Apparatus and method for optimizing loop buffer in reconfigurable processor
    1.
    发明申请
    Apparatus and method for optimizing loop buffer in reconfigurable processor 有权
    用于优化可重构处理器中循环缓冲器的装置和方法

    公开(公告)号:US20070150710A1

    公开(公告)日:2007-06-28

    申请号:US11525913

    申请日:2006-09-25

    IPC分类号: G06F9/44

    摘要: A reconfigurable processor comprising a configuration memory for storing a configuration bit for at least one loop configuration; a valid information memory for storing bit information indicating whether an operation in a loop is a delay operation; and at least one processing unit for determining whether an operation in a next cycle is the delay operation by referring to the bit information transmitted from the valid information memory, and selectively performing a change and an implementation of a configuration according to the configuration bit from the configuration memory based on the determined results.

    摘要翻译: 一种可重配置处理器,包括用于存储用于至少一个环路配置的配置位的配置存储器; 用于存储指示循环中的操作是否为延迟操作的位信息的有效信息存储器; 以及至少一个处理单元,用于通过参考从有效信息存储器发送的比特信息来确定下一个周期中的操作是否是延迟操作,并且根据来自所述有用信息存储器的配置位选择性地执行改变和配置的实现 基于确定结果的配置存储器。

    Apparatus and method for optimizing loop buffer in reconfigurable processor
    2.
    发明授权
    Apparatus and method for optimizing loop buffer in reconfigurable processor 有权
    用于优化可重构处理器中循环缓冲器的装置和方法

    公开(公告)号:US07478227B2

    公开(公告)日:2009-01-13

    申请号:US11525913

    申请日:2006-09-25

    IPC分类号: G06F9/40

    摘要: A reconfigurable processor comprising a configuration memory for storing a configuration bit for at least one loop configuration; a valid information memory for storing bit information indicating whether an operation in a loop is a delay operation; and at least one processing unit for determining whether an operation in a next cycle is the delay operation by referring to the bit information transmitted from the valid information memory, and selectively performing a change and an implementation of a configuration according to the configuration bit from the configuration memory based on the determined results.

    摘要翻译: 一种可重配置处理器,包括用于存储用于至少一个环路配置的配置位的配置存储器; 用于存储指示循环中的操作是否为延迟操作的位信息的有效信息存储器; 以及至少一个处理单元,用于通过参考从有效信息存储器发送的比特信息来确定下一个周期中的操作是否是延迟操作,并且根据来自所述有用信息存储器的配置位选择性地执行改变和配置的实现 基于确定结果的配置存储器。

    Profiler for optimizing processor architecture and application
    3.
    发明申请
    Profiler for optimizing processor architecture and application 有权
    Profiler用于优化处理器架构和应用

    公开(公告)号:US20080120493A1

    公开(公告)日:2008-05-22

    申请号:US11730170

    申请日:2007-03-29

    IPC分类号: G06F7/38

    摘要: A profiler which provides information to optimize an application specific architecture processor and a program for the processor is provided. The profiler includes: an architecture analyzer which analyzes an architecture description, and generates architecture analysis information, the architecture description describing an architecture of an application specific architecture processor which comprises a plurality of processing elements; a static analyzer which analyzes program static information that describes static information of a program, and generates static analysis information; a dynamic analyzer which analyzes program dynamic information that describes dynamic information of the program, and generates dynamic analysis information, the dynamic information of the program being generated by simulating the program; and a cross profiling analyzer which generates information for optimizing the application specific architecture processor to implement the program based on at least one of the architecture analysis information, the static analysis information, and the dynamic analysis information.

    摘要翻译: 提供了一种提供信息以优化特定于应用的架构处理器和用于处理器的程序的分析器。 分析器包括:架构分析器,其分析架构描述并生成架构分析信息,描述包括多个处理元件的应用特定架构处理器的架构的架构描述; 静态分析器,分析程序静态信息,描述程序的静态信息,并生成静态分析信息; 动态分析器,其分析描述所述程序的动态信息的程序动态信息,并且生成动态分析信息,所述程序的动态信息是通过模拟所述程序而产生的; 以及交叉分析分析器,其基于所述架构分析信息,所述静态分析信息和所述动态分析信息中的至少一个,生成用于优化所述应用专用架构处理器以实现所述程序的信息。

    Method and apparatus for interrupt handling during loop processing in reconfigurable coarse grained array
    4.
    发明授权
    Method and apparatus for interrupt handling during loop processing in reconfigurable coarse grained array 有权
    在可重构粗粒度阵列中循环处理中的中断处理方法和装置

    公开(公告)号:US07529917B2

    公开(公告)日:2009-05-05

    申请号:US11519858

    申请日:2006-09-13

    IPC分类号: G06F9/00

    CPC分类号: G06F9/4812

    摘要: A processor including a coarse grained array including a plurality of function units and a plurality of register files, wherein a loop to be executed by the coarse grained array is split into a plurality of sub-loops, and when an interrupt request occurs while executing the sub-loop in the coarse grained array, the interrupt request is processed after the executing of the sub-loop is completed.

    摘要翻译: 一种包括包括多个功能单元和多个寄存器文件的粗粒度阵列的处理器,其中由粗粒度阵列执行的循环被分割成多个子循环,并且当执行中断请求时发生中断请求 在粗粒子数组中的子循环,中断请求在子循环执行完成后被处理。

    Method and apparatus for interrupt handling in coarse grained array
    5.
    发明申请
    Method and apparatus for interrupt handling in coarse grained array 有权
    粗粒度阵列中断处理的方法和装置

    公开(公告)号:US20070162729A1

    公开(公告)日:2007-07-12

    申请号:US11519858

    申请日:2006-09-13

    IPC分类号: G06F9/44

    CPC分类号: G06F9/4812

    摘要: A processor including a coarse grained array including a plurality of function units and a plurality of register files, wherein a loop to be executed by the coarse grained array is split into a plurality of sub-loops, and when an interrupt request occurs while executing the sub-loop in the coarse grained array, the interrupt request is processed after the executing of the sub-loop is completed.

    摘要翻译: 一种包括包括多个功能单元和多个寄存器文件的粗粒度阵列的处理器,其中由粗粒度阵列执行的循环被分割成多个子循环,并且当执行中断请求时发生中断请求 在粗粒子数组中的子循环,中断请求在子循环执行完成后被处理。

    Apparatus and method of exception handling for reconfigurable architecture
    6.
    发明授权
    Apparatus and method of exception handling for reconfigurable architecture 有权
    可重构架构异常处理的装置和方法

    公开(公告)号:US09152418B2

    公开(公告)日:2015-10-06

    申请号:US11487407

    申请日:2006-07-17

    摘要: A processor including a coarse grained array including a plurality of processing elements, a central register file including a first plurality of register files, a shadow central register file including a second plurality of register files, each of the second plurality of register files corresponding to each of the first plurality of register files included in the central register file, and a plurality of shadow register files, each of the plurality of shadow register files corresponding to each of a third plurality of register files included in predetermined processing elements selected from the plurality of processing elements.

    摘要翻译: 一种处理器,包括包括多个处理元件的粗粒子阵列,包括第一多个寄存器文件的中央寄存器文件,包括第二多个寄存器文件的影子中心寄存器文件,与每个寄存器文件相对应的第二多个寄存器堆中的每一个 包括在中央寄存器文件中的第一多个寄存器文件和多个影子寄存器文件,多个影子寄存器文件中的每一个对应于包括在从多个寄存器文件中选择的预定处理元件中的第三多个寄存器文件中的每一个 处理元件。

    Method and apparatus for efficiently processing array operation in computer system
    7.
    发明授权
    Method and apparatus for efficiently processing array operation in computer system 有权
    在计算机系统中有效处理阵列运算的方法和装置

    公开(公告)号:US08024717B2

    公开(公告)日:2011-09-20

    申请号:US11492974

    申请日:2006-07-26

    IPC分类号: G06F9/45

    CPC分类号: G06F8/4452

    摘要: An apparatus and a method for processing an array in a loop in a computer system, including: applying loop unrolling to a multi-dimensional array included in a loop based on a predetermined unrolling factor to generate a plurality of unrolled multi-dimensional arrays; and transforming each of the plurality of unrolled multi-dimensional arrays into a one-dimensional array having an array subscript expression in a form of an affine function with respect to a loop counter variable.

    摘要翻译: 一种用于处理计算机系统中的循环中的阵列的装置和方法,包括:基于预定的展开因子将循环展开应用到包括在循环中的多维阵列,以生成多个展开的多维阵列; 并且将多个展开的多维数组中的每一个变换成具有相对于循环计数器变量的仿射函数形式的数组下标表达式的一维阵列。

    Method and apparatus for efficiently processing array operation in computer system
    10.
    发明申请
    Method and apparatus for efficiently processing array operation in computer system 有权
    在计算机系统中有效处理阵列运算的方法和装置

    公开(公告)号:US20070169044A1

    公开(公告)日:2007-07-19

    申请号:US11492974

    申请日:2006-07-26

    IPC分类号: G06F9/45

    CPC分类号: G06F8/4452

    摘要: An apparatus and a method for processing an array in a loop in a computer system, including: applying loop unrolling to a multi-dimensional array included in a loop based on a predetermined unrolling factor to generate a plurality of unrolled multi-dimensional arrays; and transforming each of the plurality of unrolled multi-dimensional arrays into a one-dimensional array having an array subscript expression in a form of an affine function with respect to a loop counter variable.

    摘要翻译: 一种用于处理计算机系统中的循环中的阵列的装置和方法,包括:基于预定的展开因子将循环展开应用到包括在循环中的多维阵列,以生成多个展开的多维阵列; 并且将多个展开的多维数组中的每一个变换成具有相对于循环计数器变量的仿射函数形式的数组下标表达式的一维阵列。