Replicating primitives across multiple viewports

• US 10,235,811 B2
• Filed: 12/29/2016
• Issued: 03/19/2019
• Est. Priority Date: 12/29/2016
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Bhiravabhatla, Kalyan K., Maiyuran, Subramaniam, Toth, Robert M., Janczak, Tomasz
• Primary Examiner(s): Amini, Javid A
• Application Number:
  US15/394,027
• Publication Number:
  US 20180190021A1
• Time in Patent Office: 810 Days
• Field of Search: 345633
• CPC Class Codes: G06T 15/005 General purpose rendering architectures
• CPC Class Codes: G06T 15/20 Perspective computation
• CPC Class Codes: G06T 15/40 Hidden part removal
• CPC Class Codes: G06T 15/80 Shading
• CPC Class Codes: G06T 17/10 Constructive solid geometry [CSG] using solid primitives, e.g. cylinders, cubes
• CPC Class Codes: G06T 19/006 Mixed reality object pose determination, tracking or camera calibration for mixed reality G06T7/00
• CPC Class Codes: G06T 2200/04 involving 3D image data
• CPC Class Codes: G06T 7/60 Analysis of geometric attributes

Abstract

An embodiment of a graphics processor pipeline apparatus may include a vertex fetcher to fetch vertices, a vertex shader communicatively coupled to the vertex fetcher to shade the fetched vertices, a primitive assembler communicatively coupled to the vertex shader to assemble primitives, and a primitive replicator communicatively coupled to the primitive assembler to replicate primitives for at least a first and a second viewport.

Claims

1. A graphics processor system, comprising:

a cull geometry pipeline to cull geometries, the cull geometry pipeline including;

a first vertex fetcher, implemented at least partly in one or more of configurable logic or fixed-functionality logic hardware, to fetch vertices,

a first vertex shader, implemented at least partly in one or more of configurable logic or fixed-functionality logic hardware, communicatively coupled to the first vertex fetcher to shade the fetched vertices,

a first primitive assembler, implemented at least partly in one or more of configurable logic or fixed-functionality logic hardware, communicatively coupled to the first vertex shader to assemble primitives including an original primitive,

a first primitive replicator, implemented at least partly in one or more of configurable logic or fixed-functionality logic hardware, communicatively coupled to the first primitive assembler to replicate primitives for at least a first and a second viewport, wherein the first primitive replicator is to generate replicated primitives of the original primitive and adjust respective position attributes of the replicated primitives of the original primitive for the second viewport relative to the first viewport, and

a primitive culler, implemented at least partly in one or more of configurable logic or fixed-functionality logic hardware, communicatively coupled to the first primitive replicator to cull the original primitive only in response to all of the replicated primitives of the original primitive being identified as not visible, wherein the primitive culler is to generate visibility information identifying whether the original primitive is to be culled by the primitive culler;

a replay geometry pipeline communicatively coupled to the cull geometry pipeline, the replay geometry pipeline including;

a second vertex fetcher, implemented at least partly in one or more of configurable logic or fixed-functionality logic hardware, to fetch vertices,

a second vertex shader, implemented at least partly in one or more of configurable logic or fixed-functionality logic hardware, communicatively coupled to the second vertex fetcher to shade the fetched vertices,

a second primitive assembler, implemented at least partly in one or more of configurable logic or fixed-functionality logic hardware, communicatively coupled to the second vertex shader to assemble primitives, and

a second primitive replicator, implemented at least partly in one or more of configurable logic or fixed-functionality logic hardware, communicatively coupled to the second primitive assembler to replicate primitives for at least the first and the second viewport; and

a visibility store communicatively coupled between the cull geometry pipeline and the replay geometry pipeline to store the visibility information,

wherein the replay geometry pipeline is to access the visibility store, skip a process on the original primitive in response to the original primitive being identified by the visibility information as culled, and process the original primitive in response to the original primitive being identified by the visibility information as not culled.

2. The graphics processor system of claim 1, wherein the second primitive replicator is further to adjust respective position attributes of the replicated primitives for the second viewport relative to the first viewport.

18. The graphics processor system of claim 1, wherein the cull geometry pipeline is a position-only shading pipeline that is separate from the replay geometry pipeline.

19. The graphics processor system of claim 1, wherein the primitive culler is to cull the original primitive before any process is executed on the original primitive by the replay geometry pipeline.

3. A graphics processor pipeline apparatus, comprising:

a vertex fetcher, implemented at least partly in one or more of configurable logic or fixed-functionality logic hardware, to fetch vertices;

a vertex shader, implemented at least partly in one or more of configurable logic or fixed-functionality logic hardware, communicatively coupled to the vertex fetcher to shade the fetched vertices;

a primitive assembler, implemented at least partly in one or more of configurable logic or fixed-functionality logic hardware, communicatively coupled to the vertex shader to assemble primitives including an original primitive;

a primitive replicator, implemented at least partly in one or more of configurable logic or fixed-functionality logic hardware, communicatively coupled to the primitive assembler to replicate primitives for at least a first and a second viewport, wherein the primitive replicator is to generate replicated primitives of the original primitive and adjust respective position attributes of the replicated primitives of the original primitive for the second viewport relative to the first viewport;

a primitive culler, implemented at least partly in one or more of configurable logic or fixed-functionality logic hardware, communicatively coupled to the primitive replicator to cull the original primitive only in response to all of the replicated primitives of the original primitive being identified as not visible; and

a replay geometry pipeline, implemented at least partly in one or more of configurable logic or fixed-functionality logic hardware, communicatively coupled to the primitive culler, to skip a process on the original primitive in response to the original primitive being culled by the primitive culler, and to process the original primitive in response to the original primitive not being culled by the primitive culler.

4. The graphics processor pipeline apparatus of claim 3, wherein the first viewport corresponds to a left eye viewport of a stereo virtual reality system and wherein the second viewport corresponds to a right eye viewport of the stereo virtual reality system.

5. The graphics processor pipeline apparatus of claim 3, wherein the first viewport is part of a first set of viewports which corresponds to a first face of a cube rendering system and wherein the second viewport is part of a second set of viewports which corresponds to a second face of the cube rendering system.

6. The graphics processor pipeline apparatus of any of claim 3, wherein the first viewport corresponds to a first cascade of a cascaded shadow map and wherein the second viewport corresponds to a second cascade of the cascaded shadow map.

20. The graphics processor pipeline apparatus of claim 3, wherein the primitive culler is to identify whether to cull of the original primitive before any process on the original primitive is executed by the replay geometry pipeline.

21. The graphics processor pipeline apparatus of claim 3, wherein:

the vertex shader is a position-only shader that only shades position attributes of the fetched vertices to generate shaded fetched vertices; and

the primitive assembler is to assemble the primitives based on the shaded fetched vertices.

22. The graphics processor pipeline apparatus of claim 3, wherein the vertex fetcher, the vertex shader, the primitive assembler, the primitive replicator, and the primitive culler are part of a position-only shading pipeline that is separate from the replay geometry pipeline.

7. A method of processing graphics, comprising:

fetching vertices;

shading the fetched vertices;

assembling primitives, including an original primitive, based on the shaded vertices;

replicating primitives for at least a first and a second viewport, wherein the replicating includes generating replicated primitives of the original primitive and adjusting respective position attributes of the replicated primitives of the original primitive for the second viewport relative to the first viewport;

culling the original primitive only in response to all of the replicated primitives of the original primitive being identified as not visible;

skipping, at a replay geometry pipeline, a process on the original primitive in response to the original primitive being culled; and

processing the original primitive at the replay geometry pipeline in response to the original primitive not being culled.

8. The method of processing graphics of claim 7, wherein the first viewport corresponds to a left eye viewport of a stereo virtual reality system and wherein the second viewport corresponds to a right eye viewport of the stereo virtual reality system.

9. The method of processing graphics of claim 7, wherein the first viewport is part of a first set of viewports which corresponds to a first face of a cube rendering system and wherein the second viewport is part of a second set of viewports which corresponds to a second face of the cube rendering system.

10. The method of processing graphics of claim 7, wherein the first viewport corresponds to a first cascade of a cascaded shadow map and wherein the second viewport corresponds to a second cascade of the cascaded shadow map.

11. At least one non-transitory computer readable storage media comprising a set of instructions, which when executed by a computing device, cause the computing device to:

fetch vertices;

shade the fetched vertices;

assemble primitives, including an original primitive, based on the shaded vertices;

replicate primitives for at least a first and a second viewport by a generation of replicated primitives of the original primitive and an adjustment of respective position attributes of the replicated primitives of the original primitive for the second viewport relative to the first viewport;

cull the original primitive only in response to all of the replicated primitives of the original primitive being identified as being not visible;

skip, at a replay geometry pipeline, a process on the original primitive in response to the original primitive being culled; and

process the original primitive at the replay geometry pipeline in response to the original primitive not being culled.

12. The at least one non-transitory computer readable storage media of claim 11, wherein the first viewport corresponds to a left eye viewport of a stereo virtual reality system and wherein the second viewport corresponds to a right eye viewport of the stereo virtual reality system.

13. The at least one non-transitory computer readable storage media of claim 11, wherein the first viewport is part of a first set of viewports which corresponds to a first face of a cube rendering system and wherein the second viewport is part of a second set of viewports which corresponds to a second face of the cube rendering system.

14. The at least one non-transitory computer readable storage media of claim 11, wherein the first viewport corresponds to a first cascade of a cascaded shadow map and wherein the second viewport corresponds to a second cascade of the cascaded shadow map.

15. The at least one non-transitory computer readable storage media of claim 11, wherein the cull the original primitive is to be executed by a position-only shading pipeline before any process is executed on the original primitive by the replay geometry pipeline.

16. The at least one non-transitory computer readable storage media of claim 11, wherein the shade of the fetched vertices is to only include shading position attributes.

17. The at least one non-transitory computer readable storage media of claim 11, wherein the fetch of the vertices, the shade of the fetched vertices, the assemble of the primitives, the replicate of the primitives and the cull of the original primitive are to be executed at a position-only shading pipeline that is separate from the replay geometry pipeline.