Abstract: A data processor includes a data fabric, a memory controller, a last level cache, and a traffic monitor. The data fabric is for routing requests between a plurality of requestors and a plurality of responders. The memory controller is for accessing a volatile memory. The last level cache is coupled between the memory controller and the data fabric. The traffic monitor is coupled to the last level cache and operable to monitor traffic between the last level cache and the memory controller, and based on detecting an idle condition in the monitored traffic, to cause the memory controller to command the volatile memory to enter self-refresh mode while the last level cache maintains an operational power state and responds to cache hits over the data fabric.

Abstract: A computer processing system having a first memory with a first set of memory pages resident therein and a second memory coupled to the first memory. A resource tracker provides information to instances of a long short-term memory (LSTM) recurrent neural network (RNN). A predictor identifies memory pages from the first set of memory pages for prediction by the one or more LSTM RNN instances. The system groups the memory pages of the identified plurality of memory pages into a number of patterns based on a number of memory accesses per time. An LSTM RNN instance predicts a number of page accesses for each pattern. A second set of memory pages is selected for moving from the first memory to the second memory.

Abstract: A phase training update circuit operates to perform a phase training update on individual bit lanes. The phase training update circuit adjusts a bit lane transmit phase offset forward a designated number of phase steps, transmits a training pattern, and determines a first number of errors in the transmission. It also adjusts the bit lane transmit phase offset backward the designated number of phase steps, transmits the training pattern, and determines a second number of errors in the transmission. Responsive to a difference between the first number of errors and the second number of errors, the phase training update circuits adjusts a center phase position for the bit lane transmit phase offset of the selected bit lane.

Abstract: Systems, apparatuses, and methods for reusing remote registers in processing in memory (PIM) are disclosed. A system includes at least a host processor, a memory controller, and a PIM device. When the memory controller receives, from the host processor, an operation targeting the PIM device, the memory controller determines whether an optimization can be applied to the operation. The memory controller converts the operation into N PIM commands if the optimization is not applicable. Otherwise, the memory controller converts the operation into a N?1 PIM commands if the optimization is applicable. For example, if the operation involves reusing a constant value, a copy command can be omitted, resulting in memory bandwidth reduction and power consumption savings. In one scenario, the memory controller includes a constant-value cache, and the memory controller performs a lookup of the constant-value cache to determine if the optimization is applicable for a given operation.

Abstract: Systems, apparatuses, and methods for implementing a periodic receiver clock data recovery scheme with dynamic data edge paths are disclosed. An IQ link calibration scheme performs a non-destructive data and edge path switch to determine an IQ offset without disturbing the data. A data path and an edge path pass through multiple stages of deserializers to widen the data path, with the deserializers clocked by clock divided versions of the original data and edge clocks. To initiate a calibration routine, the edge clock is aligned with the data clock, and then data and edge paths are swapped at a common point in a slower clock domain. The data path is then calibrated while the edge path carries the data signal. After the data path is calibrated, the edge and data paths are swapped back to the original configuration.

Abstract: Systems, apparatuses, and methods for detecting personal-space violations in artificial intelligence (AI) based non-player characters (NPCs) are disclosed. An AI engine creates a NPC that accompanies and/or interacts with a player controlled by a user playing a video game. During gameplay, measures of context-dependent personal space around the player and/or one or more NPCs are generated. A control circuit monitors the movements of the NPC during gameplay and determines whether the NPC is adhering to or violating the measures of context-dependent personal space. The control circuit can monitor the movements of multiple NPCs simultaneously during gameplay, keeping a separate score for each NPC. After some amount of time has elapsed, the scores of the NPCs are recorded, and then the scores are provided to a machine learning engine to retrain the AI engines controlling the NPCs.

Abstract: A data processor is for accessing a memory having a first pseudo channel and a second pseudo channel. The data processor includes at least one memory accessing agent, a memory controller, and a data fabric. The at least one memory accessing agent generates generating memory access requests including first memory access requests that access the memory. The memory controller provides memory commands to the memory in response to the first memory access requests. The data fabric routes the first memory access requests to a first downstream port in response to a corresponding first memory request accessing the first pseudo channel, and to a second downstream port in response to the corresponding first memory request accessing the second pseudo channel. The memory controller has first and second upstream ports coupled to the first and second downstream ports of the data fabric, respectively, and a downstream port coupled to the memory.

Abstract: Scheduling processing-in-memory transactions in systems with multiple memory controllers is described. In accordance with the described techniques, an addressing system segments operations of a transaction into multiple microtransactions, where each microtransaction includes a subset of the transaction operations that are scheduled by a corresponding one of the multiple memory controllers. Each transaction, and its associated microtransactions, is assigned a transaction identifier based on a current counter value maintained at the multiple memory controllers, and the multiple memory controllers schedule execution of microtransactions based on associated transaction identifiers to ensure atomic execution of operations for a transaction without interruption by operations of a different transaction.

Abstract: Methods and systems are disclosed for frequency transitioning in a memory interface system. Techniques disclosed include receiving a signal indicative of a change in operating frequency, into a new frequency, in a processing unit interfacing with memory via the memory interface system; switching the system from a normal mode of operation into a transition mode of operation; updating control and state register (CSR) banks of respective transceivers of the system through a mission bus used during the normal mode of operation; and operating the system in the new frequency.

Abstract: Various multi-die arrangements and methods of manufacturing the same are disclosed. In some embodiments, a method of manufacture includes a face-to-face process in which a first GPU chiplet and a second GPU chiplet are bonded to a temporary carrier wafer. A face surface of an active bridge chiplet is bonded to a face surface of the first and second GPU chiplets before mounting the GPU chiplets to a carrier substrate. In other embodiments, a method of manufacture includes a face-to-back process in which a face surface of an active bridge chiplet is bonded to a back surface of the first and second GPU chiplets.

Abstract: A processor determines losses of samples within an input volume that is provided to a neural network during a first epoch, groups the samples into subsets based on losses, and assigns the subsets to operands in the neural network that represent the samples at different precisions. Each subset is associated with a different precision. The processor then processes the subsets in the neural network at the different precisions during the first epoch. In some cases, the samples in the subsets are used in a forward pass and a backward pass through the neural network. A memory configured to store information representing the samples in the subsets at the different precisions. In some cases, the processor stores information representing model parameters of the neural network in the memory at the different precisions of the subsets of the corresponding samples.

Abstract: Systems, apparatuses, and methods for preemptively reserving buffer space for primitives and positions in a graphics pipeline are disclosed. A system includes a graphics pipeline frontend with any number of geometry engines coupled to corresponding shader engines. Each geometry engine launches shader wavefronts to execute on a corresponding shader engine. The geometry engine preemptively reserves buffer space for each wavefront prior to the wavefront being launched on the shader engine. When the shader engine executes a wavefront, the shader engine exports primitive and position data to the reserved buffer space. Multiple scan converters will consume the primitive and position data, with each scan converter consuming primitive and position data based on the screen coverage of the scan converter. After consuming the primitive and position data, the scan converters mark the buffer space as freed so that the geometry engine can then allocate the freed buffer space to subsequent shader wavefronts.

Abstract: Systems and methods are disclosed for reducing power consumed by capturing data from an I/O device. Techniques disclosed include receiving descriptors, by a controller of an I/O host of a system, including information associated with respective data chunks to be captured from an I/O device buffer of the I/O device. Techniques disclosed further include capturing, based on the descriptors, the data chunks. The capturing comprises pulling the data chunks from the I/O device buffer at a pulling rate, where the data chunks are transferred to a local buffer of the I/O host, and pushing segments of the pulled data chunks from the local buffer, where each segment is transferred to a data buffer of the system after a respective target time that precedes a time at which the data chunks in the segment are to be processed by an application executing on the system.

Abstract: In a cloud gaming system or other remote video streaming system, a client device and a server coordinate to introduce an adjustable delay in the frame start timing in the frame rendering pipeline at the server to reducing vertical synchronization (VSYNC) presentation latency, and thus reduce overall frame latency. In implementations, the coordination between the client device and the server includes the client device observing the current VSYNC presentation latencies in recently processed video frames reporting this observed VSYNC presentation latency to the server. The server uses this feedback to determine a frame start delay that is then used to introduce a frame start shift for an upcoming frame and subsequent frames, thereby shifting the server rendering and encoding pipeline back so that the resulting video frames are made available to present at the client device closer to their respective VSYNC signal assertions.

Abstract: Systems, apparatuses, and methods for implementing graphics pipeline optimizations are disclosed. A user interface (UI) is generated to allow a user to analyze shaders and determine resource utilization on any of multiple different target graphic devices. The UI allows the user to manipulate the state associated with the target graphics device for a given graphics pipeline. After being edited by the user, the state of the graphics pipeline is converted into a textual representation and input into a meta-app. The meta-app creates an application programming interface (API) construct from the shader source code and textual representation of the state which is compiled by a driver component into machine-level instructions. Also, resource usage statistics are generated for a simulated run of the graphics pipeline on the target graphics device. Then, the machine-level instructions and resource usage statistics are displayed in the UI for the user to analyze.

Abstract: A technique for performing ray tracing operations is provided. The technique includes, traversing through a bounding volume hierarchy for a ray to arrive at a well-fit bounding volume that is associated with first node, wherein the first node is one of a traversal node or a procedural node, and wherein the well-fit bounding volume comprises geometry other than a single axis-aligned bounding box for the first node; evaluating the ray for intersection with the well-fit bounding volume; determining whether to execute a first shader program associated with the first node based on the evaluating, wherein the first shader program comprises a traversal shader program or a procedural shader program; and executing or not executing the first shader program based on the determining.

Abstract: Cache replacement policies are described. In accordance with the described techniques, a request for data is received and a cache replacement policy controls how a controller responds to the request. The cache replacement policy assigns each cacheline a priority value, which indicates whether the cacheline should be preserved relative to other cachelines, in response to the request being a cache miss that necessitates eviction of at least one cacheline. The cache replacement policy decrements priority values until at least one cacheline achieves a minimum priority value, at which point a cacheline is evicted. The cache replacement policy designates certain cachelines as protected, either via a separate protected indicator or via the cacheline's priority value, which causes unprotected cachelines to be selected for eviction while favoring preservation of protected cachelines in the cache.

Abstract: Local and dynamic triggering of operations executed by a processing-in-memory component is described. In accordance with the described techniques, a processing-in-memory component receives a command from a host for execution by the processing-in-memory component. The processing-in-memory component references a tracking table that includes at least one entry associated with an operation performed as part of executing the command and identifies at least one additional command to be triggered locally after executing the command received from the host. Responsive to identifying that conditions associated with the at least one additional command are satisfied, the processing-in-memory component executes the at least one additional command, independent of instructions from the host.

Abstract: A technique for performing ray tracing operations is provided. The technique includes detecting intersection of a ray with a split bounding volume of an instance of a bounding volume hierarchy; determining whether the split bounding volume meets an instance traversal limiting criterion; and continuing BVH traversal based on the determining.

Abstract: An apparatus and method for efficiently managing power consumption among multiple, replicated functional blocks of an integrated circuit. An integrated circuit includes multiple, replicated functional blocks that use separate power domains. Data of a given type is stored in an interleaved manner among at least two of the multiple functional blocks. In one implementation, a prior static allocation determines that only a subset of the functional blocks store the data of the given type. In another implementation, each of the functional blocks stores the data of the given type, and when an idle state has occurred, data of the given type is moved between the multiple functional blocks until one or more functional blocks no longer store data of the given type. When a transition to the idle state has occurred, the functional blocks that do not store the data of the given type are transitioned to a sleep state.