摘要:
A method includes destructively reading bits of a spin torque magnetic random access memory and immediately writing back the original or inverted values. A detection of the majority state of the write back bits and a conditional inversion of write back bits are employed to reduce the number of write back pulses. A subsequent write command received within a specified time or before an original write operation is commenced will cause a portion of the write back pulses or the original write operation pulses to abort. Write pulses during subsequent write operations will follow the conditional inversion determined for the write back bits during destructive read.
摘要:
A method includes destructively reading bits of a spin torque magnetic random access memory, using error correcting code (ECC) for error correction, and storing inverted or non-inverted data in data-store latches. When a subsequent write operation changes the state of data-store latches, parity calculation and majority detection of the bits are initiated. A majority bit detection and potential inversion of write data minimizes the number of write current pulses. A subsequent write operation received within a specified time or before an original write operation is commenced will cause the majority detection operation to abort.
摘要:
A method includes destructively reading bits of a spin torque magnetic random access memory and immediately writing back the original or inverted values. A detection of the majority state of the write back bits and a conditional inversion of write back bits are employed to reduce the number of write back pulses. A subsequent write command received within a specified time or before an original write operation is commenced will cause a portion of the write back pulses or the original write operation pulses to abort. Write pulses during subsequent write operations will follow the conditional inversion determined for the write back bits during destructive read.
摘要:
A method includes destructively reading bits of a spin torque magnetic random access memory and immediately writing back the original or inverted values. A detection of the majority state of the write back bits and a conditional inversion of write back bits are employed to reduce the number of write back pulses. A subsequent write command received within a specified time or before an original write operation is commenced will cause a portion of the write back pulses or the original write operation pulses to abort. Write pulses during subsequent write operations will follow the conditional inversion determined for the write back bits during destructive read.
摘要:
A method includes destructively reading bits of a spin torque magnetic random access memory and immediately writing back the original or inverted values. A detection of the majority state of the write back bits and a conditional inversion of write back bits are employed to reduce the number of write back pulses. A subsequent write command received within a specified time or before an original write operation is commenced will cause a portion of the write back pulses or the original write operation pulses to abort. Write pulses during subsequent write operations will follow the conditional inversion determined for the write back bits during destructive read.
摘要:
A method includes destructively reading bits of a spin torque magnetic random access memory, using error correcting code (ECC) for error correction, and storing inverted or non-inverted data in data-store latches. When a subsequent write operation changes the state of data-store latches, parity calculation and majority detection of the bits are initiated. A majority bit detection and potential inversion of write data minimizes the number of write current pulses. A subsequent write operation received within a specified time or before an original write operation is commenced will cause the majority detection operation to abort.
摘要:
A method of applying a write current to a magnetic tunnel junction device minimizes sub-threshold leakage. NMOS- and PMOS-follower circuits are used in applying the write current, and bias signals for the follower circuits are isolated from global bias signals before the write current is applied.
摘要:
A method and memory device is provided for reading data from an ECC word of a plurality of reference bits associated with a plurality of memory device bits and determining if a double bit error in the ECC word exists. The ECC word may be first toggled twice and the reference bits reset upon detecting the double bit error.
摘要:
A memory controller and method for interleaving volatile and non-volatile memory different latencies and page sizes are described wherein a single DDR3 memory controller communicates with a number of memory modules comprising of at least non-volatile memory, e.g., spin torque magnetic random access memory, integrated in a different Rank or Channel with a volatile memory, e.g., dynamic random access memory (DRAM).
摘要:
A memory controller and method for interleaving volatile and non-volatile memory different latencies and page sizes are described wherein a single DDR3 memory controller communicates with a number of memory modules comprising of at least non-volatile memory, e.g., spin torque magnetic random access memory, integrated in a different Rank or Channel with a volatile memory, e.g., dynamic random access memory (DRAM).