METHOD TO EXPLOIT SUPERWORD-LEVEL PARALLELISM USING SEMI-ISOMORPHIC PACKING
    21.
    发明申请
    METHOD TO EXPLOIT SUPERWORD-LEVEL PARALLELISM USING SEMI-ISOMORPHIC PACKING 失效
    使用半正交包装开发超级平行平行的方法

    公开(公告)号:US20080127144A1

    公开(公告)日:2008-05-29

    申请号:US11536990

    申请日:2006-09-29

    IPC分类号: G06F9/45

    CPC分类号: G06F8/456

    摘要: A computer program product is provided for extracting SIMD parallelism. The computer program product includes instructions for providing a stream of input code comprising basic blocks; identifying pairs of statements that are semi-isomorphic with respect to each other within a basic block; iteratively combining into packs, pairs of statements that are semi-isomorphic with respect to each other, and combining packs into combined packs; collecting packs whose statements can be scheduled together for processing; and generating SIMD instructions for each pack to provide for extracting the SIMD parallelism.

    摘要翻译: 提供了一种用于提取SIMD并行性的计算机程序产品。 计算机程序产品包括用于提供包括基本块的输入代码流的指令; 识别在基本块内相对于彼此半同构的语句对; 迭代地组合成包,相对于半同构的语句对,以及将包合并成组合包; 收集包,其陈述可以一起安排处理; 并为每个包生成SIMD指令以提供SIMD并行性。

    Framework for Integrated Intra- and Inter-Loop Aggregation of Contiguous Memory Accesses for SIMD Vectorization
    23.
    发明申请
    Framework for Integrated Intra- and Inter-Loop Aggregation of Contiguous Memory Accesses for SIMD Vectorization 失效
    用于SIMD向量化的连续内存访问的集成内部和环际聚合框架

    公开(公告)号:US20080010634A1

    公开(公告)日:2008-01-10

    申请号:US11856284

    申请日:2007-09-17

    IPC分类号: G06F9/45 G06F15/00

    CPC分类号: G06F8/4452 G06F8/445

    摘要: A method, computer program product, and information handling system for generating loop code to execute on Single-Instruction Multiple-Datapath (SIMD) architectures, where the loop contains multiple non-stride-one memory accesses that operate over a contiguous stream of memory is disclosed. A preferred embodiment identifies groups of isomorphic statements within a loop body where the isomorphic statements operate over a contiguous stream of memory over the iteration of the loop. Those identified statements are then converted in to virtual-length vector operations. Next, the hardware's available vector length is used to determine a number of virtual-length vectors to aggregate into a single vector operation for each iteration of the loop. Finally, the aggregated, vectorized loop code is converted into SIMD operations.

    摘要翻译: 一种用于生成在单指令多数据路径(SIMD)架构上执行的循环码的方法,计算机程序产品和信息处理系统,其中循环包含在连续的存储器流上操作的多个非步进存储器访问 披露 优选实施例识别在循环体内同构语句的组,其中同构语句在循环的迭代上在连续的存储器流上操作。 然后将那些已识别的语句转换为虚拟长度向量操作。 接下来,使用硬件的可用向量长度来确定多个虚拟长度向量以聚合到单个向量操作中,用于循环的每次迭代。 最后,聚合的向量化循环码被转换成SIMD操作。

    Distributed processing with metadata placeholders
    24.
    发明申请
    Distributed processing with metadata placeholders 有权
    使用元数据占位符进行分布式处理

    公开(公告)号:US20070112779A1

    公开(公告)日:2007-05-17

    申请号:US11280518

    申请日:2005-11-16

    IPC分类号: G06F17/30

    CPC分类号: G06F17/30038 G06F17/30017

    摘要: Metadata placeholders are used for distributed processing. The metadata placeholders are added to digital objects. The placeholders are filled with requests to perform certain downstream processing on the objects. The placeholders are at least partially empty so they can be filled in by nodes that perform the downstream processing.

    摘要翻译: 元数据占位符用于分布式处理。 元数据占位符被添加到数字对象。 占位符填充有对对象执行某些下游处理的请求。 占位符至少部分为空,因此可以由执行下游处理的节点填充。

    System and method for SIMD code generation in the presence of optimized misaligned data reorganization
    25.
    发明申请
    System and method for SIMD code generation in the presence of optimized misaligned data reorganization 失效
    存在优化的未对齐数据重组的SIMD代码生成的系统和方法

    公开(公告)号:US20050283774A1

    公开(公告)日:2005-12-22

    申请号:US10918996

    申请日:2004-08-16

    IPC分类号: G06F9/45

    CPC分类号: G06F8/4452 G06F8/447

    摘要: A method, computer program product, and information handling system for generating loop code to execute on Single-Instruction Multiple-Datapath (SIMD) architectures, where the loop operates on datatypes having different lengths, is disclosed. Further, a preferred embodiment of the present invention includes a novel technique to efficiently realign or shift arbitrary streams to an arbitrary offset, regardless whether the alignments or offsets are known at the compile time or not. This technique enables the application of advanced alignment optimizations to runtime alignment. This allows sequential loop code operating on datatypes of disparate length to be transformed (“simdized”) into optimized SIMD code through a fully automated process.

    摘要翻译: 公开了一种用于在单指令多数据路径(SIMD)架构上生成循环码来执行循环对具有不同长度的数据类型进行操作的方法,计算机程序产品和信息处理系统。 此外,本发明的优选实施例包括一种用于有效地将任意流重新对准或将任意流移动到任意偏移的新技术,无论在编译时是否已知对准或偏移。 这种技术使得可以将高级对齐优化应用于运行时对齐。 这允许对具有不同长度的数据类型的顺序循环代码通过完全自动化的过程进行转换(“模拟化”)成优化的SIMD代码。

    Method and apparatus for identifying selected portions of a video stream
    26.
    发明申请
    Method and apparatus for identifying selected portions of a video stream 有权
    用于识别视频流的所选部分的方法和装置

    公开(公告)号:US20050257151A1

    公开(公告)日:2005-11-17

    申请号:US10844386

    申请日:2004-05-13

    申请人: Peng Wu

    发明人: Peng Wu

    摘要: A method is disclosed for identifying a selected portion of a video stream. A user interface is provided for designating a reference frame of a selected portion of a video stream. A processor is configured to compare the reference frame with other portions of the video stream to establish a similarity measure, process the similarity measure to identify a candidate region as a boundary of the selected portion of the video stream, and provide user access to the candidate region to designate the boundary for storage via the user interface.

    摘要翻译: 公开了一种用于识别视频流的选定部分的方法。 提供用于指定视频流的选定部分的参考帧的用户界面。 处理器被配置为将参考帧与视频流的其他部分进行比较以建立相似性度量,处理相似性度量以将候选区域识别为视频流的所选部分的边界,并且向候选者提供用户访问 区域通过用户界面指定存储的边界。

    Method for manufacturing nano-carbon microparticle

    公开(公告)号:US11261094B2

    公开(公告)日:2022-03-01

    申请号:US16614875

    申请日:2018-04-25

    申请人: Peng Wu Yanting Cai

    发明人: Peng Wu Yanting Cai

    IPC分类号: C01B32/90 C01B32/15

    摘要: The invention provides a method for manufacturing nano carbon micro particles, including the following steps: step one: digesting the Malvaceae plants to produce solutions containing lignin; step two: extracting lignin condensation from the solution containing lignin, and then removing salt from the lignin condensation to form material containing lignin; step three: carbonizing the material containing lignin to form carbides; step four: crushing the carbides; step five: performing high-frequency heat treatment on the crushed carbides to obtain carbon micro particles; further including step six: crushing again the carbon micro particles so that the carbon micro particles are nano-sized and finely pulverized. The high-purity carbon micro particles obtained by the present invention have excellent properties in aspects of conductivity, wear resistance, heat resistance, corrosion resistance, etc., which can be used as an electromagnetic sealing material, a wear-resistant material, a heating element, a heat-resistant material, corrosion resistant materials, the application is extremely wide.

    Voice-controlled bluetooth headset
    28.
    发明授权

    公开(公告)号:US10863279B1

    公开(公告)日:2020-12-08

    申请号:US16795409

    申请日:2020-02-19

    申请人: Peng Wu

    发明人: Peng Wu

    摘要: The disclosure is directed to a voice-controlled Bluetooth headset, which includes a receiver, a storage module, an offline voice recognition module, and a Bluetooth module. The offline voice recognition module is used to activate and recognize a preset voice when a preset activation password is received. The Bluetooth module is electrically connected to other modules, and is used for system control, Bluetooth transmission, and processing of instructions output by the preset voice recognition module and performs corresponding functions. With the above structure, the present disclosure can implement voice control of the Bluetooth headset according to a simple preset password, without manual operation, and is simple and convenient to use. The offline voice recognition module can achieve a preset voice control effect by setting a simple preset password, reducing the time required for the voice recognition device to recognize complex voice information, and the user's human-machine interaction experience is better, which further saves the cost required for voice recognition devices.