会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 44. 发明授权
    • Efficient handling of misaligned loads and stores
    • 高效处理不对齐的负载和商店
    • US09131899B2
    • 2015-09-15
    • US13177192
    • 2011-07-06
    • Hari S. KannanPradeep KanapathipillaiGreg M. Hess
    • Hari S. KannanPradeep KanapathipillaiGreg M. Hess
    • G06F12/00A61B5/00A61B5/053A61N1/05A61N1/362G06F12/08
    • A61B5/686A61B5/0538A61B5/4836A61B5/7275A61N1/05A61N1/362G06F12/0802G06F12/0862G06F12/0877
    • A system and method for efficiently handling misaligned memory accesses within a processor. A processor comprises a load-store unit (LSU) with a banked data cache (d-cache) and a banked store queue. The processor generates a first address corresponding to a memory access instruction identifying a first cache line. The processor determines the memory access is misaligned which crosses over a cache line boundary. The processor generates a second address identifying a second cache line logically adjacent to the first cache line. If the instruction is a load instruction, the LSU simultaneously accesses the d-cache and store queue with the first and the second addresses. If there are two hits, the data from the two cache lines are simultaneously read out. If the access is a store instruction, the LSU separates associated write data into two subsets and simultaneously stores these subsets in separate cache lines in separate banks of the store queue.
    • 一种用于有效地处理处理器内的未对准存储器访问的系统和方法。 处理器包括具有分组数据高速缓存(d-cache)和分组存储队列的加载存储单元(LSU)。 处理器产生对应于识别第一高速缓存行的存储器访问指令的第一地址。 处理器确定在高速缓存线边界上跨越的存储器访问未对准。 处理器生成标识逻辑上与第一高速缓存线相邻的第二高速缓存线的第二地址。 如果指令是加载指令,则LSU同时访问d-cache并存储具有第一和第二地址的队列。 如果有两个命中,则同时读出来自两条缓存行的数据。 如果访问是存储指令,则LSU将相关联的写入数据分成两个子集,并且将这些子集同时存储在存储队列的单独的存储区中的单独的高速缓存行中。
    • 45. 发明申请
    • EFFICIENT HANDLING OF MISALIGNED LOADS AND STORES
    • 有效处理缺陷货物和仓库
    • US20130013862A1
    • 2013-01-10
    • US13177192
    • 2011-07-06
    • Hari S. KannanPradeep KanapathipillaiGreg M. Hess
    • Hari S. KannanPradeep KanapathipillaiGreg M. Hess
    • G06F12/08
    • A61B5/686A61B5/0538A61B5/4836A61B5/7275A61N1/05A61N1/362G06F12/0802G06F12/0862G06F12/0877
    • A system and method for efficiently handling misaligned memory accesses within a processor. A processor comprises a load-store unit (LSU) with a banked data cache (d-cache) and a banked store queue. The processor generates a first address corresponding to a memory access instruction identifying a first cache line. The processor determines the memory access is misaligned which crosses over a cache line boundary. The processor generates a second address identifying a second cache line logically adjacent to the first cache line. If the instruction is a load instruction, the LSU simultaneously accesses the d-cache and store queue with the first and the second addresses. If there are two hits, the data from the two cache lines are simultaneously read out. If the access is a store instruction, the LSU separates associated write data into two subsets and simultaneously stores these subsets in separate cache lines in separate banks of the store queue.
    • 一种用于有效地处理处理器内的未对准存储器访问的系统和方法。 处理器包括具有分组数据高速缓存(d-cache)和分组存储队列的加载存储单元(LSU)。 处理器产生对应于识别第一高速缓存行的存储器访问指令的第一地址。 处理器确定在高速缓存线边界上跨越的存储器访问未对准。 处理器生成标识逻辑上与第一高速缓存线相邻的第二高速缓存线的第二地址。 如果指令是加载指令,则LSU同时访问d-cache并存储具有第一和第二地址的队列。 如果有两个命中,则同时读出来自两条缓存行的数据。 如果访问是存储指令,则LSU将相关联的写入数据分成两个子集,并且将这些子集同时存储在存储队列的单独的存储区中的单独的高速缓存行中。
    • 46. 发明授权
    • System and method for dynamic cooperative distributed execution of computer tasks without a centralized controller
    • 没有集中控制器的计算机任务的动态协同分布式执行的系统和方法
    • US08104043B2
    • 2012-01-24
    • US10720893
    • 2003-11-24
    • Sivaprasad PadisettyShankar ManianHari S. Narayan
    • Sivaprasad PadisettyShankar ManianHari S. Narayan
    • G06F9/46
    • G06F9/5038H04L67/10
    • A system and method is provided for cooperative execution of distributed tasks by networked computers without the use of a centralized controller to coordinate the task execution. Each computer has an execution agent that cooperates with the execution agents of the other computers to carry out the execution of a given sequence of tasks. The execution instructions for the task sequence are provided to a first computer in the selected set of computers for task execution. The first computer processes the instructions and forwards them to its peer computers so that each of them knows the tasks it is to perform in connection with the tasks of the other computers. The computers then executes the tasks assigned to them and use peer-to-peer communications to provide status update to their peer computers to synchronize and coordinate the task execution.
    • 提供了一种系统和方法,用于通过网络计算机协调执行分布式任务,而不使用集中控制器来协调任务执行。 每个计算机具有与其他计算机的执行代理协作以执行给定的任务序列的执行代理。 将任务序列的执行指令提供给用于任务执行的所选择的计算机组中的第一计算机。 第一台计算机处理指令并将其转发给对等计算机,以便每个计算机都知道与其他计算机的任务相关的任务。 然后,计算机执行分配给它们的任务,并使用对等通信来向其对等计算机提供状态更新以同步和协调任务执行。
    • 50. 发明授权
    • Electromagnetic field generator and method of operation
    • 电磁场发生器及其运行方法
    • US06545580B2
    • 2003-04-08
    • US09150274
    • 1998-09-09
    • Hari S. HegdeMihai S. RiscaAlan V. HayesAbraham J. NavyRoger P. Fremgen
    • Hari S. HegdeMihai S. RiscaAlan V. HayesAbraham J. NavyRoger P. Fremgen
    • C23C1400
    • H01J37/32623H01J37/3178H01J37/3266
    • An electromagnetic field generator and method of operation for ion beam deposition of magnetic thin-film materials is presented. A combination of open frame electromagnetic field generator elements provides precise control of magnetic field directionality. This control enables deposition of oriented magnetic films with minimal directionality error. The magnetic field direction may be oriented to enable the deposition of alternating layers of directionally oriented magnetic films. An open frame element reduces the weight of the electromagnetic field generator while truncated corners reduce diagonal clearance that may be required in a vacuum chamber. An open frame design also enables the electromagnetic field generator to surround and thus remain clear of the active deposition area; the electromagnetic field generator can thus be shielded from accumulation of sputtered material. Shielding from accumulation of sputter material reduces maintenance requirements.
    • 提出了一种用于磁性薄膜材料的离子束沉积的电磁场发生器和操作方法。 开放式电磁场发生器元件的组合提供了磁场方向性的精确控制。 该控制使得能够以最小的方向性误差沉积取向的磁性膜。 磁场方向可以被定向成能够沉积定向取向的膜的交替层。 开放式框架元件减小了电磁场发生器的重量,而截顶角减小了真空室中可能需要的对角线间隙。 开放式框架设计还使得电磁场发生器能够围绕并因此保持远离有效沉积区域; 因此电磁场发生器可以被屏蔽以免溅射材料的积聚。 防止溅射材料积聚减少维护要求。