Abstract: Method and apparatus for instruction restart processing in a microprogram-controlled data processing apparatus, wherein, in restarting an instruction execution after instruction suspension, the internal information of the data processing apparatus at the time of instruction execution suspension is saved in a memory, and after a suspension cause removal process performed the saved internal information is recovered. A check point address associated with the address of a currently executing microprogram is stored in accordance with a designation by the microprogram. After a suspension causes removal process is performed, the execution of the instruction restarts using the check point address. If a check point address has not been stored after the suspension cause removal process is performed, the execution of the instruction restarts from a read operation of the suspended instruction from the main storage.

Abstract: Method and apparatus for instruction restart processing in a microprogram - controlled data processing apparatus, wherein, in restarting an instruction execution after instruction suspension, the internal information of the data processing apparatus at the time of instruction execution suspension is saved in a memory, and after a suspension cause removal process performed the saved internal information is recovered. A check point address associated with the address of a currently executing microprogram is stored in accordance with a designation by the microprogram. After a suspension causes removal process is performed, the execution of the instruction restarts using the check point address. If a check point address has not been stored after the suspension cause removal process is performed, the execution of the instruction restarts from a read operation of the suspended instruction from the main storage.

Abstract: A data processor for execution of tagged data and tagless data has a decoder for discriminating whether the data is tagged or tagless one and in case of a tagged data, separates a tag part and uses the remaining part for address computation. The data processor also comprises a unit for evaluating the tag part and a micro program controller for multi-branching in accordance with the evaluation result of the tag part. The tag evaluating unit includes an extender eliminating part for extracting the tag part from data on a data bus, a plurality of tag part storing registers for storing the tag part from the eliminating part under the control of the micro program controller, and a tag multi-way jump encoder for generating a tag multi-way jump address to feed it to the controller on the basis of the outputs of the registers and a signal from the micro program controller, thereby enabling tag multi-way jump.

Abstract: An operation unit capable of performing round processing at a high speed in a floating point operation. A circuit for detecting an overflow on the condition of a signal representing all 1's in an output of a mantissa shifter and a signal representing round-up, a carry look-ahead circuit and a circuit for generating a round precision signal are provided. When the overflow takes place, the mantissa is produced as "1". The operation unit is compatible to single, double and extended precisions recommended by Institute of Electrical and Electronic Engineers (IEEE).

Abstract: A bit slice multiplication circuit operating to slice a multiplier, produce products for the sliced multipliers and a multiplicand and sum the products to obtain the multiplication result. The circuit includes a slicing unit for slicing the multiplicand, multiplying units corresponding in number to the number of sliced multiplicands, and adding units provided in correspondence to the multiplying units and implementing summation for multiplication results from corresponding multiplying units while shifting the sliced portions of the multiplicand at each multiplying operation for sliced multipliers and multiplicands by the multiplying units, the multiplication result being obtained by summing all summation results produced by the adding units.

Abstract: A microprogram controlled data processing apparatus executes multi-operand instructions in which one or more operand specifiers are provided for specifying the addressing for each operand independently from the operation code of the instruction. An instruction execution unit receives a top address of a microprogram from a decoding unit, a ready status signal and a signal from the decoding unit indicating whether a destination of an operand is in a general purpose register or in a memory unit, and writes an operand into a destination address of a register on the memory unit under control of a microprogram. Because the destination of the operand is indicated by the instruction decoding unit, it is not necessary to determine this information by microinstruction execution, with the result that execution of the instruction can be performed at high speed.

Abstract: A pipelined adder for adding or subtracting two floating point input data each expressed by a sign data, an exponent data and a mantissa expressed in a sign-magnitude format, in accordance with an external operation mode designation signal to produce a floating point sum or difference data in a sign-magnitude format. In a first stage of the adder, the magnitudes of the exponent data of the input data are compared by a subtractor or a comparator and the magnitudes of the mantissa data of the input data are compared by a subtractor or a comparator. An actual operation mode for the mantissa data of the input data is determined, on the basis of the compare results of the exponent data and the mantissa data and the external operation mode designation signals, so that the operation result data is always expressed in a sign-magnitude format.

Abstract: A central processing unit for executing instructions of variable length in which an operand specifier for specifying the addressing mode of an operand is independent of an operation code for ascertaining the kind of an operation and the number of operands. Each operand specifier is formed of one or more data bytes, and has a stop bit flag indicating whether or not the particular operand specifier is the last operand specifier. By adding the stop bit flag, the operand specifier can be shared, and different processing can be executed with an identical operation code. In a case where, when operation code decoding means has provided an output indicative of the last operand, the stop bit flat is not detected in the corresponding operand specifier, the corresponding instruction is detected as an error.

Abstract: An input/output control device stores variable-length data in a memory device at a high storage efficiency and without reducing the speed of data processing. The data stored in a memory are read out in the form of data of a fixed word length and then processed, the data having been processed are stored in another memory in the form of data of the fixed word length. The data stored in another memory are subjected to data organization to be outputted in the form of data of a given word length. Each of the memories is divided into a plurality of regions, and each region stores therein data of the same word length, respectively.

Abstract: In a multiprocessor system having a main memory and a plurality of processors connected through common address bus, data bus and answer bus for data transfer, a data transmission apparatus is provided for each of the main memory and the processors and includes bus request control lines for transferring bus request signals and bus control signals, and a bus controller for separately controlling selections of the address bus, the data bus and the answer bus in response to the signals on the bus request control lines and the request signal. Overlapped processing such as data write and data write answer or data read and data read answer in one cycle is possible.

Abstract: A data processing system for supporting a virtual memory is disclosed. Prior to the start of main memory write operation, a processor checks to see if a store buffer has a vacant area to store data to be written into a main memory to execute a current instruction. If a page fault occurs during the main memory write operation, the processor continues to store the subsequent write data for the current instruction and the corresponding virtual or logical addresses in the store buffer to complete execution of the current instruction.

Abstract: A virtual storage data processing system having an address translation unit shared by a plurality of processors, located in a memory control unit connected to a main memory is disclosed. One of the plurality of processors is a job processor which accesses the main memory with a virtual address to execute an instruction and includes a cache memory which is accessed with a virtual address. One of the plurality of processors is a file processor which accesses the main memory with a virtual address to transfer data between the main memory and an external memory. The cache memory receives the virtual address when the file processor writes to the main memory and if it contains a data block corresponding to the virtual address, it invalidates the corresponding data block. The address translation unit translates the address differently for the access from the file processor and the accesses from other processors.

Abstract: A data processing unit for executing variable length instructions in which operand specifiers for specifying addressing modes of operands are independent from operation codes for ascertaining operations is disclosed. An instruction fetch unit includes an instruction buffer for prefetching and retaining instructions from a memory and alignment means for aligning the instructions from the instruction buffer such that the instruction includes at least one operand specifier in one machine cycle, and provides it to a decoding unit. The decoding unit includes an operation code decoder and two operand specifier decoders to decode two operand specifiers simultaneously when the last operand specifier is a register designation mode. Each of the units executes instructions in a pipelined fashion and processes operands in a pipelined fashion.