Abstract: The disclosed system may include a processor configured to encode, using an encoding scheme that reduces a number of bits needed to represent one or more instructions from a set of instructions in an instruction buffer represented by a dependency matrix, a dependency indicating that a child instruction represented in the dependency matrix depends on a parent instruction represented in the dependency matrix. The processor may also be configured to store the encoded dependency in the dependency matrix and dispatch instructions in the instruction buffer based at least on decoding one or more dependencies stored in the dependency matrix for the instructions. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: A processor includes a plurality of execution pipes and a distributed scheduler coupled to the plurality of execution pipes. The distributed scheduler includes a first queue to buffer instruction operations from a front end of an instruction pipeline of the processor and a plurality of second queues, wherein each second queue is to buffer instruction operations allocated from the first queue for a corresponding separate subset of execution pipes of the plurality of execution pipes. The distributed scheduler further includes a queue controller to select an allocation mode from a plurality of allocation modes based on whether at least one indicator of an imbalance at the distributed scheduler is detected, and further to control the distributed scheduler to allocate instruction operations from the first queue among the plurality of second queues in accordance with the selected allocation mode.

Abstract: A data processor accesses a memory having a first pseudo channel and a second pseudo channel. The data processor includes at least one memory accessing agent for generating a memory access request, a memory controller for providing a memory command to the memory in response to a normalized request selectively using a first pseudo channel pipeline circuit and a second pseudo channel pipeline circuit, and a data fabric for converting the memory access request into the normalized request selectively for the first pseudo channel and the second pseudo channel.

Abstract: Techniques for performing shader operations are provided. The techniques include, performing pixel shading at a shading rate defined by pixel shader variable rate shading (“VRS”) data, and updating the pixel VRS data that indicates one or more shading rates for one or more tiles based on whether the tiles of the one or more tiles include triangle edges or do not include triangle edges, to generate updated VRS data.

Abstract: A virtual padding unit provides a virtual padded data structure (e.g., virtually padded matrix) that provides output values for a padded data structure without storing all of the padding elements in memory. When the virtual padding unit receives a virtual memory address of a location in the virtual padded data structure, the virtual padding unit checks whether the location is a non-padded location in the virtual padded data structure or a padded location in the virtual padded data structure. If the location is a padded location in the virtual padded data structure, the virtual padding unit outputs a padding value rather than a value stored in the virtual padded data structure. If the location is a non-padded location in the virtual padded data structure, a value stored at the location is output.

Abstract: A technique for operating a cache is disclosed. The technique includes utilizing a first portion of a cache in a directly accessed manner; and utilizing a second portion of the cache as a cache.

Abstract: A computer vision processing device is provided which comprises memory configured to store data and a processor. The processor is configured to store captured image data in a first buffer and acquire access to the captured image data in the first buffer when the captured image data is available for processing. The processor is also configured to execute a first group of operations in a processing pipeline, each of which processes the captured image data accessed from the first buffer and return the first buffer for storing next captured image data when a last operation of the first group of operations executes.

Abstract: A method includes recording a first set of consecutive memory access deltas, where each of the consecutive memory access deltas represents a difference between two memory addresses accessed by an application, updating values in a prefetch training table based on the first set of memory access deltas, and predicting one or more memory addresses for prefetching responsive to a second set of consecutive memory access deltas and based on values in the prefetch training table.

Abstract: An apparatus and method for supporting communication during error handling in a computing system. A computing system includes a first partition and a second partition, each capable of performing error management based on a respective machine check architecture (MCA). The first partition includes a host processor that executes an exception handler for managing reported errors. A message converter unit of the second partition assists in generating messages based on detected errors in the second partition. The message converter unit receives requests from hardware components of the second partition for handling errors and translates MCA addresses between the first partition and the second partition. To support the message converter unit, during an earlier bootup operation, the second partition communicates the hardware topology of the second partition to the host processor, and the host processor sends address translation information.

Abstract: A processing device and method for efficient transitioning to and from a reduced power state is provided. The processing device comprises a plurality of components having assigned registers used to store data to execute a program and a power management controller, in communication with the plurality of components. The power management controller receives an indication that the plurality of components are idle, executes a process to enter a component into a reduced power state in response to receiving an acknowledgement from the component of a request from the power management controller to remove power to the component, and executes a process to exit the component from the reduced power state in response to the component being active.

Abstract: A generator for generating artificial data, and training for the same. Data corresponding to a first label is altered within a reference labeled data set. A discriminator is trained based on the reference labeled data set to create a selectively poisoned discriminator. A generator is trained based on the selectively poisoned discriminator to create a selectively poisoned generator. The selectively poisoned generator is tested for the first label and tested for the second label to determine whether the generator is sufficiently poisoned for the first label and sufficiently accurate for the second label. If it is not, the generator is retrained based on the data set including the further altered data. The generator includes a first ANN to input first information and output a set of artificial data that is classifiable using a first label and not classifiable using a second label of the set of labeled data.

Abstract: Cache directory lookup address augmentation techniques are described. In one example, a system includes a cache system including a plurality of cache levels and a cache coherence controller. The cache coherence controller is configured to perform a cache directory lookup using a cache directory. The cache directory lookup is configured to indicate whether data associated with a memory address specified by a memory request is valid in memory. The cache directory lookup is augmented to include an additional memory address based on the memory address.

Abstract: Techniques for performing memory operations are disclosed herein. The techniques include generating a plurality of performance log entries based on observed operations; generating a plurality of memory access log entries based on the observed operations, wherein each performance log entry of the plurality of performance log entries are associated with one or more memory access log entries of the plurality of memory access log entries, wherein each performance log entry is associated with an epoch; and wherein each memory access log entry is associated with an epoch and a memory address range.

Abstract: Devices and methods for rendering curves using ray tracing are provided which include tessellating a curve, representing at least a portion of an object in a scene, into a chain of capsules each comprising two spheres and a connecting cone, generating an acceleration structure comprising the chain of capsules, casting a ray in a space comprising the curve, and performing, for a capsule of the chain of capsules, a closed-form intersection test to render the curve. In a first example, the closed-form intersection test is performed using a single quadratic equation quadratic based on coefficients from input values of the two spheres. In a second example, the closed-form intersection test is performed based on an intersection between the ray and a blended sphere generated from a smallest distance between the ray and a centerline of the capsule and an offset.

Abstract: Modifying memory page attributes using a programmable page attribute register of a core executing a process is described. In accordance with the described techniques, a host includes a core that is configured to generate a modified page table attribute for a page in system memory. The modified page table attribute represents at least one demoted permission for a page as specified by a system page table. The core is configured to maintain the modified page table attribute locally in the programmable page attribute register and execute at least one operation allocated to the page according to the modified page table attribute.

Abstract: A technique for scheduling executing items on a highly parallel processing architecture is provided. The technique includes identifying a plurality of execution items that share data, as indicated by having matching commonality metadata; identifying an execution unit for executing the plurality of execution items together; and scheduling the plurality of execution items for execution together on the execution unit.

Abstract: A memory sprint controller, responsive to an indicator of an irregular memory access phase, causes a memory controller to enter a sprint mode in which it temporarily adjusts at least one timing parameter of a dynamic random access memory (DRAM) to reduce a time in which a designated number of activate (ACT) commands are allowed to be dispatched to the DRAM.

Abstract: Task allocation with chipset attached memory and additional processing unit is described. In accordance with the described techniques, a computing device includes a main system and one or more sub-systems which are coupled to the main system via a chipset link. The main system includes at least a processing unit and a system memory. The one or more sub-systems each include at least a chipset attached processing unit and a chipset attached memory. Contents of the system memory are transferable to the chipset attached memory of the sub-system via the chipset link to enable the chipset attached processing unit to perform the one or more tasks using the contents from the chipset attached memory.

Abstract: A memory sprint controller, responsive to an indicator of an irregular memory access phase, causes a memory controller to enter a sprint mode in which it temporarily adjusts at least one timing parameter of a dynamic random access memory (DRAM) to reduce a time in which a designated number of activate (ACT) commands are allowed to be dispatched to the DRAM.

Abstract: A disclosed technique includes based on a clock pattern, determining an enable configuration for setting enable signals for one or more multi-cycle paths of a hardware logic network, setting the enable configuration for the one or more multi-cycle paths, and executing testing operations for the hardware logic network with the one or more multi-cycle paths enabled according to the enable configuration.