Fujitsu Corporation

Hidenobu Ohta

Taizo Sato

Tatsumi Nakada

Masanobu Yuhara

Kenji Ijitsu

Masahiro Yanagida

Shuntaro Fujioka

Shinichi Moriwaki

An integrated circuit includes a test pattern input unit for inputting a test pattern into cells, a clock distributing unit for distributing a clock signal to the cells, a first cell that receives the clock signal distributed by the clock distributing unit, a second cell that receives the clock signal after the clock signal is received by the first cell, a data transfer unit for transferring a data signal from the first cell to the second cell, a clock transfer unit for distributing the clock signal to the first cell and the second cell and transferring the clock signal in the same direction as the transfer direction of the data signal, and a failure detecting unit for inputting the test pattern into the cells and detecting failures of the cells on the basis of the results of the test pattern output from the cells.

Clock signal distributing circuit, information processing device and clock signal distributing method

A pipeline operation method and a pipeline operation device in which an operation result of an operation unit can be effectively written to a register. In the pipeline operation method and the pipeline operation device, a pipeline operation unit that can perform a pipeline operation, a non-pipeline operation unit that cannot perform a pipeline operation, and a register that is shared by the pipeline operation unit and the non-pipeline operation unit are arranged. To perform an operation while an operation result of each of the pipeline units is being written into the register, translating an instruction to the pipeline operation unit is interlocked when the writing of the operation result of the pipeline operation unit overlaps with the writing of the operation result of the non-pipeline operation unit. The pipeline operation method and the pipeline operation device are applicable to a computing system where plural operation units including pipeline operation units and non-pipeline operation units share one writing port to perform an arithmetic operation.

Pipeline operation method and pipeline operation device to interlock the translation of instructions based on the operation of a non-pipeline operation unit

The invention provides a data processing apparatus wherein, when a trouble of a built-in RAM is detected, stopping of the system by a comparison check of the outputs of multiple CPUs is prevented and the trouble of the built-in RAM can be removed. The data processing apparatus includes a built-in RAM error detection section for detecting that an error occurs in a built-in RAM of any of processing sections, and an inhibition section for inhibiting, when the built-in RAM error detection section detects that a built-in RAM error occurs in at least one of the processing sections, the result of comparison outputted from the comparison section originating from the occurrence of the built-in RAM error. The data processing apparatus can be applied to various computer systems wherein multiple processing sections (CPUs) perform same operation and processing is performed while the outputs of the processing sections are compared with each other to confirm that the processing sections are performing same operation.

Data processing apparatus

A cache uses A bits of an offset portion which are not subjected to the address translation of the logical address and B bits of the portion other than the offset portion, which are subjected to an address translation. It has an address monitor portion having a tag portion corresponding to the tag portion of the CPU using only A bits of the offset portion of the set address which are used as the set address in the cache and having a 2.sup.B .times.N-way set associative structure and a portion for making said tag portion of the cache correspond to said tag portion of the address monitor portion, thereby performing management of N address stored in the tag portion of the address monitor portion and transmitting the result of the management of the address to the cache and for invalidating the corresponding recording portion of the tag in the cache.

System for controlling an internally-installed cache memory

An internal address bus, which is connected between a bus control unit and a memory control unit, is formed by a bidirectional bus. The bidirectional internal address bus is connected to an external address transferring bus which is used to transfer a write address in a system bus to the memory control unit. The bidirectional internal address bus is commonly used for renewal and invalidation of first and second buffers for an instruction cache and operand cache. Therefore, in the present invention, the number of the internal bus lines can be reduced, a control constitution can be simplified, and a consistency of contents in a plurality of internal buffers and the main memory can be easily maintained.

Growing community of inventors

Kawasaki, Japan

Hidenobu Ohta