Abstract: One embodiment of the present invention includes techniques for rasterizing primitives that include edges shared between paths. For each edge, a rasterizer unit selects and applies a sample rule from multiple sample rules. If the edge is shared, then the selected sample rule causes each group of coverage samples associated with a single color sample to be considered as either fully inside or fully outside the edge. Consequently, conflation artifacts caused when the number of coverage samples per pixel exceeds the number of color samples per pixel may be reduced. In prior-art techniques, reducing such conflation artifacts typically involves increasing the number of color samples per pixel to equal the number of coverage samples per pixel. Advantageously, the disclosed techniques enable rendering using algorithms that reduce the ratio of color to coverage samples, thereby decreasing memory consumption and memory bandwidth use, without causing conflation artifacts associated with shared edges.

Abstract: One embodiment of the present invention sets forth a technique for performing a computer-implemented method that controls memory access operations. A stream of graphics commands includes at least one memory barrier command. Each memory barrier command in the stream of graphics command delays memory access operations scheduled for any command specified after the memory barrier command until all memory access operations scheduled for commands specified prior to the memory barrier command have completely executed.

Abstract: A method of binding graphics resources is provided that includes: (1) identifying graphics resources for binding, (2) generating a bind group for the graphics resources, (3) organizing the bind group into a bind group memory using a bind group layout and (4) providing bind group control for processing of the bind group. A method of organizing graphics resources and a resource organizing unit are also provided.

Abstract: Techniques for passing dependencies in an application programming interface API includes identifying a plurality of passes of execution commands. For each set of passes, wherein one pass is a destination pass and the other pass is a source pass to each other, one or more dependencies, of one or more dependency types, are determined between the execution commands of the destination pass and the source pass. Pass objects are then created for each identified pass, wherein each pass object records the execution commands and dependencies between the corresponding destination and source passes.

Abstract: One embodiment of the present invention sets forth a method for generating work to be processed by a graphics pipeline residing within a graphics processor. The method includes the steps of receiving an indication that a first graphics workload is to be submitted to a command queue associated with the graphics processor, allocating a first portion of shader accessible memory for one or more units of state information that are necessary for processing the first graphics workload, populating the first portion of shader accessible memory with the one or more units of state information, and transmitting to the command queue of the graphics processor the one or more units of state information stored within the first portion of shader accessible memory, wherein the first graphics workload is processed within the graphics pipeline based on the one or more units of state information.

Abstract: One embodiment of the present invention sets for a method for accessing data objects stored in a memory that is accessible by a graphics processing unit (GPU). The method comprises the steps of creating a data object in the memory based on a command received from an application program, transmitting a first handle associated with the data object to the application program such that data associated with different graphics commands can be accessed by the GPU, wherein the first handle includes a memory address that provides access to only a particular portion of the data object, receiving a first graphics command as well as the first handle from the application program, wherein the first graphics command includes a draw command or a compute grid launch, and transmitting the first graphics command and the first handle to the GPU for processing.

Abstract: One embodiment of the present invention sets forth a method for storing processed data within buffer objects stored in buffer object memory from within shader engines executing on a GPU. The method comprises the steps of receiving a stream of one or more shading program commands via a graphics driver, executing, within a shader engine, at least one of the one or more shading program commands to generate processed data, determining from the stream of one or more shading program commands an address associated with a first data object stored within the buffer memory, and storing, from within the shader engine, the processed data in the first data object stored within the buffer memory.

Abstract: One embodiment of the present invention sets forth a technique for compositing a rendered path object into an image buffer. A shader program executing within a graphics processing unit (GPU) performs a stenciling operation for the path object and subsequently performs a texture barrier operation, which invalidates caches configured to store texture and frame buffer data within the GPU. The shader program then performs covering operation for the path object in which the shader renders color samples for the path object and composites the color samples into an image buffer. The shader program binds to the image buffer for access as both a texture map and a writeable image. Stencil values are reset when corresponding pixels are written once per path object, and texture caches are invalidated via the texture barrier operation, which is performed after each covering operation per path object.

Abstract: One embodiment of the present invention enables threads executing on a processor to locally generate and execute work within that processor by way of work queues and command blocks. A device driver, as an initialization procedure for establishing memory objects that enable the threads to locally generate and execute work, generates a work queue, and sets a GP_GET pointer of the work queue to the first entry in the work queue. The device driver also, during the initialization procedure, sets a GP_PUT pointer of the work queue to the last free entry included in the work queue, thereby establishing a range of entries in the work queue into which new work generated by the threads can be loaded and subsequently executed by the processor. The threads then populate command blocks with generated work and point entries in the work queue to the command blocks to effect processor execution of the work stored in the command blocks.

Abstract: A method for reducing redundant rendering of frames includes receiving draw calls including state information for a frame. The method includes generating respective bounding boxes for the draw calls. The bounding box is generated based on vertex data, vertex programs and transformation matrices. The method includes comparing the draw calls of the frame to the draw calls of one or more previous frames and identifying draw calls that are not identical in the compared frames. The method includes identifying the bounding boxes containing altered regions of the frames based on the draw calls that are not identical in the compared frames. The method includes reducing the altered regions into a smaller set of clip rectangles and rendering only inside the clip rectangles.

Abstract: A raster operations (ROP) unit is configured to compress stencil values included in a stencil buffer. The ROP unit divides the stencil values into groups, subdivides each group into two halves, and selects an anchor value for each half. If the difference between each of the stencil values and the corresponding anchor lies within an offset range, and the difference between the two anchors lies within a delta range, then the group is compressible. For a compressible group, the ROP unit encodes the anchor value, offsets from anchors, and an anchor delta. This encoding enables the ROP unit to operate on the compressed group instead of the uncompressed stencil values, reducing the number of memory and computational operations associated with the stencil values. Consequently, the ROP unit reduces memory bandwidth use, reduces power consumption, and increases rendering rate compared to conventional ROP units that implement less flexible compression techniques.

Abstract: One embodiment of the present invention includes a memory management unit (MMU) that is configured to manage sparse mappings. The MMU processes requests to translate virtual addresses to physical addresses based on page table entries (PTEs) that indicate a sparse status. If the MMU determines that the PTE does not include a mapping from a virtual address to a physical address, then the MMU responds to the request based on the sparse status. If the sparse status is active, then the MMU determines the physical address based on whether the type of the request is a write operation and, subsequently, generates an acknowledgement of the request. By contrast, if the sparse status is not active, then the MMU generates a page fault. Advantageously, the disclosed embodiments enable the computer system to manage sparse mappings without incurring the performance degradation associated with both page faults and conventional software-based sparse mapping management.

Abstract: Approaches are disclosed for performing memory access operations in a texture processing pipeline having a first portion configured to process texture memory access operations and a second portion configured to process non-texture memory access operations. A texture unit receives a memory access request. The texture unit determines whether the memory access request includes a texture memory access operation. If the memory access request includes a texture memory access operation, then the texture unit processes the memory access request via at least the first portion of the texture processing pipeline, otherwise, the texture unit processes the memory access request via at least the second portion of the texture processing pipeline. One advantage of the disclosed approach is that the same processing and cache memory may be used for both texture operations and load/store operations to various other address spaces, leading to reduced surface area and power consumption.

Abstract: Techniques are disclosed for performing memory access operations. A texture unit receives a memory access operation that includes a tuple associated with a first view in a plurality of views. The texture unit retrieves a first hash value associated with a first texture header in a plurality of texture headers, where the first texture header is related to the first view. The texture unit retrieves a second hash value associated with a second texture header in the plurality of texture headers, where the second texture header is related to a second view. The texture unit determines whether the first view is potentially aliased with the second view, based on the first and second hash values. If so, then the texture unit invalidates a cache entry in a cache memory associated with the second texture header. Otherwise, the texture unit maintains the cache entry.

Abstract: A graphics processing system includes a central processing unit that processes a cubic Bezier curve corresponding to a filled cubic Bezier path. Additionally, the graphics processing system includes a cubic preprocessor coupled to the central processing unit that formats the cubic Bezier curve to provide a formatted cubic Bezier curve having quadrilateral control points corresponding to a mathematically simple cubic curve. The graphics processing system further includes a graphics processing unit coupled to the cubic preprocessor that employs the formatted cubic Bezier curve in rendering the filled cubic Bezier path. A rendering unit and a display cubic Bezier path filling method are also provided.

Abstract: One embodiment of the present invention sets forth a technique for transmitting state information associated with at least one graphics command to a graphics processor. The method includes the steps of generating a state object that specifies a set of properties that is needed to execute a first graphics command within the graphics processor, storing in the state object a value associated with a first property included in the set of properties, marking a second property included in the set of properties as a dynamic property, where a value associated with the second property is not stored in the state object and can be updated without having to modify the state object, and transmitting the state object to the graphics processor in order to execute the first graphics command.

Abstract: One embodiment of the present invention sets forth a technique for performing low latency computation on a parallel processing subsystem. A low latency functional node is exposed to an operating system. The low latency functional node and a generic functional node are configured to target the same underlying processor resource within the parallel processing subsystem. The operating system stores low latency tasks generated by a user application within a low latency command buffer associated with the low latency functional node. The parallel processing subsystem advantageously executes tasks from the low latency command buffer prior to completing execution of tasks in the generic command buffer, thereby reducing completion latency for the low latency tasks.

Abstract: One embodiment of the present invention includes techniques for rasterizing primitives that include edges shared between paths. For each edge, a rasterizer unit selects and applies a sample rule from multiple sample rules. If the edge is shared, then the selected sample rule causes each group of coverage samples associated with a single color sample to be considered as either fully inside or fully outside the edge. Consequently, conflation artifacts caused when the number of coverage samples per pixel exceeds the number of color samples per pixel may be reduced. In prior-art techniques, reducing such conflation artifacts typically involves increasing the number of color samples per pixel to equal the number of coverage samples per pixel. Advantageously, the disclosed techniques enable rendering using algorithms that reduce the ratio of color to coverage samples, thereby decreasing memory consumption and memory bandwidth use, without causing conflation artifacts associated with shared edges.

Abstract: One embodiment of the present invention includes techniques for rasterizing geometries. First, a processing unit defines a bounding primitive that covers the geometry and does not include any internal edges. If the bounding primitive intersects any enabled clip plane, then the processing unit generates fragments to fill a current viewport. Alternatively, the processing unit generates fragments to fill the bounding primitive. Because the rasterized region includes no internal edges, conflation artifacts caused when the number of coverage samples per pixel exceeds the number of color samples per pixel may be reduced. In prior-art techniques, reducing such conflation artifacts typically involves increasing the number of color samples per pixel to equal the number of coverage samples per pixel.

Abstract: One embodiment of the present invention includes techniques for rasterizing primitives that include edges shared between paths. For each edge, a rasterizer unit selects and applies a sample rule from multiple sample rules. If the edge is shared, then the selected sample rule causes each group of coverage samples associated with a single color sample to be considered as either fully inside or fully outside the edge. Consequently, conflation artifacts caused when the number of coverage samples per pixel exceeds the number of color samples per pixel may be reduced. In prior-art techniques, reducing such conflation artifacts typically involves increasing the number of color samples per pixel to equal the number of coverage samples per pixel. Advantageously, the disclosed techniques enable rendering using algorithms that reduce the ratio of color to coverage samples, thereby decreasing memory consumption and memory bandwidth use, without causing conflation artifacts associated with shared edges.