A latch is provided. The latch includes a plurality of storage nodes including a plurality of data storage nodes configured to store a data bit having one of two states and a plurality of complementary data storage nodes configured to store a complement of the data bit. The latch includes a plurality of supply voltage multi-dependency stages respectively corresponding to the plurality of storage nodes. Each supply voltage multi-dependency stage has an output coupled to a storage node and at least two control inputs respectively coupled to at least two other storage nodes of the plurality of storage nodes. The supply voltage multi-dependency stage is configured to cause a state of the data bit stored in the storage node to change from a first state to a second state in response a change in both states of two data bits respectively stored in the at least two other storage nodes.

Representative implementations of devices and techniques provide correction for a defective die in a wafer-to-wafer stack or a die stack. A correction die is coupled to a die of the stack with the defective die. The correction die electrically replaces the defective die. Optionally, a dummy die can be coupled to other die stacks of a wafer-to-wafer stack to adjust a height of the stacks.

A stack of vertically-connected, horizontally-oriented integrated circuits (ICs) may have electrical connections from the front side of one IC to the back side of another IC. Electrical signals may be transferred from the back side of one IC to the front side of the same IC by means of through substrate vias (TSVs), which may include through silicon vias. Electronic apparatus, systems, and methods may operate to test and/or replace defective TSVs. Additional apparatus, systems and methods are disclosed.

Methods, systems, and apparatuses for redundancy in a memory array are described. A memory array may include some memory cells that are redundant to other memory cells of the array. Such redundant memory cells may be used if a another memory cell is discovered to be defective in some way; for example, after the array is fabricated and before deployment, defects in portions of the array that affect certain memory cells may be identified. Memory cells may be designated as redundant cells for numerous other memory cells of the array so that a total number of redundant cells in the array is relatively small fraction of the total number of cells of the array. A configuration of switching components may allow redundant cells to be operated in a manner that supports redundancy for numerous other cells and may limit or disturbances to neighboring cells when accessing redundancy cells.

A method for correcting bit defects in an STT-MRAM memory is disclosed. The method comprises executing a read before write operation in the STT-MRAM memory, wherein the STT-MRAM memory comprises a plurality of codewords, wherein each codeword comprises a plurality of redundant bits. The read before write operation comprises reading a codeword and mapping defective bits in the codeword. Further, the method comprises replacing the defective bits in the codeword with a corresponding redundant bit and executing a write operation with corresponding redundant bits in place of the defective bits.

A latch is provided. The latch includes a plurality of storage nodes including a plurality of data storage nodes configured to store a data bit having one of two states and a plurality of complementary data storage nodes configured to store a complement of the data bit. The latch includes a plurality of supply voltage multi-dependency stages respectively corresponding to the plurality of storage nodes. Each supply voltage multi-dependency stage has an output coupled to a storage node and at least two control inputs respectively coupled to at least two other storage nodes of the plurality of storage nodes. The supply voltage multi-dependency stage is configured to cause a state of the data bit stored in the storage node to change from a first state to a second state in response a change in both states of two data bits respectively stored in the at least two other storage nodes.

A hardware forwarding element is provided that includes a group of unit memories, a set of packet processing pipelines, and an error signal fabric. Each packet processing pipeline includes several of match action stages. Each match action stage includes a set of match action tables stored in a set of unit memories. Each unit memory is configured to detect an error in the unit memory and generate an error output when an error is detected in the memory unit. The error signal fabric, for each match action stage, combines error outputs of the unit memories storing match tables into a first bit in the error signal fabric. The error signal fabric, for each match action stage, combines error outputs of the unit memories storing action tables into a second bit in the error signal fabric.

A memory-testing circuit configured to perform a test of reference bits in a memory. In a read operation, outputs of data bit columns are compared with one or more reference bit columns. The memory-testing circuit comprises: a test controller and association adjustment circuitry configurable by the test controller to associate another one or more reference bit columns or one or more data bit columns with the data bit columns in the read operation. The test controller can determine whether the original one or more reference bit columns have a defect based on results from the two different association.

A method and memory device with in-memory computing defection detection is disclosed. A memory device includes a memory including banks, wherein each bank includes a respective plurality of bit-cells, an in-memory computation (IMC) operator configured to perform an IMC operation between first data while the first data is in the bit-cells of the memory and second data received as input to the memory device, wherein the banks share the operator, and wherein the memory device is configured to: generate a first test pattern that is stored in the memory and generate a second test pattern applied to the IMC operator, and based thereon determine whether a defect has occurred in either the memory or the operator, and perform a repair based on the determination that a defect has occurred.

A memory subsystem includes memory devices with space dynamically allocated for improvement of reliability, availability, and serviceability (RAS) in the system. Error checking and correction (ECC) logic detects an error in all or a portion of a memory device. In response to error detection, the system can dynamically perform one or more of: allocate active memory device space for sparing to spare a failed memory segment; write a poison pattern into a failed cacheline to mark it as failed; perform permanent fault detection (PFD) and adjust application of ECC based on PFD detection; or, spare only a portion of a device and leave another portion active, including adjusting ECC based on the spared portion. The error detection can be based on bits of an ECC device, and error correction based on those bits and additional bits stored on the data devices.