Home | History | Annotate | Download | only in synchronization

Lines Matching full:threads

29     // There are threads of execution still in this->TimedWait() and yet the
32 // was not willing to wait for all the threads to terminate. Such hasty
38 Broadcast(); // Make sure all waiting threads have been signaled.
39 Sleep(10); // Give threads a chance to grab internal_lock_.
40 // All contained threads should be blocked on user_lock_ by now :-).
73 // Broadcast() is guaranteed to signal all threads that were waiting (i.e., had
91 // Signal() will select one of the waiting threads, and signal it (signal its
110 // objects as there are threads simultaneously using this instance's Wait()
130 // recycles it for use in future Wait() calls for this or other threads.
157 // threads to do the extraction. The iterator that identifies this extractable
288 (especially with regard to Broadcast() and the impact on threads that
290 original waiting threads have been signaled).
292 2) Why can't you use a single wait_event for all threads that call
294 event were used, then numerous threads calling Wait() could release
298 waiting(?) threads. Some number of SetEvent() calls *could* be made,
302 awaken "other" threads that hadn't even started waiting yet (oops).
396 of the internal lock. As a result, other threads, such as the thread
411 recursion counter (in a Lock) must be protected, as many threads can
421 threads have references to these otherwise leaked events. They are passed as