System, method, and computer program product for processing primitive specific attributes generated by a fast geometry shader

Patent metadata

Field	Value
Publication number	US-9836878-B2
Application number	US-201514820438-A
Country	US
Kind code	B2
Filing date	Aug 6, 2015
Priority date	Mar 15, 2013
Publication date	Dec 5, 2017
Grant date	Dec 5, 2017

Abstract

A system, method, and computer program product are provided for processing primitive-specific attributes. A portion of a graphics processor is determined to operate in a fast geometry shader mode and a vertex associated with a set of per-vertex attributes is determined to be a shared vertex. The shared vertex is determined to be a non-provoking vertex corresponding to a first primitive that is associated with a first set of per-primitive attributes and the shared vertex is determined to be a provoking vertex corresponding to a second primitive that is associated with a second set of per-primitive attributes. Only one set of the per-vertex attributes associated with the shared vertex is stored and only one of the second set of per-primitive attributes associated with the second primitive is stored.

First claim

What is claimed is:

1. A method, comprising:
determining that a portion of a graphics processor is operating in a fast geometry shader mode, wherein the graphics processor comprises at least one streaming multiprocessor configured to execute a vertex shader and a geometry shader;
determining that a first vertex associated with a first set of per-vertex attributes is a first shared vertex;
determining that the first shared vertex is a non-provoking vertex corresponding to a first primitive that is associated with a first set of per-primitive attributes;
determining that the first shared vertex is a provoking vertex corresponding to a second primitive that is associated with a second set of per-primitive attributes by examining vertices of the second primitive and vertices of a third primitive when the first primitive is received;
storing the first set of the per-vertex attributes associated with the first shared vertex; and
storing only one copy of the second set of per-primitive attributes associated with the second primitive.

2. The method of claim 1 , further comprising processing the only one copy of the second set of per-primitive attributes associated with the second primitive.

3. The method of claim 1 , wherein the first set of the per-vertex attributes associated with the first shared vertex is stored in a cache entry.

4. The method of claim 3 , wherein the only one copy of the second set of per-primitive attributes associated with the second primitive is also stored in the cache entry.

5. The method of claim 3 , further comprising setting a bit corresponding to the cache entry indicating that the first shared vertex is the provoking vertex.

6. The method of claim 3 , further comprising associating a primitive identifier corresponding to the second primitive with the cache entry.

7. The method of claim 1 , further comprising:
determining that a second vertex associated with a second set of per-vertex attributes is a second shared vertex;
determining that the second shared vertex is the provoking vertex corresponding to the first primitive and a third primitive that is associated with a third set of per-primitive attributes; and
duplicating the second set of per-vertex attributes associated with the second shared vertex to generate a duplicated set of per-vertex attributes associated with the second shared vertex.

8. The method of claim 7 , further comprising:
storing the second set of per-vertex attributes and the first set of per-primitive attributes associated with the first primitive in a first cache entry; and
storing the duplicated set of per-vertex attributes and the third set of per-primitive attributes associated with the third primitive in a second cache entry.

9. A non-transitory computer readable medium, storing instructions that, when executed by a processor, cause the processor to perform steps, comprising:
determining that a portion of a graphics processor is operating in a fast geometry shader mode, wherein the graphics processor comprises at least one streaming multiprocessor configured to execute a vertex shader and a geometry shader;
determining that a first vertex associated with a first set of per-vertex attributes is a first shared vertex;
determining that the first shared vertex is a non-provoking vertex corresponding to a first primitive that is associated with a first set of per-primitive attributes;
determining that the first shared vertex is a provoking vertex corresponding to a second primitive that is associated with a second set of per-primitive attributes by examining vertices of the second primitive and vertices of a third primitive when the first primitive is received;
storing the first set of the per-vertex attributes associated with the first shared vertex; and
storing only one copy of the second set of per-primitive attributes associated with the second primitive.

10. A system, comprising:
a graphics processor comprising at least one streaming multiprocessor configured to execute a vertex shader and a geometry shader, and configured to:
determine that a portion of a graphics processor is operating in a fast geometry shader mode;
determine that a first vertex associated with a first set of per-vertex attributes is a first shared vertex;
determine that the first shared vertex is a non-provoking vertex corresponding to a first primitive that is associated with a first set of per-primitive attributes;
determine that the first shared vertex is a provoking vertex corresponding to a second primitive that is associated with a second set of per-primitive attributes by examining vertices of the second primitive and vertices of a third primitive when the first primitive is received;
store the first set of the per-vertex attributes associated with the first shared vertex; and
store only one copy of the second set of per-primitive attributes associated with the second primitive.

11. The system of claim 10 , wherein the graphics processor comprises a cache and the first set of the per-vertex attributes associated with the first shared vertex is stored in a cache entry.

12. The system of claim 11 , wherein the only one copy of the second set of per-primitive attributes associated with the second primitive is also stored in the cache entry.

13. The system of claim 11 , wherein the graphics processor is further configured to set a bit corresponding to the cache entry indicating that the first shared vertex is the provoking vertex.

14. A method, comprising:
determining that a portion of a graphics processor is operating in a fast geometry shader mode, wherein the graphics processor comprises at least one streaming multiprocessor configured to execute a vertex shader and a geometry shader;
determining that a first vertex associated with a first set of per-vertex attributes is a first shared vertex;
determining that the first shared vertex is a non-provoking vertex corresponding to a first primitive that is associated with a first set of per-primitive attributes;
determining that the first shared vertex is a provoking vertex corresponding to a second primitive that is associated with a second set of per-primitive attributes;
determining that a second vertex associated with a second set of per-vertex attributes is a second shared vertex;
determining that the second shared vertex is the provoking vertex corresponding to the first primitive;
duplicating the second set of per-vertex attributes associated with the second shared vertex to generate a duplicated set of per-vertex attributes associated with the second shared vertex;
storing the first set of the per-vertex attributes associated with the first shared vertex; and
storing only one copy of the second set of per-primitive attributes associated with the second primitive.

15. The method of claim 14 , wherein the first set of the per-vertex attributes associated with the first shared vertex is stored in a cache entry.

Assignees

• Nvidia Corp

Inventors

• Hakura Ziyad Sami
• Nelson Peter

Classifications

• G06T15/87 Primary

  Gouraud shading · CPC title

• G06T2200/04

  involving 3D image data · CPC title

• G06T2200/28

  involving image processing hardware · CPC title

• G06T15/005 Primary

  General purpose rendering architectures · CPC title

• G06T15/80

  Shading · CPC title