Abstract: Method and apparatus monitor eviction conflicts among cache directory entries in a cache directory and produce cache directory victim entry information for a memory manager. In some examples, the memory manager reduces future cache directory conflicts by changing a page level physical address assignment for a page of memory based on the produced cache directory victim entry information. In some examples, a scalable data fabric includes hardware control logic that performs the monitoring of the eviction conflicts among cache directory entries in the cache directory and produces the cache directory victim entry information.

Abstract: A system and method for efficiently creating layout for memory bit cells are described. In various implementations, a memory bit cell uses Cross field effect transistors (FETs) that include vertically stacked gate all around (GAA) transistors with conducting channels oriented in an orthogonal direction between them. The channels of the vertically stacked transistors use opposite doping polarities. The memory bit cell includes one of a read bit line and a write word line routed in no other metal layer other than a local interconnect layer. In addition, a six transistor (6T) random access data storage of the given memory bit cell consumes a planar area above a silicon substrate of four transistors.

Abstract: A memory package includes first, second, third, and fourth channels arranged consecutively in a clockwise direction on the memory package, each of the first, second, third, and fourth channels having access circuitry and memory arrays. In a first mode, the first channel controls access to the memory arrays in the second channel and the fourth channel controls access to the memory arrays in the third channel.

Abstract: Concurrently with performing a visibility pass to generate visibility data for two or more bins of an image, a processing system determines whether a primitive to be rendered covers at least a predetermined threshold percentage of a tile of the image. In response to the primitive coving at least the predetermined threshold percentage of the tile, the processing system stores the depth data of the primitive in a depth buffer for pixel-based rendering. In response to the primitive not covering at least the predetermined threshold percentage of the tile, the processing system fuses the primitive with one or more preceding primitives sharing an edge with the primitive in the tile to generate a fused primitive. In response to the fused primitive being valid in the tile, the processing system passes the depth data of the fused primitive to the depth buffer.

Abstract: In accordance with the described techniques for data compression using reconfigurable hardware based on data redundancy patterns, a computing device includes a memory, processing-in-memory units, a host processing unit, and a compression unit having reconfigurable logic for performing multiple compression algorithms. The host processing unit issues processing-in-memory requests instructing the processing-in-memory units to scan a block of the memory for one or more data redundancy patterns, and to identify a compression algorithm of the multiple compression algorithms based on the one or more data redundancy patterns. Further, the host processing unit issues a memory request to access a memory address in the block of the memory. The memory request causes data of the memory address to be communicated from the block of the memory to the compression unit to be compressed using the compression algorithm.

Abstract: The disclosed device includes a processor and an interconnect connecting the processor to a memory. The interconnect includes an interconnect agent that can forward memory requests from the processor to the memory and receive requested data returned by the memory. The requested data can include information for a next memory request such that the interconnect agent can send, to the memory, a speculative memory request using information for the next memory request that was received in response to the memory request. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: A technique for rendering is provided. The technique includes performing a visibility pass that designates portions of shade space textures visible in a scene, wherein the visibility pass generates tiles that cover the shade space textures visible in the scene; performing a temporal rate controller operation; performing a shade space shading operation on the tiles that cover the shade space textures visible in the scene based on a temporal shading rate output by the temporal rate controller operation, wherein only a subset of samples in the tiles that cover the shade space textures visible in the scene are shaded in the shade space shading operation; and performing a reconstruction operation using output from the shade space shading operation to produce a final scene.

Abstract: In accordance with the described techniques, a host processor receives a task graph including tasks and indicating dependencies between the task graph. The host processor formats the task graph, in part, by sorting the tasks of the task graph in an order based on the dependencies between the tasks. Further, the host processor submits the formatted task graph to a scalable input/output virtualization (SIOV) device, which directs the SIOV device to process the tasks of the task graph based on the order.

Abstract: Method and apparatus for collaborative memory accesses is described. A system includes a memory controller that receives a command from a host. The command is associated with at least one of a plurality of data elements. The memory controller causes execution of data casting operations that adjust a bit size of the plurality of data elements to generate casted data elements. The system includes an interface for communicating data between the host and a memory.

Abstract: The disclosed device includes power circuits that can communicate with a control circuit. In response to a power circuit communicating a low efficiency state, the control circuit can redistribute at least a portion of a load of the power circuit to one or more other power circuits. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: A technique for building a bounding volume hierarchy is disclosed. The technique includes for a subject node, selecting a dimension along which to perform a split to form child nodes of the subject node; assigning primitives of the subject node to the child nodes; and updating bounds for the child nodes in a next split dimension and not in the other dimensions.

Abstract: A hybrid bonding method includes fabricating plural semiconductor devices in a region of a bottom wafer adjacent to a front surface thereof, fusion bonding the front surface to a carrier substrate, thinning the bottom wafer opposite to the front surface to expose conductive regions of the semiconductor devices, forming a dielectric layer over a backside of the semiconductor devices, forming openings in the dielectric layer to expose the conductive regions, forming metal pads within the openings, dicing the bottom wafer and the carrier substrate to singulate the plural semiconductor devices, bonding the dielectric layer overlying the backside of the semiconductor devices to a dielectric layer overlying a front surface of a top wafer, bonding the metal pads within the openings in the dielectric layer to metal pads overlying the front surface of the top wafer, and removing the carrier substrate from the front surface of the bottom wafer.

Abstract: The disclosed device includes a cache organized by sets and ways and a control circuit that selects a first way for a cache replacement from a first half of a set of ways. The control circuit also selects another way from a second half of the set of ways, and uses the second way for the cache replacement when the first way is unavailable. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: Selectively bypassing cache directory lookups for processing-in-memory instructions is described. In one example, a system maintains information describing a status—clean or dirty—of a memory address, where a dirty status indicates that the memory address is modified in a cache and thus different than the memory address as represented in system memory. A processing-in-memory request involving the memory address is assigned a cache directory bypass bit based on the status of the memory address. The cache directory bypass bit for a processing-in-memory request controls whether a cache directory lookup is performed after the processing-in-memory request is issued by a processor core and before the processing-in-memory request is executed by a processing-in-memory component.

Abstract: An apparatus can include: a processor; a voltage regulator configured to provide a processor voltage and a processor current to the processor; and a voltage regulator controller that can include a current sensor comprising an analog-to-digital converter (ADC) having an ADC input range and configured to provide current data based on an ADC input voltage, and a configuration manager configured to receive processor power data and adjust the ADC input range based on the processor power data. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: Systems and techniques for selectively transferring one or more portions of a cache block in response to a request are described. Computing system components are informed as to instances where data transfer operations involve moving less than an entirety of data included in a cache block cache block. In one example, executable code for a computational task includes hints that identify when memory requests involve accessing and transmitting less than an entirety of a cache block and cause system components to communicate a subset of the cache block during a memory access. In another example, a data differentiator unit is implemented to analyze a cache block and return a portion of the cache block that is selected based on one or more criteria specified for a computational task. The described techniques thus overcome conventional drawbacks facing systems that transmit an entire cache block when only a portion is needed.

Abstract: A technique for rendering is provided. The technique includes performing a visibility operation to generate shade space visibility information and reconstruction information; performing a shade space shading operation based on the shade space visibility information generate shaded shade space textures; and performing a reconstruction operation based on the reconstruction information and the shaded shade space textures.

Abstract: Efficient memory operation using a destructive read memory array is described. In accordance with the described techniques, a system may include a memory configured to store data of a first logic state in a ferroelectric capacitor when an electric polarization of the ferroelectric capacitor is in a first direction. A system may include a controller configured to erase the data from the memory by commanding the electric polarization of the ferroelectric capacitor in a second direction, opposite of the first direction and skipping a subsequent write operation of a null value to the memory.

Abstract: Scratchpad memory translation lookaside buffer techniques are described. In an implementation, the techniques described herein relate to a device including a memory management unit implemented in hardware of an integrated circuit to receive a mapping instruction from a mapping instruction source, the mapping instruction specifying a mapping between a virtual memory address and a physical memory address of a scratchpad memory and store a virtual-to-physical mapping entry in a translation lookaside buffer based on the mapping instruction.

Abstract: Preemptive flushing of data involved in executing a processing-in-memory command, from a cache system to main memory that is accessible by a processing-in-memory component, is described. In one example, a system includes an asynchronous flush controller that receives an indication of a subsequent processing-in-memory command to be executed as part of performing a computational task. While earlier commands of the computational task are executed, the asynchronous flush controller evicts or invalidates data elements involved in executing the subsequent processing-in-memory command from the cache system, such that the processing-in-memory command can proceed without stalling.