R5700XT and Ubuntu 19.10 - 5.3.13 with Navi patch v2 works well.
@gnif, thanks for your good work!

That’s depends on VM OS/AMD driver VERSION! The reset bug wasn’t totally fixed yet, the patch don’t cover all situations as gnif pointed out before . Some AMD drivers leave the GPU on a state that the patch can’t handle yet, so it fails!. I’ve run Window 10 with the last Adrenaline driver(26.20.13031.20012), that’s stable if the VM is properly(GPU reset can fail otherwise) shutdown/restarted. The reset patch fail on Ubuntu 18.04 VM with the last Adrenaline driver(17.12.1). This particularly AMD driver version 17.12.1 leave the GPU on a state that the patch cannot handle. I don’t have tested Ubuntu 19.10 yet, but as @Jano posted, it is working. Probably the AMD driver version on ubuntu 19.10 leave the GPU on a state that the patch can handle.

( Someone correct me if I wrong)

That’s not the full story. The error @Benjamin_Klettbach and I experience is not related (directly) to the driver version as for me not even the mentioned AMD Adrenaline Driver leads to a usable gpu reset using Windows 10 as a guest.

1 Like

Patch works great on Windows VM (guest) but I have no luck getting this to work on Linux VM (guest) on 5700. I am getting the following errors:

Navi10: SMU error 0xfd (line 4023)
Navi10: SMU error 0xfd (line 4029)
Navi10: SMU error 0xfd (line 4035)
Navi10: SMU error 0xfd (line 4043)
Navi10: sol register = 0x3c0636c6

Anyone here managed to get this working on a Linux guest?

Thanks @gnif for your hard work.