公开(公告)号：US11086967B2

公开(公告)日：2021-08-10

申请号：US15907356

申请日：2018-02-28

Applicant: Texas Instruments Incorporated

Inventor： Arthur John Redfern ,  Timothy David Anderson ,  Kai Chirca ,  Chenchi Luo ,  Zhenhua Yu

IPC: G06F17/16 ,  G06F17/14 ,  G06N3/04 ,  G06N3/063

Abstract: A method for performing a fundamental computational primitive in a device is provided, where the device includes a processor and a matrix multiplication accelerator (MMA). The method includes configuring a streaming engine in the device to stream data for the fundamental computational primitive from memory, configuring the MMA to format the data, and executing the fundamental computational primitive by the device.

公开(公告)号：US20180253402A1

公开(公告)日：2018-09-06

申请号：US15907356

申请日：2018-02-28

Applicant: Texas Instruments Incorporated

Inventor： Arthur John Redfern ,  Timothy David Anderson ,  Kai Chirca ,  Chenchi Luo ,  Zhenhua Yu

IPC: G06F17/16 ,  G06F17/14

CPC classification number: G06F17/16 ,  G06F17/141 ,  G06N3/0454 ,  G06N3/063

Abstract: A method for performing a fundamental computational primitive in a device is provided, where the device includes a processor and a matrix multiplication accelerator (MMA). The method includes configuring a streaming engine in the device to stream data for the fundamental computational primitive from memory, configuring the MMA to format the data, and executing the fundamental computational primitive by the device.

公开(公告)号：US20240265062A1

公开(公告)日：2024-08-08

申请号：US18633703

申请日：2024-04-12

Applicant: Texas Instruments Incorporated

Inventor： Arthur John Redfern ,  Timothy David Anderson ,  Kai Chirca ,  Chenchi Luo ,  Zhenhua Yu

IPC: G06F17/16 ,  G06F17/14 ,  G06N3/045 ,  G06N3/063

CPC classification number: G06F17/16 ,  G06F17/141 ,  G06N3/045 ,  G06N3/063

Abstract: A method for performing a fundamental computational primitive in a device is provided, where the device includes a processor and a matrix multiplication accelerator (MMA). The method includes configuring a streaming engine in the device to stream data for the fundamental computational primitive from memory, configuring the MMA to format the data, and executing the fundamental computational primitive by the device.

公开(公告)号：US11960567B2

公开(公告)日：2024-04-16

申请号：US17367389

申请日：2021-07-04

Applicant: TEXAS INSTRUMENTS INCORPORATED

Inventor： Arthur John Redfern ,  Timothy David Anderson ,  Kai Chirca ,  Chenchi Luo ,  Zhenhua Yu

IPC: G06F17/16 ,  G06F17/14 ,  G06N3/045 ,  G06N3/063

CPC classification number: G06F17/16 ,  G06F17/141 ,  G06N3/045 ,  G06N3/063

Abstract: A method for performing a fundamental computational primitive in a device is provided, where the device includes a processor and a matrix multiplication accelerator (MMA). The method includes configuring a streaming engine in the device to stream data for the fundamental computational primitive from memory, configuring the MMA to format the data, and executing the fundamental computational primitive by the device.

公开(公告)号：US20210334337A1

公开(公告)日：2021-10-28

申请号：US17367389

申请日：2021-07-04

Applicant: TEXAS INSTRUMENTS INCORPORATED

Inventor： Arthur John Redfern ,  Timothy David Anderson ,  Kai Chirca ,  Chenchi Luo ,  Zhenhua Yu

IPC: G06F17/16 ,  G06F17/14 ,  G06N3/04 ,  G06N3/063

Abstract: A method for performing a fundamental computational primitive in a device is provided, where the device includes a processor and a matrix multiplication accelerator (MMA). The method includes configuring a streaming engine in the device to stream data for the fundamental computational primitive from memory, configuring the MMA to format the data, and executing the fundamental computational primitive by the device.