Global spreader and method for a parallel graphics processor

• US 20070030280A1
• Filed: 08/08/2005
• Published: 02/08/2007
• Est. Priority Date: 08/08/2005
• Status: Abandoned Application

• Current Assignee: VIA Technologies Incorporated (VIA Technologies)
• Original Assignee: VIA Technologies Incorporated (VIA Technologies)
• Inventors: Prokopenko, Boris, Gladding, Derek, Paltashev, Timour
• Application Number:
  US11/199,459
• Publication Number:
  US 20070030280A1
• Time in Patent Office: Days
• US Class Current: 345/506
• CPC Class Codes: G06T 1/20 Processor architectures; Processor configuration, e.g. pipelining

Abstract

A parallel graphics processor having a spreader coupled to a plurality of execution components is disclosed. The spreader maintains status information for each of the plurality of execution components and establishes a priority for each of the plurality of execution blocks to receive a graphics entity to be processed. The priorities are arranged in accordance with the maintained status information and a type of graphics entity to be processed. The spreader communicates a request to a selected execution component to allocate the graphics entity to be processed in its entity descriptor table and copies graphics entity data to the selected execution component. The spreader indexes assignment of the graphics entity in its logical table and subsequently receives indication from the selected instruction execution component that the graphics entity has been processed. Thereafter, graphics images may be presented on a display.

Claims

1. A parallel graphics processor having a spreader coupled to a plurality of instruction execution components, the spreader comprising:

logic configured to maintain status information for each of the plurality of instruction execution components;

logic configured to establish a priority for each of the plurality of execution components to receive a graphics entity to be processed in accordance with the maintained status information and a type of graphics entity to be processed;

logic configured to communicate a request to a selected instruction execution component of the plurality of instruction execution components to allocate the graphics entity to be processed in an entity descriptor table of the selected instruction execution component;

logic configured to copy data associated with the graphics entity to be processed to the selected instruction execution component;

logic configured to index assignment of the graphics entity to be processed to the selected instruction execution component in a logical table; and

logic configured to receive indication from the selected instruction execution component that the graphics entity to be processed has been processed, wherein graphics images are subsequently presented on a display and resources associated with the graphics entity are released.

2. The processor of claim 1, wherein the maintained status information includes data associated with an availability of the instruction execution component for accepting an assignment from the spreader.

3. The processor of claim 1, further comprising:

logic configured to determine an amount of resources to be accessed for processing the graphics entity to be processed.

4. The processor of claim 1, wherein the graphics entity to be processed is a packet of vertexes, the spreader further comprising:

logic configured to allocate an entry in a vertex descriptor table in the spreader; and

logic configured to fill the allocated entry in the vertex descriptor table with an index and a number associated with the graphics entity to be processed, the number being allocated by the selected instruction execution component.

5. The processor of claim 4, wherein the vertex descriptor table contains records associated with a logical location for a group of vertices contained in the packet of vertexes, the logical location corresponding with each of the one or more instruction execution components processing the a group of vertices of the packet of vertexes.

6. The processor of claim 4, wherein the vertex descriptor table retains data corresponding to destination information for a packet of vertexes.

7. The processor of claim 1, wherein the graphics entity to be processed is a packet of triangles, the spreader further comprising:

logic configured to use indexes of a plurality of vertices associated with the packet of triangles;

logic configured to retrieve entity numbers for the plurality of vertices;

logic configured to extract vertical element numbers associated with the plurality of vertices; and

logic configured to communicate the entity numbers for the plurality of vertices and the element numbers to the selected instruction execution component.

8. The processor of claim 7, wherein the spreader communicates to the selected instruction execution component assigned to process the graphics entity address information corresponding to one or more other instruction execution components assigned to process vertex data associated with the one or more packets of triangles.

9. The processor of claim 1, wherein the graphics entity to be processed is a packet of pixels, the spreader further comprising:

logic configured to communicate entity and element numbers for one or more triangles associated to the packet of pixels to the selected instruction execution component.

10. The processor of claim 1, wherein spreader communicates the request to the selected instruction execution component that is a first candidate execution component according to the predetermined priority, and wherein the spreader communicates the request to the selected execution component that is a subsequent candidate execution component if the first candidate execution component does not accept the request.

11. The processor of claim 1, wherein spreader communicates the request to one or more instruction execution components that is a next candidate execution component according to the predetermined priority until at least one of the one or more instruction execution components accepts the request.

12. A method for a parallel graphics processor having a spreader coupled to a plurality of instruction execution components to process graphics entities, comprising the steps of:

collecting status information for each of the plurality of instruction execution components;

establishing a priority for each of the plurality of execution blocks to receive a graphics entity to be processed in accordance with the maintained status information and a type of graphics entity to be processed;

communicating a request to a selected instruction execution component of the plurality of instruction execution units to allocate the graphics entity to be processed in an entity descriptor table of the selected instruction execution component;

copying data associated with the graphics entity to be processed to the selected instruction execution component;

indexing assignment of the graphics entity to be processed to the selected instruction execution component in a logical table; and

receiving indication from the selected instruction execution component that the graphics entity to be processed has been processed, wherein graphics images are subsequently presented on a display.

13. The method of claim 12, wherein the collected status information includes data associated with an availability of the instruction execution component for accepting an assignment-from the spreader.

14. The method of claim 12, further comprising the step of:

determining an amount of resources to be accessed for processing the graphics entity to be processed.

15. The method of claim 12, wherein the graphics entity to be processed is a packet of vertexes, the method further comprising the steps of:

allocating an entry in a vertex descriptor table in the spreader; and

filling the allocated entry in the vertex descriptor table with an index and a number associated with the graphics entity to be processed, the number being allocated by the selected instruction execution component.

16. The method of claim 15, wherein the vertex descriptor table contains records associated with a logical location for a group of vertices contained in the packet of vertexes, the logical location corresponding with each of the one or more instruction execution components processing the a group of vertices of the packet of vertexes.

17. The method of claim 12, wherein the graphics entity to be processed is a packet of triangles, the method further comprising the steps of:

using indexes of a plurality of vertices associated with the packet of triangles;

retrieving entity numbers for the plurality of vertices;

extracting vertical element numbers associated with the plurality of vertices; and

communicating the entity numbers for the plurality of vertices and the element numbers to the selected instruction execution component.

18. The method of claim 17, wherein the spreader communicates to the selected instruction execution component assigned to process the graphics entity address information corresponding to one or more other instruction execution components assigned to process vertex data associated with the one or more packets of triangles.

19. The method of claim 12, wherein the graphics entity to be processed is a packet of pixels, the method further comprising the steps of:

communicating entity and element numbers for one or more triangles associated to the packet of pixels to the selected instruction execution component.

20. The method of claim 12, further comprising the step of:

communicating the request to one or more selected instruction execution components that is a next candidate execution component according to the established priority until at least one of the one or more selected instruction execution components accepts the request.