In general, if anything goes wrong in the startup script, we
should fail entirely because things might not work as expected.
In particular, this prevents that we have to press Ctrl+C twice if the
simulator start call is hung waiting for the simulator to appear and
start communicating. We now press Ctrl+C once and exit straightaway
whereas before we would press it once to get the warning:
"Startup script returned with return value: 2",
and then finally exit on the second press.
- nuttx in PX4/Firmware (327c7f3672ee5f066ec53dfe1554a02d1aa610da): 2e55ce8ec9
- nuttx current upstream: 9331fda0d7
- Changes: 2e55ce8ec9...9331fda0d7
9331fda0d7 2019-12-10 David Sidrane - [BACKPORT] IMXRT106x USDHC: Support regular GPIO for CD and inversion.
91c80c708b 2019-12-05 David Sidrane - [BACKPORT] imxrt106x:pinout add ALT 8 GPIO_GPT1_CAPTURE[1|2]
32f758f0e9 2019-12-03 David Sidrane - [BACKPORT] stm32f7:Add Serial Tx DMA
- added gazebo VTOL tailsitter debug target
- improved jmavsim debug target and helper tasks to work properly (build and launch directly instead of through the jmavsim_run.sh script)
- improved output console handling for helper tasks (eg gazebo and jmavsim build and run, followed by cleanup after debug)
- added miniterm.py as a task
- nuttx in PX4/Firmware (22a005c9f422180f0c0c5f96c20b16deb66a674c): 2d7920055f
- nuttx current upstream: 398a59aaa4
- Changes: 2d7920055f...398a59aaa4
398a59aaa4 2019-11-29 David Sidrane - [BACKPORT] stm32f7:ethernet: Add some delays so that ifup() does not hog the CPU.
Change PID to 0x4b:
Holybro obtained their own PID and VID but APM did not follow
the PX4 convention of makeing the board_id (0x8b) match the PID)
Incorporated the Upstream Bootloader state sequencing checking change.
Change the usb cout to send all chars in 1 write.
The Addition of PX4_ARCH_DCACHE_LINESIZE is fundamentally wrong.
It asserst that an STM32F4 has a cache and it does not.
This should be #if defined() on CONFIG_ARMV7M_DCACHE
Main UAVCAN protocol handling and ESC updates run on the same thread/wq as
before. There are 2 WorkItems for separate scheduling of the 2, so that
ESC updates run in sync with actuator_control updates. UAVCAN is scheduled
at a fixed rate of 3ms (previously the poll timeout) and on each UAVCAN
bus event.
This leads to roughly the same behavior as before. CPU & RAM usage are
pretty much the same (tested on Pixhawk 4).
Testing done: Motors still work (with feedback), param changes and a
UAVCAN optical flow sensor.
getLockGuard relies on copy elision to work correctly, which the compiler
is not required to do (only with C++17).
If no copy elision happens, the mutex ends up being unlocked twice, and the
CS is executed with the mutex unlocked.
The patch also ensures that the same pattern cannot be used again.
