Yes, I’ve had the same issue pretty consistently. I wish Nvidia would fix this bug. I can’t remember the last time I encountered a computer where the USB ports couldn’t consistently negotiate the proper connection speed:
> lsusb -t | grep Storage
|__ Port 001: Dev 002, If 0, Class=Mass Storage, Driver=uas, 480M
I don’t know if it’s useful info, but FWIW I ran the command above after a reboot because I have an external drive (Crucial X10, USB 3.2x2), but it doesn’t seem to have this issue:
danny@toad:~$ lsusb -t | grep Storage
|__ Port 001: Dev 002, If 0, Class=Mass Storage, Driver=uas, 20000M/x2
I am guessing that it depends on the chipset in the external USB enclosure. If you run lsusb -tvv, it might give us the chipset of your usb storage device.
Assuming that was aimed at me, this is what I see (to be clear, it’s not an enclosure, it’s just an external drive, Cruxial X10):
/: Bus 002.Port 001: Dev 001, Class=root_hub, Driver=xhci-hcd/1p, 20000M/x2
ID 1d6b:0003 Linux Foundation 3.0 root hub
/sys/bus/usb/devices/usb2 /dev/bus/usb/002/001
|__ Port 001: Dev 002, If 0, Class=Mass Storage, Driver=uas, 20000M/x2
ID 0634:5607 Micron Technology, Inc.
/sys/bus/usb/devices/2-1 /dev/bus/usb/002/002
I have connected 3 USB storage before powering on.
|__ Port 001: Dev 002, If 0, Class=Mass Storage, Driver=uas, 20000M/x2
ID 174c:2364 ASMedia Technology Inc.
|__ Port 001: Dev 002, If 0, Class=Mass Storage, Driver=uas, 20000M/x2
ID 174c:2364 ASMedia Technology Inc.
|__ Port 001: Dev 002, If 0, Class=Mass Storage, Driver=uas, 20000M/x2
ID 152d:0581 JMicron Technology Corp. / JMicron USA Technology Corp.
I believe it depends on USB storage device. I have another USB storage with 3.2 gen 2x2 / ASMedia , but is couldn’t connect 20000M/x2.
The issue only appears after rebooting the machine with the device plugged in. If I unplug/replug, it negotiates back to the correct speed. Maybe there is an issue only with certain devices, but it does happen with the NVMe enclosure I’ve been trying to use.
Bus 006 Device 002: ID 1e91:de79 Other World Computing Express 1M2
It can be cable problem too so I will try to get some new cables and test it to give more feedback. I think it is a driver issue. Most of these chip has internal driver writting in embeded C, they might not test all scenario.
They might test some motherboards then call it a day.