During JVM shutdown, some JNI calls may not return, so it's not possible to
reliably perform a join() during static variable destruction (which occurs
as the JVM unloads the JNI module).
Also, due to static variable destruction, it's not safe to use any members
of a static class instance from a separate thread of execution.
SafeThread is a templated thread class and a related owner class that's
designed for safe operation and shutdown of threads in the presence of
callbacks that may not return. It also passes ownership of variables from
the static instance to the thread, so the thread can safely operate until
it exits (the last operation of the thread being to destroy its instance).
Notifiers, RpcServer, and Logger now use SafeThread to ensure race-free
destruction in both C++ and Java.
All Java callback threads are now marked as Java daemon threads so they
don't keep the JVM running after main() terminates.
All Java callback threads are now named so their purpose is more easily
identified in a debugger.
Add SetRpcServerOnStart and SetRpcServerOnExit (similar to Listener).
To publish the simulation zip, run ./gradlew publish -PmakeSim
Targeting Ubuntu 14.04 and 15.10 for now, with 14.04 being the
currently best supported.
Two scripts have been drafted for installing, for 14.04 and 15.10
It currently publishes to ~/releases/maven/development/simulation
There is a known bug that gz_msgs for 15.10 must be built using
protobuf 2.6, which is not the default on 14.04.
Change-Id: I6cccd601671553d30fd05bbbc79c2b7dc1efbf1d
fixed a few incorrect statements
added structure and organization to main README to help people find
stuff when they're first starting
Change-Id: I6148f2e9bc63cbeaf5f2e40566bf9d274ed36aaa
including grabbing ntcore sources, for both
Java and C++. This will need changes made in
the wpilib promotion tasks to copy the generatd
documentation to the correct places.
Change-Id: I64590b5eda001da2cc8ae498b2b1c0fd298da284
This prevents issues building wpilib if you have ntcore but haven't
built desktop
If you want to build sim, use -PmakeSim
This will require the desktop build of ntcore, which can be done with:
./gradlew build publish -PmakeDesktop
Change-Id: I20f379267a226dbdfcff8fe6223670257725965a
The previous use of a timeout resulting in thread detach instead of thread
join resulted in a race condition on Mac between destruction and thread
closeout. This commit removes the detach functionality and uses dup2() to
on Linux/Mac and connecting to itself on Windows to try to ensure accept()
exits.
Each call to AttachCurrentThread results in a new Java thread object being
created. This is inefficient and also causes debugging issues with Eclipse
due to constant creation and removal of threads. Now AttachCurrentThread is
only called once for (all) listeners and once for logging (if used).
DriverStation was being added to the singletons and also as a const
member variable. This made it so that on program exit, it was
getting double freed.
Change-Id: I87f8260615dc31d57ce7c7204c1dfde22973ad51
Reverted to old driverstation and joystick code because we're not ready
for windows drive station yet
updated paths to reflect new wpilib organization
fixed name of gazebo topic (if you want /gazebo/frc/time use ~/time)
included network tables in wpilibJavaSim
Added ds script, and improved frcsim script
always start gazebo with verbose
Change-Id: I3c54b7000019a5985079a88200896a8069e69b86
Previously this would always return false due to how explicit bool is
evaluated in a return context.
Also add a test for this function.
Reported by: jcreigh
the task allcsim will build everything
requires ntcore desktop to be built
also fixed Driverstation to match non-sim C++ API
Conflicts:
wpilibc/simulation/CMakeLists.txt
Change-Id: Id38141a5b48ed7fe064c7e6c8d2f618481b7e298
- CANJaguar also touched up to report it can't do the new control mode (just like with follower).
- New third optional param for talon c'tor to speed up enable control frame.
- The pulse width routines have been moved to where the script generator puts them. No actual changes there but should help Peter integrate the latest code generator.
Last feature additions for TalonSRX HAL for FRC2015-FRC-2016 season.
-HAL driver uses control_5 frame if firmware supports it. This allows teams to see/confirm control settings taking effect before enabling the robot.
For example selecting the sensor type and going to web-dash to check sensor values now works without having to enable the robot.
-Motion profile HAL routines added. Tested on Single-Speed Double reduction (with slave Talon too).
-Start moving ctre frame defs into a new common header (better then shoving a bunch of struct defs at top of module).
-New child class in CANTalonSRX for buffering motion profile points. Not sure it would be best to leave it as is or make another module. It's trivial now so I thought that was acceptable, (in future it will likely possess compression strategies => no longer trivial).
Change-Id: I803680c1a6669ca3f5157d7875942def6f75b540
Also fix DigitalGlitchFilter usage reporting for C++ (only report on
construction rather than on each call) and add for Java.
Change-Id: I73758b16d81c40442c4acec43e0aa6804b2ba250
The previous commit missed the MindsensorsSD540 value.
Remove wpilibc/Athena/include/NetworkCommunication/UsageReporting.h copy; it
was not actually used (all references except the one in DriverStation.cpp
had been commented out), and has been superseded by the one in HAL.hpp.
Change-Id: I51e2eafa85ee5e40050d702aa0bf69171e887a60
