GPU self throttling

摘要

Techniques for GPU self throttling are described. In one or more embodiments, timing information for GPU frame processing is obtained using a timeline for the GPU. This may occur by inserting callbacks into the GPU processing timeline. An elapsed time for unpredictable work that is inserted into the GPU workload is determined based on the obtained timing information. A decision is then made regarding whether to “throttle” designated optional/non-critical portions of the work for a frame based on the amount of elapsed time. In one approach the elapsed time is compared to a configurable timing threshold. If the elapsed time exceeds the threshold, work is throttled by performing light or no processing for one or more optional portions of a frame. If the elapsed time is less than the threshold, heavy processing (e.g., “normal” work) is performed for the frame.

主权项

1. A computer-implemented method comprising:
processing a frame by a graphics processing unit (GPU) of a computing device; and during the processing for the frame:
obtaining timing information for the processing of the frame using a timeline for the GPU;computing an elapsed time for unpredictable work inserted into the processing performed by the GPU for the frame based on the obtained timing information;determining according to the elapsed time computed for the unpredictable work whether to throttle work performed by the GPU later in processing of the frame by reducing the work for one or more designated portions of the frame that can be throttled; andselectively performing light processing or heavy processing for the one or more designated portions later in the frame based on the determination.