公开(公告)号：US20090063775A1

公开(公告)日：2009-03-05

申请号：US12293312

申请日：2006-09-20

申请人： Jeroen Molema ,  Wilko Westerhof ,  Bartele Henrik De Vries ,  Reinier Niels Lap ,  Olaf Martin De Jong ,  Bart-Jan Zwart ,  Johannes Rogier De Vrind

发明人： Jeroen Molema ,  Wilko Westerhof ,  Bartele Henrik De Vries ,  Reinier Niels Lap ,  Olaf Martin De Jong ,  Bart-Jan Zwart ,  Johannes Rogier De Vrind

IPC分类号： G06F12/08

CPC分类号： B60K15/04 ,  A61C17/36 ,  B26B19/40 ,  Y10T137/87957

摘要： The present invention provides a system and a method for a cache partitioning technique for application tasks based on the scheduling information in multiprocessors. Cache partitioning is performed dynamically based on the information of the pattern of task scheduling provided by the task scheduler (405). Execution behavior of the application tasks is obtained from the task scheduler (405) and partitions are allocated (415) to only a subset of application tasks, which are going to be executed in the upcoming clock cycles. The present invention will improve the cache utilization by avoiding unnecessary reservation of the cache partitions for the executing application tasks during the entire duration of their execution and hence an effective utilization of the cache is achieved.

摘要翻译： 本发明提供一种基于多处理器中的调度信息的用于应用任务的高速缓存分配技术的系统和方法。 基于由任务调度器（405）提供的任务调度的模式的信息来动态地执行高速缓存分区。 从任务调度器（405）获得应用任务的执行行为，并将分配（415）分配给仅在即将到来的时钟周期中执行的应用任务的子集。 本发明将通过在执行应用任务的整个持续时间期间避免对执行的应用任务的高速缓存分区的不必要的预留来提高高速缓存利用率，从而实现高速缓存的有效利用。