ACS override Broke Array

I had some issues with getting Nvidia GPU passthrough working i got an error starting the VM.

I discovered turning on pcie ACS override and it looks like that split out the 1070 from the IOMMU group 1.

Which means i can now start my VM with the graphics card..

However i started seeing these errors after i rebooted.

Jan  9 16:11:50 Vault kernel: DMAR: DRHD: handling fault status reg 3
Jan  9 16:11:50 Vault kernel: DMAR: [DMA Write] Request device [04:00.1] fault addr 39d9c0000 [fault reason 02] Present bit in context entry is clear
Jan  9 16:11:50 Vault kernel: DMAR: DRHD: handling fault status reg 3
Jan  9 16:11:50 Vault kernel: DMAR: [DMA Write] Request device [04:00.1] fault addr 39d9c0000 [fault reason 02] Present bit in context entry is clear
Jan  9 16:11:50 Vault kernel: DMAR: DRHD: handling fault status reg 3
Jan  9 16:11:50 Vault kernel: DMAR: [DMA Write] Request device [04:00.1] fault addr 39d9c0000 [fault reason 02] Present bit in context entry is clear
Jan  9 16:11:50 Vault kernel: DMAR: DRHD: handling fault status reg 3
Jan  9 16:11:50 Vault kernel: DMAR: [DMA Write] Request device [04:00.1] fault addr 39d9c0000 [fault reason 02] Present bit in context entry is clear
Jan  9 16:11:50 Vault kernel: DMAR: DRHD: handling fault status reg 3
Jan  9 16:11:50 Vault kernel: DMAR: [DMA Write] Request device [04:00.1] fault addr 39d9c0000 [fault reason 02] Present bit in context entry is clear
Jan  9 16:11:50 Vault kernel: DMAR: DRHD: handling fault status reg 3
Jan  9 16:11:50 Vault kernel: DMAR: [DMA Write] Request device [04:00.1] fault addr 39d9c0000 [fault reason 02] Present bit in context entry is clear
Jan  9 16:11:50 Vault kernel: DMAR: DRHD: handling fault status reg 3
Jan  9 16:11:50 Vault kernel: DMAR: [DMA Write] Request device [04:00.1] fault addr 39d9c0000 [fault reason 02] Present bit in context entry is clear
Jan  9 16:11:50 Vault kernel: DMAR: DRHD: handling fault status reg 3
Jan  9 16:11:50 Vault kernel: DMAR: [DMA Write] Request device [04:00.1] fault addr 39d9c0000 [fault reason 02] Present bit in context entry is clear
Jan  9 16:11:50 Vault kernel: DMAR: DRHD: handling fault status reg 3
Jan  9 16:11:50 Vault kernel: DMAR: [DMA Write] Request device [04:00.1] fault addr 39d9c0000 [fault reason 02] Present bit in context entry is clear
Jan  9 16:11:50 Vault kernel: DMAR: DRHD: handling fault status reg 3
Jan  9 16:11:50 Vault kernel: DMAR: [DMA Write] Request device [04:00.1] fault addr 39d9c0000 [fault reason 02] Present bit in context entry is clear

I left it like this for 24 hours and then tonight, i lost a Parity Drive and Disk 1.

I tried moving the data sata cables around and still had the same drives failed in the system.

I have turned off ACS and ended up doing a new config to get the disk 1 back and then forced a party rebuild on the failed parity drive.

There is no physical error it looks like ACS broke something here.

Now i can't start my VM with GPU passthrough is there another option here ?

Edited January 10, 20188 yr by Maticks