An apparatus is described. The apparatus includes a memory management unit. The memory management unit is to receive a memory access request from an accelerator, wherein the memory access request includes a virtual address of a payload provided by an application that invokes the accelerator to perform a function on the payload, wherein. The memory access request also includes an identifier of the application's CPU process. The memory management unit is to translate the virtual address to a physical address to fetch the payload from a location allocated to the application within a memory.

Methods, systems, and devices for memory bus drive defect detection and related operations are described. A controller coupled with a memory array may receive a command for data. The memory array may include one or more pins for communicating data to and from the memory array, in response to the command. The controller may transmit to the memory array, over a bus that is coupled with the controller and the pins, the command. The controller may detect, based at least in part on a resistor coupled with the bus and a power supply of the memory array, that the bus is operating in a first state after transmitting the command. The first state may comprise a voltage that is relatively higher than a voltage of the second state. The controller may determine a defect associated with the bus or the pin based on detecting the bus in the first state.

A device may include an array of data processing engines (DPEs) on a die and an event broadcast network. Each of the DPEs includes a core, a memory module, event logic in at least one of the core or the memory module, and an event broadcast circuitry coupled to the event logic. The event logic is capable of detecting an occurrence of one or more events in the core or the memory module. The event broadcast circuitry is capable of receiving an indication of a detected event detected by the event logic. The event broadcast network includes interconnections between the event broadcast circuitry of the DPEs. Detected events can trigger or initiate various responses, such as debugging, tracing, and profiling.

The present disclosure relates to asymmetric read/write architectures for enhanced throughput and reduced latency. One example embodiment includes an integrated circuit. The integrated circuit includes a network interface. The integrated circuit also includes a communication bus interface. The integrated circuit is configured to establish a communication link with a processor of the host computing device over the communication bus interface, which includes mapping to memory addresses associated with the processor of the host computing device. The integrated circuit is also configured to receive payload data for transmission over the network interface in response to the processor of the host computing device writing payload data to the mapped memory addresses using one or more programmed input-outputs (PIOs). Further, the integrated circuit is configured to write payload data received over the network interface to the memory of the host computing device using direct memory access (DMA).

According to some example embodiments according to the present disclosure, a device includes a printed circuit board (PCB); a solid state drive (SSD) connected at a first side of the PCB via at least one SSD connector; at least one field programmable gate array (FPGA) attached to the PCB at a second side of the PCB; and at least one front end connector attached to the PCB at a third side of the PCB, wherein the device is configured to process data stored in the SSD based on a command received via the at least one front end connector.

A memory controller includes a host interface circuit connectable to a host device by a bus conforming to a memory card system specification, a data buffer circuit including a buffer memory, a tag information generation circuit configured to generate tag information associated with a command received by the host interface circuit, and a first register in which the tag information generated by the tag information generation circuit is stored, and a second register into which the tag information stored in the first register is copied after the command is fetched from the host interface circuit for processing. When a read request is made from the host interface circuit to the data buffer circuit, the data buffer circuit returns read data stored in the buffer memory upon confirming that the tag information stored in the first register and the tag information stored in the second register match each other.

Systems and methods for time control for a data interface between a source device and a receiving device are provided. In one example, a method can include performing a capture time sweep process at the receiving device. The capture time sweep process includes performing a plurality of test data transfers at the receiving device at a plurality of different capture time settings. The method can include determining a capture time window based at least in part on the capture time sweep process. The capture time window can be defined as a duration between a first capture time and a second capture time. The method can include determining a selected capture time between the first capture time and the second capture time. The method can include controlling data transfer across the data interface based at least in part on the selected capture time.

According to some example embodiments according to the present disclosure, a device includes a printed circuit board (PCB); a solid state drive (SSD) connected at a first side of the PCB via at least one SSD connector; at least one field programmable gate array (FPGA) attached to the PCB at a second side of the PCB; and at least one front end connector attached to the PCB at a third side of the PCB, wherein the device is configured to process data stored in the SSD based on a command received via the at least one front end connector.

A method of operating a storage device includes receiving, from a host, a first packet containing a buffer address indicating a location of a data buffer selected from among a plurality of data buffers in the host, parsing the buffer address from the first packet, and transmitting a second packet containing the buffer address to the host in response to the first packet.

A data storage system includes a storage medium including a plurality of columns of memory cells, a storage controller coupled to the storage medium, and data path circuitry including a data bus coupled to the storage controller, the data bus configured to receive a plurality of bytes of data to be written to the plurality of columns of memory cells; a block of data latches having a pitch equal to a first number of bit lines of the plurality of columns of memory cells; and column redundancy circuitry configured to pass the plurality of bytes of data to the block of data latches via the plurality of columns in accordance with a nonconsecutive mapping scheme. The nonconsecutive mapping scheme includes mapping each group of three bytes to two columns by splitting one byte of each group of three bytes into two nibbles.