Advanced Micro Devices Corporation

Alok Garg

Paul S Keltcher

John Kalamatianos

Matthew T Sobel

Furkan Eris

Mayank Chhablani

Marius Evers

Sharad Dilip Bade

Carson Donahue Henrion

Michael K Ciraula

Chetana Nagendra Keltcher

Neil N Marketkar

Eric W Busta

Donald A Priore

Chitresh Chandra Narasimhaiah

Gregg Donley

Alice Danielle Kivity

Scott Andrew McLelland

A method for reducing cache fills can include training a filter, by at least one processor and in response to at least one of eviction or rewrite of one or more entries of a cache, the filter indicating one or more cache loads from which the one or more entries were previously filled. The method can also include preventing, by the at least one processor and based on the trained filter, one or more subsequent fills to the cache from the one or more cache loads. Various other methods and systems are also disclosed.

Systems and methods for reducing cache fills

Techniques are described for a hardware processor to dynamically configure a component that improves a processor function with a configuration setting based on invariant statistics. The invariant statistics are generated by execution of the instructions from one or more applications and are independent of the performance metrics of the processor function for the execution. In an embodiment, the configuration setting for the component is generated using a machine learning model.

Invariant statistics-based configuration of processor components

Data reuse cache techniques are described. In one example, a load instruction is generated by an execution unit of a processor unit. In response to the load instruction, data is loaded by a load-store unit for processing by the execution unit and is also stored to a data reuse cache communicatively coupled between the load-store unit and the execution unit. Upon receipt of a subsequent load instruction for the data from the execution unit, the data is loaded from the data reuse cache for processing by the execution unit.

Data reuse cache

Array of pointers prefetching is described. In accordance with described techniques, a pointer target instruction is detected by identifying that a destination location of a load instruction is used in an address compute for a memory operation and the load instruction is included in a sequence of load instructions having addresses separated by a step size. An instruction for fetching data of a future load instruction is injected in an instruction stream of a processor. The data of the future load instruction is stored in a temporary register. An additional instruction is injected in the instruction stream for prefetching a pointer target based on an address of the memory operation and the data of the future load instruction.

Array of pointers prefetching

Techniques for identifying a hardware configuration for operation are disclosed. The techniques include applying feature measurements to a trained model; obtaining output values from the trained model, the output values corresponding to different hardware configurations; and operating according to the output values, wherein the output values include one of a certainty score, a ranking, or a regression value.

Hardware configuration selection using machine learning model

Techniques for generating a model for predicting when different hybrid prefetcher configurations should be used are disclosed. Techniques for using the model to predict when different hybrid prefetcher configurations should be used are also disclosed. The techniques for generating the model include obtaining a set of input data, and generating trees based on the training data. Each tree is associated with a different hybrid prefetcher configuration and the trees output certainty scores for the associated hybrid prefetcher configuration based on hardware feature measurements. To decide on a hybrid prefetcher configuration to use, a prefetcher traverses multiple trees to obtain certainty scores for different hybrid prefetcher configurations and identifies a hybrid prefetcher configuration to used based on a comparison of the certainty scores.

Multi-class multi-label classification using clustered singular decision trees for hardware adaptation

Systems, apparatuses, and methods for implementing scheduler queue assignment burst mode are disclosed. A scheduler queue assignment unit receives a dispatch packet with a plurality of operations from a decode unit in each clock cycle. The scheduler queue assignment unit determines if the number of operations in the dispatch packet for any class of operations is greater than a corresponding threshold for dispatching to the scheduler queues in a single cycle. If the number of operations for a given class is greater than the corresponding threshold, and if a burst mode counter is less than a burst mode window threshold, the scheduler queue assignment unit dispatches the extra number of operations for the given class in a single cycle. By operating in burst mode for a given operation class during a small number of cycles, processor throughput can be increased without starving the processor of other operation classes.

Scheduler queue assignment burst mode

Systems, apparatuses, and methods for implementing scheduler queue assignment logic are disclosed. A processor includes at least a decode unit, scheduler queue assignment logic, scheduler queues, pickers, and execution units. The assignment logic receives a plurality of operations from a decode unit in each clock cycle. The assignment logic includes a separate logical unit for each different type of operation which is executable by the different execution units of the processor. For each different type of operation, the assignment logic determines which of the possible assignment permutations are valid for assigning different numbers of operations to scheduler queues in a given clock cycle. The assignment logic receives an indication of how many operations to assign in the given clock cycle, and then the assignment logic selects one of the valid assignment permutations for the number of operations specified by the indication.

Scheduler queue assignment

A processing system includes a shadow tag memory, which stores a plurality of entries containing coherency information for the cachelines residing at the various levels of private caches. If a cache miss occurs at a private cache, or if coherency information for a cacheline requires updating, a probe is sent to the shadow tag memory maintained at the shared cache to determine whether the requested (or affected) cacheline is stored at another private cache. The probe includes a tag which can be divided into two or more portions. To more efficiently compare the probe tag to the shadow tag entries, the comparison is performed in multiple stages based on the portions of the probe tag.

Predictive multistage comparison for associative memory

Methods and systems for prefetching data for a processor are provided. A system is configured for and a method includes selecting one of a first prefetching control logic and a second prefetching control logic of the processor as a candidate feature, capturing the performance metric of the processor over an inactive sample period when the candidate feature is inactive, capturing a performance metric of the processor over an active sample period when the candidate feature is active, comparing the performance metric of the processor for the active and inactive sample periods, and setting a status of the candidate feature as enabled when the performance metric in the active period indicates improvement over the performance metric in the inactive period, and as disabled when the performance metric in the inactive period indicates improvement over the performance metric in the active period.

Dynamic evaluation and reconfiguration of a data prefetcher

A method, an apparatus, and a non-transitory computer readable medium for tracking prefetches generated by a stride prefetcher are presented. Responsive to a prefetcher table entry for an address stream locking on a stride, prefetch suppression logic is updated and prefetches from the prefetcher table entry are suppressed when suppression is enabled for that prefetcher table entry. A stride is a difference between consecutive addresses in the address stream. A prefetch request is issued from the prefetcher table entry when suppression is not enabled for that prefetcher table entry.

Growing community of inventors

Maynard, MA, United States of America

Alok Garg