公开(公告)号：US06687818B1

公开(公告)日：2004-02-03

申请号：US09362388

申请日：1999-07-28

申请人： Penny L. Svenkeson ,  Philip J. Erickson ,  Peter C. Wilson ,  John A. Miller ,  Doug E. Morrissey

发明人： Penny L. Svenkeson ,  Philip J. Erickson ,  Peter C. Wilson ,  John A. Miller ,  Doug E. Morrissey

IPC分类号： G06F15177

CPC分类号： G06F15/177 ,  G06F9/4405

摘要： A method and apparatus is disclosed for initiating execution of selected application processors in a clustered multiprocessor system without disrupting the addressing mode of the processors. In one embodiment, this is accomplished by initializing the processors in a conventional way, setting the addressing mode, and then leaving each of the processors in an active mode spinning on a designated Startup Address Location. To start an application processor, the OS simply write a target processor ID and a desired start-up address into the Startup Address Location. The application processor with the matching processor ID reads the Startup Address Location, jumps to the address specified in the Start-Up Address code section, and begins execution at that location. In another embodiment, a processor is provided that does not switch addressing modes when switched from a halt state.

摘要翻译： 公开了用于在群集多处理器系统中启动所选择的应用处理器的执行而不中断处理器的寻址模式的方法和装置。 在一个实施例中，这是通过以常规方式初始化处理器，设置寻址模式，然后使处于主动模式的每个处理器在指定的启动地址位置上旋转来实现的。 要启动应用处理器，操作系统只需将目标处理器ID和所需的启动地址写入启动地址位置。 具有匹配处理器ID的应用处理器读取启动地址位置，跳转到启动地址代码部分中指定的地址，并在该位置开始执行。 在另一个实施例中，提供一种处理器，当从停止状态切换时不处理寻址模式。

公开(公告)号：US06665761B1

公开(公告)日：2003-12-16

申请号：US09362389

申请日：1999-07-28

申请人： Penny L. Svenkeson ,  Robert J. Gulick ,  Doug E. Morrissey

发明人： Penny L. Svenkeson ,  Robert J. Gulick ,  Doug E. Morrissey

IPC分类号： G06F1332

CPC分类号： G06F13/24

摘要： A method and apparatus for increasing the routing bandwidth of interrupts between cluster manager devices in a clustered multiprocessor system is disclosed. This is accomplished by providing special cluster manager devices that can convert “N” serial messages received from a local APIC to “M” parallel messages, wherein M is less than N. The special cluster manager device then transfers the “M” parallel messages to a receiving cluster manager device. The receiving cluster manager device converts the “M” parallel messages into the original “N” serial messages, and sends the “N” serial messages to the appropriate local APIC within the receiving cluster.

摘要翻译： 公开了一种用于增加集群多处理器系统中的集群管理器设备之间的中断的路由带宽的方法和装置。 这是通过提供可以将从本地APIC接收的“N”个串行消息转换为“M”并行消息的特殊集群管理器设备来实现的，其中M小于N.然后，专用集群管理器设备将“M”个并行消息传送到 接收集群管理器设备。 接收集群管理器将“M”个并行消息转换成原来的“N”个串行消息，并将“N”个串行消息发送到接收集群内的适当本地APIC。