d0251581d3
samv7: fix include
2018-12-22 10:32:18 +01:00
1b685de70d
simulator: raise timeout
...
This is not suposed to ever trigger anyway.
2018-12-22 10:32:18 +01:00
8b2c47523f
simulator: set socket options after opening socket
2018-12-22 10:32:18 +01:00
e4af97b11b
simulator: fix printf format
2018-12-22 10:32:18 +01:00
564cae86b5
simulator: remove doubled tcgetattr call
2018-12-22 10:32:18 +01:00
8a5e0558e4
simulator: improve a couple of error messages
2018-12-22 10:32:18 +01:00
1165739a56
sensors: check other gyros on timeout
...
@bkueng reviewed this and found that it is dangerous to continue on
timeout because we don't update other gyro sensors in the timeout case.
2018-12-22 10:32:18 +01:00
8e4296d000
parameters: remove unneeded usleep
...
According to @bkueng this was to schedule other processes but can now be
removed.
2018-12-22 10:32:18 +01:00
4efe4b0d15
platforms: reorg after rebase on new board config
...
The new board config changed everything, so I needed to move the
lockstep_scheduler repo.
2018-12-22 10:32:18 +01:00
b4382d4a71
DriverFramework: updated submodule
2018-12-22 10:32:18 +01:00
185ddb2ab6
lockstep_scheduler: format to PX4 style
...
Consistency over taste :(.
2018-12-22 10:32:18 +01:00
fb9f72d029
lockstep_scheduler: modernize warnings
2018-12-22 10:32:18 +01:00
85486e0469
platforms: let's try clockid_t as int
...
Who knows what this really is.
2018-12-22 10:32:18 +01:00
f3aad419ac
DriverFramework: update submodule
...
Hopefully, we can fix the defines for macOS.
2018-12-22 10:32:18 +01:00
7b43e96896
platforms: copied files from lockstep-scheduler
...
The files are copied from:
https://github.com/julianoes/lockstep-scheduler/
2018-12-22 10:32:18 +01:00
3cbb2d8661
platforms: removed lockstep_scheduler submodule
2018-12-22 10:32:18 +01:00
f36f85010c
simulator: fix casting warning on Windows
2018-12-22 10:32:18 +01:00
393a665424
DriverFramework: update submodule
...
Fixes the unit test on macOS.
2018-12-22 10:32:18 +01:00
e046a6255f
DriverFramework: update submodule
...
This brings a couple more build fixes.
2018-12-22 10:32:18 +01:00
e41518a08b
platforms: clean up various build/linking issues
...
This makes sure lockstep is only built for SITL builds and fixes the
existing builds.
2018-12-22 10:32:18 +01:00
23264fe579
kinetis: add missing include
2018-12-22 10:32:18 +01:00
8cbe26c6f6
platforms: fix linking issues for NuttX build
2018-12-22 10:32:18 +01:00
a193e4117c
drv_hrt: update signature of ts_to_abstime
2018-12-22 10:32:18 +01:00
d75389b2e4
platforms: consolidate latency_buckets
...
These symbols were duplicated all over the place.
2018-12-22 10:32:18 +01:00
b13600ed1a
platforms: remove TODO
...
It turns out Mac OS X v10.11.6 (El Capitan) is still covered by security
updates at this time, so not end-of-lifed, so we might as well keep
these legacy things.
2018-12-22 10:32:18 +01:00
22b584b3b4
simulator: fix TCP connection init on macOS
...
When `connect()` fails, the state of the socket is unspecified and we
need to close the it and create it again.
It turns out retrying connect worked on Linux but it didn't on macOS.
2018-12-22 10:32:18 +01:00
b2f1a59164
DriverFramework: update submodule
...
This fixes the build on macOS.
2018-12-22 10:32:18 +01:00
6ed6941aba
DriverFramework: update submodule
...
This fixes the SITL speedup for macOS.
2018-12-22 10:32:18 +01:00
546f89aee7
lockstep_scheduler: update submodule
2018-12-22 10:32:18 +01:00
876d4ebd76
posix: remove appalling errno hack
2018-12-22 10:32:18 +01:00
66e7b4d578
posix: remove random lies
2018-12-22 10:32:18 +01:00
11b7411bf8
platforms: cleaned up defines
2018-12-22 10:32:18 +01:00
107606e757
simulator: set device_ids when publishing directly
...
Without setting the device_id, the preflight checks fail because the
calibration is not found.
2018-12-22 10:32:18 +01:00
843033d530
platforms: remove unused function
2018-12-22 10:32:18 +01:00
6d7fb232dd
simulator: added TCP, general cleanup, lockstep
...
This adds the option to connect over UDP or TCP for mavlink to the
SITL simulator.
Also, this includes a bunch of general cleanup and refactoring of the
simulator interface code. For instance sending of mavlink messages was
put into separate functions and unneccessary comments were removed.
Also, this now sets the timestamp sent by the SITL simulator in the
HIL_SENSOR message in order to enable lockstep.
2018-12-22 10:32:18 +01:00
cf39fdd37b
simulator: use usual type of timestamp
2018-12-22 10:32:18 +01:00
9b7857315c
sensors: handle timeout
2018-12-22 10:32:18 +01:00
22bd06e5d5
mc_pos_control: don't do orb_check on polled topic
...
When we poll on vehicle_local_position we shouldn't do yet another
orb_check but copy it first up.
2018-12-22 10:32:18 +01:00
0e04d3e9a9
commander: remove lying comment
2018-12-22 10:32:18 +01:00
e36e0b6ad0
posix: print warning before backtrace
2018-12-22 10:32:18 +01:00
4ef0d74ea3
platforms: remove old lockstep functionality
...
These functions won't be used any longer with the new implementation.
2018-12-22 10:32:18 +01:00
14642edfda
gps/devices: updated submodule
2018-12-22 10:32:18 +01:00
cc7cae4e7f
DriverFramework: updated submodule
...
This fixes a race/lockup happening in the DriverFramework scheduler.
2018-12-22 10:32:18 +01:00
a7e93bc2da
simulator: removed leftover (wrong) comment
2018-12-22 10:32:18 +01:00
09595a1de9
dataman: left FIXME comment
...
We shouldn't be using module here.
2018-12-22 10:32:18 +01:00
3f508c99fd
parameters: commented out strange usleep
2018-12-22 10:32:18 +01:00
da9029274c
pwm_out_sim: always publish actuators
...
This is required so that the SITL simulator receives a feedback from
PX4.
2018-12-22 10:32:18 +01:00
d1514cd667
POSIX: add defines system clock calls
2018-12-22 10:32:18 +01:00
98ae0186e9
px4_sem: use px4_sem on all POSIX system
...
This uses the "fake" px4_sem based on mutex and condition_variable on
all POSIX system, not just macOS and Cygwin. This means that we can
change px4_sem_timedwait under the hood and inject the simulated time.
2018-12-22 10:32:18 +01:00
3e6e1f5c2b
POSIX: use lockstep_scheduler to fake time
...
This integrates the lockstep_scheduler, so that the system time is set
by the mavlink HIL_SENSOR message.
This means that the speed factor is removed and the speed is entirely
given by the simulator.
2018-12-22 10:32:18 +01:00
Julian Oes