Reassigning geometry based on timing analysis when rendering an image frame

Patent metadata

Field	Value
Publication number	US-12400286-B2
Application number	US-202418440793-A
Country	US
Kind code	B2
Filing date	Feb 13, 2024
Priority date	Feb 3, 2020
Publication date	Aug 26, 2025
Grant date	Aug 26, 2025

Abstract

A method for graphics processing. The method including rendering graphics for an application using graphics processing units (GPUs). The method including using the plurality of GPUs in collaboration to render an image frame including a plurality of pieces of geometry. The method including during the rendering of the image frame, subdividing one or more of the plurality of pieces of geometry into smaller pieces, and dividing the responsibility for rendering these smaller portions of geometry among the plurality of GPUs, wherein each of the smaller portions of geometry is processed by a corresponding GPU. The method including for those pieces of geometry that are not subdivided, dividing the responsibility for rendering the pieces of geometry among the plurality of GPUs, wherein each of these pieces of geometry is processed by a corresponding GPU.

First claim

What is claimed is:

1. A method for graphics processing, comprising:
using a plurality of graphics processing units (GPUs) in collaboration to render an image frame including a plurality of geometry, wherein the plurality of geometry includes one or more pieces of geometry;
generating information during a Z pre-pass phase of rendering by at least determining an overlap of each of the plurality of geometry for the image frame with each of a plurality of screen regions;
assigning the plurality of screen regions to the plurality of GPUs for performing a geometry pass phase of rendering for the image frame based on the information;
determining timing information for each of the plurality of GPUs during execution of the geometry pass phase of rendering for the image frame by the plurality of GPUs, wherein the timing information is related to execution of the geometry pass phase of rendering for the one or more pieces of geometry for the image frame by the each of plurality of GPUs; and
dynamically reassigning the plurality of screen regions to the plurality of GPUs based on the timing information during execution of the geometry pass phase of rendering for the image frame.

2. The method of claim 1 , wherein the dynamically reassigning the plurality of screen regions includes:
reassigning a first screen region initially assigned to a first GPU to a second GPU during the geometry pass phase of rendering.

3. The method of claim 1 , further comprising:
reassigning a geometry from a first GPU to a second GPU for processing during the geometry pass phase of rendering based on the timing information.

4. The method of claim 1 , wherein the timing information, for a corresponding GPU, tracks progress of completing the geometry pass phase of rendering on one or more corresponding screen regions assigned to the corresponding GPU.

5. The method of claim 4 , further comprising:
determining a first GPU is behind a second GPU based on the timing information when performing the geometry pass phase of rendering; and
reassigning less screen regions to the first GPU than the second GPU when performing the dynamically reassigning the plurality of screen regions to the plurality of GPUs,
wherein a first progress of the first GPU completing the geometry pass phase of rendering is behind a second progress of the second GPU completing the geometry pass phase of rendering.

6. The method of claim 1 , wherein one or more of the plurality of GPUs are portions of a larger GPU that is configured as a plurality of virtual GPUs.

7. A computer system comprising:
a processor; and
memory coupled to the processor and having stored therein instructions that, if executed by the computer system, cause the computer system to execute a method for graphics processing, comprising:
using a plurality of graphics processing units (GPUs) in collaboration to render an image frame including a plurality of geometry, wherein the plurality of geometry includes one or more pieces of geometry;
generating information during a Z pre-pass phase of rendering by at least determining an overlap of each of the plurality of geometry for the image frame with each of a plurality of screen regions;
assigning the plurality of screen regions to the plurality of GPUs for performing a geometry pass phase of rendering for the image frame based on the information;
determining timing information for each of the plurality of GPUs during execution of the geometry pass phase of rendering for the image frame by the plurality of GPUs, wherein the timing information is related to execution of the geometry pass phase of rendering for the one or more pieces of geometry for the image frame by the each of plurality of GPUs; and
dynamically reassigning the plurality of screen regions to the plurality of GPUs based on the timing information during execution of the geometry pass phase of rendering for the image frame.

8. The computer system of claim 7 , wherein in the method the dynamically reassigning the plurality of screen regions includes:
reassigning a first screen region initially assigned to a first GPU to a second GPU during the geometry pass phase of rendering.

9. The computer system of claim 7 , the method further comprising:
reassigning a geometry from a first GPU to a second GPU for processing during the geometry pass phase of rendering based on the timing information.

10. The computer system of claim 7 , wherein in the method the timing information, for a corresponding GPU, tracks progress of completing the geometry pass phase of rendering on one or more corresponding screen regions assigned to the corresponding GPU.

11. The computer system of claim 10 , the method further comprising:
determining a first GPU is behind a second GPU based on the timing information when performing the geometry pass phase of rendering; and
reassigning less screen regions to the first GPU than the second GPU when performing the dynamically reassigning the plurality of screen regions to the plurality of GPUs,
wherein a first progress of the first GPU completing the geometry pass phase of rendering is behind a second progress of the second GPU completing the geometry pass phase of rendering.

12. The computer system of claim 7 , wherein in the method one or more of the plurality of GPUs are portions of a larger GPU that is configured as a plurality of virtual GPUs.

13. A non-transitory computer-readable medium storing a computer program for graphics processing, the computer readable medium comprising:
program instructions for using a plurality of graphics processing units (GPUs) in collaboration to render an image frame including a plurality of geometry, wherein the plurality of geometry includes one or more pieces of geometry;
program instructions for generating information during a Z pre-pass phase of rendering by at least determining an overlap of each of the plurality of geometry for the image frame with each of a plurality of screen regions;
program instructions for assigning the plurality of screen regions to the plurality of GPUs for performing a geometry pass phase of rendering for the image frame based on the information;
program instructions for determining timing information for each of the plurality of GPUs during execution of the geometry pass phase of rendering for the image frame by the plurality of GPUs, wherein the timing information is related to execution of the geometry pass phase of rendering for the one or more pieces of geometry for the image frame by the each of plurality of GPUs; and
program instructions for dynamically reassigning the plurality of screen regions to the plurality of GPUs based on the timing information during execution of the geometry pass phase of rendering for the image frame.

14. The non-transitory computer-readable medium of claim 13 , wherein the program instructions for dynamically reassigning the plurality of screen regions includes:
program instructions for reassigning a first screen region initially assigned to a first GPU to a second GPU during the geometry pass phase of rendering.

15. The non-transitory computer-readable medium of claim 13 , further comprising:
program instructions for reassigning a geometry from a first GPU to a second GPU for processing during the geometry pass phase of rendering based on the timing information.

16. The non-transitory computer-readable medium of claim 13 , wherein in the program instructions for reassigning the geometry from a first GPU to a second GPU, the timing information, for a corresponding GPU, tracks progress of completing the geometry pass phase of rendering on on

Assignees

• Sony Interactive Entertainment Inc

Inventors

• Cerny Mark E

Classifications

• G06T17/10

  Constructive solid geometry [CSG] using solid primitives, e.g. cylinders, cubes · CPC title

• G06F9/3877

  using a secondary processor, e.g. coprocessor (peripheral processor G06F13/12) · CPC title

• G06F9/3836

  Instruction issuing, e.g. dynamic instruction scheduling or out of order instruction execution · CPC title

• G06F9/4806

  Task transfer initiation or dispatching · CPC title

• G06F9/544

  Buffers; Shared memory; Pipes · CPC title