Proxmox 8.4.17. Problem with PCI passthrough for ASM1064 SATA controller

H

horkryzard

New Member
Jul 30, 2025
1
0
1
Hello

Kernel 6.8.12-16 broke my PCI passthrough for the ASM1064 SATA controller. I thought I would wait a bit, hoping that it would be fixed in newer releases. However, the problem still persists, as the same issue occurs on versions 6.8.12-16, 6.8.12-17, 6.8.12-18, and 6.8.12-19.

journalctl
Mar 05 13:38:38 pve pve-guests[1786]: VM 105 started with PID 1824.
Mar 05 13:38:39 pve pvesh[1601]: Waiting for 300 seconds (startup delay)
Mar 05 13:38:39 pve kernel: pcieport 0000:00:1c.0: DPC: containment event, status:0x1f11: unmasked uncorrectable error detected
Mar 05 13:38:39 pve QEMU[1824]: kvm: vfio_err_notifier_handler(0000:09:00.0) Unrecoverable error detected. Please collect any data possible and then kill the guest
Mar 05 13:38:39 pve kernel: pcieport 0000:00:1c.0: AER: device recovery successful

lspci -nn | grep 09:00
09:00.0 SATA controller [0106]: ASMedia Technology Inc. ASM1064 Serial ATA Controller [1b21:1064] (rev 02)

dmesg -T
[Thu Mar 5 14:16:12 2026] fwbr105i0: port 2(tap105i0) entered blocking state
[Thu Mar 5 14:16:12 2026] fwbr105i0: port 2(tap105i0) entered forwarding state
[Thu Mar 5 14:16:17 2026] pcieport 0000:00:1c.0: DPC: containment event, status:0x1f11: unmasked uncorrectable error detected
[Thu Mar 5 14:16:17 2026] pcieport 0000:00:1c.0: AER: device recovery successful
[Thu Mar 5 14:21:18 2026] tap102i0: entered promiscuous mode
[Thu Mar 5 14:21:18 2026] vmbr2: port 2(fwpr102p0) entered blocking state
[Thu Mar 5 14:21:18 2026] vmbr2: port 2(fwpr102p0) entered disabled state

Any thoughts on what changed and how it can be fixed or I'm stuck at Kernel 6.8.12-15?
 
Hi,

the possible workaround is disabling DPC/AER.
But there a cons, you're losing advanced error handling.
 
Last edited:
You must log in or register to reply here.
Top Bottom
Back