Infrastructure for supporting SCons build.
The upload functions are split into two parts:
- Independent upload logics reside in
scons_upload.py; - Board specific upload configs reside in
upload/*.json.
A common upload process is:
- Start upload with
aos upload ...; - Parse board configs from
upload/*.json; - Run board specific commands to finish upload process.
By using upload configs, users can add upload support easily. They only need to focus on board specific informations, such as flash commands, arguments, binaries location ..., but don't need to care about how the common logic runs.
For more details about upload configs, please refer to upload/README.md.
After build done, run below command to upload images to board:
$ aos upload appname@board
Add upload support for boards, just creating upload configs for them simply. We strongly recommend that using gen_upload_configs.py to update or create upload configs:
Step 1: Update gen_upload_configs.py
# Register board with format:
# { 'board name': [upload_config_file], }
# For example:
registry_board = {
'esp32': ['esp32.json'],
...
}
# Create python dict for upload_config_file, for example:
# Flash configs for esp32* board
esp32 = {
'cmd': [
'esptool.py',
'--chip',
'esp32',
'--port', '@PORT@',
'--baud', '921600',
'--before', 'default_reset',
'--after', 'hard_reset',
'write_flash',
'-z',
'--flash_mode',
'dio',
'--flash_freq',
'40m',
'--flash_size',
'detect',
'0x1000', '@AOSROOT@/platform/mcu/esp32/bsp/bootloader.bin',
'0x8000', '@AOSROOT@/platform/mcu/esp32/bsp/custom_partitions.bin',
'0x10000', '@AOSROOT@/out/@TARGET@/binary/@TARGET@.bin'
],
}
flash_configs['esp32'] = esp32
Step 2: Run gen_upload_configs.py to create upload/*.json
$ cd build; ./gen_upload_configs.py
Step 3: Run aos upload appname@board to test the upload process.
Similiar with upload, the debug functions are split into:
- Independent debug logic reside in
scons_debug.py; - Board specific debug configs reside in
debug/*.json.
For more details about debug configs, please refer to debug/README.md.
Start GDB server first:
$ aos debug appname@board
Open a new terminal and start GBD client:
$ aos debug application@board -c
# Connect to server and start debug:
gdb) target remote localhost:<port>
...
Use gen_debug_configs.py to update or create debug configs for boards:
Step 1: Update gen_debug_configs.py
# Register board name with format:
# { 'board name': [ debug_config_file ], }
# For example:
registry_board = {
'developerkit': ['stm32l4x.json'],
}
# Create python dicts for debug_config_file, for example:
# Debug configs for stm32l4xx
stm32l4x = {
'start_server': [
{
'Linux64': {
'cmd': '@AOSROOT@/build/openocd_master/bin/openocd_linux',
'url': '', },
'OSX': {
'cmd': '@AOSROOT@/build/openocd_master/bin/openocd_osx',
'url': '', },
},
'-f',
'@AOSROOT@/build/openocd_master/share/openocd/scripts/board/stm32l4discovery.cfg',
],
'start_client': [
{
'Linux64': {
'cmd': '@AOSROOT@/build/compiler/gcc-arm-none-eabi/Linux64/bin/arm-none-eabi-gdb',
'url': '', },
'OSX': {
'cmd': '@AOSROOT@/build/compiler/gcc-arm-none-eabi/OSX/bin/arm-none-eabi-gdb',
'url': 'https://gdb-archive.oss-cn-shanghai.aliyuncs.com/gdb-arm-none-eabi-osx.tar.gz', },
},
'@AOSROOT@/out/@TARGET@/binary/@TARGET@.elf'
]
}
debug_configs['stm32l4x'] = stm32l4x
Step 2: Run gen_debug_configs.py to create debug/*.json
$ cd build; ./gen_debug_configs.py
Step 3: Run aos debug ... to test debug process