Nvidia Orin PCIe device, there is no PCIe device in "virsh nodedev-list", does Orin IOMMU supports passthrough mode?

Hi,

when I assign pcie device to my VM, I experienced this issue:

I enabled SR-IOV on NVMe device, and created 4 VFs, I can use lspci command to list these PCIe devices, but there is no device in the command “virsh nodedev-list”. such as:

root@orin1:~# lspci

0001:00:00.0 PCI bridge: NVIDIA Corporation Device 229e (rev a1)

0001:01:00.0 Network controller: Realtek Semiconductor Co., Ltd. RTL8822CE 802.11ac PCIe Wireless Network Adapter

0005:00:00.0 PCI bridge: NVIDIA Corporation Device 229a (rev a1)

0005:01:00.0 PCI bridge: PMC-Sierra Inc. Device 4028

0005:01:00.1 Memory controller: PMC-Sierra Inc. Device 4028

0005:02:00.0 PCI bridge: PMC-Sierra Inc. Device 4028

0005:03:00.0 Non-Volatile memory controller: Samsung Technology Inc Device 6003

0005:03:01.0 Non-Volatile memory controller: Samsung Technology Inc Device 6003

0005:03:01.1 Non-Volatile memory controller: Samsung Technology Inc Device 6003

0005:03:01.2 Non-Volatile memory controller: Samsung Technology Inc Device 6003

0005:03:01.3 Non-Volatile memory controller: Samsung Technology Inc Device 6003

but:

root@orin1:~# virsh nodedev-list

block_mmcblk1_0x8d47007a

computer

net_dummy0_a6_ec_9d_ce_0c_69

net_eth0_48_b0_2d_7b_8a_28

net_lo_00_00_00_00_00_00

net_virbr0_nic_52_54_00_c5_1c_30

net_wlan0_90_e8_68_84_3d_23

usb_1_0_1_0

usb_1_3

usb_1_3_1_0

usb_1_3_1_1

usb_1_4

usb_1_4_1

usb_1_4_1_0

usb_1_4_1_1

usb_1_4_1_1_0

usb_1_4_1_1_1_0

usb_1_4_1_2

usb_1_4_1_2_1_0

usb_1_4_1_2_1_1

usb_1_4_1_2_1_2

usb_1_4_1_3

usb_1_4_1_3_1_0

usb_1_4_1_3_1_1

usb_1_4_1_3_1_2

there is no PCIe device here. also I tried to used vrit-manager, in the pcie host device, it shows that no available device.

who knows that if this is the Orin board issue or SW issue???

thanks so much for your suggestion!!!

Note:

I googled some information about this issue, seems it is related to IOMMU:

PCI(e) Passthrough - Proxmox VE

IOMMU Passthrough Mode

If your hardware supports IOMMU passthrough mode, enabling this mode might increase performance. This is because VMs then bypass the (default) DMA translation normally performed by the hyper-visor and instead pass DMA requests directly to the hardware IOMMU. To enable these options

the question is that does Orin IOMMU support this feature?

Martin