PERFORMANCE OVERHEAD OPTIMIZATION IN GPU SCOPING

    公开(公告)号:US20230009205A1

    公开(公告)日:2023-01-12

    申请号:US17373718

    申请日:2021-07-12

    Abstract: The present disclosure relates to methods and devices for graphics processing including an apparatus, e.g., a GPU. The apparatus may process a first workload of a plurality of workloads at each of multiple clusters in a GPU pipeline. The apparatus may also increment a plurality of performance counters during the processing of the first workload at each of the multiple clusters. Further, the apparatus may determine, at each of the multiple clusters, whether the first workload is finished processing. The apparatus may also read, upon determining that the first workload is finished processing, a value of each of the multiple clusters for each of the plurality of performance counters. Additionally, the apparatus may transmit an indication of the read value of each of the multiple clusters for all of the plurality of performance counters.

    ELIMINATION CACHE
    2.
    发明公开
    ELIMINATION CACHE 审中-公开

    公开(公告)号:US20240289912A1

    公开(公告)日:2024-08-29

    申请号:US18175480

    申请日:2023-02-27

    CPC classification number: G06T1/20 G06T1/60

    Abstract: This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for an elimination cache. A graphics processor may obtain an indication of at least one state update from at least one CP associated with a graphics processor, where the at least one state update corresponds to one or more states in a set of states associated with the graphics processor. The graphics processor may determine whether the one or more states are stored in a cache associated with the graphics processor. The graphics processor may discard the at least one state update based on a determination that the one or more states are stored in the cache or update the cache based on a determination that the one or more states are not stored in the cache.

Patent Agency Ranking