Method and apparatus for performing early branch prediction in a microprocessor
    1.
    发明授权
    Method and apparatus for performing early branch prediction in a microprocessor 失效
    用于在微处理器中执行早期分支预测的方法和装置

    公开(公告)号:US06185676B2

    公开(公告)日:2001-02-06

    申请号:US08940435

    申请日:1997-09-30

    IPC分类号: G06F1500

    CPC分类号: G06F9/3804 G06F9/3844

    摘要: A pipelined microprocessor having a branch prediction unit implemented in an instruction pointer generation stage of the microprocessor. The branch prediction unit includes a memory device having at least a first entry configured to hold at least a part of a memory address of a pre-selected branch instruction and at least a part of a memory address of a branch target corresponding to the pre-selected branch instruction. The branch prediction unit compares an instruction pointer of an instruction to be executed with the memory address of the pre-selected branch instruction. In response to a match between the instruction pointer and the memory address of pre-selected branch instruction, the unit causes the microprocessor to fetch an instruction corresponding to the branch target. In one embodiment, the instruction pointer generation stage of the microprocessor is implemented as a first stage of the pipelined microprocessor. In addition, the branch prediction unit is compares the instruction pointer and the memory address of the pre-selected branch instruction during a single clock cycle in which the instruction pointer is generated.

    摘要翻译: 具有在微处理器的指令指针生成级中实现的分支预测单元的流水线微处理器。 分支预测单元包括具有至少第一条目的存储器件,其被配置为保持预选择的转移指令的存储器地址的至少一部分以及与预选择的转移指令对应的分支目标的存储器地址的至少一部分, 选择分支指令。 分支预测单元将要执行的指令的指令指针与预先选择的分支指令的存储器地址进行比较。 响应于指令指针和预先选择的分支指令的存储器地址之间的匹配,单元使微处理器获取与分支目标相对应的指令。 在一个实施例中,微处理器的指令指针生成级被实现为流水线微处理器的第一级。 此外,分支预测单元在生成指令指针的单个时钟周期期间比较指令指针和预先选择的分支指令的存储器地址。

    Utilizing an advanced load address table for memory disambiguation in an out of order processor
    6.
    发明申请
    Utilizing an advanced load address table for memory disambiguation in an out of order processor 失效
    利用高级加载地址表,在一个乱序处理器中消除内存消歧

    公开(公告)号:US20050149703A1

    公开(公告)日:2005-07-07

    申请号:US10750150

    申请日:2003-12-31

    IPC分类号: G06F9/30 G06F9/38

    摘要: Embodiments include a system for minimizing storage space required for tracking load instructions through a pipeline in a processor. Store instructions are tracked in a separate queue and only load instructions that require speculation on data or addresses are tracked in a load table and flagged in the reorder buffer. This system improves system performance by reducing energy and space requirements.

    摘要翻译: 实施例包括用于最小化通过处理器中的流水线跟踪加载指令所需的存储空间的系统。 存储指令在单独的队列中进行跟踪,并且仅加载需要对数据或地址进行推测的指令,并在加载表中跟踪并在重新排序缓冲区中标记。 该系统通过减少能源和空间需求来提高系统性能。