Destroy a thread immediately
#include <sys/neutrino.h> int ThreadDestroy( int tid, int priority, void* status ); int ThreadDestroy_r( int tid, int priority, void* status );
libc
Use the -l c option to qcc to link against this library. This library is usually included automatically.
These kernel calls terminate the thread specified by tid. If tid is 0, the calling thread is assumed. If tid is -1, all of the threads in the process are destroyed. When multiple threads are destroyed, the destruction is scheduled one thread at a time at the priority specified by the priority argument. If priority is -1, then the priority of the calling thread is used.
The ThreadDestroy() and ThreadDestroy_r() functions are identical, except in the way they indicate errors. See the Returns section for details.
If a terminated thread isn't detached, it makes the value specified by the status argument available to any successful join on it (through pthread_join()). Until another thread retrieves this value, the thread ID tid isn't reused, and a small kernel resource (a thread object) is held in the system. If the thread is detached, then status is ignored, and all thread resources are immediately released.
(QNX Neutrino 7.0.1 or later) Any event that you registered for the thread with the _NTO_TCTL_ADD_EXIT_EVENT ThreadCtl() command is delivered when the thread is terminated.
When the last thread in a process is destroyed, the process terminates and all thread resources are released, even for threads that haven't been detached or joined.
Blocking states
If these calls return, they don't block.
If the calling thread is destroyed, ThreadDestroy() and ThreadDestroy_r() don't return.
The only difference between these functions is the way they indicate errors:
Safety: | |
---|---|
Cancellation point | No |
Interrupt handler | No |
Signal handler | Yes |
Thread | Yes |