Device and method for assembling retainer and cotter
    1.
    发明授权
    Device and method for assembling retainer and cotter 有权
    装配和装配的装置和方法

    公开(公告)号:US08646159B2

    公开(公告)日:2014-02-11

    申请号:US12937576

    申请日:2009-01-14

    IPC分类号: B23P11/00

    摘要: A large-diameter-side end surface (104a) of a cotter (104) can be supported by a cotter holder (32) from below through inserting a shaft member (102) along an inner periphery of the cotter (104) from above the cotter (104). In this state, a protrusion (104b) of the cotter (104) and an annular groove (102a) of the shaft member (102) are fitted to each other, and the cotter (104) and a retainer (103) are taper-fitted to each other. Consequently, the cotter is prevented from being unstable, and hence it is possible to assemble the retainer (103) and the cotter (104) to the shaft member (102) with good accuracy.

    摘要翻译: 通过沿着所述开口(104)的内周缘从所述开口(104)的内周方向插入轴构件(102),所述开口(104)的大直径侧端面(104a)可以由开口托架(32)从下方支撑, 开口(104)。 在这种状态下,将开口(104)的突起(104b)和轴构件(102)的环形槽(102a)彼此嵌合,并且开口(104)和保持器(103) 相互配合 因此,可防止开口不稳定,因此能够以良好的精度将保持器(103)和开口(104)组装到轴构件(102)。

    Floating point collect and operate
    2.
    发明授权
    Floating point collect and operate 有权
    浮点收集和运行

    公开(公告)号:US08595467B2

    公开(公告)日:2013-11-26

    申请号:US12648527

    申请日:2009-12-29

    IPC分类号: G06F9/00

    摘要: Mechanisms are provided for performing a floating point collect and operate for a summation across a vector for a dot product operation. A routing network placed before the single instruction multiple data (SIMD) unit allows the SIMD unit to perform a summation across a vector with a singe stage of adders. The routing network routes the vector elements to the adders in a first cycle. The SIMD unit stores the results of the adders into a results vector register. The routing network routes the summation results from the results vector register to the adders in a second cycle. The SIMD unit then stores the results from the second cycle in the results vector register.

    摘要翻译: 提供了用于执行浮点收集和操作用于点产品操作的矢量的求和的机制。 放置在单指令多数据(SIMD)单元之前的路由网络允许SIMD单元在具有单个加法器的一个阶段的矢量上执行求和。 路由网络在第一个周期中将向量元素路由到加法器。 SIMD单元将加法器的结果存储到结果向量寄存器中。 路由网络将求和结果从结果向量寄存器路由到第二个周期中的加法器。 然后,SIMD单元将来自第二周期的结果存储在结果向量寄存器中。

    Systems and methods for transferring data to maintain preferred slot positions in a bi-endian processor
    4.
    发明授权
    Systems and methods for transferring data to maintain preferred slot positions in a bi-endian processor 有权
    用于传送数据以维持双端处理器中优选插槽位置的系统和方法

    公开(公告)号:US08145804B2

    公开(公告)日:2012-03-27

    申请号:US12563756

    申请日:2009-09-21

    IPC分类号: G06F13/28

    CPC分类号: G06F9/30007 G06F9/3824

    摘要: A bi-endian multiprocessor system having multiple processing elements, each of which includes a processor core, a local memory and a memory flow controller. The memory flow controller transfers data between the local memory and data sources external to the processing element. If the processing element and the data source implement data representations having the same endian-ness, each multi-word line of data is stored in the local memory in the same word order as in the data source. If the processing element and the data source implement data representations having different endian-ness, the words of each multi-word line of data are transposed when data is transferred between local memory and the data source. The processing element may incorporate circuitry to add doublewords, wherein the circuitry can alternately carry bits from a first word to a second word or vice versa, depending upon whether the words in lines of data are transposed.

    摘要翻译: 一种具有多个处理元件的双端式多处理器系统,每个处理单元包括处理器核心,本地存储器和存储器流控制器。 存储器流控制器在本地存储器和处理元件外部的数据源之间传送数据。 如果处理元件和数据源实现具有相同字节数的数据表示,则每个多字数据行以与数据源中相同的字顺序存储在本地存储器中。 如果处理元件和数据源实现具有不同端点的数据表示,则当数据在本地存储器和数据源之间传送时,每个多字数据行的字被转置。 处理元件可以包括用于添加双字的电路,其中,根据数据行中的字是否被转置,电路可以交替地将位从第一个字运送到第二个字,反之亦然。

    Information processing apparatus and computer-readable storage medium
    5.
    发明授权
    Information processing apparatus and computer-readable storage medium 失效
    信息处理装置和计算机可读存储介质

    公开(公告)号:US08037277B2

    公开(公告)日:2011-10-11

    申请号:US12071966

    申请日:2008-02-28

    IPC分类号: G06F12/14 G06F12/16

    CPC分类号: G06F12/0802 G06F12/126

    摘要: A computer-readable storage medium stores a program for causing a processor to perform a process including: acquiring a first address that specifies a start address of a first area on the main memory where a target data to be cached is stored and range information that specifies a size of the first area on the main memory; converting the first address into a second address that specifies a start address of a second area on the local memory, the second area having a one-to-n correspondence (n=positive integer) to a part of a bit string of the first address; copying the target data stored in the first area specified by the first address and the range information onto the second area specified by the second address and the range information; and storing the second address to allow accessing the target data copied onto the local memory.

    摘要翻译: 计算机可读存储介质存储用于使处理器执行处理的程序,包括:获取指定主存储器上存储要存储的目标数据的第一区域的起始地址的第一地址,以及指定 主存储器上第一个区域的大小; 将所述第一地址转换为指定所述本地存储器上的第二区域的起始地址的第二地址,所述第二区域具有与所述第一地址的位串的一部分的一对数对应关系(n =正整数) ; 将由第一地址和范围信息指定的第一区域中存储的目标数据复制到由第二地址和范围信息指定的第二区域上; 并存储第二地址以允许访问复制到本地存储器上的目标数据。

    CACHE MEMORY CONTROL CIRCUIT AND CACHE MEMORY CONTROL METHOD
    7.
    发明申请
    CACHE MEMORY CONTROL CIRCUIT AND CACHE MEMORY CONTROL METHOD 审中-公开
    高速缓存存储器控制电路和高速缓存存储器控制方法

    公开(公告)号:US20110099336A1

    公开(公告)日:2011-04-28

    申请号:US12882588

    申请日:2010-09-15

    IPC分类号: G06F12/08 G06F12/00

    摘要: A cache memory control circuit has a plurality of counters, each of which is provided per set and per memory space and configured to count how many pieces of data of a corresponding memory space is stored in a corresponding set. The cache memory control circuit controls activation of a tag memory and a data memory of each of a plurality of sets according to a count value of each of the plurality of counters.

    摘要翻译: 高速缓冲存储器控制电路具有多个计数器,每个计数器按照每个存储空间和每个存储器空间提供,并且被配置为对相应存储器空间的数据数量存储在对应的集合中。 高速缓冲存储器控制电路根据多个计数器中的每一个的计数值来控制多个集合中的每一个的标签存储器和数据存储器的激活。

    Maintaining level heat emission in multiprocessor by rectifying dispatch table assigned with static tasks scheduling using assigned task parameters
    8.
    发明授权
    Maintaining level heat emission in multiprocessor by rectifying dispatch table assigned with static tasks scheduling using assigned task parameters 失效
    通过使用分配的任务参数整理分配有静态任务调度的调度表来维持多处理器中的发热量

    公开(公告)号:US07877751B2

    公开(公告)日:2011-01-25

    申请号:US11232984

    申请日:2005-09-23

    IPC分类号: G06F9/46 G06F1/00

    摘要: According to an aspect of the present invention, heat emissions of processors are level among the processors, and it is possible to suppress occurrence of stop of process due to overheating. The control IC assigns tasks to the processors, and thereafter rectifies an assignment result such that temperatures of the processors become almost level among the processors, on the basis of the temperatures of the processors obtained by temperature sensors. This structure enables level heat emissions among the processors, and suppresses occurrence of stop of process due to overheating.

    摘要翻译: 根据本发明的一个方面,处理器的热排放是处理器之间的水平,并且可以抑制由于过热而导致的处理停止的发生。 控制IC将任务分配给处理器,然后基于由温度传感器获得的处理器的温度,对处理器中的处理器的温度几乎达到一定程度的分配结果进行整流。 这种结构使得处理器之间能够发生热量排放,并且抑制由于过热而导致的过程停止的发生。

    Real-time schedulability determination method and real-time system
    9.
    发明授权
    Real-time schedulability determination method and real-time system 失效
    实时可调度确定方法和实时系统

    公开(公告)号:US07797703B2

    公开(公告)日:2010-09-14

    申请号:US11085532

    申请日:2005-03-22

    IPC分类号: G06F9/46 G06G7/00 G06G7/32

    CPC分类号: G06F9/4887

    摘要: Schedulability determination method of determining whether real-time scheduling of tasks is possible using processors, includes calculating Lk and Σi=1 . . . NMi*Uk, i, (1≦k, i≦N; k, i: integer) where Lk corresponds to task-k, Mi represents number of the one or more processors simultaneously used by task-i, Uk, i corresponds to task-k and task-i, and N represents number of tasks, and determining that real-time scheduling of tasks is possible using processors, if tasks all satisfy conditions, Σi=1 . . . NMi*Uk, i≦Lk for all k (1≦k≦N), wherein Lk corresponds to task-k which is to be detected to satisfy corresponding one of conditions, Lk being expressed as follows if Mk≦(M+1)/2, Lk=(M−Mk+1)−(M−2Mk+1)Ck/Dk, and Uk, i is expressed as follows if Mk≦(M+1)/2, Mi

    摘要翻译: 使用处理器确定任务的实时调度是否可行的调度确定方法包括计算Lk和&Sgr; i = 1。 。 。 NMi * Uk,i,(1&nlE; k,i&nlE; N; k,i:整数)其中Lk对应于task-k,Mi表示task-i同时使用的一个或多个处理器的数量,Uk,i对应于 任务k和任务i,N表示任务数,并且如果任务满足条件,则确定任务的实时调度是可能的,并且Sgr i = 1。 。 。 对于所有的k(1&nlE; k&nlE; N),其中Lk对应于待检测以满足相应条件的任务k,如果Mk&nlE;(M + 1) 如果Mk&nlE;(M + 1)/ 2,Mi <(M + 1)/ 2,则Lk =(M-Mk + 1) - (M-2Mk + 1)Ck / Dk, 如果Mk&nlE;(M + 1)/ 2,Mi((i-1)/ 2,Xk,i&nlE; 0,(1)Uk,i = Ci / Ti {1+(Ti-Di)/ Dk} M + 1)/ 2,0

    Multiprocessor computer and program
    10.
    发明授权
    Multiprocessor computer and program 失效
    多处理器计算机和程序

    公开(公告)号:US07770176B2

    公开(公告)日:2010-08-03

    申请号:US11233026

    申请日:2005-09-23

    IPC分类号: G06F9/46 G06F1/00

    摘要: According to an aspect of the present invention, the processor temperatures can be leveled among processors, thereby suppressing the occurrence of stop of processing due to overheating. For example, on the basis of the temperatures of the processors sensed by temperature sensors, the control IC assigns the processor whose temperature is the lowest to the task whose heat emission is the highest. This makes it possible to level the processor temperatures among processors and suppress occurrence of stop of processing due to overheating.

    摘要翻译: 根据本发明的一个方面,可以在处理器之间调整处理器温度,从而抑制由于过热而导致的处理停止的发生。 例如,基于由温度传感器感测到的处理器的温度,控制IC将温度最低的处理器分配给发热最高的任务。 这使得可以对处理器之间的处理器温度进行调整,并抑制由于过热而导致的处理停止的发生。