![]() |
LambLisp 01 Red Fox Alpha
From AI to Actuator
|
Public Attributes | |
| int | bcm |
| int | board |
| int | offset |
| const char * | chip |
| const char * | name |
PIN NUMBERING IS THE ACTUAL WORK, not the libgpiod calls. "Raspberry-Pi compatible" cannot mean passing a number straight to gpiod: the Jetson's lines are named PA.00, PQ.05, PAC.06 – Tegra SoC naming – not GPIO17, and not header positions. The Orin Nano's 40-pin header IS physically RPi-compatible, so the mapping exists and simply has to be written down.
THIS TABLE IS GENERATED, NOT TRANSCRIBED. It comes from NVIDIA's own jetson-gpio data (/usr/lib/python3/dist-packages/Jetson/GPIO/gpio_pin_data.py, JETSON_ORIN_NANO), read on the machine and emitted as C. Hand-copying 22 rows of pin numbers is exactly the kind of transcription that produces a table which looks right and drives the wrong pin – and driving the wrong pin on a board with motors attached is not a typo you get to take back. Regenerate rather than edit: python3 -c "import importlib.util; spec=importlib.util.spec_from_file_location('g', \ '/usr/lib/python3/dist-packages/Jetson/GPIO/gpio_pin_data.py'); m=importlib.util.module_from_spec(spec); \ spec.loader.exec_module(m); [print(r) for r in m.jetson_gpio_data[m.JETSON_ORIN_NANO][0]]" (Note: importing Jetson.GPIO itself FAILS on this unit -- "Could not determine Jetson model" on the Orin Nano Engineering Reference Developer Kit Super – so load the data module directly.)