Changes

Lower priority values give the thread higher priority. For userland apps, priorities between 0x18 and 0x3F are allowed. The priority of the app's main thread seems to be 0x30.

Lower priority values give the thread higher priority. For userland apps, priorities between 0x18 and 0x3F are allowed. The priority of the app's main thread seems to be 0x30.

The [[Glossary#appcore|appcore]] thread scheduler, in typical real-time operating system fashion, implements a simple preemptive algorithm based around multiple thread priority levels. This algorithm guarantees that the currently executing thread is always the highest priority runnable thread (also known as SCHED_FIFO). In other words, a thread will be interrupted (preempted) if and only if a higher priority thread is woken up, by means of an event (i.e. svcSendSyncRequest) or similar. Contrary to typical desktop operating systems, no timeslice-based scheduling is performed, which means that if a thread uses up all available CPU time (for example if it enters an endless loop), all other threads with equal or lower priority that run on the same CPU core won't get a chance to run. Address arbiters can be used to implement process-local synchronization primitives.

The [[Glossary#appcore|appcore]] thread scheduler, in typical real-time operating system fashion, implements a simple preemptive algorithm based around multiple thread priority levels. This algorithm guarantees that the currently executing thread is always the highest priority runnable thread (also known as SCHED_FIFO). In other words, a thread will be interrupted (preempted) if and only if a higher priority thread is woken up, by means of an event (i.e. svcSendSyncRequest) or similar. Contrary to typical desktop operating systems, no timeslice-based scheduling is performed, which means that if a thread uses up all available CPU time (for example if it enters an endless loop), all other threads with equal or lower priority that run on the same CPU core won't get a chance to run. Yielding to other threads is otherwise done by means of synchronization primitives (thread sleep, mutex, address arbiter, etc.). Address arbiters can be used to implement process-local synchronization primitives.

0xFFFF8000 is a handle alias for the currently active thread.

0xFFFF8000 is a handle alias for the currently active thread.