META PREDICTOR RESTORATION UPON DETECTING MISPREDICTION
    2.
    发明申请
    META PREDICTOR RESTORATION UPON DETECTING MISPREDICTION 有权
    META预测恢复检测故障

    公开(公告)号:US20130036297A1

    公开(公告)日:2013-02-07

    申请号:US13647153

    申请日:2012-10-08

    IPC分类号: G06F9/38

    CPC分类号: G06F9/3861 G06F9/3848

    摘要: Methods and apparatus for restoring a meta predictor system upon detecting a branch or binary misprediction, are disclosed. An example apparatus may include a base misprediction history register to store a set of misprediction history values each indicating whether a previous branch prediction taken by a previous branch instruction was predicted correctly or incorrectly. The apparatus may comprise a meta predictor to detect a branch misprediction of a current branch prediction based at least in part on an output of the base misprediction history register. The meta predictor may restore the base misprediction history register based on the detecting of the branch misprediction. Additional apparatus, systems, and methods are disclosed.

    摘要翻译: 公开了在检测到分支或二进制错误预测时恢复元预测系统的方法和装置。 一个示例性装置可以包括一个基本错误预测历史寄存器,用于存储一组错误预测历史值,每个错误预测历史值指示是否正确或不正确地预测了由先前的分支指令取得的先前分支预测。 该装置可以包括元预测器,用于至少部分地基于基本错误预测历史寄存器的输出来检测当前分支预测的分支错误预测。 元预测器可以基于检测到分支错误预测来恢复基本的错误预测历史寄存器。 公开了附加装置,系统和方法。

    Meta predictor restoration upon detecting misprediction

    公开(公告)号:US08572358B2

    公开(公告)日:2013-10-29

    申请号:US13647153

    申请日:2012-10-08

    IPC分类号: G06F9/00 G06F9/44

    CPC分类号: G06F9/3861 G06F9/3848

    摘要: Methods and apparatus for restoring a meta predictor system upon detecting a branch or binary misprediction, are disclosed. An example apparatus may include a base misprediction history register to store a set of misprediction history values each indicating whether a previous branch prediction taken by a previous branch instruction was predicted correctly or incorrectly. The apparatus may comprise a meta predictor to detect a branch misprediction of a current branch prediction based at least in part on an output of the base misprediction history register. The meta predictor may restore the base misprediction history register based on the detecting of the branch misprediction. Additional apparatus, systems, and methods are disclosed.

    Efficient method and apparatus for employing a micro-op cache in a processor
    4.
    发明授权
    Efficient method and apparatus for employing a micro-op cache in a processor 有权
    在处理器中采用微操作高速缓存的高效方法和装置

    公开(公告)号:US08103831B2

    公开(公告)日:2012-01-24

    申请号:US12060239

    申请日:2008-03-31

    IPC分类号: G06F12/00

    摘要: Methods and apparatus for using micro-op caches in processors are disclosed. A tag match for an instruction pointer retrieves a set of micro-op cache line access tuples having matching tags. The set is stored in a match queue. Line access tuples from the match queue are used to access cache lines in a micro-op cache data array to supply a micro-op queue. On a micro-op cache miss, a macroinstruction translation engine (MITE) decodes macroinstructions to supply the micro-op queue. Instruction pointers are stored in a miss queue for fetching macroinstructions from the MITE. The MITE may be disabled to conserve power when the miss queue is empty-likewise for the micro-op cache data array when the match queue is empty. Synchronization flags in the last micro-op from the micro-op cache on a subsequent micro-op cache miss indicate where micro-ops from the MITE merge with micro-ops from the micro-op cache.

    摘要翻译: 公开了在处理器中使用微操作高速缓存的方法和装置。 指令指针的标签匹配检索一组具有匹配标签的微操作高速缓存行访问元组。 该集合存储在匹配队列中。 来自匹配队列的线路访问元组用于访问微操作高速缓存数据阵列中的高速缓存行以提供微操作队列。 在微操作缓存未命中时,宏指令转换引擎(MITE)解码宏指令以提供微操作队列。 指令指针存储在从MITE获取宏指令的小队列中。 当缺席队列为空时,MITE可能会被禁用以节省电力,而当匹配队列为空时,也可以为微操作高速缓存数据阵列。 随后微操作高速缓存未命中的微操作高速缓存中的最后一个微操作中的同步标志指示来自MITE的微操作与微操作高速缓存的微操作合并。

    EFFICIENT METHOD AND APPARATUS FOR EMPLOYING A MICRO-OP CACHE IN A PROCESSOR
    5.
    发明申请
    EFFICIENT METHOD AND APPARATUS FOR EMPLOYING A MICRO-OP CACHE IN A PROCESSOR 有权
    在处理器中使用微型高速缓存的有效方法和设备

    公开(公告)号:US20090249036A1

    公开(公告)日:2009-10-01

    申请号:US12060239

    申请日:2008-03-31

    IPC分类号: G06F9/30

    摘要: Methods and apparatus for using micro-op caches in processors are disclosed. A tag match for an instruction pointer retrieves a set of micro-op cache line access tuples having matching tags. The set is stored in a match queue. Line access tuples from the match queue are used to access cache lines in a micro-op cache data array to supply a micro-op queue. On a micro-op cache miss, a macroinstruction translation engine (MITE) decodes macroinstructions to supply the micro-op queue. Instruction pointers are stored in a miss queue for fetching macroinstructions from the MITE. The MITE may be disabled to conserve power when the miss queue is empty-likewise for the micro-op cache data array when the match queue is empty. Synchronization flags in the last micro-op from the micro-op cache on a subsequent micro-op cache miss indicate where micro-ops from the MITE merge with micro-ops from the micro-op cache.

    摘要翻译: 公开了在处理器中使用微操作高速缓存的方法和装置。 指令指针的标签匹配检索一组具有匹配标签的微操作高速缓存行访问元组。 该集合存储在匹配队列中。 来自匹配队列的线路访问元组用于访问微操作高速缓存数据阵列中的高速缓存行以提供微操作队列。 在微操作缓存未命中时,宏指令转换引擎(MITE)解码宏指令以提供微操作队列。 指令指针存储在从MITE获取宏指令的小队列中。 当缺席队列为空时,MITE可能会被禁用以节省电力,而当匹配队列为空时,也可以为微操作高速缓存数据阵列。 随后微操作高速缓存未命中的微操作高速缓存中的最后一个微操作中的同步标志指示来自MITE的微操作与微操作高速缓存的微操作合并。

    Method and apparatus for partitioned pipelined execution of multiple execution threads
    6.
    发明授权
    Method and apparatus for partitioned pipelined execution of multiple execution threads 有权
    分割流水线执行多个执行线程的方法和装置

    公开(公告)号:US09146745B2

    公开(公告)日:2015-09-29

    申请号:US11479245

    申请日:2006-06-29

    IPC分类号: G06F9/38

    摘要: Methods and apparatus for partitioning a microprocessor pipeline to support pipelined branch prediction and instruction fetching of multiple execution threads. A thread selection stage selects a thread from a plurality of execution threads. In one embodiment, storage in a branch prediction output queue is pre-allocated to a portion of the thread in one branch prediction stage in order to prevent stalling of subsequent stages in the branch prediction pipeline. In another embodiment, an instruction fetch stage fetches instructions at a fetch address corresponding to a portion of the selected thread. Another instruction fetch stage stores the instruction data in an instruction fetch output queue if enough storage is available. Otherwise, instruction fetch stages corresponding to the selected thread are invalidated and refetched to avoid stalling preceding stages in the instruction fetch pipeline, which may be fetching instructions of another thread.

    摘要翻译: 用于分割微处理器流水线以支持流水线分支预测和多个执行线程的指令获取的方法和装置。 线程选择阶段从多个执行线程中选择线程。 在一个实施例中,分支预测输出队列中的存储被预分配给一个分支预测阶段中的线程的一部分,以便防止分支预测流水线中后续阶段的停顿。 在另一个实施例中,指令提取阶段在与所选线程的一部分相对应的获取地址处获取指令。 如果有足够的存储可用,另一个指令获取阶段将指令数据存储在指令提取输出队列中。 否则,与所选线程相对应的指令获取阶段无效并被重新设计,以避免在指令提取流水线中停止前进阶段,这可能是获取另一线程的指令。

    Vector completion mask handling
    8.
    发明授权
    Vector completion mask handling 有权
    矢量完成掩码处理

    公开(公告)号:US08510536B2

    公开(公告)日:2013-08-13

    申请号:US13535685

    申请日:2012-06-28

    IPC分类号: G06F15/00 G06F15/76

    摘要: Techniques for vector completion mask (VCM) handling are provided. A data structure includes a mask field for each operand of a particular operation. A processor attempts to execute the operation with multiple operands, which are identified in the data structure by the mask fields. If operands are successfully retrieved for execution with the operation, then the corresponding mask field within the data structure is cleared. The processor can reset if any field remains set within the data structure and can re-process the operation with operands that were not previously handled with the operation.

    摘要翻译: 提供矢量完成掩码(VCM)处理技术。 数据结构包括用于特定操作的每个操作数的掩码字段。 处理器尝试通过掩码字段在数据结构中标识的多个操作数来执行操作。 如果成功检索操作数以执行操作,则数据结构中的相应掩码字段将被清除。 如果任何字段在数据结构中保持设置,并且可以使用以前未被操作的操作数重新处理操作,则处理器可以重置。

    Vector Completion Mask Handling
    9.
    发明申请
    Vector Completion Mask Handling 有权
    矢量完成面具处理

    公开(公告)号:US20120272046A1

    公开(公告)日:2012-10-25

    申请号:US13535685

    申请日:2012-06-28

    IPC分类号: G06F9/302

    摘要: Techniques for vector completion mask (VCM) handling are provided. A data structure includes a mask field for each operand of a particular operation. A processor attempts to execute the operation with multiple operands, which are identified in the data structure by the mask fields. If operands are successfully retrieved for execution with the operation, then the corresponding mask field within the data structure is cleared. The processor can reset if any field remains set within the data structure and can re-process the operation with operands that were not previously handled with the operation.

    摘要翻译: 提供矢量完成掩码(VCM)处理技术。 数据结构包括用于特定操作的每个操作数的掩码字段。 处理器尝试通过掩码字段在数据结构中标识的多个操作数来执行操作。 如果成功检索操作数以执行操作,则数据结构中的相应掩码字段将被清除。 如果任何字段在数据结构中保持设置,并且可以使用以前未被操作的操作数重新处理操作,则处理器可以重置。