Abstract: A method and system for optimizing graphics program execution by allowing the sharing of shader resources is disclosed. The method includes accessing a graphics program using a shader pipeline. If a texture projective instruction is included in the graphics program, a determination is made as to whether a texture projective parameter q indicates a non-projective texture. If the texture projective parameter indicates a non-projective texture, the texture projective instruction is demoted and a resulting demoted texture instruction is executed using a plurality of interpolators of the shader pipeline, which requires fewer shader resources.

Abstract: A graphics processing subsystem includes a vertex processing unit that allows vertex shader programs to arbitrarily access data stored in vertex texture maps. The vertex processing unit includes a vertex texture fetch unit and vertex processing engines. The vertex processing engines operate in parallel to execute vertex shader programs that specify operations to be performed on vertices. In response to a vertex texture load instruction, a vertex processing engine dispatches a vertex texture request to the vertex texture fetch unit. The vertex texture fetch unit retrieves the corresponding vertex texture map data. While the vertex texture fetch unit is processing a vertex texture request, the requesting vertex processing engine is adapted to evaluate whether instructions that follow the vertex texture load instruction are dependent on the vertex texture map data, and if the instructions are not dependent on the vertex texture map data, to execute the additional instructions.

Abstract: A vertex texture cache unit enables vertex shader programs to arbitrarily access array data while minimizing pipeline stalls due to memory latency. The vertex texture cache unit receives vertex texture requests from multiple vertex processing engines, each executing multiple vertex shader programs. The vertex texture cache unit stores frequently accessed vertex texture map data in a cache memory. When a cache miss occurs, the vertex texture cache unit continues to process subsequent vertex texture requests while data is being retrieved from memory for the cache miss. Because the vertex texture cache unit may output vertex texture map data in a different order than the corresponding vertex texture requests are received, the vertex texture cache unit maintains the association between vertex texture map data and its set of attributes, so that the vertex texture map data is formatted correctly and returned to the appropriate vertex processing engine and vertex shader program.