Fujitsu Corporation

Shigeki Itou

Yasuhiko Nakano

Yuta Toyoda

Ryuji Kan

Takumi Maruyama

Noriko Itani

Instruction-execution processors each execute a first instruction. A control processor converts a second instruction to be emulated into the first instruction, and enters the converted first instruction into the instruction-execution processors. In a parallel-execution period, each instruction-execution processor executes a writing-access instruction or a reading-access instruction to a memory, suspends writing of data into the memory caused by the writing-access instruction, and retains an execution history of the writing-access instruction and the reading-access instruction. The control processor selects one of instruction-execution processors in which addresses of the memory access instructions conflict with each other, causes the selected instruction-execution processor to complete the writing of the data into the memory, enters first instructions to be executed in a next parallel-execution period into the selected instruction-execution processor, and causes another instruction-execution processor to re-execute first instructions executed by the other instruction-execution processors in the parallel-execution period, in the next parallel-execution period.

Apparatus and method to emulate a sequence of instructions via parallel processors

A technique realizes execution of various combinations of arithmetic operations in, for example, SIMD floating-point multiply-add arithmetic operation, with less instruction kind codes. An arithmetic operating apparatus sets, in one or more unused bits of a single arithmetic instruction, particular instruction information to instruct at least one of arithmetic operators to perform a process different from an ordinary process.

Apparatus and method for processing an instruction that selects between single and multiple data stream operations with register specifier field control

A data compression apparatus includes a memory and a processor. The processor extracts a second character string as a matching string from a character string after a first character string in a character string of data before compression that is stored in the memory, the second character string being identical with the first character string, and identifies a length of the matching string, and a relative position indicating how many addresses the first character string precedes the second character string by. The processor extracts a third character string having a length that is less than the relative position from the extracted second character string. The processor encodes a length of the third character string. The processor encodes the relative position.

Data compression apparatus and method

An information processing device includes a plurality of barrier banks, and one or more processors including at least one of the plurality of barrier banks. Each of barrier banks includes one or more hardware threads and a barrier synchronization mechanism. The barrier synchronization mechanism includes a bottom unit having a barrier state, and a bitmap indicating that each of the one or more hardware threads has arrived at a synchronization point, and a top unit having a non-arrival counter indicating the number of barrier banks yet to be synchronized. The bottom unit notifies of bottom unit synchronization completion when all the one or more hardware threads have arrived at a barrier synchronization point. The non-arrival counter decrements its value by 1 upon receipt of the bottom unit synchronization completion, and the top unit sets the barrier state to a value indicating synchronization completion when the non-arrival counter decrements to 0.

Growing community of inventors

Kawasaki, Japan

Shigeki Itou