Abstract: Detection logic of a memory subsystem obtains a threshold for a memory device that indicates a number of accesses within a time window that causes risk of data corruption on a physically adjacent row. The detection logic obtains the threshold from a register that stores configuration information for the memory device, and can be a register on the memory device itself and/or can be an entry of a configuration storage device of a memory module to which the memory device belongs. The detection logic determines whether a number of accesses to a row of the memory device exceeds the threshold. In response to detecting the number of accesses exceeds the threshold, the detection logic can generate a trigger to cause the memory device to perform a refresh targeted to a physically adjacent victim row.

Abstract: A memory device is configured to provide internal or on-die ECC (error checking and correction or error correction coding). In such a system, the code matrix can be managed as four quadrants of (N/4) bits, with two adjacent quadrants in an (N/2)-bit segment or portion. The N codes of the matrix correspond to the N bits of a data word to be protected by the ECC. The code matrix includes M codes corresponding to the M ECC check bits. The memory device includes internal ECC circuitry to perform ECC in the DRAM device with the ECC bits and code matrix in response to a request to access the data word. The codes in a quadrant steer an aliased bit to a quadrant other than an adjacent quadrant.

Abstract: A memory controller issues a targeted refresh command. A specific row of a memory device can be the target of repeated accesses. When the row is accessed repeatedly within a time threshold (also referred to as “hammered” or a “row hammer event”), physically adjacent row (a “victim” row) may experience data corruption. The memory controller receives an indication of a row hammer event, identifies the row associated with the row hammer event, and sends one or more commands to the memory device to cause the memory device to perform a targeted refresh that will refresh the victim row.

Abstract: Devices, systems, and methods include an active mode to accommodate read/write operations of a memory device and a self-refresh mode to accommodate recharging of voltage levels representing stored data when read/write operations are idle. At least one register source provides a first voltage level and a second voltage level that is less than the first voltage level. With such a configuration, during the active mode, the memory device operates at the first voltage level as provided by the at least one register source, and during the self-refresh mode, the memory device operates at the second voltage level as provided by the at least one register source.

Abstract: In a memory system a multichip memory provides data redundancy for error recovery. The multichip memory can be an integrated circuit package with multiple memory dies or memory devices integrated with a common package. The multiple memory dies are coupled in a daisy chain, and can be a vertical stack or in a planar formation. The memory chip or chips at the end of the chain store parity data, and the other devices store data. The multichip memory includes XOR (exclusive OR) logic to compute parity to store in the redundant parity chips.

Abstract: A stacked memory with interface providing offset interconnects. An embodiment of memory device includes a system element and a memory stack coupled with the system element, the memory stack including one or more memory die layers. Each memory die layer includes first face and a second face, the second face of each memory die layer including an interface for coupling data interface pins of the memory die layer with data interface pins of a first face of a coupled element. The interface of each memory die layer includes connections that provide an offset between each of the data interface pins of the memory die layer and a corresponding data interface pin of the data interface pins of the coupled element.

Abstract: A memory subsystem enables a refresh abort command. A memory controller can issue an abort for an in-process refresh command sent to a memory device. The refresh abort enables the memory controller to more precisely control the timing of operations executed by memory devices in the case where a refresh command causes refresh of multiple rows of memory. The memory controller can issue a refresh command during active operation of the memory device, which is active operation refresh as opposed to self-refresh when the memory device controls refreshing. The memory controller can then issue a refresh abort during the refresh, and prior to completion of the refresh. The memory controller thus has deterministic control over both the start of refresh as well as when the memory device can be made available for access.

Abstract: Memory refresh includes timing offsets for different memory devices, to initiate refresh of different memory devices at different times. A memory controller sends a refresh command to cause refresh of multiple memory devices. In response to the refresh command, the multiple memory devices initiate refresh with timing offsets relative to another of the memory devices. The timing offsets reduce the instantaneous power surge associated with all memory devices starting refresh simultaneously.

Abstract: Techniques and mechanisms to provide selective access to data error information by a memory controller. In an embodiment, a memory device stores a first value representing a baseline number of data errors determined prior to operation of the memory device with the memory controller. Error detection logic of the memory device determines a current count of data errors, and calculates a second value representing a difference between the count of data errors and the baseline number of data errors. The memory device provides the second value to the memory controller, which is unable to identify that the second value is a relative error count. In another embodiment, the memory controller is restricted from retrieving the baseline number of data errors.

Abstract: Embodiments are generally directed to performance of additional refresh operations during self-refresh mode. An embodiment of a memory device includes one or more memory banks, a mode register set, the mode register set including a first set of mode register bits, and a control logic to provide control operations for the memory device, the operations including refresh operations for the one or more memory banks in a refresh credit mode. The control logic is to perform one or more extra refresh cycles in response to receipt of a self-refresh command, the self-refresh command to provide current refresh status information, and is to store information in the first set of mode register bits regarding a modified refresh status after the performance of the one or more extra refresh cycles.

Abstract: Memory subsystem refresh management enables commands to access one or more identified banks across different bank groups with a single command. Instead of sending commands identifying a bank or banks in separate bank groups by each bank group individually, the command can cause the memory device to access banks in different bank groups. The command can be a refresh command. The command can be a precharge command.

Abstract: Error correction in a memory subsystem includes a memory device generating internal check bits after performing internal error detection and correction, and providing the internal check bits to the memory controller. The memory device performs internal error detection to detect errors in read data in response to a read request from the memory controller. The memory device selectively performs internal error correction if an error is detected in the read data. The memory device generates check bits indicating an error vector for the read data after performing internal error detection and correction, and provides the check bits with the read data to the memory controller in response to the read request. The memory controller can apply the check bits for error correction external to the memory device.

Abstract: A memory controller issues a targeted refresh command. A specific row of a memory device can be the target of repeated accesses. When the row is accessed repeatedly within a time threshold (also referred to as “hammered” or a “row hammer event”), physically adjacent row (a “victim” row) may experience data corruption. The memory controller receives an indication of a row hammer event, identifies the row associated with the row hammer event, and sends one or more commands to the memory device to cause the memory device to perform a targeted refresh that will refresh the victim row.

Abstract: Techniques and mechanisms to provide write access to a memory device. In an embodiment, a memory controller sends commands to a memory device which comprises multiple memory banks. The memory controller further sends a signal specifying that the commands include back-to-back write commands each to access the same memory bank. In response to the signal, the memory device buffers first data of a first write command, wherein the first data is buffered at least until the memory device receives second data of a second write command. Error correction information is calculated for a combination of the first data and second data, and the combination is written to the memory bank. In another embodiment, buffering of the first data and combining of the first data with the second data is performed, based on the signal from the memory controller, in lieu of read-modify-write processing of the first data.

Abstract: Detection logic of a memory subsystem obtains a threshold for a memory device that indicates a number of accesses within a time window that causes risk of data corruption on a physically adjacent row. The detection logic obtains the threshold from a register that stores configuration information for the memory device, and can be a register on the memory device itself and/or can be an entry of a configuration storage device of a memory module to which the memory device belongs. The detection logic determines whether a number of accesses to a row of the memory device exceeds the threshold. In response to detecting the number of accesses exceeds the threshold, the detection logic can generate a trigger to cause the memory device to perform a refresh targeted to a physically adjacent victim row.

Abstract: A check bit read mode enables a memory device to provide internal check bits to an associated host. A memory controller of a memory subsystem can generate one or more read commands for memory devices of the memory subsystem. The read command can include address location information. The memory devices include memory arrays with memory locations addressable with the address location information. The memory locations have associated data and internal check bits, where the check bits are generated internally by the memory for error correction. If the memory device is configured for check bit read mode, in response to the read command, it sends the internal check bits associated with the identified address location. If the memory device is not configured check bit read mode, it returns the data in response to the read command without exposing the internal check bits.

Abstract: A memory subsystem enables a refresh abort command. A memory controller can issue an abort for an in-process refresh command sent to a memory device. The refresh abort enables the memory controller to more precisely control the timing of operations executed by memory devices in the case where a refresh command causes refresh of multiple rows of memory. The memory controller can issue a refresh command during active operation of the memory device, which is active operation refresh as opposed to self-refresh when the memory device controls refreshing. The memory controller can then issue a refresh abort during the refresh, and prior to completion of the refresh. The memory controller thus has deterministic control over both the start of refresh as well as when the memory device can be made available for access.

Abstract: A memory subsystem enables managing error correction information. A memory device internally performs error detection for a range of memory locations, and increments an internal count for each error detected. The memory device includes ECC logic to generate an error result indicating a difference between the internal count and a baseline number of errors preset for the memory device. The memory device can provide the error result to an associated host of the system to expose only a number of errors accumulated without exposing internal errors from prior to incorporation into a system. The memory device can be made capable to generate internal addresses to execute commands received from the memory controller. The memory device can be made capable to reset the counter after a first pass through the memory area in which errors are counted.

Abstract: In a system where a memory device performs on-die ECC, the ECC operates on N-bit data words as two (N/2)-bit segments, with a code matrix having a corresponding N codes that can be operated on as a first portion of (N/2) codes and a second portion of (N/2) codes to compute first and second error checks for first and second (N/2)-bit segments of the data word, respectively. In the code matrix, a bitwise XOR of any two codes in the first portion of the code matrix or any two codes in the second portion of the code matrix results in a code that is either not in the code matrix or is in the other portion of the code matrix. Thus, a miscorrected double bit error in one portion causes a bit to be toggled in the other portion instead of creating a triple bit error.

Abstract: Error correction in a memory subsystem includes a memory device generating internal check bits after performing internal error detection and correction, and providing the internal check bits to the memory controller. The memory device performs internal error detection to detect errors in read data in response to a read request from the memory controller. The memory device selectively performs internal error correction if an error is detected in the read data. The memory device generates check bits indicating an error vector for the read data after performing internal error detection and correction, and provides the check bits with the read data to the memory controller in response to the read request. The memory controller can apply the check bits for error correction external to the memory device.