Meta Platforms, Inc.

Facebook, Inc.

Thomas Mark Ulrich

Krishnakumar Narayanan Nair

Ehsan Khish Ardestani Zadeh

Abdulkadir Utku Diril

Yuchen Hao

Rakesh Komuravelli

Olivia Wu

Anup Ramesh Kadkol

Martin Schatz

Amin Firoozshahian

Zhao Wang

Aravind Kalaiah

A device includes a matrix transpose component, a matrix processing component, a data alignment component, and a data reduction component. The matrix transpose component is configured to transpose an input matrix of elements to output an output matrix of the elements that have been transposed. The matrix processing component is configured to multiply a first multiplication input matrix with a second multiplication input matrix, wherein the output matrix of the matrix transpose component is utilized as the first multiplication input matrix and a mask vector is utilized as the second multiplication input matrix. The data alignment component is configured to modify at least a portion of elements of a result of the matrix processing component. The data reduction component is configured to sum at least the elements of the modified result of the matrix processing component to determine a sum of the group of values.

Device and method for flexibly summing matrix values

A device (e.g., integrated circuit chip) includes a first operand register, a second operand register, a multiplication unit, and a hardware logic component. The first operand register is configured to store a first operand value. The second operand register is configured to store a second operand value. The multiplication unit is configured to at least multiply the first operand value with the second operand value. The hardware logic component is configured to detect whether a zero value is provided and in response to a detection that the zero value is being provided: cause an update of at least the first operand register to be disabled, and cause a result of a multiplication of the first operand value with the second operand value to be a zero-value result.

Bypassing zero-value multiplications in a hardware multiplier

A computer-implemented method may include (1) maintaining (a) a filter matrix in a filter cache included in a local memory device (LMD) included in a hardware accelerator, and (b) a plurality of activation matrices corresponding to different rows of an activation volume in an activation cache included in the LMD, (2) for each activation matrix, directing a matrix multiplication unit (MMU) included in the hardware accelerator to execute a matrix multiplication operation (MMU) using the filter matrix and the activation matrix, (3) loading an additional filter matrix into the filter cache, and (4) directing the MMU to execute a plurality of additional MMOs, each additional MMO using one filter matrix included in the filter cache and one activation matrix included in the activation cache, such that the MMU reuses the filter matrix for at least one additional MMO and uses the additional filter matrix for a different additional MMO.

Systems and methods for reducing power consumption of convolution operations of artificial neural networks

A processor system comprises a plurality of processing elements. Each processing element includes a corresponding convolution processor unit configured to perform a portion of a groupwise convolution. The corresponding convolution processor unit determines multiplication results by multiplying each data element of a portion of data elements in a convolution data matrix with a corresponding data element in a corresponding groupwise convolution weight matrix. The portion of data elements in the convolution data matrix that are multiplied belong to different channels and different groups. For each specific channel of the different channels, the corresponding convolution processor unit sums together at least some of the multiplication results belonging to the same specific channel to determine a corresponding channel convolution result data element. The processing elements sum together a portion of the channel convolution result data elements from a group of different convolution processor units to determine a groupwise convolution result data element.

Grouped convolution using point-to-point connected channel convolution engines

A processor system comprises a first and second group of registers and a hardware channel convolution processor unit. The first group of registers is configured to store data elements of channels of a portion of a convolution data matrix. Each register stores at least one data element from each channel. The second group of registers is configured to store data elements of convolution weight matrices including a separate convolution weight matrix for each channel. Each register stores at least one data element from each convolution weight matrix. The hardware channel convolution processor unit is configured to multiply each data element in the first group of registers with a corresponding data element in the second group of registers and sum together the multiplication results for each specific channel to determine corresponding channel convolution result data elements in a corresponding channel convolution result matrix.

Mapping convolution to a channel convolution engine

A processor system comprises two groups of registers and a hardware channel convolution processor unit. The first group of registers is configured to store data elements of channels of a portion of a convolution data matrix. Each register stores at least one data element from each channel. The second group of registers is configured to store data elements of convolution weight matrices including a separate matrix for each channel. Each register stores at least one data element from each matrix. The hardware channel convolution processor unit is configured to multiply each data element in a first and second portion of the first group of registers with a corresponding data element in the second group of registers to determine corresponding multiplication results and sum together the multiplication results for each specific channel to determine two corresponding channel convolution result data elements in a corresponding channel convolution result matrix.

Mapping convolution to a partition channel convolution engine

A first group of elements is element-wise multiplied with a second group of elements using a plurality of multipliers belonging to a matrix multiplication hardware unit. Results of the plurality of multipliers are added together using a hierarchical tree of adders belonging to the matrix multiplication hardware unit and a final result of the hierarchical tree of adders or any of a plurality of intermediate results of the hierarchical tree of adders is selectively provided for use in determining an output result matrix. A control unit is used to instruct the matrix multiplication hardware unit to perform a plurality of different matrix multiplications in parallel by using a combined matrix that includes elements of a plurality of different operand matrices and utilize one or more selected ones of the intermediate results of the hierarchical tree of adders for use in determining the output result matrix that includes different groups of elements representing different multiplication results corresponding to different ones of the different operand matrices.

Support for different matrix multiplications by selecting adder tree intermediate results

A device (e.g., an integrated circuit chip) includes a dot product processing component, a data alignment component, and an accumulator. The dot product processing component is configured to calculate a dot product of a first group of elements stored in a first storage unit with a second group of elements, wherein: each element of the first group of elements is represented using a first number of bits, each value of a group of values stored in the first storage unit is represented using a second number of bits greater than the first number of bits, and each value of the group of values is stored as split segments across more than one element of the elements of the first group of elements. The data alignment component is configured to receive results of the dot product processing component and modify one or more of the results of the dot product processing component. The accumulator is configured to sum outputs of the data alignment component to at least in part determine a sum of the group of values.

Using a low-bit-width dot product engine to sum high-bit-width numbers

A processor system comprises a hardware channel convolution processor unit and dot product processor unit. The channel convolution processor unit is configured to perform depthwise convolution, including by multiplying each data element of a first group of data elements of a convolution data matrix with a corresponding data element of a second group of data elements of a plurality of depthwise convolution weight matrices and summing together, for each specific channel, multiplication results corresponding to the specific channel to determine one corresponding result data element in a corresponding channel convolution result matrix to calculate a portion of depthwise convolution results. The dot product processor unit is configured to perform pointwise convolution, including applying pointwise weight matrices to the portion of depthwise convolution results to determine a portion of separable convolution results while at least another portion of the depthwise convolution results is being calculated by the processor system.

Pipelined pointwise convolution using per-channel convolution operations

A system comprises a data input vector unit, a weight input vector unit, and a plurality of calculation units of a matrix processor unit. The data input vector unit is configured to concurrently receive elements of different rows of a first and second data matrix. The weight input vector unit is configured to receive a combined weight vector and at least in part concurrently provide obtained weight elements of a first and second weight matrix to a corresponding first and second group of calculation units. Each calculation unit of the first and second group of calculation units is configured to multiply elements from the data input vector unit with elements of the corresponding weight matrix from the weight input vector unit and sum together multiplication results of the corresponding calculation unit to at least in part determine a corresponding element in a first or second convolution result matrix.

High throughput matrix processor with support for concurrently processing multiple matrices

A processor system comprises a shared memory and a processing element. The processing element includes a matrix processor unit and is in communication with the shared memory. The processing element is configured to receive a processor instruction specifying a data matrix and a matrix manipulation operation. A manipulation matrix based on the processor instruction is identified. The data matrix and the manipulation matrix are loaded into the matrix processor unit and a matrix operation is performed to determine a result matrix. The result matrix is outputted to a destination location.

Matrix processing instruction with optional up/down sampling of matrix

A floating-point number in a first format representation is received. Based on an identification of a floating-point format type of the floating-point number, different components of the first format representation are identified. The different components of the first format representation are placed in corresponding components of a second format representation of the floating-point number, wherein a total number of bits of the second format representation is larger than a total number of bits of the first format representation. At least one of the components of the second format representation is padded with one or more zero bits. The floating-point number in the second format representation is stored in a register. A multiplication using the second format representation of the floating-point number is performed.

Hardware for floating-point arithmetic in multiple formats

A processor system comprises one or more logic units configured to receive a processor instruction identifying a first floating point number to be multiplied with a second floating point number. The floating point numbers are each decomposed into a group of a plurality of component numbers, wherein a number of bits used to represent each floating point number is greater than a number of bits used to represent any component number in each group of the plurality of component numbers. The component numbers of the first group are multiplied with the component numbers of the second group to determine intermediate multiplication results that are summed together to determine an effective result that represents a result of multiplying the first floating point number with the second floating point number.

Floating point multiply hardware using decomposed component numbers

A first group of modulo result matrices corresponding to modulo of elements of a first matrix by each of a plurality of moduli is stored. A second group of modulo result matrices corresponding to modulo of elements of a second matrix by each of the plurality of moduli is stored. It is determined whether an element operation of a multiplication of the first matrix with the second matrix can be performed using a first hardware multiplication module rather than a second hardware multiplication module. In response to a determination that the element operation can be performed using the first hardware multiplication module, the element operation is performed using the first hardware multiplication module including by multiplying one or more corresponding elements from the first group of modulo result matrices with one or more corresponding elements from the second group of modulo result matrices.

Growing community of inventors

Mountain View, CA, United States of America

Thomas Mark Ulrich