Xilinx, Inc.

Xiao Teng

Elliott Delaye

Aaron Ng

Ashish Sirasao

Ehsan Ghasemi

Jindrich Zejda

Sean Settle

Rajeev Patwari

Jorn Tuyls

Sanket Pandit

Ephrem C Wu

Sonal Santan

Soren T Soe

Yongjun Wu

Satyaprakash Pareek

Ling Liu

Zhao Ma

Yifei Zhou

Tejus Siddagangaiah

Chuanhua Song

Bryan Lozano

Pramod Peethambaran

Hardware acceleration of machine learning (ML) designs includes translating an ML primitive into an intermediate representation. The intermediate representation is subdivided to specify a functional compute block. The functional compute block is sized according to a compute node primitive adapted for implementing the ML primitive on target hardware. An overlay is generated for the ML primitive, at least in part, by mapping the functional compute block to the compute node primitive. The overlay is synthesizable to implement the ML primitive on the target hardware. The overlay can be scheduled for operation within the target hardware as part of an ML design including the ML primitive.

Hardware acceleration of machine learning designs

Controlling a data processing (DP) array includes creating a replica of a register address space of the DP array based on the design and the DP array. A sequence of instructions, including write instructions and read instructions, is received. The write instructions correspond to buffer descriptors specifying runtime data movements for a design for a DP array. The write instructions are converted into transaction instructions and the read instructions are converted into wait instructions based on the replica of the register address space. The transaction instructions and the wait instructions are included in an instruction buffer. The instruction buffer is provided to a microcontroller configured to execute the transaction instructions and the wait instructions to implement the runtime data movements for the design as implemented in the DP array. In another aspect, the instruction buffer is stored in a file for subsequent execution by the microcontroller.

Instruction set architecture for data processing array control

Examples herein describe techniques for performing parallel processing using a plurality of processing elements (PEs) and a controller for data that has data dependencies. For example, a calculation may require an entire row or column to be summed, or to determine its mean. The PEs can be assigned different chunks of a data set (e.g., a tensor set, a column, or a row) for processing. The PEs can use one or more tokens to inform the controller when they are done with partial processing of their data chunks. The controller can then gather the partial results and determine an intermediate value for the data set. The controller can then distribute this intermediate value to the PEs which then re-process their respective data chunks using the intermediate value to generate final results.

Scalable acceleration of reentrant compute operations

An integrated circuit includes a plurality of kernels and a virtual machine coupled to the plurality of kernels. The virtual machine is configured to interpret instructions directed to different ones of the plurality of kernels. The virtual machine is configured to control operation of the different ones of the plurality of kernels responsive to the instructions.

Reconfigurable neural engine with extensible instruction set architecture

Embodiments herein describe techniques for interfacing a neural network application with a neural network accelerator using a library. The neural network application may execute on a host computing system while the neural network accelerator executes on a massively parallel hardware system, e.g., a FPGA. The library operates a pipeline for submitting the tasks received from the neural network application to the neural network accelerator. In one embodiment, the pipeline includes a pre-processing stage, an FPGA execution stage, and a post-processing stage which each correspond to different threads. When receiving a task from the neural network application, the library generates a packet that includes the information required for the different stages in the pipeline to perform the tasks. Because the stages correspond to different threads, the library can process multiple packets in parallel which can increase the utilization of the neural network accelerator on the hardware system.

Machine learning runtime library for neural network acceleration

In the disclosed methods and systems for processing in a neural network system, a host computer system writes a plurality of weight matrices associated with a plurality of layers of a neural network to a memory shared with a neural network accelerator. The host computer system further assembles a plurality of per-layer instructions into an instruction package. Each per-layer instruction specifies processing of a respective layer of the plurality of layers of the neural network, and respective offsets of weight matrices in a shared memory. The host computer system writes input data and the instruction package to the shared memory. The neural network accelerator reads the instruction package from the shared memory and processes the plurality of per-layer instructions of the instruction package.

Multi-layer neural network processing by a neural network accelerator using host communicated merged weights and a package of per-layer instructions

A disclosed neural network processing system includes a host computer system, a RAMs coupled to the host computer system, and neural network accelerators coupled to the RAMs, respectively. The host computer system is configured with software that when executed causes the host computer system to write input data and work requests to the RAMS. Each work request specifies a subset of neural network operations to perform and memory locations in a RAM of the input data and parameters. A graph of dependencies among neural network operations is built and additional dependencies added. The operations are partitioned into coarse grain tasks and fine grain subtasks for optimal scheduling for parallel execution. The subtasks are scheduled to accelerator kernels of matching capabilities. Each neural network accelerator is configured to read a work request from the respective RAM and perform the subset of neural network operations on the input data using the parameters.

Neural network processing system having host controlled kernel acclerators

At least one neural network accelerator performs operations of a first subset of layers of a neural network on an input data set, generates an intermediate data set, and stores the intermediate data set in a shared memory queue in a shared memory. A first processor element of a host computer system provides input data to the neural network accelerator and signals the neural network accelerator to perform the operations of the first subset of layers of the neural network on the input data set. A second processor element of the host computer system reads the intermediate data set from the shared memory queue, performs operations of a second subset of layers of the neural network on the intermediate data set, and generates an output data set while the neural network accelerator is performing the operations of the first subset of layers of the neural network on another input data set.

Neural network processing system having multiple processors and a neural network accelerator

An example multiply accumulate (MACC) circuit includes: a multiply-accumulator having an accumulator output register; a quantizer, coupled to the multiply accumulator; and a control circuit coupled to the multiply-accumulator and the quantizer, the control circuit configured to provide control data to the quantizer, the control data indicative of a most-significant bit (MSB) to least significant bit (LSB) range for selecting bit indices from the accumulator output register.

Software-driven design optimization for fixed-point multiply-accumulate circuitry

Disclosed approaches for multiplying a sparse matrix by dense a vector or matrix include first memory banks for storage of column indices, second memory banks for storage of row indices, and third memory banks for storage of non-zero values of a sparse matrix. A pairing circuit distributes an input stream of vector elements across first first-in-first-out (FIFO) buffers according to the buffered column indices. Multiplication circuitry multiplies vector elements output from the first FIFO buffers by corresponding ones of the non-zero values from the third memory banks, and stores products in second FIFO buffers. Row-aligner circuitry organize the products output from the second FIFO buffers into third FIFO buffers according to row indices in the second memory banks. Accumulation circuitry accumulates respective totals from products output from the third FIFO buffers.

Sparse matrix processing circuitry

An example multiply accumulate (MACC) circuit includes a multiply-accumulator having an accumulator output register, a scaler, coupled to the multiply accumulator, and a control circuit coupled to the multiply-accumulator and the scaler. The control circuit is configured to provide control data to the scaler, the control data indicative of: a most-significant bit (MSB) to least significant bit (LSB) range for selecting bit indices from the accumulator output register for implementing a first right shift; a multiplier; and a second right shift.

Growing community of inventors

Cupertino, CA, United States of America

Xiao Teng