Hitachi, Ltd.

Kenichi Wada

Akira Yamaoka

Kazunori Kuriyama

Yooichi Shintani

Tsuguo Shimizu

Kanji Kubo

Naoki Yamada

Masahiro Hashimoto

Mamoru Hinai

Takahiro Yamada

Tsuyoshi Watanabe

Yoshihiro Takiyasu

Kuniaki Tabata

Takanori Miyamoto

Shigeki Morimoto

Takashige Kubo

Taminori Tomita

Chikahiko Izumi

Masami Yamagishi

Seiichi Ozaki

Tomoharu Maehara

Masahumi Hino

Tomoshisa Kohiyama

Makoto Noumi

Yoichi Shintani

A multimedia system capable of readily permitting expansion of its display memory as well as its compression/expansion function and other features as desired. The multimedia system is implemented by an information processing apparatus which includes a controller which controls the apparatus, an extended input/output bus for connecting with the controller, and interface device which interfaces with the extended input/output bus, at least one function extending device which extends the functionality of the controller and a local bus for permitting data exchanges between the interface device and the function extending device.

Motion image display apparatus

In a memory device shared among processors, a communication buffer having a size requested by the processing program of the origin of sending is dynamically secured. After the communication buffer has been secured, the send program writes a message to be conveyed to the receive program into the above described communication buffer and asks the send OS to perform sending. The send OS sends a communication ID having '1' set in the bit position corresponding to the receive program. On the basis of the above described bit position, the receive OS specifies a receive program and informs the receive program of that fact. The receive program reads a message from the communication buffer. Communication between the send program and the receive program is thus realized. A send instruction and a receive instruction respectively for exclusive use of sending and reading out a communication ID are prepared beforehand.

Multiprocessors including means for communicating with each other

A data processing system which has a plurality of sets of sub-systems, with each set including: a plurality of processors; a main storage; and a controller for controlling the transfer between at least each of the processors and the main storage. A shared storage apparatus is shared between the sub-systems to store exclusive control information, information on the processor-to-processor communications and an instruction to be transferred between the main storages and the shared storage apparatus when the information is accessed by each sub-system. The instruction designates a main storage address, a transfer data length and specified information on accessing the location of the shared storage apparatus and is decoded by the processors to that the main storage address is transferred to the main storage, whereas the specified information such as a data identifier and a relative address is transferred to the shared storage apparatus.

Multiprocessor system having subsystems which are loosely coupled

A level machine information system is provided with an instruction register for storing an address of an instruction read out from an instruction memory, an instruction decoder for producing control signals complying with the instruction read out, a calculator for producing a new address of an instruction to be executed next on the basis of the address outputted from the instruction register, an instruction length code supplied from the decoder as the control signals, a plurality of first registers for saying the address outputted from the instruction register, a plurality of second registers for saving the new address, a controller and a selector for supplying the instruction register with the new address, the saved new address and fixed addresses each indicating the head of interruption programs in the instruction memory. The first and second registers correspond to all but the highest interruption priority level. Upon an acknowledged interruption request, the controller controls the first and second registers and selector to save the new address of the calculator and the old address of the instruction register in one of the first and second registers corresponding to the priority levels of an interrupted process, respectively. One of the fixed addresses corresponding to a process to be executed for the interruption request is transferred in the instruction register. If necessary, the contents of the first and second registers are referenced by the process to detect the reasons for the interruption.

Information processing system with instruction address saving function

An instruction processor effecting operations for register operands and for processing branch instructions to perform address calculations for branch destination instructions, comprising general-purpose registers storing data including results of operations of said instruction processor, address adders calculating the address of branch destination instructions by using data read out from the general-purpose register and an ALU performing arithmetical or logical operations on the data read out from the general-purpose register in the decode cycle of the instructions. The result of the arithmetical or logical operation is inputted into the address adder but not from the general-purpose register, in the case where the result of the arithmetical or logical operation is utilized for address calculation in the execution of a succeeding instruction.

Instruction processor for processing branch instruction at high speed

Address calculation adders and a buffer storages are each independently provided for each operand of an instruction requiring two or more operands. In the translation instruction processing, the address calculations and operand fetch operations on the first and second operands are substantially asynchronously conducted. Consequently, the overhead that takes place one every n second operand fetch operations can be removed by independently and asynchronously performing the address calculations and operand fetch operations by use of a plurality of address adders. Moreover, the circuit for separating and obtaining a byte from the operand buffer can be dispensed with by adopting an operation procedure in which a byte of the first operand is fetched and is stored in temporary store means that supplies the address adder the data stored therein.

Pipelined instruction processor capable of reading dependent operands in

A data processing apparatus of processing first instruction of a type in which the result of operation of the first instruction is stored in at least one storage location designated by operands of the first instruction and second instruction of a type which succeeds to the first instruction and makes use of the result of operation of the first instruction as operand data. The apparatus comprises an OSC control circuit for detecting whether at least a part of the result of operation of the first instruction is to be used or not as the operand data for the second instruction, and an arithmetic unit for allowing the result of operation of the first instruction to be directly used as the operand data for the second instruction when the OSC control circuit detects the given condition is fulfilled.

Pipelined parallel data processing apparatus for directly transferring

In the pipeline data processing system, a branch instruction is preliminarily detected from a string of instructions to be executed before execution thereof and the branch address is calculated. When the branch address is thereafter changed due to a change in the content of the register used to calculate the branch address, the calculated branch address is invalidated and a new branch address is calculated and the branch is executed. The change of the branch address is checked and performed on a basis of the comparison between the fields used for address calculation and the field of the address into which the data is to be written.

Branch advanced control apparatus for advanced control of a branch

A method of multi-address communication in a loop network formed by connecting a plurality of nodes each provided with a data processor to one another through a transmission line in which, in order for a transmitting node having transmitted a multicast information frame to be able to confirm the delivery of the multicast information frame, the transmitting node transmits a multicast response frame for the delivery of confirmation of the multicast information frame, immediately after the transmission of the multicast information frame, and receiving nodes present information for indicating the reception of the multicast information frame at the address part of the multicast response frame, in the order of node arrangement in the loop network in the direction from the transmitting node toward the downstream thereof, to check the information by each of the receiving nodes and the transmitting node.

Method of multi-address communication

An information processing apparatus having a buffer register for pre-fetching a plurality of instructions and executing one instruction after another by reading them from the buffer registers, is provided with a first instruction decode start determination unit for register type instructions and a second instruction decode start determination unit for non-register type instructions, provided separately from the first unit, whereby 0.5 cycle after a register type instruction starts being decoded, or 1 cycle after a non-register type instruction starts being decoded, the next instruction starts to be decoded. By decoding a register type instruction at high speed, it becomes possible to execute a branch instruction at high speed.

Information processing apparatus

A decimal multiplier device including a register A storing the multiplier, a register B storing the multiplicand, a shifter for outputting the output of the register A as it is or after having been shifted, based on a first signal, a gate for outputting the output of the register B or '0', based on a second signal, an adder/subtractor for adding the output of the shifter and that of the gate and storing the result thus obtained in the register A, and a decoder for receiving the value of a selected digit of the content of the register A and controlling the gate and the shifter by generating the first signal and the second signal based on the received value so that the multiplicand B is added n times, n corresponding to the received value, to the content of the register A or substracted (10-n) times therefrom. The register A, the shifter and the adder/subtractor form a single loop. Decimal multiplication is performed by controlling the shifter, when signals pass through the loop repeatedly.

Decimal multiplier device and method therefor

A data processing system for executing an instruction in a plurality of stages in a pipeline mode comprises a main operation unit capable of executing all instructions to be executed in the data processing system, a pre-operation unit capable of executing instructions which occurs at a high frequency and can be executed with a small number of circuit components, general purpose registers for storing operation results of the instructions, and a control unit for controlling the writing of the operation results by the main operation unit and the pre-operation unit into the general purpose registers.

Data processing system

This invention employs a construction in which subtraction processing and digit shift processing in decimal division are carried out in parallel with each other to shorten the time required for decimal division. A dividend is stored in a register B and a divisor, in a register C. A selector 6 selects register B when the result of subtraction by an adder/subtracter 1 is positive or zero, and selects register A at other times. Both adder/subtracter 1 and a shifter 2 receive the signal from the selector 6 in the same way, and execute the subtraction processing and the shift processing, respectively. The results of these processings are stored in the registers B and A', respectively. The division time can be shortened because the adder/subtracter 1 and the shifter 2 can be actuated simultaneously.

Coded decimal non-restoring divider

An arithmetic system includes an arithmetic unit of a pipeline structure for executing arithmetic operations for instructions which require different arithmetic cycles. The arithmetic unit executes N arithmetics in pipeline for N instruction at maximum. Initiation of arithmetic operation for a new instruction in the arithmetic unit is indicated by an indicator which detects that each of the instruction executed in the arithmetic is N cycles before completion of the execution and allows arithmetic operation for the new instruction to be initiated in the succeeding cycle.

Arithmetic system having pipeline structure arithmetic means

In a storage hierarchy system, a part of data stored in a main storage in held as a copy by a buffer storage of a smaller capacity and higher speed than the main storage. A processor fetches data from the buffer storage or stores data in the buffer storage at a high speed. A buffer storage control system includes a first buffer directory and a second buffer directory. The first buffer directory and the buffer storage are accessed by an address for a fetch request, while the second buffer directory and the buffer storage are accessed by an address for store request, whereby competition for the access to the buffer storage between the fetch and store operations is reduced.

Growing community of inventors

Zama, Japan

Kenichi Wada