You are not logged in.

#1 2023-11-09 05:41:06

solskog
Member
Registered: 2020-09-05
Posts: 462

deduce iommu group

If it's possible to deduce iommu groups from hardware logs without knowing /sys/kernel/iommu_groups/ ?
I am trying to find out if Genesys Logic and ASM1074 controllers are in separate and standalone groups.

https://www.asrock.com/mb/Intel/N100M/index.us.asp#Specification
https://linux-hardware.org/?probe=a52836d33c

Offline

#2 2023-11-09 08:51:43

Lone_Wolf
Administrator
From: Netherlands, Europe
Registered: 2005-10-04
Posts: 14,989

Re: deduce iommu group

The info is usually present in dmesg & journal output, but the dmesg attached to the probe don't show it.

https://ark.intel.com/content/www/us/en … 0-ghz.html shows (at the bottom) the processor does support VT-d / VT-x .

$ lspci -v should show the iommu group, but the lspci outputs of the probe don't.

try adding intel_iommu=on to the kernel command line.

Last edited by Lone_Wolf (2023-11-09 08:52:24)


Disliking systemd intensely, but not satisfied with alternatives so focusing on taming systemd.

clean chroot building not flexible enough ?
Try clean chroot manager by graysky

Offline

#3 2023-11-09 09:05:54

seth
Member
From: Won't reply 2 private help req
Registered: 2012-09-03
Posts: 75,428

Re: deduce iommu group

Afaiu you don't have access to the HW and probably want to inquire their options before purchasing it?

The concerned devices don't even show up in those lspci, only lsusb.
genesys and one ASM are on the same bus, I'd lower my expectations for splitting those
Only 174c:3074 ASMedia is on a different bus so with some luck those are split between 00:0d.0 and 00:14.0 and those *might* be in different IOMMU groups

Offline

#4 2023-11-09 12:26:56

solskog
Member
Registered: 2020-09-05
Posts: 462

Re: deduce iommu group

seth wrote:

Afaiu you don't have access to the HW and probably want to inquire their options before purchasing it?

Yes, exactly

genesys and one ASM are on the same bus

That is also what I could deduce from Log:hwinfo where both share the same device ID: 00:14.0

between 00:0d.0 and 00:14.0 and those *might* be in different IOMMU groups

Intel USB Controller device id: 00:0d.0 and ASMedia ASM1074 device id:00:14.0 looks like been in different IOMMU groups.
As for the standalone part of the iommu groups, I think Intel RAM memory device id: 00:14.2 is in the same group as 00:14.0.
And 00:0d.0 is probably in an standalone iommu group.
The question is now which of these USB controllers I can pass through?

Offline

#5 2023-11-09 13:34:45

seth
Member
From: Won't reply 2 private help req
Registered: 2012-09-03
Posts: 75,428

Re: deduce iommu group

Bus 003 Device 002: ID 1997:2433 Shenzhen Riitek Technology Co., Ltd wireless mini keyboard with touchpad

So probably not that one, but that's really everyones guess only.
They might still be in the same group - or grouped w/ other devices you want to keep on the host.

To answer your original question: yes, but, as Lone_Wolf pointed out, not from those logs sad

Offline

Board footer

Powered by FluxBB