I have a question regarding the LS1021A USB3.0 port when set up as a gadget.
Here is what I'm observing:
When connecting my LS1021A based gadget to a USB 3.0 Host Port, I observe the LS1021A attempt enumeration using both 3.0 signals and 2.0 signals concurrently. This results in the host responding as if two devices are trying to connect to the USB bus. From a Microsoft Windows perspective, the device enumerates properly at Superspeed but displays an error message regarding the enumeration attempt using 2.0 since the driver has already attached at Superspeed.
The behaviour above is observed using both our in-house gadget driver and with the standard NXP gadget driver.
From a hardware perspective, our device uses a Type-C connector. The USB 3.0 signals from the Type-C connector are routed through a Type-C controller and on to the Superspeed diff pairs of the LS1021A (port USB1). The USB 2.0 signals are routed directly from the connector to the LS1021A USB 2.0 diff pair (port USB1). Unfortunately I can't share schematics due to security reasons.
It's my understanding that a USB 3.0 controller in gadget mode should attempt enumerating at Superspeed (USB 3) prior to falling back to HS/FS (USB 2). Does anyone have any information on setting up the LS1021A so that it follows this sequence?
Thanks and best regards,