Abstract: Various resistor circuits and methods of making and using the same are disclosed. In one aspect, a method of manufacturing is provided that includes forming a resistor onboard an interposer. The resistor is adapted to dampen a capacitive network. The capacitive network has at least one capacitor positioned external to the interposer.

Abstract: Techniques are provided for using a translation lookaside buffer to provide low latency memory address translations for data streams. Clients of a memory system first prepare the address translation cache hierarchy by requesting that a translation pre-fetch stream is initialized. After the translation pre-fetch stream is initialized, the cache hierarchy returns an acknowledgment of completion to the client, which then begins to access memory. Pre-fetch streams are specified in terms of address ranges and are performed for large contiguous portions of the virtual memory address space.

Abstract: Techniques are provided for managing address translation request traffic where memory access requests can be made with differing quality-of-service levels, which specify latency and/or bandwidth requirements. The techniques involve translation lookaside buffers. Within the translation lookaside buffers, certain resources are reserved for specific quality-of-service levels. More specifically, translation lookaside buffer slots, which store the actual translations, as well as finite state machines in a work queue, are reserved for specific quality-of-service levels. The translation lookaside buffer receives multiple requests for address translation. The translation lookaside buffer selects requests having the highest quality-of-service level for which an available finite state machine is available.

Abstract: A method and apparatus for processing video utilize individually collected image enhancement statistic information from differing processor cores for a same frame or multi-view that are then either shared between the processor cores or used by a third processor core to combine the statistical information that has been individually collected to generate global image-enhancement control information. The global image enhancement control information is based on a global analysis of both left and right eye views for example using the independently generated statistic information for a pair of frames. Respective image output information is produced by each of the plurality of processor cores based on the global image enhancement control information, for display on one or more displays.

Abstract: A virtualized DisplayPort (DP) configuration data (DPCD) for multi-stream transport (MST) logical DP end points and non-DP end points allows DPCD configuration for links within a DisplayPort topology which are not configurable using DPCD. A virtualized DPCD may configure a link to an internal display of a MST sink device or a non-DP display to receive data using a dynamic refresh rate (DRR), display stream compression (DSC), panel self-refresh (PSR) and other DPCD configurable features.

Abstract: A system, method and computer program product to traverse a polygon mesh, partition the polygon mesh into a set of polygon fans based on the traversal order, and tessellate the set of polygon fans into triangles based on the traversal order. This transformation of the polygon mesh into a triangle mesh enables the polygon mesh to be compressed and decompressed using the SC3DMC standard.

Abstract: A method for determining a macroblock (MB) coding mode for a current MB in a dependent view. A window around a co-located MB in a base view is determined, wherein the co-located MB is a MB in the base view having a same location as the current MB in the dependent view. A coding mode complexity value (CMCV) is determined for each MB in the window, wherein the CMCV is based on a coding mode used to encode the MB. Rate distortion optimization (RDO) is performed for the current MB using a reduced number of coding modes if a total CMCV for all MBs in the window is less than a threshold, or using all supported coding modes if the total CMCV for all MBs in the window is greater than the threshold. A coding mode for the current MB is determined based on the RDO results.

Abstract: A method and device of over training a connection is provided. Noise is intentionally supplied and added to a signal that is subjected to a link training operation. The link training operation is used to obtain a link between a source device and a receiving device. The device includes a noise source from which noise is obtained and added to a signal to aid in link over-training.

Abstract: A method and apparatus of asynchronous scheduling in a graphics device includes sending one or more instructions from an instruction scheduler to one or more instruction first-in/first-out (FIFO) devices. An instruction in the one or more FIFO devices is selected for execution by a single-instruction/multiple-data (SIMD) pipeline unit. It is determined whether all operands for the selected instruction are available for execution of the instruction, and if all the operands are available, the selected instruction is executed on the SIMD pipeline unit. The self-timed arithmetic pipeline unit (SIMD pipeline unit) is effectively encapsulated in a synchronous, (e.g., clocked by global clock), scheduler and register file environment.

Abstract: The present invention provides for page table access and dirty bit management in hardware via a new atomic test[0] and OR and Mask. The present invention also provides for a gasket that enables ACE to CCI translations. This gasket further provides request translation between ACE and CCI, deadlock avoidance for victim and probe collision, ARM barrier handling, and power management interactions. The present invention also provides a solution for ARM victim/probe collision handling which deadlocks the unified northbridge. These solutions includes a dedicated writeback virtual channel, probes for IO requests using 4-hop protocol, and a WrBack Reorder Ability in MCT where victims update older requests with data as they pass the requests.

Abstract: Methods and apparatus for facilitating processing a reference frame to produce an output frame. Motion vector data for a block of reference frame pels estimates the displacement of the reference frame pels from corresponding pels in a prior input frame. Comparison metrics are produced for a pel of the reference frame with respect to that pel and a plurality of neighboring reference frame pels. A first comparison metric is based on a comparison with corresponding pels of a prior output frame that corresponds to the prior input frame as previously processed. A second comparison metric is based on a comparison with corresponding pels of a motion compensated prior output frame derived from applying motion vector data to the pels of the prior output frame. A pel of the output frame that corresponds to the reference frame pel is determined using the first and second comparison metrics.

Abstract: Techniques for performing cache invalidates and write-backs in an accelerated processing device (e.g., a graphics processing device that renders three-dimensional graphics) are disclosed. The techniques involve receiving requests from a “master” (e.g., the central processing unit). The techniques involve invalidating virtual-to-physical address translations in an address translation request. The techniques include splitting up the requests based on whether the requests target virtually or physically tagged caches. Addresses for the portions of a request that target physically tagged caches are translated using invalidated virtual-to-physical address translations for speed. The split up request is processed to generate micro-transactions for individual caches targeted by the request. Micro-transactions for physically and virtually tagged caches are processed in parallel. Once all micro-transactions for a request have been processed, the unit that made the request is notified.

Abstract: Described herein is a method and system for directly accessing and transferring data between a first memory architecture and a second memory architecture associated with a graphics processing unit (GPU) or a discrete GPU (dGPU). In particular, a method is described for transferring data between the first memory architecture and the second memory architecture that bypasses interaction with a system memory of a processor and a root complex. A transfer command is sent from the processor, (or a host agent in the GPU or dGPU), to a first memory architecture controller. The first memory architecture controller initiates the transfer of the data directly between the first memory architecture and the second memory architecture. The method bypasses: 1) a host root complex; and 2) storing the data in the system memory and then having to transfer the data to the second memory architecture or the first memory architecture.

Abstract: Described is a solid state graphics (SSG) subsystem including a die and a package, where the die includes a memory hub, graphics processing unit(s) (GPU(s)) connected to the memory hub, first memory architecture controller(s) connected to the memory hub and directly controlling access to first memory architecture(s), second memory architecture controller associated with each GPU and each second memory architecture controller connected to the memory hub and second memory architecture(s), an expansion bus first memory architecture controller connected to the memory hub and being an endpoint for a host system and an expansion bus controller coupled to the expansion bus first memory architecture controller and capable of connecting to the host system. The first memory architecture(s) and the second memory architecture(s) are either located on the SSG subsystem, located on the package, or a combination thereof.

Abstract: Described herein are apparatus for connecting a first memory architecture locally to a graphics processing unit (GPU) through a local switch, where the first memory architecture can be a non-volatile memory (NVM) or other similarly used memories, for example, along with associated controllers. The apparatus includes the GPU(s) or discrete GPU(s) (dGPU(s)) (collectively GPU(s)), second memory architectures associated with the GPU(s), the local switch, first memory architecture(s), first memory architecture controllers or first memory architecture connector(s). In an implementation, the local switch is part of the GPU. The apparatus can also include a controller for distributing a large transaction among multiple first memory architectures. In an implementation, the first memory architectures can be directly connected to the GPU. In an implementation, the apparatus is user configurable. In an implementation, the apparatus is a solid state graphics (SSG) card.

Abstract: Described herein is a method and system for accessing a block addressable input/output (I/O) device, such as a non-volatile memory (NVM), as byte addressable memory. A front end processor connected to a Peripheral Component Interconnect Express (PCIe) switch performs as a front end interface to the block addressable I/O device to emulate byte addressability. A PCIe device, such as a graphics processing unit (GPU), can directly access the necessary bytes via the front end processor from the block addressable I/O device. The PCIe compatible devices can access data from the block I/O devices without having to go through system memory and a host processor. In an implementation, a system can include block addressable I/O, byte addressable I/O and hybrids thereof which support direct access to byte addressable memory by the host processor, GPU and any other PCIe compatible device.

Abstract: A method for rendering a scene across N number of processors is provided. The method includes evaluating performance statistics for each of the processors and establishing load rendering boundaries for each of the processors, the boundaries defining a respective portion of the scene. The method also includes dynamically adjusting the boundaries based upon the establishing and the evaluating.

Abstract: Techniques for removing or identifying overlapping fragments in a fragment stream after z-culling are disclosed. The techniques include maintaining a first-in-first-out buffer that stores post-z-cull fragments. Each time a new fragment is received at the buffer, the screen position of the fragment is checked against all other fragments in the buffer. If the screen position of the fragment matches the screen position of a fragment in the buffer, then the fragment in the buffer is removed or marked as overlapping. If the screen position of the fragment does not match the screen position of any fragment in the buffer, then no modification is performed to fragments already in the buffer. In either case, he fragment is added to the buffer. The contents of the buffer are transmitted to the pixel shader for pixel shading at a later time.

Abstract: Briefly, methods and apparatus provide image content to, and display image content on, displays with a variable refresh rate that reduce frame delays and avoid display image flickering problems. In one example, the methods and apparatus are operative to vary a display's refresh rate by varying a current frame's vertical blanking period by re-providing the current frame for display prior to providing a new frame for display. In this fashion, the displaying of a new frame may be advanced by assuring that a new frame can be provided for display as soon as it has been rendered and available for display. In addition, by re-providing the current frame for display prior to providing a new frame for display, new frames may be provided for display at rates within a safe rate range such that display image flickering issues are avoided or reduced.

Abstract: A processing apparatus is provided which includes memory configured to store hardware parameter settings for each of a plurality of applications. The processing apparatus also includes a processor in communication with the memory configured to store, in the memory, the hardware parameter settings, identify one of the plurality of applications as a currently executing application and control an operation of hardware by tuning a plurality of hardware parameters according to the stored hardware parameter settings for the identified application.