Parallel duplicate primitive vertex detection and batching

A non-transitory computer-readable storage medium storing instructions that, when executed by a processor generates a batch specifying multiple primitives defined by vertex indices, by performing the steps of:

• receiving a set of input vertex indices that reference vertex data;
• inserting the set of input vertex indices into a sequence of previously received vertex indices to update a history list of vertex indices;
• simultaneously comparing each input vertex index in the set of input vertex indices with each previously received vertex index in the history list of vertex indices and each preceding input vertex in the set of input vertex indices to produce history list hit/miss results for each input vertex index;
• identifying a number of the input vertex indices that define primitives fitting in the batch based on the history list hit/miss results;
• assembling the number of the input vertex indices into slots to produce the batch, wherein any duplicate input vertex indices reference a slot storing a previous instance of the duplicate input vertex;
• and
• outputting the batch for processing to produce rasterized pixel data for each primitive defined by the number of input vertex indices in the batch.

A system for generating a batch specifying multiple primitives defined by vertex indices, the system comprising:

• a slot assignment unit configured to:
• receive a set of input vertex indices that reference vertex data;
• insert the set of input vertex indices into a sequence of previously received vertex indices to update a history list of vertex indices;
• simultaneously compare each input vertex index in the set of input vertex indices with each previously received vertex index in the history list of vertex indices and each preceding input vertex in the set of input vertex indices to produce history list hit/miss results for each input vertex index;
• identify a number of the input vertex indices that define primitives fitting in the batch based on the history list hit/miss results;
• assemble the number of the input vertex indices into slots to produce the batch, wherein any duplicate input vertex indices reference a slot storing a previous instance of the duplicate input vertex;
• and
• output the batch for processing to produce rasterized pixel data for each primitive defined by the number of input vertex indices in the batch.