Hi guys - looking for a bit of advice on how to check what the problem here is.  I thought i had a shutdown (the VM I left running looks to have restarted), a quick check in the logs gives me repeated incidences of this message:

 

"Aug 14 09:00:06 FILE-SERVER kernel: nvme 0000:04:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0000 address=0x00000000ffc75000 flags=0x0000]"

 

and mixed in with these are:

 

"Aug 14 09:00:07 FILE-SERVER kernel: amd_iommu_report_page_fault: 1617 callbacks suppressed"

 

I am running new hardware, Ryzen 3900x on Asus x570, and the NVIDIA UNRAID version - the offending hardware is my Crucial NVME SSD running under unassigned devices, it is inside it's own IOMMU group and doesnt share this group with any other devices:

 

"IOMMU group 21:[c0a9:2263] 04:00.0 Non-Volatile memory controller: Micron/Crucial Technology Device 2263 (rev 03)"

 

Looking back I can see this is appearing in the logs since the new installation, but am now at a bit of a loss on what I can do, any advice or guidance would be appreciated!

 

 

Edited August 14, 20196 yr by fireb1adex

Hmmm, found this - possibly related?

 

https://level1techs.com/article/patch-npt-ryzen-better-performance 

npt patch was added a long time ago so that won't be it. Have you tried running it without the SSD to see what happens in the log? 

Looks like there is a new BIOS for my board, AGESA 1.0.0.3 Patch ABB that solves PCI/IOMMU issue for some people, might give that a go first and report back (there is no current firmware upgrade for the NVME drive).

Ok, well new agesa installed (ABBA), still seeing this issue. The NVME is used solely for 3 Windows VMs and I have notices no issues with running them.

 

Not really sure if this error is anything to worry about or not...the VMs are in a production environment so I need them to remain error free and stable, is there a risk to my machines or their data with this error?

 

Here's what I am seeing at the moment (device 04:00:00 is the NVME drive on its own IOMMU group 21:[c0a9:2263] 04:00.0 Non-Volatile memory controller: Micron/Crucial Technology Device 2263 (rev 03)):

 

Sep 30 13:00:09 FILE-SERVER kernel: amd_iommu_report_page_fault: 1516 callbacks suppressed
Sep 30 13:00:09 FILE-SERVER kernel: AMD-Vi: Event logged [IO_PAGE_FAULT device=04:00.0 domain=0x0000 address=0x00000000fe507000 flags=0x0000]