EDITORIAL
jetson nano boards at the preset mode 10W 5v4A of the DC power supply need.
5v2A with a DC power supply or micro-usb 5W mode is recommended.
Insufficient power supply will cause power failure shut down.
The following is a learning jetson nano, the translation of the document supply mode.
nvidia Original: https: //docs.nvidia.com/jetson/l4t/index.html#page/Tegra%20Linux%20Driver%20Package%20Development%20Guide%2Fpower_management_nano.html%23wwpID0E0YI0HA
Power supply mode
Jetson Nano Efficient power management integrated circuit (PMIC), regulator and power source tree design to optimize power efficiency. It supports two power modes, e.g. 5W and MaxN (10W). Each mode allows a variety of configurations, and having various line frequencies CPU cores.
You can limit by a predefined value memory, CPU and GPU frequency and number of cores, the module is configured to a predefined limit.
The following table shows the upper module and associated power mode resources used NVIDIA predefined.
The default mode is: MaxN (10W) (corresponding to ID 0).
Switch modes:
•input the command:
$ sudo /usr/sbin/nvpmodel -m <x>
Which <x>
corresponds to the mode ID, such as 0 or 1.
After setting the power mode, the module will remain in that mode until you change it. The pattern persist between power cycles and SC7.
Displays the current power mode:
•input the command:
$ sudo /usr/sbin/nvpmodel -q
Learn about other options:
•input the command:
$ /usr/sbin/nvpmodel -h
Define Custom mode:
• To define your own custom mode, add the schema definition to the file:
<top>/l4t/release/rfs/etc/nvpmodel/nvpmodel_t210_jetson-nano.conf
This is an example of mode 1:
< POWER_MODEL ID=1 NAME=5W >
CPU_ONLINE CORE_0 1
CPU_ONLINE CORE_1 1
CPU_ONLINE CORE_2 0
CPU_ONLINE CORE_3 0
CPU_A57 MIN_FREQ 0
CPU_A57 MAX_FREQ 918000
GPU_POWER_CONTROL_ENABLE GPU_PWR_CNTL_EN on
GPU MIN_FREQ 0
GPU MAX_FREQ 640000000
GPU_POWER_CONTROL_DISABLE GPU_PWR_CNTL_DIS auto
EMC MAX_FREQ 1600000000
CPU frequency units are kilohertz (KHz). GPU and EMMC is measured in hertz (Hz). You must assign a unique ID for each custom mode in the ID field.
Test your custom mode to determine:
number of active cores to be used •
• the frequency GPU, EMC and each CPU cluster set of
constraints limit of 0 MaxN defined by the mode you set the frequency.