Abstract: A non-transitory computer-readable recording medium stores an arithmetic processing program for causing a computer to execute a process including: setting, in a mask register used for a mask operation, to each of a plurality of mask bits that indicates a bit corresponding to each element of each row of a sparse matrix, each mask pattern for designating the mask operation; and expanding the plurality of mask bits to which the respective mask patterns are set to different areas of a physical register, respectively.

Abstract: An arithmetic processing device including: request issuing units configured to issue an access request to a storage; and banks each of which includes: a first cache area including first entries; a second cache area including second entries; a control unit; and a determination unit that determines a cache hit or a cache miss for each of the banks, wherein the control unit performs: in response that the access requests simultaneously received from the request issuing units make the cache miss, storing the data, which is read from the storage device respectively by the access requests, in one of the first entries and one of the second entries; and in response that the access requests simultaneously received from the request issuing units make the cache hit in the first and second cache areas, outputting the data retained in the first and second entries, to each of issuers of the access requests.

Abstract: An arithmetic processing device includes one or more lanes configured to execute at most a single element operation of an instruction for each cycle, and an element operation issuing processor configured to issue the element operations to the one or more lanes. Each lane is separated into a plurality of sections by a buffer that has a plurality of entries. While the one or more sections that are not able to continue processing of the element operations stop the processing, another section stores an element operation that proceeds to each downstream section in an immediately subsequent buffer and continues processing, and at a horizontal addition processing, a lane in which addition results are finally aggregated is set to be variable, and a target lane that waits for synchronization is set to be a lane adjacent to its own lane.

Abstract: An information processor includes a memory; and a processor coupled to the memory and the processor configured to: acquire first statistical information about distribution of most significant bit position that is not a sign or least significant bit position that is not zero for each of a plurality of first fixed-point number data, the data being a computation result of the computation in the first layer; execute computation on a plurality of output data of the first layer according to a predetermined rule, in the computation in the second layer; and acquire second statistical information based on the predetermined rule and the first statistical information, and determine a bit range for limiting a bit width when a plurality of second fixed-point number data, the data being a computation result of the computation in the second layer, are stored in a register, based on the second statistical information.

Abstract: An arithmetic processing apparatus includes: a first determiner that determines, when a given learning model is repeatedly learned, an offset amount for correcting a decimal point position of fixed-point number data used in the learning in accordance with a degree of progress of the learning; and a second determiner that determines, based on the offset amount, the decimal point position of the fixed-point number data to be used in the learning. This configuration avoids lowering of the accuracy of a learning result of a learning model.

Abstract: An arithmetic processing device including: request issuing units configured to issue an access request to a storage; and banks each of which includes: a first cache area including first entries; a second cache area including second entries; a control unit; and a determination unit that determines a cache hit or a cache miss for each of the banks, wherein the control unit performs: in response that the access requests simultaneously received from the request issuing units make the cache miss, storing the data, which is read from the storage device respectively by the access requests, in one of the first entries and one of the second entries; and in response that the access requests simultaneously received from the request issuing units make the cache hit in the first and second cache areas, outputting the data retained in the first and second entries, to each of issuers of the access requests.

Abstract: An information processing apparatus includes a processor including a first operation circuit that executes a product-sum operation, a second operation circuit that executes a certain operation, and a resister. The processor executes a first operation including the certain operation in a first layer in a neural network. The processor executes the first operation by a second method of calculating the certain operation by the second operation circuit, in a case where second operation time necessary for the first operation when the certain operation is executed by the second operation circuit is less than memory transfer time. Or the processor executes the first operation by a first method of calculating the certain operation by an approximate calculation by the first operation circuit, in a case where first operation time necessary for the first operation when executed by the first method is less than the memory transfer time.

Abstract: A non-transitory computer-readable recording medium stores therein a program for causing a computer to execute a process for, in repeatedly training a given training model, repeatedly training the training model a given number of times by using a numerical value of a floating-point number, the numerical value being a parameter of the training model or training data of the training model, or any combination thereof; and, after the training by using the numerical value of the floating-point number, repeatedly training the training model by using a numerical value of a fixed-point number corresponding to a numerical value of the floating-point number obtained by the training.

Abstract: An operation processing device includes: a memory; and a processor coupled to the memory and configured to: acquire statistical information on distribution of bits in fixed point number data after execution of an instruction on the fixed point number data; and update a decimal point position of the fixed point number data.

Abstract: An arithmetic processing device includes a memory and a processor coupled to the memory. The processor configured to calculate statistical information of a first operation result by executing the predetermined operation using input data as a first fixed-point number with a first decimal point at a first decimal point position, determine a second decimal point position using the statistical information, and calculate a second operation result when the predetermined operation is executed using the input data as a second fixed-point number with a second decimal point at the second decimal point position.

Abstract: An apparatus for data output control includes: an encryption executing circuit configured to receive first data from a processor with a control signal indicating whether the first data is to be encrypted, and encrypt the first data when the control signal indicates that the first data is to be encrypted; a selection circuit configured to output any of the encrypted first data and second data; and an output control unit configured to set a frequency of second timing to be smaller than a frequency of first timing, and transmit a signal to the selection circuit instructing that the second data be outputted at the second timing, in a case where the second data is received from the processor.

Abstract: An arithmetic processing device includes a memory and a processor coupled to memory. The processor configured to acquire a first operation result of a first operation executing by using a candidate decimal point position, determine a specific decimal point position determined based on statistical information of the first operation result, and acquires, as a final operation result, either the first operation result or a second operation result of a second operation executing by using the specific decimal point position, based on the candidate decimal point position and the specific decimal point position.

Abstract: A simulation method performed by a computer for simulating operations by a plurality of cores based on resource access operation descriptions on the plurality of cores, the method includes steps of: extracting a resource access operation description on at least one core of the plurality of cores by executing simulation for the one core; and, under a condition where the one core and a second core among the plurality of cores have a specific relation in execution processing, generating a resource access operation description on the second core from the resource access operation description on the one core by reflecting an address difference between an address of a resource to which the one core accesses and an address of a resource to which the second core accesses.

Abstract: An arithmetic processor includes a plurality of arithmetic circuits that individually execute an arithmetic operation for fixed point data; and at least one of first and second statistical information is acquired regarding a plurality of fixed point data that are results of arithmetic operation executed by the plurality of arithmetic circuits. The first statistical information is obtained by accumulating a bit pattern, which is obtained by setting a flag bit to each of bit positions corresponding to a range from a least-significant-bit position to a highest-order bit position for each of the digits corresponding to the bit positions, and the second statistical information is obtained by accumulating a bit pattern, which is obtained by setting a flag bit to each of bit positions corresponding to a range from the position of the sign bit to a lowest-order-bit position for each of the digits corresponding to the bit positions.

Abstract: An arithmetic processing apparatus includes: a first determiner that determines, when a given learning model is repeatedly learned, an offset amount for correcting a decimal point position of fixed-point number data used in the learning in accordance with a degree of progress of the learning; and a second determiner that determines, based on the offset amount, the decimal point position of the fixed-point number data to be used in the learning. This configuration avoids lowering of the accuracy of a learning result of a learning model.

Abstract: A method implemented by an arithmetic processing device configured to repeatedly execute similar fixed-point arithmetic operations a plurality of times, the process includes: acquiring, in each of iterations, decimal point position information of the next iteration from statistical information on the arithmetic operations of each of the iterations; calculating an adjustment amount based on a result of comparing the decimal point position information of the next iteration with the already acquired decimal point position information of the previous iteration; and using the decimal point position information of the next iteration and the adjustment amount to execute the fixed-point arithmetic operations in the next iteration.

Abstract: An information processor includes a memory; and a processor coupled to the memory and the processor configured to: acquire first statistical information about distribution of most significant bit position that is not a sign or least significant bit position that is not zero for each of a plurality of first fixed-point number data, the data being a computation result of the computation in the first layer; execute computation on a plurality of output data of the first layer according to a predetermined rule, in the computation in the second layer; and acquire second statistical information based on the predetermined rule and the first statistical information, and determine a bit range for limiting a bit width when a plurality of second fixed-point number data, the data being a computation result of the computation in the second layer, are stored in a register, based on the second statistical information.

Abstract: A processor circuit includes: multiple processor cores; multiple individual memories; multiple shared memories; multiple memory control circuits; multiple selectors; and a control core; wherein when an address of the read request from the first processor associated with a specific memory control circuit is identical to the transfer source address, the specific memory control circuit controls the transfer data based on the read request to be transferred to the transfer destination address via a specific selector of the multiple selectors in which the transfer selection information is set, wherein, when the control core sets read selection information in each of the multiple selectors, read data is read by one of the first processor core and the first adjacent processor core from the associated shared memory via a specific selector of the multiple selectors in which the read selection information is set.

Abstract: A processor includes: a plurality of processor cores; and an internal memory configured to be accessed from the plurality of processor cores, wherein an arithmetic circuit provided in any of the plurality of processor cores includes: a plurality of first registers provided in a first stage of the arithmetic circuit, a regular addition circuit including a first adder and a second register, the first adder being configured to add a plurality of outputs of the plurality of first registers, the second register being configured to be provided in a second stage and latch an output of the first adder, an overtaking addition circuit including a second adder, the second adder being configured to add a plurality of outputs of the plurality of first registers, and a synthesis circuit including a third adder and a third register.

Abstract: A processor includes: a first memory configured to store image data including pixel data of a plurality of pixels that are two-dimensionally arranged; a second memory configured to store neighborhood matrix image data including pixel data of a neighborhood matrix; and a format converter that includes (a) a readout circuit configured to read out the image data from the first memory, (b) a padding arithmetic unit configured to receive the read-out image data, select pixel data of the received read-out image data and padding data inserted at periphery of the plurality of pixels in accordance with mask values of a padding mask, and generate the neighborhood matrix image data including the pixel data and the padding data, and (c) a writing circuit configured to write the neighborhood matrix image data to the second memory.