The present disclosure relates to devices and methods for using a banked memory structure with accelerators. The devices and methods may segment and isolate dataflows in datapath and memory of the accelerator. The devices and methods may provide each data channel with its own register memory bank. The devices and methods may use a memory address decoder to place the local variables in the proper memory bank.

A system-on-chip is connected to a first memory device and a second memory device. The system-on-chip comprises a memory controller configured to control an interleaving access operation on the first and second memory devices. A modem processor is configured to provide an address for accessing the first or second memory devices. A linear address remapping logic is configured to remap an address received from the modem processor and to provide the remapped address to the memory controller. The memory controller performs a linear access operation on the first or second memory device in response to receiving the remapped address.

Described herein are systems, methods, and products utilizing a cache coherent switch on chip. The cache coherent switch on chip may utilize Compute Express Link (CXL) interconnect open standard and allow for multi-host access and the sharing of resources. The cache coherent switch on chip provides for resource sharing between components while independent of a system processor, removing the system processor as a bottleneck. Cache coherent switch on chip may further allow for cache coherency between various different components. Thus, for example, memories, accelerators, and/or other components within the disclose systems may each maintain caches, and the systems and techniques described herein allow for cache coherency between the different components of the system with minimal latency.

Methods, systems, and devices for array access with receiver masking are described. A first device may issue to a second device a first sequence of write commands for a set of data. The first sequence of write commands may indicate different memory addresses in an order. After issuing the first sequence of write commands, the first device may issue to the second device a second sequence of read commands for the set of data. The second sequence of read commands may indicate the different memory addresses in the same order as the first sequence of write commands. Based on issuing the second sequence of read commands, the first device may receive the set of data from the second device.

A memory controller includes: a map data storage for storing map data; and a read operation controller for receiving, from a host, a read request and a target logical address corresponding to the read request, acquiring a first physical address mapped to the target logical address, based on the map data, and obtaining data stored at the first physical address. When an uncorrectable error is present in the data stored at the first physical address, the read operation controller acquires a second physical address previously mapped to the target logical address before the first physical address, obtains data stored at the second physical address, and provides the host with the data stored at the second physical address and information representing occurrence of the uncorrectable error.

A method of managing a garbage collection (GC) operation on a flash memory includes: setting a GC starting threshold, wherein the GC starting threshold indicates a predetermined spare block number that is higher than a target spare block number of spare blocks maintained by a flash translation layer (FTL) of the flash memory; determining whether to start the GC operation according to a current number of spare blocks in the flash memory and the GC starting threshold; and performing the GC operation on a source block in the flash memory when the current number of spare blocks is lower than or equal to the GC starting threshold.

Systems and methods for memory management for virtual machines. An example method may comprise running, by a host computer system, a Level 0 hypervisor managing a Level 1 virtual machine running a Level 1 hypervisor which manages a Level 2 virtual machine. The Level 1 hypervisor may detecting execution of an operation that prevents modification to a set of entries in a Level 2 page table and generate a shadow page table where each shadow page table entry of the plurality of shadow page table entries maps a Level 2 guest virtual address of a Level 2 address space associated with the Level 2 virtual machine to a corresponding Level 1 guest physical address of a Level 1 address space associated with the Level 1 virtual machine. The Level 0 hypervisor may generate a Level 0 page table.

In various examples, a VPU and associated components may be optimized to improve VPU performance and throughput. For example, the VPU may include a min/max collector, automatic store predication functionality, a SIMD data path organization that allows for inter-lane sharing, a transposed load/store with stride parameter functionality, a load with permute and zero insertion functionality, hardware, logic, and memory layout functionality to allow for two point and two by two point lookups, and per memory bank load caching capabilities. In addition, decoupled accelerators may be used to offload VPU processing tasks to increase throughput and performance, and a hardware sequencer may be included in a DMA system to reduce programming complexity of the VPU and the DMA system. The DMA and VPU may execute a VPU configuration mode that allows the VPU and DMA to operate without a processing controller for performing dynamic region based data movement operations.

Examples of the present disclosure describe systems and methods for sharing memory using a multi-ring shared, traversable and dynamic database. In aspects, the database may be synchronized and shared between multiple processes and/or operation mode protection rings of a system. The database may also be persisted to enable the management of information between hardware reboots and application sessions. The information stored in the database may be view independent, traversable, and resizable from various component views of the database. In some aspects, an event processor is additionally described. The event processor may use the database to allocate memory chunks of a shared heap to components/processes in one or more protection modes of the operating system.

Embodiments of the present disclosure relate to a memory system and an operating method of the memory system. According to embodiments of the present disclosure, a memory system may include a buffer memory for storing tail doorbell information for N submission queues capable of storing a command fetched from the host or head doorbell information for N completion queues capable of storing an execution result of the command fetched from the host.