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
    • 4. 发明专利
    • CONTROL SYSTEM FOR CACHE MEMORY
    • JPS6437640A
    • 1989-02-08
    • JP19389987
    • 1987-08-03
    • MITSUBISHI ELECTRIC CORP
    • KODAMA TAKASHI
    • G06F12/08
    • PURPOSE:To set up block size for maximizing the performance of a system by providing the titled system with a switching means for switching the block size of a cache memory. CONSTITUTION:Data to be complicatedly accessed from a processing part 1A to a main storage device 1B are temporarily stored in the cache memory 1C of a cache part 11D and the data are successively updated in each block so that the data to be complicatedly accessed by the processing part 1A are always stored. In case of accessing data, the processing part 1a checks whether data exist in the memory 1C or not, and when the data exist, the memory 1C is accessed. At the time of absence of data, the main storage device 1B is accessed. The control part 1D in the cache part 11D is provided with the switching means 10D for switching the block size of the memory 1C and the block size for maximizing the performance of the system is set up in the switching means 10D.
    • 6. 发明专利
    • CACHE MEMORY DEVICE
    • JPH0421048A
    • 1992-01-24
    • JP12363890
    • 1990-05-14
    • MITSUBISHI ELECTRIC CORP
    • KODAMA TAKASHI
    • G06F12/12
    • PURPOSE:To effectively utilize the memory bus, and also, to omit the drawback register group by executing alternately a data transfer from other memory based on a request address and a transfer of replace data to other memory, when a dirty flag of the replace data is turned on. CONSTITUTION:A comparing part 107 detects a mishit by comparing a request address with a directory 104, a cache controller 108 inspects an address part of a replace block due to a fact that the comparing part detects the mishit, and also, when a dirty flag of this address part is turned on, a data transfer from other memory based on a request address and a transfer of replace data to other memory are executed alternately. Subsequently, an address generating part 109 outputs each address of a data transfer request from other memory and a transfer request of the replace data to other memory. In such a way, memory buses 21, 22 are utilized effectively, and also, a drawback register group is omitted.
    • 7. 发明专利
    • BRANCH PREDICTING ALGORITHM
    • JPH038023A
    • 1991-01-16
    • JP14328189
    • 1989-06-06
    • MITSUBISHI ELECTRIC CORP
    • KODAMA TAKASHI
    • G06F9/38
    • PURPOSE:To increase a branch predicting condition and to improve the rate of success for predicting a branch by providing more than three kinds of branch status for a branch predicting algorithm and increasing the history of the branch. CONSTITUTION:The branch is expressed by T and a negative branch is expressed by N. For example, in the case of NTNT, the branch and negative branch are repeated. In such a case, since it can be predicted that the branch and negative branch are repeated in following operation, such a state is predicted as N. The case of TNTN is samely predicted as T. Next, in the case of NNNT, concerning a branch instruction after arithmetic calculation, since the instruction is defined to execute the branch only when a digit flows over, such a state is predicted as N. In the case of TTTN, when a counter is not 0, such a state can be predicted as T and in the case of NNTN, such a state can be predicted as N similary as the case of NNNT. When it is failed to predict NTTN and TNNT, NTTN is transited to TTNN. Accordingly, the next state can be predicted as N and accuracy for predicting is improved in the following operation. Samely, since TNNT is transited to NNTT, the next state is predicted as T and the accuracy for predicting the following branch is improved.
    • 8. 发明专利
    • Logical circuit
    • 逻辑电路
    • JPS6120424A
    • 1986-01-29
    • JP14026684
    • 1984-07-06
    • Mitsubishi Electric Corp
    • KODAMA TAKASHI
    • H03K19/0175H03K19/00H03K19/01
    • H03K19/00
    • PURPOSE:To process a signal at a high speed by generating in advance by a logical operation the first output signal in case when it is assumed that logic of a binary signal is ''0'', and the second output signal in case when it is assumed that logic of the binary signal is ''1'', with respect to the binary signal which is delayed and inputted. CONSTITUTION:In logical circuits 1a, 1b, in the circuit 1a, a signal of logic ''0'' and logic ''L'' is connected to a terminal 4a corresponding to a terminal 4 to which a signal C is inputted by a circuit 1, and in the circuit 1b, a signal of logic ''1'' and logic ''H'' is connected to a terminal 4b. Accordingly, after a processing time tau from a time point t0 when signals A and B have arrived, output signals of terminals 5a, 5b are settled and inputted to a selector 6. Therefore, at the time point t1 when the signal C has arrived, one of the signals from the terminal 5a or 5b is outputted as an output signal D, therefore, the time between the input time point t0 of the signals A, B and the output time point t1 of the signal D goes to t1-t0, and it is shortened comparing with the conventional (t1-t0)+tau.
    • 目的:通过在假设二进制信号的逻辑为“0”的情况下通过逻辑运算产生第一输出信号来高速处理信号,而在第二输出信号 假设二进制信号的逻辑对于被延迟和输入的二进制信号是“1”。 构成:在逻辑电路1a,1b中,在电路1a中,将逻辑“0”和逻辑“L”的信号连接到对应于输入信号C的端子4的端子4a 电路1中,并且在电路1b中,将逻辑“1”和逻辑“H”的信号连接到端子4b。 因此,在信号A,B到达时间点t0的处理时间τ以后,端子5a,5b的输出信号被置位并输入到选择器6.因此,在信号C到达时刻t1, 输出端子5a或5b的信号之一作为输出信号D,因此信号A,B的输入时刻t0与信号D的输出时刻t1之间的时间变为t1-t0, 并且与常规(t1-t0)+ tau相比缩短。
    • 10. 发明专利
    • PARALLEL PROCESSING SYSTEM
    • JPH0475159A
    • 1992-03-10
    • JP18899790
    • 1990-07-17
    • MITSUBISHI ELECTRIC CORP
    • KODAMA TAKASHI
    • G06F15/16G06F9/38G06F15/177G06F15/80
    • PURPOSE:To efficiently perform parallel processing by dividing an instruction group into several blocks in unit of column, and performing the sorting of the block representing the maximum value and that of the block representing the minimum value out of execution predictive count information on a pre-fetch string alternately. CONSTITUTION:The processing elements 2a-2e of an arithmetic processing part provided at a parallel computer and instruction queues 4a-4c which supply instructions to the elements are provided. Sort circuits 5a-5e, 6a-6e which attach the execution predictive count information of the instruction on an instruction field part 30 and furthermore, re-arrange the instruction group based on the execution predictive count information are provided. At this time, the instruction group is divided into the several blocks in unit of column as pre-fetch processing, and the sorting of the block representing the maximum value and that of the block representing the minimum value out of the execution predictive count information are performed on the pre-fetch string alternately by the sort circuits 5a-5e, 6a-6e. Thereby, the instruction group can be re-arranged so as to average instruction execution time, and the overhead of processing can be reduced, and the parallel processing with high efficiency can be executed.