Method and system for organizing coherence directories in shared memory systems
    1.
    发明授权
    Method and system for organizing coherence directories in shared memory systems 失效
    在共享内存系统中组织一致目录的方法和系统

    公开(公告)号:US06792512B2

    公开(公告)日:2004-09-14

    申请号:US10214085

    申请日:2002-08-06

    IPC分类号: G06F1200

    CPC分类号: G06F12/0826

    摘要: A method and structure for a “dynamic CCR/sparse directory implementation,” includes maintaining state information of the main memory cached in the shared caches of the other compute nodes, organizing a cache directory so that the state information can be stored in a first area efficient CCR directory format, switching to a second sparse directory format if the entry is shared by more than one other compute node, and dynamically switching between formats so as to maximize the number of entries stored in the directory.

    摘要翻译: “动态CCR /稀疏目录实现”的方法和结构包括维护缓存在其他计算节点的共享高速缓存中的主存储器的状态信息,组织高速缓存目录,使得状态信息可以存储在第一区域 高效的CCR目录格式,如果条目由多个其他计算节点共享,则切换到第二稀疏目录格式,并且动态地在格式之间切换,以便最大化存储在目录中的条目数量。

    Real time emulation of coherence directories using global sparse directories
    2.
    发明授权
    Real time emulation of coherence directories using global sparse directories 失效
    使用全局稀疏目录的实时目录的实时仿真

    公开(公告)号:US06965972B2

    公开(公告)日:2005-11-15

    申请号:US10254745

    申请日:2002-09-25

    IPC分类号: G06F9/455 G06F12/00

    CPC分类号: G06F9/45537 G06F12/082

    摘要: A method and structure for an emulation system comprises of a plurality of field programmable gate arrays adapted to emulate nodes of a multi-node shared memory system, a plurality of cache directories, each connected to one of the arrays, and a plurality of global coherence directories, each connected to one of the arrays. Each of the global coherence directories maintain information on all memory lines remotely cached by each of the cache directories.

    摘要翻译: 用于仿真系统的方法和结构包括适于模拟多节点共享存储器系统的节点的多个现场可编程门阵列,多个高速缓存目录,每个高速缓存目录连接到阵列之一,以及多个全局相干性 目录,每个连接到一个阵列。 每个全局一致性目录都保留了每个缓存目录远程高速缓存的所有内存条的信息。

    List based prefetch
    4.
    发明授权
    List based prefetch 有权
    基于列表的预取

    公开(公告)号:US08806141B2

    公开(公告)日:2014-08-12

    申请号:US13593838

    申请日:2012-08-24

    IPC分类号: G06F12/00 G06F12/08

    CPC分类号: G06F12/0862

    摘要: A list prefetch engine improves a performance of a parallel computing system. The list prefetch engine receives a current cache miss address. The list prefetch engine evaluates whether the current cache miss address is valid. If the current cache miss address is valid, the list prefetch engine compares the current cache miss address and a list address. A list address represents an address in a list. A list describes an arbitrary sequence of prior cache miss addresses. The prefetch engine prefetches data according to the list, if there is a match between the current cache miss address and the list address.

    摘要翻译: 列表预取引擎提高并行计算系统的性能。 列表预取引擎接收当前高速缓存未命中地址。 列表预取引擎评估当前缓存未命中地址是否有效。 如果当前高速缓存未命中地址有效,则列表预取引擎将比较当前高速缓存未命中地址和列表地址。 列表地址表示列表中的地址。 列表描述了先前高速缓存未命中地址的任意序列。 如果当前缓存未命中地址和列表地址之间存在匹配,则预取引擎将根据列表预取数据。

    South bridge system and method
    5.
    发明授权
    South bridge system and method 失效
    南桥系统及方法

    公开(公告)号:US07624222B2

    公开(公告)日:2009-11-24

    申请号:US11539211

    申请日:2006-10-06

    IPC分类号: G06F13/00 G06F3/00 G06F13/36

    CPC分类号: G06F13/4031 G06F13/1657

    摘要: A system including a south bridge, a first processor connected to the south bridge, and a second processor connected to the south bridge. The system further includes at least one device connected to the south bridge, and a resource manager coupled to the south bridge that allocates use of the at least one device between the first processor and the second processor.

    摘要翻译: 一种包括南桥,连接到南桥的第一处理器和连接到南桥的第二处理器的系统。 该系统还包括连接到南桥的至少一个设备和耦合到南桥的资源管理器,其分配在第一处理器和第二处理器之间的至少一个设备的使用。

    SOUTH BRIDGE SYSTEM AND METHOD
    6.
    发明申请
    SOUTH BRIDGE SYSTEM AND METHOD 失效
    南桥系统与方法

    公开(公告)号:US20080086583A1

    公开(公告)日:2008-04-10

    申请号:US11539211

    申请日:2006-10-06

    IPC分类号: G06F13/36

    CPC分类号: G06F13/4031 G06F13/1657

    摘要: A system including a south bridge, a first processor connected to the south bridge, and a second processor connected to the south bridge. The system further includes at least one device connected to the south bridge, and a resource manager coupled to the south bridge that allocates use of the at least one device between the first processor and the second processor.

    摘要翻译: 一种包括南桥,连接到南桥的第一处理器和连接到南桥的第二处理器的系统。 该系统还包括连接到南桥的至少一个设备和耦合到南桥的资源管理器,其分配在第一处理器和第二处理器之间的至少一个设备的使用。

    TESTING AND OPERATING A MULTIPROCESSOR CHIP WITH PROCESSOR REDUNDANCY
    7.
    发明申请
    TESTING AND OPERATING A MULTIPROCESSOR CHIP WITH PROCESSOR REDUNDANCY 有权
    测试和操作具有处理器冗余的多处理器芯片

    公开(公告)号:US20130031418A1

    公开(公告)日:2013-01-31

    申请号:US13196459

    申请日:2011-08-02

    IPC分类号: G06F11/28

    CPC分类号: G06F11/2242 G06F11/202

    摘要: A system and method for improving the yield rate of a multiprocessor semiconductor chip that includes primary processor cores and one or more redundant processor cores. A first tester conducts a first test on one or more processor cores, and encodes results of the first test in an on-chip non-volatile memory. A second tester conducts a second test on the processor cores, and encodes results of the second test in an external non-volatile storage device. An override bit of a multiplexer is set if a processor core fails the second test. In response to the override bit, the multiplexer selects a physical-to-logical mapping of processor IDs according to one of: the encoded results in the memory device or the encoded results in the external storage device. On-chip logic configures the processor cores according to the selected physical-to-logical mapping.

    摘要翻译: 一种用于提高包括主处理器核心和一个或多个冗余处理器核心的多处理器半导体芯片的产率的系统和方法。 第一个测试人员对一个或多个处理器内核进行第一次测试,并在片上非易失性存储器中对第一次测试的结果进行编码。 第二个测试者对处理器核进行第二次测试,并将外部非易失性存储设备的第二次测试结果进行编码。 如果处理器核心故障第二次测试,则设置多路复用器的覆盖位。 响应于覆盖位,多路复用器根据以下之一选择处理器ID的物理到逻辑映射:存储器件中的编码结果或外部存储器件中的编码结果。 片上逻辑根据所选的物理到逻辑映射配置处理器内核。

    PROCESSOR RESUME UNIT
    9.
    发明申请
    PROCESSOR RESUME UNIT 审中-公开
    处理器修复单元

    公开(公告)号:US20110173420A1

    公开(公告)日:2011-07-14

    申请号:US12684852

    申请日:2010-01-08

    IPC分类号: G06F9/30

    摘要: A system for enhancing performance of a computer includes a computer system having a data storage device. The computer system includes a program stored in the data storage device and steps of the program are executed by a processor. An external unit is external to the processor for monitoring specified computer resources. The external unit is configured to detect a specified condition using the processor. The processor including one or more threads. The thread resumes an active state from a pause state using the external unit when the specified condition is detected by the external unit.

    摘要翻译: 一种用于增强计算机性能的系统包括具有数据存储装置的计算机系统。 计算机系统包括存储在数据存储装置中的程序,并且程序的步骤由处理器执行。 处理器外部的外部单元用于监视指定的计算机资源。 外部单元配置为使用处理器检测指定的条件。 处理器包括一个或多个线程。 当外部单元检测到指定的条件时,线程将使用外部单元从暂停状态恢复活动状态。

    PROGRAMMABLE STREAM PREFETCH WITH RESOURCE OPTIMIZATION
    10.
    发明申请
    PROGRAMMABLE STREAM PREFETCH WITH RESOURCE OPTIMIZATION 失效
    可编程流程资源优化

    公开(公告)号:US20110173397A1

    公开(公告)日:2011-07-14

    申请号:US12684693

    申请日:2010-01-08

    IPC分类号: G06F12/08 G06F12/00

    CPC分类号: G06F12/0862 G06F2212/6026

    摘要: A stream prefetch engine performs data retrieval in a parallel computing system. The engine receives a load request from at least one processor. The engine evaluates whether a first memory address requested in the load request is present and valid in a table. The engine checks whether there exists valid data corresponding to the first memory address in an array if the first memory address is present and valid in the table. The engine increments a prefetching depth of a first stream that the first memory address belongs to and fetching a cache line associated with the first memory address from the at least one cache memory device if there is not yet valid data corresponding to the first memory address in the array. The engine determines whether prefetching of additional data is needed for the first stream within its prefetching depth. The engine prefetches the additional data if the prefetching is needed.

    摘要翻译: 流预取引擎在并行计算系统中执行数据检索。 引擎从至少一个处理器接收加载请求。 引擎评估在加载请求中请求的第一个内存地址是否存在,并且在表中有效。 如果第一个存储器地址在表中存在且有效,引擎将检查是否存在与数组中的第一个存储器地址对应的有效数据。 如果还没有对应于第一存储器地址的有效数据,则引擎增加第一存储器地址所属的第一流的预取深度并从至少一个高速缓冲存储器设备获取与第一存储器地址相关联的高速缓存行 阵列。 该引擎确定在其预取深度内的第一个流是否需要预取附加数据。 如果需要预取,引擎将预取附加数据。