• sales

    +86-0755-88291180

Milk-V Duo linux Compile kernel Tutorial

[] Preparatory work


Local environment windows11 64 bit


Compilation environment: VMware Ubuntu 22.04.4.2 LTS

(If you don't have Ubuntu installed, you need to install one locally or in a virtual machine.)


Prepare an SD card

milk-v duo


[] Installation:

sudo apt-get update


sudo apt install dialog python3-dev make git bc gcc flex bison ninja-build libssl-dev rsync pkg-config device-tree-compiler squashfs-tools parted dosfstools


Installation cmake


sudo apt-get install cmake


Create a development environment directory:

mkdir duo && cd duo


[] Download the development package cvi_mmf_sdk.zip:

wget -c https://link.zhihu.com/?target=https%3A//codeload.github.com/sophgo/cvi_mmf_sdk/zip/refs/heads/v4.1.0 -O cvi_mmf_sdk.zip


Then decompress:

unzip cvi_mmf_sdk.zip

mv cvi_mmf_sdk-4.1.0/ cvi_mmf_sdk/


[] Download tool chain:

wget https://link.zhihu.com/?target=https%3A//sophon-file.sophon.cn/sophon-prod-s3/drive/23/03/07/16/host-tools.tar.gz


Then decompress:

tar xvf host-tools.tar.gz

Move to the development package:

mv host-tools/ cvi_mmf_sdk/


[] Start compiling in a single step as follows:

cd cvi_mmf_sdk/

source build/cvisetup.sh

defconfig cv1800b_sophpi_duo_sd

clean_all

build_all

pack_sd_image


Note: If you are prompted that libssl.so.1.1 is missing during compilation, install it as follows.

wget

http://security.ubuntu.com/ubuntu/pool/main/o/openssl/libssl1.1_1.1.1-1ubuntu2.1~18.04.23_amd64.deb

(If the link fails, you can download it yourself here http://security.ubuntu.com/ubuntu/pool/main/o/openssl/)


sudo dpkg -i libssl1.1_1.1.1-1ubuntu2.1_18.04.21_amd64.deb


After the final compilation is completed, an install folder will be generated under the folder.


install/soc_cv1800b_sophpi_duo_sd The folder directory is as follows:



[] Write system to SD card:


cd install/soc_cv1800b_sophpi_duo_sd/


You can find the *.img file. SD card and insert it into the computer.


Write image to SD card using command line.

sudo dd if=./sophpi-duo-*.img of=/dev/sdx bs=32M status=progress oflag=direct

("sophpi-duo-*.img" and "sdx" Change it to your name.)


Or copy the *.img file. Use the software to directly burn the image to the SD card, refer to Milk-V Duo Image writing


[] Start the system:


The SD card is plugged into the board, connected to the board's serial port 0 by TTL serial port (no power supply is needed), and then the serial port is opened on the computer with serial debugging tools, and the board is connected to the computer with TYPEC cable, and the system can be entered after a while.



Note that RNDIS protocol is not added to this version of the system kernel, so the DUO development board cannot be recognized in the device manager after it is plugged into USB!

TAG: AI Machine Vision Kit OAK-D-POE JPEG Encoder 12MP IP67 Waterproof 4TOPS OpenCV Camera Industrial grade isolated USB to RS232/485 converter supports USB to 2-way RS232 + 2-way RS232/485 original FT4232HL ESP32 C6 AI 0.85inch Development Board 0.85 inch DeepSeek RGB surround light Dual mike LuckFox SC3336 Jetson Nano PWM Fan ESP32 S3 Development Board 4 inch LCD Capacitive Touch Screen 4inch Display 480×480 N16R8 Raspberry Pi 4/5 UPS Uninterruptible Power Supply HAT Lithium Battery Module PoE Power Supply ESP32 S3 Development Board 4.3 inch LCD Capacitive Touch Screen 4.3inch Display 480×272 N16R8 X1005 Raspberry Pi 5 PCIe to M.2 NVMe Dual SSD Adapter Board HAT Pi5 Double 2230/2242/2260/2280 Raspberry Pi 5 PCIe to M.2 Mini NVMe SSD Board (E) With Customized Active Cooler Together Industrial Isolated RS485 TO ETH (C) RJ45 Converter Wall/Rail-Mount For Modbus OLED-LCD-HAT-A ESP32 S3 3.5 inch LCD Development Board 480x320 Display Capacitive TouchScreen Tablet /MP3 Video /Weather Clock LVGL For Arduino Raspberry Pi Pico 2 RP2350 Core2350B Core Board PSRAM ESP32-S3 2.8inch Round LCD USB Monitor Computer Monitor Display Screen USB Type C Secondary Screen PC CPU GPU Raspberry Pi 15.6 inch FHD Monitor (C) LCD 1080P Capacitive TouchScreen Display Type-C/HDMI For Computer PC Integrated capacitive fingerprint module Cortex core UART serial port High-precision fingerprint recognition Raspberry Pi 5 Camera