Fixes the error:
Aborting due to missing @type tag in file: 'build/px4_fmu-v5_test/etc/init.d/airframes/11001_hexa_cox'
This can happen due to a change to e.g. board_serial_ports, which changes
the CLI command and triggers a re-execution of the airframe processing.
- cmake NuttX build wrapper compile in place instead of copying source tree to build directory
- slightly faster skipping necessary copying (depending on system)
- allows debugging in place
- easier to work directly in NuttX following official documentation
- simplifies overall build which should make it easier to resolve any remaining NuttX dependency issues in the build system
- the downside is switching back and forth between different builds always require rebuilding NuttX, but I think this is worth the improved developer experience
- also no longer builds px4io and bootloader in every single build, for most users these rarely change and we're wasting a lot of build time
This avoids incremental build errors when switching between branches with
a different set of airframes.
E.g:
Aborting due to missing @type tag in file: 'Firmware/build/px4_fmu-v5_default/etc/init.d/airframes/13030_generic_vtol_quad_tiltrotor'
- moves process airframe deps to the tar generation, to ensure whenever
airframes are processed, the tar is extracted first.
This is required, because if the airframe processing is done on pruned
ROMFS, it's missing the tags.
- avoid using init.d/rcS in dependencies, use a separate file instead.
This is because pruning modifies the rcS, and the next rebuild was
seeing a changed rcS, resulting in unnecessary regeneration.
-moved rc.mavlink to the boards optional rc additions (now it's called rc.board_mavlink) to handle board specific mavlink needs (mavlink over usb, ethernet, additional streams, etc.)
-mavlink module will be responsible to usb defaults, therefore less args are needed to be passed to mavlink module if one wants to use mavlink over usb.
-the way to check if connection is usb is by it's designated variable and not by config mode.
The startup file rc.board is copied to the build/../genromfs folder
after the pruning of the files happened. Thus, about 2000 bytes of
comments and whitespace were included in the ROMFS.
This commit moves pruning of the files in ROMFS to a separate cmake
custom command which happens after the ROMFS files have been copied and
the extra files have been added.
Testing showed that this change only affected the rc.board file as
expected.
Starting with cmake version 3.12 the cmake -E copy_directory
command fails if the source directory does not exist. This results
in a build failure. This fix creates the source directory before
the copy which does not harm if the source directory exists.
Closes: #10368
- every incremental rebuild extended ${config_romfs_files_list}, the
chached variable was never reset
- cmake -E copy_directory did not remove deleted files
remaining issue: removing a file from the ROMFS & px4_add_romfs_files()
does not trigger the px_romfs_pruner.py COMMANDs to be re-executed.
* Lists ROMFS files explicity instead of using GLOB_RECURSE
Previously, when ROMFS files that were not airframes were touched, the ROMFS
would not be rebuilt. The ROMFS files are now specified explicityl in a
CMakeLists.txt file that is located in the root ROMFS directory.
Now when one of the ROMFS files is touched the whole ROMFS is rebuilt.
When new files are added to the ROMFS, they need to be explicity added to
the CMakeLists in the ROMFS root directory.
* ROMFS: adds individual CMakeLists files in each subdirectory
Also moves the temporary ROMFS build directory to ${PX4_BINARY_DIR}/ROMFS/genromfs
so that the cmake_install.cmake files and the CMakeFiles directories (generated whenever
are not add_subdirectory() is called) are not generated in the temporary ROMFS directory
from which the ROMFS binary is created.
* cmake ROMFS generate add px4_add_romfs_files function
* ROMFS CMakeLists: adds explanatory comment to px4_add_romfs_files function
* ROMFS CMakeLists: updates copyright headers
* NuttX cmake
* px4_macros:Pass the stringified predicate as second arg to static assert
CC_ASSERT mapes to the c++ static_assert or provides the same
funtionality for c via the other macros. The c++ static assert
takes 2 argumants the prdicate and a message. This fixes the
lacking second argument.
* Updated nuttx and apps submodule to upstream nuttx 7.21+==master
This is the latest uptake of upstream nuttx and apps.
* ROMFS generate with xxd instead of objcopy
* delete nuttx-patches
* NuttX update submodules to latest px4_nuttx-master
* fix nuttx apps and board dependency
* docker_run update to latest container 2017-08-29
* cmake ROMFS portable sed usage
* NuttX update submodules to latest px4_nuttx-master
Daniel Agar
Beat Küng
garfieldG
David Sidrane
Julien Lecoeur
David Sidrane
Julian Oes
fredowski
ksschwabe