Abstract: The present disclosure provides an operation apparatus operating based on the Winograd algorithm for multiplying a first matrix by a second matrix to generate a third matrix, including a plurality of second accumulated value calculation units, of which one second accumulated value calculation unit is configured to accumulate second multiplication values obtained by multiplying each of paired element values of the second matrix, a second accumulated value output unit outputting selecting and outputting one of output values of adjacent second accumulated value calculation unit and an accumulated second multiplication value as a second accumulated value, a third accumulated value output unit including a plurality of third accumulated value calculation units and generating third accumulated value, and one or more row element value calculation units, of which one row element value calculation unit is configured to accumulate first matrix element multiplication values obtained by multiplying each of the paired elem

Abstract: A swizzle pattern generator is provided to reduce an overhead due to execution of a swizzle instruction in vector processing. The swizzle pattern generator is configured to provide swizzle patterns with respect to data sets of at least one vector register or vector processing unit. The swizzle pattern generator may be reconfigurable to generate various swizzle patterns for different vector operations.

Abstract: A swizzle pattern generator is provided to reduce an overhead due to execution of a swizzle instruction in vector processing. The swizzle pattern generator is configured to provide swizzle patterns with respect to data sets of at least one vector register or vector processing unit. The swizzle pattern generator may be reconfigurable to generate various swizzle patterns for different vector operations.

Abstract: An apparatus for generating a test case includes a constrained description generator configured to define a plurality of constrained verification spaces in a total verification space, and generate a constrained description for each of the plurality of constrained verification spaces; and a test case generator configured to generate a test case using the constrained description.

Abstract: A latency management apparatus and method are provided. A latency management apparatus for a multiprocessor system having a plurality of processors and shared memory, when the shared memory and each of the processors is configured to generate a delayed signal, includes a delayed signal detector configured to detect the generated delayed signal; and one or more latency managers configured to manage an operation latency of any one of the processors upon detection of the delayed signal.

Abstract: A computing apparatus is provided. The computing apparatus includes a memory unit configured to have an address space defined as a multidimensional space having at least two axes, and a memory access unit configured to include a first pointer register storing a first pointer pointing to a row corresponding to the first axis and a second pointer register storing a second pointer pointing to a column corresponding to the second axis.

Abstract: A coarse-grained reconfigurable processor having an improved code compression rate and a code decompression method thereof are provided to reduce a capacity of a configuration memory and reduce power consumption in a processor chip. The coarse-grained reconfigurable processor includes a configuration memory configured to store reconfiguration information including a header storing a compression mode indicator and a compressed code for each of a plurality of units and a body storing at least one uncompressed code, a decompressor configured to specify a code corresponding to each of the plurality of units among the at least one uncompressed code within the body based on the compression mode indicator and the compressed code within the header, and a reconfigurator including a plurality of PEs and configured to reconfigure data paths of the plurality of PEs based on the code corresponding to each unit.

Abstract: A method of compiling a program to be executed on a multicore processor is provided. The method may include generating an initial solution by mapping a task to a source processing element (PE) and a destination PE, and selecting a communication scheme for transmission of the task from the source PE to the destination PE, approximately optimizing the mapping and communication scheme included in the initial solution, and scheduling the task, wherein the communication scheme is designated in a compiling process.

Abstract: An apparatus for generating a test case includes a constrained description generator configured to define a plurality of constrained verification spaces in a total verification space, and generate a constrained description for each of the plurality of constrained verification spaces; and a test case generator configured to generate a test case using the constrained description.

Abstract: Provided is a multi-port cache memory apparatus and a method of the multi-port cache memory apparatus. The multi-port memory apparatus may divide an address space into address regions and allocate the divided memory regions to cache banks, thereby preventing the concentration of access to a particular cache.

Abstract: An apparatus and method for supporting a multi-mode. The apparatus for supporting a multi-mode may include an instruction distributor configured to select, according to a current execution mode, at least one instruction from among a plurality of received instructions that each include an operand and an opcode, and transfer the opcode included in each of at least one selected instruction to the plurality of functional units; an operand switch controller configured to generate, based on the operand included in each of the selected at least one instruction, switch configuration information for routing in order to execute the selected at least one instruction; and an operand switch configured to route, based on the switch configuration information, a functional unit output or a register file output to either a functional unit input or a register file input.

Abstract: A processor for batch thread processing includes a central register file, and one or more function unit batches each including two or more function units and one or more ports to access the central register file. The function units of the function unit batches execute an instruction batch including one or more instructions to sequentially execute the one or more instructions in the instruction batch.

Abstract: A simulation apparatus and method for a multicore system are provided. The simulation apparatus may prevent the occurrence of a data collision during the communication between modules and may reduce the overhead generated during simulation. The simulation apparatus may select a plurality of modules to be synchronized in terms of function execution timing based on timing information and may configure a multicore system architecture model using the selected modules. The simulation apparatus may acquire function execution timing information of the modules, control the execution of functions by the modules based on the acquired function execution timing information, and output the results of the control of the execution of functions by the modules.

Abstract: A routing apparatus and a network apparatus that are capable of improving general system performance by compressing/decompressing data and transmitting the result of the compression/decompression, are provided. The routing apparatus may compress and/or decompress input data, and may transmit the compressed and/or decompressed input data.

Abstract: A method of compiling a program to be executed on a multicore processor is provided. The method may include generating an initial solution by mapping a task to a source processing element (PE) and a destination PE, and selecting a communication scheme for transmission of the task from the source PE to the destination PE, approximately optimizing the mapping and communication scheme included in the initial solution, and scheduling the task, wherein the communication scheme is designated in a compiling process.

Abstract: A coarse-grained reconfigurable processor having an improved code compression rate and a code decompression method thereof are provided to reduce a capacity of a configuration memory and reduce power consumption in a processor chip. The coarse-grained reconfigurable processor includes a configuration memory configured to store reconfiguration information including a header storing a compression mode indicator and a compressed code for each of a plurality of units and a body storing at least one uncompressed code, a decompressor configured to specify a code corresponding to each of the plurality of units among the at least one uncompressed code within the body based on the compression mode indicator and the compressed code within the header, and a reconfigurator including a plurality of PEs and configured to reconfigure data paths of the plurality of PEs based on the code corresponding to each unit.

Abstract: There is provided a vector processing apparatus and method allowing for the parallel processing of a plurality of different instructions while maintaining vector processing architecture. The vector processing apparatus includes an instruction memory storing a multiple instruction group including one or more instructions; an instruction fetch unit reading the multiple instruction group from the instruction memory; and a plurality of instruction processing units each receiving the multiple instruction group through the instruction fetch unit, selecting a single instruction from the multiple instruction group according to a previous arithmetic result, and performing a arithmetic operation.

Abstract: A video decoding apparatus and method based on a data and function splitting scheme are disclosed. The video decoding apparatus based on a data and function splitting scheme includes a variable length decoding unit performing variable length decoding and parsing on a bit stream to acquire residual data and a decoding parameter, and splitting the residual data and the decoding parameter by row; and N (N is a natural number of 2 or larger) number of clusters splitting dequantization and inverse discrete cosine transform (IDCT), motion vector prediction, intra prediction and motion compensation, video restoration, and deblocking function into M number of functions, acquiring the residual data, the decoding parameter, and macroblock (MB) processing information of an upper cluster by column, and splitting the information acquired by column into M number of functions to process the same.

Abstract: Provided are a network load reducing method and a node structure for a multiprocessor system with a distributed memory. The network load reducing method uses a multiprocessor system including a node having a distributed memory and an auxiliary memory storing a sharer history table. The network load reducing method includes recording the history of a sharer node in the sharer history table of the auxiliary memory, requesting share data with reference to the sharer history table of the auxiliary memory, and deleting share data stored in the distributed memory and updating the sharer history table of the auxiliary memory.

Abstract: A core cluster includes a cache memory, a core, and a cluster cache controller. The cache memory stores and provides instructions and data. The core accesses the cache memory or a cache memory provided in an adjacent core cluster, and performs an operation. The cluster cache controller allows the core to access the cache memory when the core requests memory access. The cluster cache controller allows the core to access the cache memory provided in the adjacent core cluster when the core requests a clustering to the adjacent core cluster. The cluster cache controller allows a core provided in the adjacent core cluster to access the cache memory when the core receives a clustering request from the adjacent core cluster.