The present disclosure advantageously provides a system cache and a method for storing coherent data and non-coherent data in a system cache. A transaction is received from a source in a system, the transaction including at least a memory address, the source having a location in a coherent domain or a non-coherent domain of the system, the coherent domain including shareable data and the non-coherent domain including non-shareable data. Whether the memory address is stored in a cache line is determined, and, when the memory address is not determined to be stored in a cache line, a cache line is allocated to the transaction including setting a state bit of the allocated cache line based on the source location to indicate whether shareable or non-shareable data is stored in the allocated cache line, and the transaction is processed.

Methods, systems, and devices for signal development caching in a memory device are described. In one example, a memory device in accordance with the described techniques may include a memory array, a sense amplifier array, and a signal development cache configured to store signals (e.g., cache signals, signal states) associated with logic states (e.g., memory states) that may be stored at the memory array (e.g., according to various read or write operations). In various examples, accessing the memory device may include accessing information from the signal development cache, or the memory array, or both, based on various mappings or operations of the memory device.

A method is provided that includes searching tags in a tag group comprised in a tagged memory system for an available tag line during a clock cycle, wherein the tagged memory system includes a plurality of tag lines having respective tags and wherein the tags are divided into a plurality of non-overlapping tag groups, and searching tags in a next tag group of the plurality of tag groups for an available tag line during a next clock cycle when the searching in the tag group does not find an available tag line.

An apparatus and method for handling stash requests are described. The apparatus has a processing element with an associated storage structure that is used to store data for access by the processing element, and an interface for coupling the processing element to interconnect circuitry. Stash request handling circuitry is also provided that, in response to a stash request targeting the storage structure being received at the interface from the interconnect circuitry, causes a block of data associated with the stash request to be stored within the storage structure. The stash request identifies a given address that needs translating into a corresponding physical address in memory, and also identifies an address space key. Address translation circuitry is used to convert the given address identified by the stash request into the corresponding physical address by performing an address translation that is dependent on the address space key identified by the stash request. The stash request handling circuitry is then responsive to the corresponding physical address determined by the address translation circuitry to cause the block of data to be stored at a location within the storage structure associated with the physical address.

A method includes accessing a first memory component of a memory sub-system via a first interface, accessing a second memory component of the memory sub-system via a second interface, and transferring data between the first memory component and the second memory component via the first interface. The method further includes initially writing data in the first memory component via a first address window and accessing data in the second memory component via a second address window in response to caching the data in first memory component to the second memory component, wherein caching the data in the first memory component to the second component includes changing an address for the data from the first address window to the second address window.

Systems, apparatuses, and methods for implementing flexible dictionary sharing techniques for caches are disclosed. A set-associative cache includes a dictionary for each data array set. When a cache line is to be allocated in the cache, a cache controller determines to which set a base index of the cache line address maps. Then, a selector unit determines which dictionary of a group of dictionaries stored by those sets neighboring this set would achieve the most compression for the cache line. This dictionary is then selected to compress the cache line. An offset is added to the base index of the cache line to generate a full index in order to map the cache line to the set corresponding to this chosen dictionary. The compressed cache line is stored in this set with the chosen dictionary, and the offset is stored in the corresponding tag array entry.

According to one general aspect, an apparatus may include a memory configured to store both data and metadata, such that for portions of data associated with the metadata, the data and metadata are interleaved such that a unit of metadata succeeds each power of two contiguous units of data. The apparatus may also include a memory manager circuit. The memory management circuit may be configured to receive a data access to the memory, wherein the data access includes a public memory address. The memory management circuit may be configured to determine if the public memory address is associated with metadata. The memory management circuit may be configured to, if so, convert the public memory address to a private memory address. The memory management circuit may be configured to complete the data access at the private memory address.

A memory-adaptive processing method for a convolutional neural network includes a feature map counting step, a size relation counting step and a convolution calculating step. The feature map counting step is for counting a number of a plurality of input channels of a plurality of input feature maps, an input feature map tile size, a number of a plurality of output channels of a plurality of output feature maps and an output feature map tile size for a convolutional layer operation. The size relation counting step is for obtaining a cache free space size in a feature map cache and counting a size relation. The convolution calculating step is for performing the convolutional layer operation with the input feature maps to produce the output feature maps according to a memory-adaptive processing technique, and the memory-adaptive processing technique includes a dividing step and an output-group-first processing step.

Disclosed in some examples are memory devices which feature customizable Single Level Cell (SLC) and Multiple Level Cell (MLC) configurations. The configuration (e.g., the size and position) of the SLC cache may have an impact on power consumption, speed, and other performance of the memory device. An operating system of an electronic device to which the memory device is installed may wish to achieve different performance of the device based upon certain conditions detectable by the operating system. In this way, the performance of the memory device can be customized by the operating system through adjustments of the performance characteristics of the SLC cache.

A storage device includes a nonvolatile memory device, a memory controller, and a buffer memory. The memory controller determines a first memory block of the nonvolatile memory device, which is targeted for a read reclaim operation, and reads target data from a target area of the first memory block. The target data are stored in the buffer memory. The memory controller reads at least a portion of the target data stored in the buffer memory in response to a read request corresponding to at least a portion of the target area.