Parallelization of convolution operations

    公开(公告)号:US12198295B2

    公开(公告)日:2025-01-14

    申请号:US17565301

    申请日:2021-12-29

    Abstract: A technique for performing convolution operations is disclosed. The technique includes performing a first convolution operation based on a first convolutional layer input image to generate at least a portion of a first convolutional layer output image; while performing the first convolution operation, performing a second convolution operation based on a second convolutional layer input image to generate at least a portion of a second convolutional layer output image, wherein the second convolutional layer input image is based on the first convolutional layer output image; storing the portion of the first convolutional layer output image in a first memory dedicated to storing image data for convolution operations; and storing the portion of the second convolutional layer output image in a second memory dedicated to storing image data for convolution operations.

    HARDWARE SUPPORT FOR CONVOLUTION OPERATIONS
    4.
    发明公开

    公开(公告)号:US20230206395A1

    公开(公告)日:2023-06-29

    申请号:US17565301

    申请日:2021-12-29

    CPC classification number: G06T3/4046 G06F17/153

    Abstract: A technique for performing convolution operations is disclosed. The technique includes performing a first convolution operation based on a first convolutional layer input image to generate at least a portion of a first convolutional layer output image; while performing the first convolution operation, performing a second convolution operation based on a second convolutional layer input image to generate at least a portion of a second convolutional layer output image, wherein the second convolutional layer input image is based on the first convolutional layer output image; storing the portion of the first convolutional layer output image in a first memory dedicated to storing image data for convolution operations; and storing the portion of the second convolutional layer output image in a second memory dedicated to storing image data for convolution operations.

    DISABLING SELECTED IP
    5.
    发明公开

    公开(公告)号:US20230206368A1

    公开(公告)日:2023-06-29

    申请号:US17565409

    申请日:2021-12-29

    CPC classification number: G06Q50/184 G06Q2220/18

    Abstract: A technique for operating a processing device is disclosed. The method includes configuring at least one switch to interconnect one or more selected IP to the processing device, receiving an activation signal associated with the at least one switch based on the one or more selected IP, in response to the activation signal, causing the at least one switch to disable connection to the one or more selected IP, and verifying access to the one or more selected IP is disabled.

    CONVOLUTIONAL NEURAL NETWORK OPERATIONS

    公开(公告)号:US20230097279A1

    公开(公告)日:2023-03-30

    申请号:US17489734

    申请日:2021-09-29

    Abstract: Methods and systems are disclosed for executing operations on single-instruction-multiple-data (SIMD) units. Techniques disclosed perform a dot product operation on input data during one computer cycle, including convolving the input data, generating intermediate data, and applying one or more transitional operations to the intermediate data to generate output data. Aspects described, wherein the input data is an input to a layer of a convolutional neural network and the generated output data is the output of the layer.

    FLEXIBLE MEMORY SYSTEM
    7.
    发明申请

    公开(公告)号:US20240402907A1

    公开(公告)日:2024-12-05

    申请号:US18805083

    申请日:2024-08-14

    Abstract: A technique for operating a memory system is disclosed. The technique includes performing a first request, by a first memory client, to access data at a first memory address, wherein the first memory address refers to data in a first memory section that is coupled to the first memory client via a direct memory connection; servicing the first request via the direct memory connection; performing a second request, by the first client, to access data at a second memory address, wherein the second memory address refers to data in a second memory section that is coupled to the first client via a cross connection; and servicing the second request via the cross connection.

    SYSTEM AND METHOD FOR EXECUTING A TASK
    8.
    发明公开

    公开(公告)号:US20240355044A1

    公开(公告)日:2024-10-24

    申请号:US18762389

    申请日:2024-07-02

    CPC classification number: G06T15/80 G06F9/5016

    Abstract: A method, system, and computer-readable medium for executing a task is disclosed. The method includes receiving input data and computing instructions, launching a workgroup including wavefronts to execute the task, wherein the launching causes the wavefronts to process the input data by sharing intermediate results and resources, and adjusting the operation based on characteristics of the wavefronts. The characteristics include data dependencies, computational load, memory usage, and execution timing requirements. The wavefronts execute the task in stages, where each stage processes portions of input data and data generated by other wavefronts.

    Flexible memory system
    9.
    发明授权

    公开(公告)号:US12067237B2

    公开(公告)日:2024-08-20

    申请号:US17565315

    申请日:2021-12-29

    Abstract: A technique for operating a memory system is disclosed. The technique includes performing a first request, by a first memory client, to access data at a first memory address, wherein the first memory address refers to data in a first memory section that is coupled to the first memory client via a direct memory connection; servicing the first request via the direct memory connection; performing a second request, by the first client, to access data at a second memory address, wherein the second memory address refers to data in a second memory section that is coupled to the first client via a cross connection; and servicing the second request via the cross connection.

Patent Agency Ranking