Hello,
I have an issue with the USB mode in my jetson nano. When I connect the USB to one of my computer (an Ubuntu 18.04) , the computer detect it ( I can find the jetson in /dev/ttyACM0 ) but when I do this with my other computer (an Ubuntu 20.04) it can’t detect it.
Does anyone knows why ?
In case it helps this is my dmesg in my second computer right after plug the USB :
[ 7740.324755] usb 3-3.2: new high-speed USB device number 13 using xhci_hcd
[ 7740.416763] usb 3-3.2: device descriptor read/64, error -71
[ 7740.616817] usb 3-3.2: device descriptor read/64, error -71
[ 7740.816681] usb 3-3.2: new high-speed USB device number 14 using xhci_hcd
[ 7740.908811] usb 3-3.2: device descriptor read/64, error -71
[ 7741.112826] usb 3-3.2: device descriptor read/64, error -71
[ 7741.220832] usb 3-3-port2: attempt power cycle
[ 7741.848686] usb 3-3.2: new high-speed USB device number 15 using xhci_hcd
[ 7741.848824] usb 3-3.2: Device not responding to setup address.
[ 7742.056974] usb 3-3.2: Device not responding to setup address.
[ 7742.264690] usb 3-3.2: device not accepting address 15, error -71
[ 7742.356805] usb 3-3.2: new high-speed USB device number 16 using xhci_hcd
[ 7742.356979] usb 3-3.2: Device not responding to setup address.
[ 7742.564865] usb 3-3.2: Device not responding to setup address.
[ 7742.772790] usb 3-3.2: device not accepting address 16, error -71
[ 7742.773225] usb 3-3-port2: unable to enumerate USB device
I can’t provide anything particularly useful, but that is detecting the device mode Nano. Then it fails for USB reasons. Perhaps signal quality, don’t know. But you might want to try a HUB between the Nano and PC as a kind of buffer. Or if it is already on a HUB, then direct wire without a HUB.
I tried both with and without HUB and nothing change…
I would say then that this is either a signal issue, or else a quirk needing some sort of driver adjustment (because of the testing with and without a HUB odds shift towards quirk, but it isn’t a guarantee). A USB protocol analyzer would go straight to the answer, but those are expensive. I don’t have any way of knowing what the actual issue is, but perhaps it has been fixed in an upstream kernel. If you find a fix in a newer kernel release, then you could post it here and it could be back-ported into the current kernel.
I have some updates. I re-tried to plug with and without HUB but this time I do dmesg after every USB that I have tested.
And when I tried on the USB 3.0 of my PC dmesg give me this :
[ 414.338182] usb 3-3: new high-speed USB device number 28 using xhci_hcd
[ 414.491017] usb 3-3: unable to get BOS descriptor or descriptor too short
[ 414.492132] usb 3-3: New USB device found, idVendor=0955, idProduct=7020, bcdDevice= 0.02
[ 414.492135] usb 3-3: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[ 414.492137] usb 3-3: Product: Linux for Tegra
[ 414.492138] usb 3-3: Manufacturer: NVIDIA
[ 414.492139] usb 3-3: SerialNumber: 1424320061364
[ 414.497534] rndis_host 3-3:1.0: RNDIS init failed, -71
[ 414.497550] rndis_host: probe of 3-3:1.0 failed with error -71
[ 414.498803] cdc_acm 3-3:1.2: ttyACM0: USB ACM device
[ 414.499481] usb-storage 3-3:1.4: USB Mass Storage device detected
[ 414.499775] scsi host0: usb-storage 3-3:1.4
[ 414.518847] cdc_ncm 3-3:1.5: bind() failure
[ 415.658570] usb 3-3: reset high-speed USB device number 28 using xhci_hcd
[ 415.806633] usb 3-3: device firmware changed
[ 415.806702] usb 3-3: USB disconnect, device number 28
[ 415.994192] usb 3-3: new high-speed USB device number 29 using xhci_hcd
[ 416.122427] usb 3-3: Device not responding to setup address.
[ 416.352503] usb 3-3: New USB device found, idVendor=0955, idProduct=7020, bcdDevice= 0.02
[ 416.352506] usb 3-3: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[ 416.352508] usb 3-3: Product: Linux for Tegra
[ 416.352509] usb 3-3: Manufacturer: NVIDIA
[ 416.357487] usb 3-3: can’t set config #1, error -71
NVIDIA may have seen this before, but I can’t answer how to work with this short of a logic analyzer. To me the error -71 is too generic, but it is how Ubuntu does it (this kernel code is not specific to the Jetson even if there is some part of the Jetson-specific code which is a problem). However, that last log is useful since it confirms recognition of the device even if data is failing.
Hi,
It seems to be an issue in signal quality. Do you have other type-A to micro-B cables for a try.
And please share the release version( $ head -1 /etc/nv_tegra_release ) for reference.