This is needed in case the bootloader configured the MPU, which could interfere with the app as the bootloader memory map / configuration is different.
When the last WorkItem is deleted, it is removed from a work queue and the
queue is being stopped. But, the queue itself might get deleted in the middle,
in a higher priority thread than where the WorkItem deletion was performed from
If the WorkQueue::Detach accesses the member variables after this, there is memory
corruption
This happens in particular when launching i2c or spi devices in
I2CSPIDriverBase::module_start:
- The "initializer" is deleted when the instance is not found and the iterator
while loop continues.
- The workqueue is deleted in the middle of "initializer" deletion when the
WorkQueueRunner returns.
This prevents deletion of the WorkQueue before the Detach has been finished,
in the specific case that the ::Detach triggers the deletion
Signed-off-by: Jukka Laitinen <jukkax@ssrc.tii.ae>
- fix yaw spin on ground after large mag field change
- change covariance prediction for better stability
- fix yaw alignment in air for mag-less mode
- improve tilt alignment for non-static applications
- add momentum drag model
This reverts the behavior for offboard velocity setpoint.
Back in v1.11, the velocity input in NED_BODY was assumed to be in the
world frame but rotated by yaw to the vehicle frame. With the current
state the frame is completely fixed to the body. While this might be
technically correct, it doesn't seem much intuitive for multicopters
and breaks the MAVSDK offboard velocity API.
So as an example, with a velocity setpoint of 5 m/s forward, the drone
would in
- v1.11: fly forward with 5 m/s
- v1.12: start to fly forward by pitching down but then descend rapidly
as the forward velocity would translate to a setpoint in Z into the
ground as it is pitched down.
This commit restores the behavior to what we had previously.
