Adding to the endless list of “nobody knows why the hell this happens” workarounds:
I found that Intel 8265NGW, widely recommended online, works for Wi-Fi but causes all USB buses to disappear in Xavier. Use 3168NGW which does not mess up the USB. It also worked out of the box with no fuss.
We only knew 8260 would cause the i2c bus to pulldown and thus usb type c is going down too.
May I have your kernel log to check why usb is gone?
This is the log gathered from Xavier with Intel 8265NGW & USB Stick work in the same time.
nvidia@N:~$ iwconfig
lo no wireless extensions.
usb0 no wireless extensions.
l4tbr0 no wireless extensions.
dummy0 no wireless extensions.
wlan0 IEEE 802.11 ESSID:off/any
Mode:Managed Access Point: Not-Associated Tx-Power=0 dBm
Retry short limit:7 RTS thr:off Fragment thr:off
Power Management:on
eth0 no wireless extensions.
rndis0 no wireless extensions.
nvidia@N:~$ lspci
0001:00:00.0 PCI bridge: NVIDIA Corporation Device 1ad2 (rev a1)
0001:01:00.0 SATA controller: Marvell Technology Group Ltd. Device 9171 (rev 13)
0003:00:00.0 PCI bridge: NVIDIA Corporation Device 1ad2 (rev a1)
0003:01:00.0 Network controller: Intel Corporation Wireless 8265 / 8275 (rev 78)
nvidia@N:~$ dmesg | tail -10
[ 391.890705] scsi host2: usb-storage 1-2.3:1.0
[ 393.077926] scsi 2:0:0:0: Direct-Access USB DISK 1100 PQ: 0 ANSI: 4
[ 393.078999] sd 2:0:0:0: [sda] 15730688 512-byte logical blocks: (8.05 GB/7.50 GiB)
[ 393.079714] sd 2:0:0:0: [sda] Write Protect is off
[ 393.079821] sd 2:0:0:0: [sda] Mode Sense: 43 00 00 00
[ 393.080344] sd 2:0:0:0: [sda] No Caching mode page found
[ 393.080456] sd 2:0:0:0: [sda] Assuming drive cache: write through
[ 393.088320] sda: sda4
[ 393.090810] sd 2:0:0:0: [sda] Attached SCSI removable disk
[ 427.961243] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready
My logs with a 8265NGW are in:
https://devtalk.nvidia.com/default/topic/1051726/usb-stopped-working-all-of-a-sudden/?offset=3#5338719
Note: I did not send it in for RMA since it is being actively used for something critical, so I could not RMA it without getting a temporary one first to bridge the RMA service time gap.
I changed it to 3168NGW and the USB started working again.
Please always share the full log instead of partial one.
BTW, your description in that thread just indicates it is more like a hardware issue… please RMA it if possible…
It might be some delay but I’ll try to get it when there is a convenient time. I need to swap the 8265NGW back in to get the full log.
I did post that thread earlier but later realized that using 3168NGW reliably made USB work and 8265NGW reliably made USB fail.
I would be happy to send for RMA but would need a rental unit as it’s running something that can’t tolerate much downtime …