Hi everyone! I’m trying to use the microusb on the jetson tx2, is there any procedure that I need to do to have the microusb as usb 2.0. I’m trying to connect a gps in that port.
There is no special change required. Do know though that this is a micro-OTG (“On the Go”) port, and this kind of connector accepts either a type-A or a type-B connector. With a type-B (which is what flash uses and the Jetson comes with a cable for this) this port is useful only with the Jetson acting as a device, not as a host. A type-A cable connector is needed for the case of plugging devices into the port (the Jetson is a host instead of a device for type-A).
This is unrelated to @davidnet’s problem, but using the micro-USB to full size USB adapter included in Jetson dev board pack without any device plugged in leads to spurious kernel messages (2 each 10 seconds):
[11588.227092] extcon-gpio-states external-connection:extcon@1: Cable state 2
[11588.234263] tegra-xotg xotg: update_id_state: ID grounded
[11588.251712] extcon-gpio-states external-connection:extcon@1: Cable state 2
[11597.275092] extcon-gpio-states external-connection:extcon@1: Cable state 2
[11598.275073] extcon-gpio-states external-connection:extcon@1: Cable state 2
[11607.299106] extcon-gpio-states external-connection:extcon@1: Cable state 2
[11608.299075] extcon-gpio-states external-connection:extcon@1: Cable state 2
It is easy to fix, it is not a common use case, but last day I got kernel logs filled with this after a few hours letting my Jetson with the cable adapter only…
Is there really a need to repeat this message ?
I have one adapter for micro-A to full size which works, one which fails. I don’t remember which one is the one which arrived with the Jetson versus the other.
I have one adapter from a TX1 dev board kit and another from TX2 kit (seems to be same model), both give that.
Do these cables have something wrong ?
I was thinking it was just a spurious trace from kernel.
There probably shouldn’t be a spurious trace. I have not looked closely at this since installing R28.2, and saw no such spurious trace in R28.1. There could be some issue with changes to how OTG detect pin and related code work together such that cables matter…I don’t know. I have many cables though, I’m not sure where they all came from :P I do see there is some device tree setup tied to “extcon” drivers, but I probably won’t get a chance to look at any of it until tomorrow.
For @davidnet, are you seeing an error? Regardless, it would be useful to see what the system thinks of your device by running “dmesg --follow” prior to connecting to the micro-USB port and looking at what is reported. I’m not sure if you tried it and it failed, or if you saw an outright error. It could be that the USB was working for you, but you didn’t have a driver…more information is needed.