专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US5495591A Method and system for cache miss prediction based on previous cache access requests 失效
标题翻译：基于先前缓存访问请求的高速缓存未命中预测方法和系统
公开(公告)号：US5495591A
公开(公告)日：1996-02-27
申请号：US906618
申请日：1992-06-30
申请人： Charles P. Ryan
发明人： Charles P. Ryan
IPC分类号： G06F12/08
CPC分类号： G06F12/0862 , G06F2212/6026 , Y10S707/99936
摘要： For a data processing system which employs a cache memory, the disclosure includes both a method for lowering the cache miss ratio for requested operands and an example of special purpose apparatus for practicing the method. Recent cache misses are stored in a first in, first out miss stack, and the stored addresses are searched for displacement patterns thereamong. Any detected pattern is then employed to predict a succeeding cache miss by prefetching from main memory the signal identified by the predictive address. The apparatus for performing this task is preferably hard wired for speed purposes and includes subtraction circuits for evaluating variously displaced addresses in the miss stack and comparator circuits for determining if the outputs from at least two subtraction circuits are the same, indicating a pattern which yields information which can be combined with an address in the stack to develop a predictive address. The efficiency of the apparatus is improved by placing a series of "select pattern" values representing the search order for trying patterns into a register stack and providing logic circuitry by which the most recently found "select pattern" value is placed at the top of the stack with the remaining "select pattern" values pushed down accordingly.
摘要翻译：对于采用高速缓冲存储器的数据处理系统，本公开包括用于降低所请求的操作数的高速缓存未命中率的方法和用于实施该方法的专用设备的示例。最近的高速缓存未命中被存储在先入先出的堆栈中，并且存储的地址被搜索到位移模式。然后，通过从主存储器预取由预测地址识别的信号，随后采用任何检测到的模式来预测随后的高速缓存未命中。用于执行该任务的装置优选地是为了速度目的而进行硬连接的，并且包括用于评估未命令堆栈中的各种移位地址的减法电路和用于确定来自至少两个减法电路的输出是否相同的比较器电路，指示产生信息的模式其可以与栈中的地址组合以开发预测地址。通过将表示用于尝试图案的搜索顺序的一系列“选择图案”值放置到寄存器堆栈中并提供逻辑电路来改善装置的效率，通过该逻辑电路将最近发现的“选择图案”值放置在堆栈与剩余的“选择模式”值相应地下推。

2. 发明授权

US4314331A Cache unit information replacement apparatus 失效
标题翻译：缓存单元信息更换装置
公开(公告)号：US4314331A
公开(公告)日：1982-02-02
申请号：US968048
申请日：1978-12-11
申请人： Marion G. Porter , Robert W. Norman, Jr. , Charles P. Ryan
发明人： Marion G. Porter , Robert W. Norman, Jr. , Charles P. Ryan
IPC分类号： G06F12/08 , G06F12/12 , G06F9/30
CPC分类号： G06F12/126
摘要： A cache unit includes a cache store organized into a number of levels to provide a fast access to instructions and data words. Directory circuits, associated with the cache store, contain address information identifying those instructions and data words stored in the cache store. The cache unit has at least one instruction register for storing address and level signals for specifying the location of the next instruction to be fetched and transferred to the processing unit. Replacement circuits are included which, during normal operation, assign cache locations sequentially for replacing old information with new information. The cache unit further includes detection apparatus for detecting a conflict condition resulting in an improper assignment. The detection apparatus, upon detecting such a condition, advances the relacement circuits forward for assigning the next sequential group of locations or level inhibiting it from making its normal location assignment. It also inhibits the directory circuits from writing the necessary information therein required for making the location assignment and prevents the information which produced the conflict from being written into cache store when received from memory.
摘要翻译：缓存单元包括组织成多个级别的缓存存储器，以提供对指令和数据字的快速访问。与缓存存储相关联的目录电路包含标识存储在高速缓存存储器中的那些指令和数据字的地址信息。高速缓存单元具有至少一个用于存储地址和电平信号的指令寄存器，用于指定要提取并传送到处理单元的下一个指令的位置。包括替换电路，其在正常操作期间顺序地分配高速缓存位置以用新信息替换旧信息。高速缓存单元还包括用于检测导致不正确分配的冲突状态的检测装置。检测装置在检测到这样的状况时，前进用于分配下一个连续的位置组或级别的替换电路以防止其正常的位置分配。它还禁止目录电路在进行位置分配时写入其中所需的必要信息，并且在从存储器接收时防止产生冲突的信息被写入高速缓存存储器。

3. 发明授权

US06249880B1 Method and apparatus for exhaustively testing interactions among multiple processors 有权
标题翻译：用于彻底测试多个处理器之间的交互的方法和装置
公开(公告)号：US06249880B1
公开(公告)日：2001-06-19
申请号：US09156378
申请日：1998-09-17
申请人： William A. Shelly , Charles P. Ryan
发明人： William A. Shelly , Charles P. Ryan
IPC分类号： H02H305
CPC分类号： G06F11/24 , G06F11/2242
摘要： Interactions among multiple processors (92) are exhaustively tested. A master processor (92) retrieves test information for a set of tests from a test table (148). It then enters a series of embedded loops, with one loop for each of the tested processors (92). A cycle delay count for each of the tested processors (92) is incremented (152, 162, 172) through a range specified in the test table entry. For each combination of cycle delay count loop indices, a single test is executed (176). In each such test (176), the master processor (92) sets up (182) each of the other processors (92) being tested. This setup (182) specifies the delay count and the code for that processor (92) to execute. When each processor (92) is setup (182), it waits (192) for a synchronize interrupt (278). When all processors (92) have been setup (182), the master processor (92) issues (191) the synchronize interrupt signal (276). Each processor (92) then starts traces (193) and delays (194) the specified number of cycles. After the delay, the processor (92) executes its test code (195).
摘要翻译：多处理器之间的相互作用（92）进行了详尽的测试。主处理器（92）从测试表（148）检索一组测试的测试信息。然后，它进入一系列嵌入式循环，每个测试处理器（92）有一个循环。每个测试处理器（92）的周期延迟计数通过测试表条目中指定的范围递增（152,162,172）。对于循环延迟计数循环指标的每个组合，执行单个测试（176）。在每个这样的测试（176）中，主处理器（92）建立（182）被测试的每个其他处理器（92）。该设置（182）指定延迟计数和该处理器（92）执行的代码。当每个处理器（92）被建立（182）时，它等待（192）同步中断（278）。当所有处理器（92）已经建立（182）时，主处理器（92）发出（191）同步中断信号（276）。每个处理器（92）然后开始指定数量的循环的迹线（193）和延迟（194）。在延迟之后，处理器（92）执行其测试代码（195）。

4. 发明授权

US06230263B1 Data processing system processor delay instruction 有权
标题翻译：数据处理系统处理器延时指令
公开(公告)号：US06230263B1
公开(公告)日：2001-05-08
申请号：US09156376
申请日：1998-09-17
申请人： Charles P. Ryan , Ronald W. Yoder , William A. Shelly
发明人： Charles P. Ryan , Ronald W. Yoder , William A. Shelly
IPC分类号： G06F930
CPC分类号： G06F9/30079
摘要： A processor (92) in a data processing system (80) provides a DELAY instruction. Executing the DELAY instruction causes the processor (92) to a specified integral number of clock (98) cycles before continuing. Delays are guaranteed to have a linear relationship with a constant slope with the specified number of clock cycles. Incrementing the specified delay through a range allows exhaustive testing of interactions among multiple processors.
摘要翻译：数据处理系统（80）中的处理器（92）提供DELAY指令。执行DELAY指令使处理器（92）在指定的整数时钟（98）周期之前继续。延迟保证与具有指定时钟周期数的恒定斜率具有线性关系。通过一个范围增加指定的延迟允许对多个处理器之间的交互进行详尽的测试。

5. 发明授权

US5694572A Controllably operable method and apparatus for predicting addresses of future operand requests by examination of addresses of prior cache misses 失效
标题翻译：可控制的可操作的方法和装置，用于通过检查先前的高速缓存未命中的地址来预测未来的操作数请求的地址
公开(公告)号：US5694572A
公开(公告)日：1997-12-02
申请号：US841687
申请日：1992-02-26
申请人： Charles P. Ryan
发明人： Charles P. Ryan
IPC分类号： G06F12/10 , G06F9/38 , G06F12/08 , G06F12/12 , G06F13/00
CPC分类号： G06F9/383 , G06F12/0862 , G06F9/3832 , G06F2212/6026
摘要： In a data processing system which employs a cache memory feature, a method and exemplary special purpose apparatus for practicing the method are disclosed to lower the cache miss ratio for called operands. Recent cache misses are stored in a first in, first out miss stack, and the stored addresses are searched for displacement patterns thereamong. Any detected pattern is then employed to predict a succeeding cache miss by prefetching from main memory the signal identified by the predictive address. The apparatus for performing this task is preferably hard wired for speed purposes and includes subtraction circuits for evaluating variously displaced addresses in the miss stack and comparator circuits for determining if the outputs from at least two subtraction circuits are the same indicating a pattern yielding information which can be combined with an address in the stack to develop a predictive address. The cache miss prediction mechanism is only selectively enabled during cache "in-rush" following a process change to increase the recovery rate; thereafter, it is disabled, based upon timing-out a timer or reaching a hit ratio threshold, in order that normal procedures allow the hit ratio to stabilize at a higher percentage than if the cache miss prediction mechanism were operated continuously.
摘要翻译：在采用高速缓冲存储器特征的数据处理系统中，公开了一种用于实施该方法的方法和示例性专用设备，用于降低被叫操作数的高速缓存未命中率。最近的高速缓存未命中被存储在先入先出的堆栈中，并且存储的地址被搜索到位移模式。然后，通过从主存储器预取由预测地址识别的信号，随后采用任何检测到的模式来预测随后的高速缓存未命中。用于执行该任务的装置优选地用于速度目的是硬连线的，并且包括用于评估未命令堆栈中的各种移位的地址的减法电路和用于确定来自至少两个减法电路的输出是否相同的指示可以与堆栈中的地址组合以开发预测地址。高速缓存未命中预测机制仅在过程变化缓存“急速”中有选择地启用，以提高恢复速率; 此后，基于定时器的定时器或达到命中率阈值来禁用它，以便正常程序允许命中率稳定在比高速缓存未命中预测机制连续运行的更高的百分比。

6. 发明授权

US5450561A Cache miss prediction method and apparatus for use with a paged main memory in a data processing system 失效
标题翻译：用于数据处理系统中的分页主存储器的缓存未命中预测方法和装置
公开(公告)号：US5450561A
公开(公告)日：1995-09-12
申请号：US921825
申请日：1992-07-29
申请人： Charles P. Ryan
发明人： Charles P. Ryan
IPC分类号： G06F12/08 , G06F12/02
CPC分类号： G06F12/0862 , G06F2212/6026
摘要： In a data processing system which employs a cache memory feature, a method and exemplary special purpose apparatus for practicing the method are disclosed to lower the cache miss ratio for called operands. Recent cache misses are stored in a first in, first out miss stack, and the stored addresses are searched for displacement patterns thereamong. Any detected pattern is then employed to predict a succeeding cache miss by prefetching from main memory the signal identified by the predictive address. The apparatus for performing this task is preferably hard wired for speed purposes and includes subtraction circuits for evaluating variously displaced addresses in the miss stack and comparator circuits for determining if the outputs from at least two subtraction circuits are the same indicating a pattern yielding information which can be combined with an address in the stack to develop a predictive address. The efficiency of the apparatus operating in an environment incorporating a paged main memory is improved, according to the invention, by the addition of logic circuitry which serves to inhibit prefetch if a page boundary would be encountered.
摘要翻译：在采用高速缓冲存储器特征的数据处理系统中，公开了一种用于实施该方法的方法和示例性专用设备，用于降低被叫操作数的高速缓存未命中率。最近的高速缓存未命中被存储在先入先出的堆栈中，并且存储的地址被搜索到位移模式。然后，通过从主存储器预取由预测地址识别的信号，随后采用任何检测到的模式来预测随后的高速缓存未命中。用于执行该任务的装置优选地用于速度目的是硬连线的，并且包括用于评估未命令堆栈中的各种移位的地址的减法电路和用于确定来自至少两个减法电路的输出是否相同的指示可以与堆栈中的地址组合以开发预测地址。根据本发明，在包含分页主存储器的环境中操作的设备的效率得到改善，通过添加用于在遇到页边界时用于禁止预取的逻辑电路。

7. 发明授权

US5367656A Controlling cache predictive prefetching based on cache hit ratio trend 失效
标题翻译：基于缓存命中率趋势控制缓存预测预取
公开(公告)号：US5367656A
公开(公告)日：1994-11-22
申请号：US850713
申请日：1992-03-13
申请人： Charles P. Ryan
发明人： Charles P. Ryan
IPC分类号： G06F12/08 , G06F13/00
CPC分类号： G06F12/0862 , G06F2212/6026
摘要： In a data processing system which employs a cache memory feature, a method and exemplary special purpose apparatus for practicing the method are disclosed to lower the cache miss ratio for called operands. Recent cache misses are stored in a first in, first out miss stack, and the stored addresses are searched for displacement patterns thereamong. Any detected pattern is then employed to predict a succeeding cache miss by prefetching from main memory the signal identified by the predictive address. The apparatus for performing this task is preferably hard wired for speed purposes and includes subtraction circuits for evaluating variously displaced addresses in the miss stack and comparator circuits for determining if the outputs from at least two subtraction circuits are the same indicating a pattern yielding information which can be combined with an address in the stack to develop a predictive address. The cache miss prediction mechanism is adaptively selectively enabled by an adaptive circuit that develops a short term operand cache hit ratio history and responds to ratio improving and ratio deteriorating trends by accordingly enabling and disabling the cache miss prediction mechanism.
摘要翻译：在采用高速缓冲存储器特征的数据处理系统中，公开了一种用于实施该方法的方法和示例性专用设备，用于降低被叫操作数的高速缓存未命中率。最近的高速缓存未命中被存储在先入先出的堆栈中，并且存储的地址被搜索到位移模式。然后，通过从主存储器预取由预测地址识别的信号，随后采用任何检测到的模式来预测随后的高速缓存未命中。用于执行该任务的装置优选地用于速度目的是硬连线的，并且包括用于评估未命令堆栈中的各种移位的地址的减法电路和用于确定来自至少两个减法电路的输出是否相同的指示可以与堆栈中的地址组合以开发预测地址。高速缓存未命中预测机制由自适应电路自适应地选择性地使能，该自适应电路产生短期操作数高速缓存命中率历史，并且通过相应地启用和禁用高速缓存未命中机制来响应比率改进和比例恶化趋势。

8. 发明授权

US4521850A Instruction buffer associated with a cache memory unit 失效
标题翻译：与高速缓冲存储器单元相关联的指令缓冲器
公开(公告)号：US4521850A
公开(公告)日：1985-06-04
申请号：US433569
申请日：1982-10-04
申请人： John E. Wilhite , William A. Shelly , Charles P. Ryan
发明人： John E. Wilhite , William A. Shelly , Charles P. Ryan
IPC分类号： G06F9/38 , G06F9/12
CPC分类号： G06F9/3804
摘要： Apparatus and method for providing an improved instruction buffer associated with a cache memory unit. The instruction buffer is utilized to transmit to the control unit of the central processing unit a requested sequence of data groups. In the current invention, the instruction buffer can store two sequences of data groups. The instruction buffer can store the data group sequence for the procedure currently in execution by the data processing unit and can simultaneously store data groups to which transfer, either conditional or unconditional, has been identified in the sequence currently being executed. In addition, the instruction buffer provides signals for use by the central processing unit defining the status of the instruction buffer.
摘要翻译：用于提供与高速缓冲存储器单元相关联的改进的指令缓冲器的装置和方法。指令缓冲器用于向中央处理单元的控制单元发送所请求的数据组序列。在本发明中，指令缓冲器可以存储两个数据组序列。指令缓冲器可以存储数据处理单元当前正在执行的过程的数据组序列，并且可以同时存储在当前执行的序列中已经被识别的有条件的或无条件的传输的数据组。此外，指令缓冲器提供由定义指令缓冲器的状态的中央处理单元使用的信号。

9. 发明授权

US4371927A Data processing system programmable pre-read capability 失效
标题翻译：数据处理系统可编程预读功能
公开(公告)号：US4371927A
公开(公告)日：1983-02-01
申请号：US131739
申请日：1980-03-20
申请人： John E. Wilhite , William A. Shelly , Charles P. Ryan
发明人： John E. Wilhite , William A. Shelly , Charles P. Ryan
IPC分类号： G06F9/30 , G06F9/308 , G06F9/38 , G06F12/08 , G06F9/28
CPC分类号： G06F9/30018 , G06F12/0862 , G06F9/30145 , G06F9/30167 , G06F9/3824 , G06F9/383 , G06F2212/6028
摘要： A data processing system includes a cache store to provide an interface with a main storage unit for a central processing unit. The central processing unit includes a microprogram control unit in addition to control circuits for establishing the sequencing of the processing unit during the execution of program instructions. Both the microprogram control unit and control circuits include means for generating pre-read commands to the cache store in conjunction with normal processing operations during the processing of certain types of instructions. In response to pre-read commands, the cache store, during predetermined points of the processing of each such instruction, fetches information which is required by such instruction at a later point in the processing thereof.
摘要翻译：数据处理系统包括高速缓存存储器，用于向中央处理单元提供与主存储单元的接口。中央处理单元除了用于在执行程序指令期间建立处理单元的排序的控制电路之外还包括微程序控制单元。微程序控制单元和控制电路都包括用于在处理某些类型的指令期间结合正常处理操作来向高速缓存存储器生成预读命令的装置。响应于预读命令，高速缓存存储器在每个这样的指令的处理的预定点期间，在其处理的稍后点获取这种指令所需的信息。

10. 发明授权

US06760811B2 Gateword acquisition in a multiprocessor write-into-cache environment 有权
标题翻译：在多处理器写入缓存环境中的门字获取
公开(公告)号：US06760811B2
公开(公告)日：2004-07-06
申请号：US10219644
申请日：2002-08-15
申请人： Wayne R. Buzby , Charles P. Ryan
发明人： Wayne R. Buzby , Charles P. Ryan
IPC分类号： G06F1200
CPC分类号： G06F12/0811 , G06F12/084
摘要： In a multiprocessor data processing system including: a memory, first and second shared caches, a system bus coupling the memory and the shared caches, first, second, third and fourth processors having, respectively, first, second, third and fourth private caches with the first and second private caches being coupled to the first shared cache, and the third and fourth private caches being coupled to the second shared cache, gateword hogging is prevented by providing a gate control flag in each processor. Priority is established for a processor to next acquire ownership of the gate control word by: broadcasting a “set gate control flag” command to all processors such that setting the gate control flags establishes delays during which ownership of the gate control word will not be requested by another processor for predetermined periods established in each processor. Optionally, the processor so acquiring ownership broadcasts a “reset gate control flag” command to all processors when it has acquired ownership of the gate control word.
摘要翻译：一种多处理器数据处理系统，包括：存储器，第一和第二共享高速缓存，耦合存储器和共享高速缓存的系统总线，第一，第二，第三和第四处理器分别具有第一，第二，第三和第四专用高速缓存，所述第一和第二专用高速缓冲存储器耦合到所述第一共享高速缓存，并且所述第三和第四专用高速缓冲存储器耦合到所述第二共享高速缓冲存储器，通过在每个处理器中提供门控制标志来防止门字锁定。为处理器建立下一个优先级，即：通过以下方式获得门控制字的所有权：向所有处理器广播“设置门控制标志”命令，使得设置门控制标志建立延迟，在该延迟期间不会请求门控制字的所有权由另一处理器在每个处理器中建立预定时段。可选地，获得所有权的处理器在获得了门控制字的所有权时向所有处理器广播“复位门控制标志”命令。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式