Abstract: A technique for generating a trained discriminator is provided. The technique includes applying one or more of a glitched image or an unglitched image to a discriminator; receiving classification output from the discriminator; adjusting weights of the discriminator to improve classification accuracy of the discriminator; applying noise to a generator; receiving an output image from the generator; applying the output image to the discriminator to obtain a classification; and adjusting weights of one of the discriminator or the generator to improve ability of the generator to reduce classification accuracy of the discriminator, based on the classification.

Abstract: Systems, apparatuses, and methods for implementing a depth buffer pre-pass are disclosed. A rendering application uses a binning approach to render primitives of a virtual scene on a tile-by-tile basis, with each tile corresponding to a portion of the screen. The application causes a depth buffer pre-pass to be performed for the primitives of the tile before a pixel shader is invoked. During the depth buffer pre-pass, only the depth part of the virtual scene is rendered to determine which pixel samples are visible and which pixel samples are hidden. Then, the scene is redrawn, but the pixel samples that are hidden are not sent to the pixel shader. In cases where a relatively large percentage of primitives overlap, this technique increases the efficiency of the rendering application since pixel shading can be avoided for the pixel samples that are hidden.

Abstract: Some implementations provide systems, devices, and methods for implementing a cache replacement policy. A memory request is issued for attribute information associated with a node in an acceleration data structure. The attribute information associated with the node is inserted into a cache entry of the cache and an age associated with the cache entry is set to a value based on the attribute information, in response to the memory request causing a cache miss.

Abstract: A power supply circuit is provided for supplying power from multiple peripheral power supplies to a data processor. The power supply circuit includes a power bus, a plurality of load voltage converters each including an input coupled to the power bus and an output coupled to a respective one of multiple subsystems of the data processor, a plurality of input voltage converters each including an input for coupling to a respective one of multiple peripheral power supply voltages and an output coupled to the power bus, and a feedback control circuit having an input coupled to the power bus and a plurality of outputs coupled to respective ones of the input voltage converters for controlling a current draw of the respective input voltage converter.

Abstract: An integrated circuit includes a TSV extending from a first surface of a semiconductor substrate to a second surface of the semiconductor substrate and having a first end and a second end, and a non-volatile repair circuit. The non-volatile repair circuit includes a one-time programmable (OTP) element having a programming terminal, wherein in response to an application of a fuse voltage to the programming terminal, the OTP element electrically couples the first end of the TSV to the second end of the TSV.

Abstract: Methods are provided for creating objects in a way that permits an API client to explicitly participate in memory management for an object created using the API. Methods for managing data object memory include requesting memory requirements for an object using an API and expressly allocating a memory location for the object based on the memory requirements. Methods are also provided for cloning objects such that a state of the object remains unchanged from the original object to the cloned object or can be explicitly specified.

Abstract: Systems, apparatuses, and methods for implementing a safety monitor framework for a safety-critical inference application are disclosed. A system includes a safety-critical inference application, a safety monitor, and an inference accelerator engine. The safety monitor receives an input image, test data, and a neural network specification from the safety-critical inference application. The safety monitor generates a modified image by adding additional objects outside of the input image. The safety monitor provides the modified image and neural network specification to the inference accelerator engine which processes the modified image and provides outputs to the safety monitor. The safety monitor determines the likelihood of erroneous processing of the original input image by comparing the outputs for the additional objects with a known good result. The safety monitor complements the overall fault coverage of the inference accelerator engine and covers faults only observable at the network level.

Abstract: Devices, methods, and systems for secure communications on a computing device. A host operating system (OS) runs on a host processor in communication with a host memory. A secure OS runs on a coprocessor in communication with a secure memory. The coprocessor receives information from an external device over a secure peer-to-peer (P2P) connection. The secure P2P connection is managed by the secure OS and is not accessible by the host OS.

Abstract: A graphics processing architecture in one example performs vertex manipulation operations and pixel manipulation operations by transmitting vertex data to a general purpose register block, and performing vertex operations on the vertex data by a processor unless the general purpose register block does not have enough available space therein to store incoming vertex data; and continues pixel calculation operations that are to be or are currently being performed by the processor based on instructions maintained in an instruction store until enough registers within the general purpose register block become available.

Abstract: In a system with a master processor and slave processors, sync points are used in boot instructions. While executing the boot instructions, the slave processor determines whether the sync point is enabled. In response to determining the sync point is enabled, the slave processor pauses execution of the boot instructions, waits for commands from the master processor, receives commands from the master processor, executes the received commands until a release command is received, and then continues to execute boot instructions. In response to determining the sync point is not enabled, the slave processor continues to execute boot instructions.

Abstract: Systems and methods are disclosed that automatically generating a gameplay recording from an application. Techniques are provided to extract data from a buffer, the extracted data are associated with the application; to detect, based on a signature associated with the extracted data, the occurrence of an event; and upon detection of the occurrence of the event, to generate the gameplay recording from an output of the application.

Abstract: An encoding method is provided which includes receiving a plurality of images, obtaining values of elements in a portion of the images, sorting the elements according to different values of the elements, sorting the elements according to a number of occurrences of the different values and encoding the elements using a subset of the different values having corresponding numbers of occurrences that are higher than corresponding numbers of occurrences of other values. Examples also include a processing device and method for use with palette mode encoding in which the elements are a portion of pixels in images and the values are color values of the portion of pixels in the images.

Abstract: A data processor includes a plurality of requestors, a plurality of responders, and a data fabric. The data fabric is for routing requests between the plurality of requestors and the plurality of responders and has a plurality of non-operational power states including a normal C-state and a light-weight C-state. The light-weight C-state has lower entry and exit latencies than the normal C-state. The data fabric monitors traffic through the data fabric and places the data fabric in the light-weight C-state in response to detecting an idle traffic state.

Abstract: A disclosed technique includes transmitting data in a first buffer associated with a first display pipe to a first display associated with the first display pipe; transmitting data in a second buffer associated with a second display pipe to the first display; requesting wake-up of a memory; and refilling one or both of the first buffer and the second buffer from the memory.

Abstract: An electronic device includes a memory, a processor that executes a software entity, a page migration engine (PME), and an input-output memory management unit (IOMMU). The software entity and the PME perform operations for preparing to migrate a page of memory that is accessible by at least one IO device in the memory, the software entity and the PME set migration state information in a page table entry for the page of memory and information in reverse map table (RMT) entries involved with migrating the page of memory based on the operations being performed. The IOMMU controls usage of information from the page table entry and controls performance of memory accesses of the page of memory based on the migration state information in the page table entry and information in the RMT entries. When the operations for preparing to migrate the page of memory are completed, the PME migrates the page of memory in the memory.

Abstract: A processing device for executing a machine learning neural network operation includes memory and a processor. The processor is configured to receive input data at a layer of the machine learning neural network operation, receive a plurality of sorted filters to be applied to the input data, apply the plurality of sorted filters to the input data to produce a plurality of different feature maps, compress the plurality of different feature maps according to a similarity of the feature maps relative to each other and store the plurality of different feature maps in the memory.

Abstract: An electronic device that includes a decompression engine that includes N decoders and a decompressor decompresses compressed input data that includes N streams of data. Upon receiving a command to decompress compressed input data, the decompression engine causes each of the N decoders to decode a respective one of the N streams from the compressed input data separately and substantially in parallel with others of the N decoders. Each decoder outputs a stream of decoded data of a respective type for generating commands associated with a compression standard for decompressing the compressed input data. The decompressor next generates, from the streams of decoded data output by the N decoders, commands for decompressing the data using the compression standard to recreate the original data. The decompressor next executes the commands to recreate the original data and stores the original data in a memory or provides the original data to another entity.

Abstract: A technique for determining an adaptive quantization parameter offset for a block of encoded video includes obtaining a rate control factor for the quantization parameter, determining a content-based quantization parameter factor for the quantization parameter, determining an adaptive variance based quantization offset based on content-based quantization parameter factors for a frame prior to the current frame, and combining the rate control factor, the content-based quantization parameter factor, and the adaptive offset to generate the quantization parameter.

Abstract: A technique for detecting a glitch in an image is provided. The technique includes providing an image to a plurality of individual classifiers to generate a plurality of individual classifier outputs and providing the plurality of individual classifier outputs to an ensemble classifier to generate a glitch classification.

Abstract: Systems, apparatuses, and methods for implementing light volume rendering techniques are disclosed. A processor is coupled to a memory. A processor renders the geometry of a scene into a geometry buffer. For a given light source in the scene, the processor initiates two shader pipeline passes to determine which pixels in the geometry buffer to light. On the first pass, the processor renders a front-side of a light volume corresponding to the light source. Any pixels of the geometry buffer which are in front of the front-side of the light volume are marked as pixels to be discarded. Then, during the second pass, only those pixels which were not marked to be discarded are sent to the pixel shader. This approach helps to reduce the overhead involved in applying a lighting effect to the scene by reducing the amount of work performed by the pixel shader.