Future scheduling by direct representation of possible dependencies
    1.
    发明授权
    Future scheduling by direct representation of possible dependencies 有权
    通过直接表示可能依赖的未来调度

    公开(公告)号:US08225326B2

    公开(公告)日:2012-07-17

    申请号:US12049914

    申请日:2008-03-17

    IPC分类号: G06F9/46 G06F9/52

    CPC分类号: G06F9/524

    摘要: A method for evaluating objects in a data structure is provided. The method includes assigning one or more objects to one or more nodes in a data structure having at least a root node, in which the objects are assigned to the nodes in accordance with a first order to maintain pre-existing dependencies between the objects and to allow the objects to be evaluated in a serial manner to avoid deadlock when concurrently executing threads to evaluate the objects, and selecting a first object for evaluation, in response to determining that the current object is unevaluated.

    摘要翻译: 提供了一种用于评估数据结构中的对象的方法。 该方法包括将一个或多个对象分配给具有至少根节点的数据结构中的一个或多个节点,其中根据第一顺序将对象分配给节点,以保持对象之间的预先存在的依赖关系,以及 允许以串行方式评估对象,以在同时执行线程以评估对象时选择第一对象以进行评估,以响应于确定当前对象未被评估而避免死锁。

    FUTURE SCHEDULING BY DIRECT REPRESENTATION OF POSSIBLE DEPENDENCIES
    2.
    发明申请
    FUTURE SCHEDULING BY DIRECT REPRESENTATION OF POSSIBLE DEPENDENCIES 有权
    未来可能直接表示可能的依赖性

    公开(公告)号:US20090235273A1

    公开(公告)日:2009-09-17

    申请号:US12049914

    申请日:2008-03-17

    IPC分类号: G06F9/46

    CPC分类号: G06F9/524

    摘要: A method for evaluating objects in a data structure is provided. The method comprises assigning one or more objects to one or more nodes in a data structure having at least a root node, wherein the objects are assigned to the nodes in accordance with a first order to maintain pre-existing dependencies between said objects and to allow the objects to be evaluated in a serial manner to avoid deadlock when concurrently executing threads to evaluate the objects; and selecting a first object for evaluation, in response to determining that the current object is unevaluated.

    摘要翻译: 提供了一种用于评估数据结构中的对象的方法。 该方法包括将一个或多个对象分配给具有至少根节点的数据结构中的一个或多个节点,其中根据第一顺序将对象分配给节点,以维持所述对象之间的预先存在的依赖关系并允许 要以串行方式评估的对象,以在同时执行线程来评估对象时避免死锁; 以及响应于确定当前对象未被评估,选择用于评估的第一对象。

    Concurrent management of adaptive programs
    3.
    发明授权
    Concurrent management of adaptive programs 有权
    并发管理适应性方案

    公开(公告)号:US08627301B2

    公开(公告)日:2014-01-07

    申请号:US11750441

    申请日:2007-05-18

    IPC分类号: G06F9/44 G06F9/45

    CPC分类号: G06F9/5066

    摘要: A method for concurrent management of adaptive programs is disclosed wherein changes in a set of modifiable references are initially identified. A list of uses of the changed references is next computed using records made in structures of the references. The list is next inserted into an elimination queue. Comparison is next made of each of the uses to the other uses to determine independence or dependence thereon. Determined dependent uses are eliminated and the preceding steps are repeated for all determined independent uses until all dependencies have been eliminated.

    摘要翻译: 公开了一种用于并行管理自适应程序的方法,其中初始地识别一组可修改参考文献中的改变。 接下来使用在引用的结构中作出的记录来计算更改的引用的使用列表。 列表接下来插入消除队列。 下一步将对其他用途的各种用途进行比较,以确定其独立性或依赖性。 消除确定的依赖用途,并且对于所有确定的独立使用重复前述步骤,直到所有依赖关系被消除。

    Method and apparatus for supporting scalable coherence on many-core products through restricted exposure
    4.
    发明授权
    Method and apparatus for supporting scalable coherence on many-core products through restricted exposure 有权
    通过限制曝光来支持多核产品上的可扩展一致性的方法和装置

    公开(公告)号:US08312225B2

    公开(公告)日:2012-11-13

    申请号:US13156777

    申请日:2011-06-09

    IPC分类号: G06F12/00

    CPC分类号: G06F12/0817 G06F12/0822

    摘要: In one embodiment, a multi-core processor having cores each associated with a cache memory, can operate such that when a first core is to access data owned by a second core present in a cache line associated with the second core, responsive to a request from the first core, cache coherency state information associated with the cache line is not updated. A coherence engine associated with the processor may receive the data access request and determine that the data is of a memory page owned by the first core and convert the data access request to a non-cache coherent request. Other embodiments are described and claimed.

    摘要翻译: 在一个实施例中,具有每个与高速缓存存储器相关联的核心的多核处理器可以操作,使得当第一核心要访问存在于与第二核心相关联的高速缓存行中的第二核心拥有的数据时,响应于请求 从第一核心,与高速缓存行相关联的高速缓存一致性状态信息不被更新。 与处理器相关联的相干引擎可以接收数据访问请求,并确定数据是由第一核心拥有的存储器页面,并将数据访问请求转换为非高速缓存一致性请求。 描述和要求保护其他实施例。

    Method And Apparatus For Supporting Scalable Coherence On Many-Core Products Through Restricted Exposure
    5.
    发明申请
    Method And Apparatus For Supporting Scalable Coherence On Many-Core Products Through Restricted Exposure 有权
    通过限制性曝光支持多核产品的可扩展一致性的方法和装置

    公开(公告)号:US20110238926A1

    公开(公告)日:2011-09-29

    申请号:US13156777

    申请日:2011-06-09

    IPC分类号: G06F12/08

    CPC分类号: G06F12/0817 G06F12/0822

    摘要: In one embodiment, a multi-core processor having cores each associated with a cache memory, can operate such that when a first core is to access data owned by a second core present in a cache line associated with the second core, responsive to a request from the first core, cache coherency state information associated with the cache line is not updated. A coherence engine associated with the processor may receive the data access request and determine that the data is of a memory page owned by the first core and convert the data access request to a non-cache coherent request. Other embodiments are described and claimed.

    摘要翻译: 在一个实施例中,具有每个与高速缓存存储器相关联的核心的多核处理器可以操作,使得当第一核心要访问存在于与第二核心相关联的高速缓存行中的第二核心拥有的数据时,响应于请求 从第一核心,与高速缓存行相关联的高速缓存一致性状态信息不被更新。 与处理器相关联的相干引擎可以接收数据访问请求,并确定数据是由第一核心拥有的存储器页面,并将数据访问请求转换为非高速缓存一致性请求。 描述和要求保护其他实施例。

    Method and apparatus for supporting scalable coherence on many-core products through restricted exposure
    6.
    发明授权
    Method and apparatus for supporting scalable coherence on many-core products through restricted exposure 有权
    通过限制曝光来支持多核产品上的可扩展一致性的方法和装置

    公开(公告)号:US07984244B2

    公开(公告)日:2011-07-19

    申请号:US12005785

    申请日:2007-12-28

    IPC分类号: G06F12/00

    CPC分类号: G06F12/0817 G06F12/0822

    摘要: In one embodiment, a multi-core processor having cores each associated with a cache memory, can operate such that when a first core is to access data owned by a second core present in a cache line associated with the second core, responsive to a request from the first core, cache coherency state information associated with the cache line is not updated. A coherence engine associated with the processor may receive the data access request and determine that the data is of a memory page owned by the first core and convert the data access request to a non-cache coherent request. Other embodiments are described and claimed.

    摘要翻译: 在一个实施例中,具有每个与高速缓存存储器相关联的核心的多核处理器可以操作,使得当第一核心要访问存在于与第二核心相关联的高速缓存行中的第二核心拥有的数据时,响应于请求 从第一核心,与高速缓存行相关联的高速缓存一致性状态信息不被更新。 与处理器相关联的相干引擎可以接收数据访问请求,并确定数据是由第一核心拥有的存储器页面,并将数据访问请求转换为非高速缓存一致性请求。 描述和要求保护其他实施例。

    Method and apparatus for supporting scalable coherence on many-core products through restricted exposure
    7.
    发明申请
    Method and apparatus for supporting scalable coherence on many-core products through restricted exposure 有权
    通过限制曝光来支持多核产品上的可扩展一致性的方法和装置

    公开(公告)号:US20090172294A1

    公开(公告)日:2009-07-02

    申请号:US12005785

    申请日:2007-12-28

    IPC分类号: G06F12/00

    CPC分类号: G06F12/0817 G06F12/0822

    摘要: In one embodiment, a multi-core processor having cores each associated with a cache memory, can operate such that when a first core is to access data owned by a second core present in a cache line associated with the second core, responsive to a request from the first core, cache coherency state information associated with the cache line is not updated. A coherence engine associated with the processor may receive the data access request and determine that the data is of a memory page owned by the first core and convert the data access request to a non-cache coherent request. Other embodiments are described and claimed.

    摘要翻译: 在一个实施例中,具有每个与高速缓存存储器相关联的核心的多核处理器可以操作,使得当第一核心要访问存在于与第二核心相关联的高速缓存行中的第二核心拥有的数据时,响应于请求 从第一核心,与高速缓存行相关联的高速缓存一致性状态信息不被更新。 与处理器相关联的相干引擎可以接收数据访问请求,并确定数据是由第一核心拥有的存储器页面,并将数据访问请求转换为非高速缓存一致性请求。 描述和要求保护其他实施例。

    Concurrent Management of Adaptive Programs
    8.
    发明申请
    Concurrent Management of Adaptive Programs 有权
    自适应程序并发管理

    公开(公告)号:US20080288950A1

    公开(公告)日:2008-11-20

    申请号:US11750441

    申请日:2007-05-18

    IPC分类号: G06F9/50 G06F9/46

    CPC分类号: G06F9/5066

    摘要: A method for concurrent management of adaptive programs is disclosed wherein changes in a set of modifiable references are initially identified. A list of uses of the changed references is next computed using records made in structures of the references. The list is next inserted into an elimination queue. Comparison is next made of each of the uses to the other uses to determine independence or dependence thereon. Determined dependent uses are eliminated and the preceding steps are repeated for all determined independent uses until all dependencies have been eliminated.

    摘要翻译: 公开了一种用于并行管理自适应程序的方法,其中初始地识别一组可修改参考文献中的改变。 接下来使用在引用的结构中作出的记录来计算更改的引用的使用列表。 列表接下来插入消除队列。 下一步将对其他用途的各种用途进行比较,以确定其独立性或依赖性。 消除确定的依赖用途,并且对于所有确定的独立使用重复前述步骤,直到所有依赖关系被消除。