Document link:
RHEL10:
https://docs.redhat.com/en/documentation/red_hat_enterprise_linux/10/html-single/configuring_and_managing_linux_virtual_machines/index#setting-up-nvidia-vgpu-devices
or
RHEL9:
https://docs.redhat.com/en/documentation/red_hat_enterprise_linux/9/html/configuring_and_managing_virtualization/assembly_managing-gpu-devices-in-virtual-machines_configuring-and-managing-virtualization
or
RHEL8
https://docs.redhat.com/en/documentation/red_hat_enterprise_linux/8/html/configuring_and_managing_virtualization/assembly_managing-gpu-devices-in-virtual-machines_configuring-and-managing-virtualization
Section number and name:
Take the RHEL9 as an example:
RHEL9 Configuring and managing virtualization => Chapter 16. Managing GPU devices in virtual machines =>16.2.1. Setting up NVIDIA vGPU devices => Procedure 11. Set the vGPU device to start automatically after the host reboots
Describe the issue:

Set the vGPU device to start automatically after the host reboots

virsh nodedev-autostart mdev_30820a6f_b1a5_4503_91ca_0c10ba58692a_0000_01_00_0
Device mdev_d196754e_d8ed_4f43_bf22_684ed698b08b_0000_9b_00_0 marked as autostarted

The mdev device identifier is different between command line and output.

Impact of this issue:
Medium

Suggestions for improvement:
Correct the output of "virsh nodedev-autostart $mdev", like

# virsh nodedev-autostart mdev_30820a6f_b1a5_4503_91ca_0c10ba58692a_0000_01_00_0
Device mdev_30820a6f_b1a5_4503_91ca_0c10ba58692a_0000_01_00_0 marked as autostarted