Warning: UPDATING these packages broke my pci passthrough.
- Thread starter AWP Technologies
- Start date
Sorry to borrow this topic. After updating the latest pve related components, compared with the original KVM that could not be started, now KVM can be started, but there are warnings.
References are as follows:
error writing '1' to '/sys/bus/pci/devices/0000:01:00.0/reset': Inappropriate ioctl for device
failed to reset PCI device '0000:01:00.0', but trying to continue as not all devices need a reset
swtpm_setup: Not overwriting existing state file.
kvm: warning: host doesn't support requested feature: CPUID.01H:ECX.pcid [bit 17]
Among them, the swtpm and warnings exist on all virtual machines. The PCI device is my NV graphics card, which currently does not affect the normal startup of KVM and device passthrough.
References are as follows:
error writing '1' to '/sys/bus/pci/devices/0000:01:00.0/reset': Inappropriate ioctl for device
failed to reset PCI device '0000:01:00.0', but trying to continue as not all devices need a reset
swtpm_setup: Not overwriting existing state file.
kvm: warning: host doesn't support requested feature: CPUID.01H:ECX.pcid [bit 17]
Among them, the swtpm and warnings exist on all virtual machines. The PCI device is my NV graphics card, which currently does not affect the normal startup of KVM and device passthrough.
yes the pci warnings are new and intentional. Previously we tried e.g. to reset the device, but failed silently not knowing if it worked or not
those warnings are not bad per se, but could indicate a problem if e.g. something in the guest is not working right
Just wanted to let the community know that the workaround Fiona suggested worked in my case.
I experienced issues on hosts with 2 CPU sockets and passing through two Nvidia L40(s) cards to a VM where only one of the GPUs was recognised in the VM. I do not exactly know which update broke the passtrough, but adding the testing apt repostiory, updating the two packages Fiona suggested and reverting back to the subscription repository, followed by an apt update and finally a host reboot, helped. Both GPUs show up in the VM now.
same here with proxmox 8.3.0:
Code:
error writing '1' to '/sys/bus/pci/devices/0000:c6:00.0/reset': Inappropriate ioctl for device
failed to reset PCI device '0000:c6:00.0', but trying to continue as not all devices need a reset
TASK ERROR: timeout waiting on systemdseems that RadeonResetBugFix does not work anymore https://github.com/inga-lovinde/RadeonResetBugFix/tree/master
how can I fix that?
Hi,
please share the output of
pveversion -v and the VM configuration qm config <ID> as well as an excerpt of the system logs/journal around the time the issue happened. Does it work if you start the VM with a higher timeout, e.g. qm start <ID> --timeout 900? If yes, how long does it take to actually start?Hello,
thank you for your support:
pveversion -v:
qm config <ID> (Win11 vm with iGPU pass):
the "excerpt of the system logs/journal around the time the issue" it's only about this message as before, even if I start it with 900 timeout:
with a HDMI connected to the machine, after win vm go off, monitor stay blank, does not return to the host...
thank you again for your time
thank you for your support:
pveversion -v:
Code:
proxmox-ve: 8.3.0 (running kernel: 6.8.12-4-pve)
pve-manager: 8.3.0 (running version: 8.3.0/c1689ccb1065a83b)
proxmox-kernel-helper: 8.1.0
proxmox-kernel-6.8: 6.8.12-4
proxmox-kernel-6.8.12-4-pve-signed: 6.8.12-4
proxmox-kernel-6.8.4-2-pve-signed: 6.8.4-2
amd64-microcode: 3.20240820.1~deb12u1
ceph-fuse: 17.2.7-pve3
corosync: 3.1.7-pve3
criu: 3.17.1-2
glusterfs-client: 10.3-5
ifupdown2: 3.2.0-1+pmx11
ksm-control-daemon: 1.5-1
libjs-extjs: 7.0.0-5
libknet1: 1.28-pve1
libproxmox-acme-perl: 1.5.1
libproxmox-backup-qemu0: 1.4.1
libproxmox-rs-perl: 0.3.4
libpve-access-control: 8.2.0
libpve-apiclient-perl: 3.3.2
libpve-cluster-api-perl: 8.0.10
libpve-cluster-perl: 8.0.10
libpve-common-perl: 8.2.9
libpve-guest-common-perl: 5.1.6
libpve-http-server-perl: 5.1.2
libpve-network-perl: 0.10.0
libpve-rs-perl: 0.9.1
libpve-storage-perl: 8.2.9
libspice-server1: 0.15.1-1
lvm2: 2.03.16-2
lxc-pve: 6.0.0-1
lxcfs: 6.0.0-pve2
novnc-pve: 1.5.0-1
proxmox-backup-client: 3.3.0-1
proxmox-backup-file-restore: 3.3.0-1
proxmox-firewall: 0.6.0
proxmox-kernel-helper: 8.1.0
proxmox-mail-forward: 0.3.1
proxmox-mini-journalreader: 1.4.0
proxmox-offline-mirror-helper: 0.6.7
proxmox-widget-toolkit: 4.3.3
pve-cluster: 8.0.10
pve-container: 5.2.2
pve-docs: 8.3.1
pve-edk2-firmware: 4.2023.08-4
pve-esxi-import-tools: 0.7.2
pve-firewall: 5.1.0
pve-firmware: 3.14-1
pve-ha-manager: 4.0.6
pve-i18n: 3.3.2
pve-qemu-kvm: 9.0.2-4
pve-xtermjs: 5.3.0-3
qemu-server: 8.3.0
smartmontools: 7.3-pve1
spiceterm: 3.3.0
swtpm: 0.8.0+pve1
vncterm: 1.8.0
zfsutils-linux: 2.2.6-pve1qm config <ID> (Win11 vm with iGPU pass):
Code:
agent: 1
args: -cpu 'host,-hypervisor,kvm=off'
bios: ovmf
boot: order=scsi0;net0
cores: 16
cpu: host
efidisk0: local-lvm:vm-100-disk-0,efitype=4m,pre-enrolled-keys=1,size=4M
hostpci0: 0000:c6:00.0,pcie=1,romfile=vbios_8845.bin,x-vga=1
hostpci1: 0000:c6:00.1,pcie=1,romfile=AMDGopDriver_8845hs.rom
machine: pc-q35-9.0
memory: 16384
meta: creation-qemu=9.0.2,ctime=1732920848
name: win11
net0: virtio=BC:24:11:46:0D:C3,bridge=vmbr0,firewall=1
numa: 0
ostype: win11
scsi0: local-lvm:vm-100-disk-1,cache=writeback,discard=on,iothread=1,size=200G,ssd=1
scsihw: virtio-scsi-single
smbios1: uuid=3be659fd-57b6-4496-9f8b-37b4dc200bad
sockets: 1
tpmstate0: local-lvm:vm-100-disk-2,size=4M,version=v2.0
vga: none
vmgenid: 9cdc5539-d08b-473d-b970-a2b6bbc76522the "excerpt of the system logs/journal around the time the issue" it's only about this message as before, even if I start it with 900 timeout:
Code:
error writing '1' to '/sys/bus/pci/devices/0000:c6:00.0/reset': Inappropriate ioctl for device
failed to reset PCI device '0000:c6:00.0', but trying to continue as not all devices need a reset
timeout waiting on systemdthank you again for your time
That does not look like
journalctl output, e.g. timestamp and process IDs are missing. There also should be log lines mentioning the qmstart task.oh sorry, I didn't undestand, I guess this:
Code:
Dec 02 10:52:57 proxmox login[166633]: ROOT LOGIN on '/dev/pts/0'
Dec 02 10:52:59 proxmox qm[166665]: <root@pam> starting task UPID:proxmox:00028B0A:0047DD96:674D837B:qmstart:100:root@pam:
Dec 02 10:52:59 proxmox qm[166666]: start VM 100: UPID:proxmox:00028B0A:0047DD96:674D837B:qmstart:100:root@pam:
Dec 02 10:53:00 proxmox systemd[1]: 100.scope: Deactivated successfully.
Dec 02 10:53:00 proxmox systemd[1]: Stopped 100.scope.
Dec 02 10:53:00 proxmox systemd[1]: 100.scope: Consumed 5h 41min 49.973s CPU time.
Dec 02 10:53:20 proxmox qm[166666]: timeout waiting on systemd
Dec 02 10:53:20 proxmox qm[166665]: <root@pam> end task UPID:proxmox:00028B0A:0047DD96:674D837B:qmstart:100:root@pam: timeout waiting on systemd
Dec 02 10:54:22 proxmox qm[166950]: <root@pam> starting task UPID:proxmox:00028C27:0047FDF6:674D83CE:qmstart:100:root@pam:
Dec 02 10:54:22 proxmox qm[166951]: start VM 100: UPID:proxmox:00028C27:0047FDF6:674D83CE:qmstart:100:root@pam:
Dec 02 10:54:42 proxmox qm[166951]: timeout waiting on systemd
Dec 02 10:54:42 proxmox qm[166950]: <root@pam> end task UPID:proxmox:00028C27:0047FDF6:674D83CE:qmstart:100:root@pam: timeout waiting on systemdJust checking in to see if there has been any progress on this? I am also experiencing this issue - happy to help anyway I can.
I'm experiencing the same issue.
I’m using an AMD Ryzen 7 8745HS with a 780M iGPU, running PVE 8.3.2.
After applying the RadeonResetBugFix, every time I reboot or shut down the Windows VM, it seems the iGPU isn't released. The VM cannot restart, and the status shows the following error message:
Code:
error writing '1' to '/sys/bus/pci/devices/0000:65:00.0/reset': Inappropriate ioctl for deviceThe only way I can get the VM running again is by restarting the host machine.
For me I am passing through an AMG 7900 GRE, it passes through fine initially, then at some point the VM will freeze. When I try and start the VM again, I get:
Only way to start the VM again is to reboot the host.
Code:
error writing '1' to '/sys/bus/pci/devices/0000:26:00.0/reset': Inappropriate ioctl for device
failed to reset PCI device '0000:26:00.0', but trying to continue as not all devices need a reset
kvm: ../hw/pci/pci.c:1633: pci_irq_handler: Assertion `0 <= irq_num && irq_num < PCI_NUM_PINS' failed.
TASK ERROR: start failed: QEMU exited with code 1Only way to start the VM again is to reboot the host.
same here, do you think is a proxmox's problem o ResetBugFix on AMD does not work anymore?