Abstract: An apparatus and method for selecting a storage location in a memory device including receiving at least one of a pre-decoded location address signal, a match signal, and a redundant location address enable signal, enabling one of a decoder and a redundant decoder in response to the match signal, wherein the decoder is operable to generate a location select signal for selecting a first location, the decoder being responsive to the pre-decoded location address signal, and wherein the redundant decoder is operable to generate a redundant location select signal for selecting a second location, the redundant decoder being responsive to the redundant location address enable signal, and terminating one of the generation of a location select signal and the generation of a redundant location select signal in response to a precharge signal.

Abstract: A data strobe synchronization circuit includes first and second logic circuits receiving global data strobe pulses and respective enable signal. A control circuit initially applies an enable signal to the first logic circuit so that the first logic circuit generates a first data strobe pulse responsive to each global data strobe pulse. The control circuit receives a write control signal. When the write control signal becomes active, the control circuit terminates the enable signal applied to the first logic circuit and applies an enable signal to the second logic circuit. The second logic circuit then generates a second data strobe pulse responsive to the next global data strobe pulse. The first and second data strobe pulses are used to latch a data signal in respective flip-flops. The data strobe pulses may latch the data signal in pairs of flip-flops on the leading and trailing edges of the data strobe pulses.

Abstract: A method and apparatus to convert parallel data to serial data. More specifically, there is provided a parallel-to-serial converter comprising a data pipeline configured to receive parallel data, and binary sort logic comprising a plurality of switches arranged to receive the parallel data from the data pipeline, and configured to output the parallel data serially.

Abstract: An apparatus and method for selecting a storage location in a memory device including receiving at least one of a pre-decoded location address signal, a match signal, and a redundant location address enable signal, enabling one of a decoder and a redundant decoder in response to the match signal, wherein the decoder is operable to generate a location select signal for selecting a first location, the decoder being responsive to the pre-decoded location address signal, and wherein the redundant decoder is operable to generate a redundant location select signal for selecting a second location, the redundant decoder being responsive to the redundant location address enable signal, and terminating one of the generation of a location select signal and the generation of a redundant location select signal in response to a precharge signal.

Abstract: A calibration circuit for calibrating the input data path of a digital circuit is disclosed. A simple string of a repeating data pattern such as, e.g., “1100,” is sent on the data path. The digital circuit captures the data using a clock signal, examines the data signal for the predetermined pattern and adjusts a delay applied to the data signal until the predetermined pattern is recognized. Then the delay is further adjusted until the predetermined pattern is no longer recognized indicating that an edge of the eye of the data is near a clocking edge of the clocking signal. The delay applied to the data signal is then further adjusted by a predetermined amount to position the clock edge near the center of the data eye.

Abstract: An apparatus and method for selecting a storage location in a memory device including receiving at least one of a pre-decoded location address signal, a match signal, and a redundant location address enable signal, enabling one of a decoder and a redundant decoder in response to the match signal, wherein the decoder is operable to generate a location select signal for selecting a first location, the decoder being responsive to the pre-decoded location address signal, and wherein the redundant decoder is operable to generate a redundant location select signal for selecting a second location, the redundant decoder being responsive to the redundant location address enable signal, and terminating one of the generation of a location select signal and the generation of a redundant location select signal in response to a precharge signal.

Abstract: Disclosed herein are exemplary embodiments of an improved write address shift register structure useful for example in a DDR3 DRAM having read/write latency. The disclosed shift register structure propagates write addresses from an address bus outside the device to array decoders to allow latent data to be written into the cells in the memory array at a proper time. The register structure comprises a reduced number of registers (e.g., four) thus eliminating the need for extraneous registers which might otherwise be used to propagate “don't care” addresses. The registers are clocked, and the addresses propagated though the registers, in accordance with a latency bus through which a user defines the desired read/write latency in accordance with user preferences and the desired clock speed of the device. The clock for each register is preferably decoded from the latency bus and hence each register preferably has its own unique clock.

Abstract: A calibration circuit for calibrating the input data path of a digital circuit is disclosed. A simple string of a repeating data pattern such as, e.g., “1100,” is sent on the data path. The digital circuit captures the data using a clock signal, examines the data signal for the predetermined pattern and adjusts a delay applied to the data signal until the predetermined pattern is recognized. Then the delay is further adjusted until the predetermined pattern is no longer recognized indicating that an edge of the eye of the data is near a clocking edge of the clocking signal. The delay applied to the data signal is then further adjusted by a predetermined amount to position the clock edge near the center of the data eye.

Abstract: A digital counter allows the provision of start and stop values in order to allow it to be configurable to any length. The counter rolls over to zero at a maximum value and proceeds to count until it reaches an indicated stop count. At that point, it proceeds to the start count and continues counting. The counter counts half bytes of a cell of data for transmission to and from a multiport DRAM in accordance with communication protocols, such as asynchronous transfer mode (ATM.

Abstract: The present invention provides a clock signal input circuit that is able to provide inverse internal clock signals generated by the same input buffer as the address and data signals which exhibit reduced skew. When a skewed external noninverse clock signal and a corresponding external inverse clock signal are passed through respective reference voltage input buffers there is no reduction in skew between the two internal signals. In a preferred embodiment, the invention provides back to back inverters connected to both lines carrying the noninverted and inverted internal clock signals. The slower internal clock signal has an extra inverter driving it when it switches states and the faster internal clock signal has an extra inverter fighting it when it switches states. The skew of the two signals is reduced, allowing for faster operation of the integrated circuit and a reduction in misread data signals.

Abstract: The disclosed system and method significantly reduce or eliminate DQS postamble ringing problem in modern high-speed memory chips, allowing the memory chips to be operated at significantly faster clock speeds. The external strobe signal (XDQS) may be used to generate at least two derivative strobe signals therefrom. Instead of the XDQS signal, the derivative strobe signals are then used, in a predetermined order, to clock in or strobe the data to be written into memory cells. The last generated derivative strobe signal may be used to finally transfer the data bits into memory cells. Once the last of the derivative strobe signals is activated, and so long as there are no more data writes pending in the command pipe for the next clock cycle, the rising or falling edge of the last derivative strobe signal can be detected to turn off further generation of the strobe signals prior to any onset of postamble ringing on the XDQS signal.

Abstract: A memory device includes a memory cell array, an addressing circuit, a data communication circuit and a control circuit. The addressing circuit receives first signals that are indicative of an address associated with a write command, decodes the address to provide column select signals that are indicative of a column address in the memory cell array, and uses the column address to perform a column redundancy check. The data communication circuit latches data signals that are associated with the write command in response to a data strobe signal. The control circuit causes the addressing circuit to perform the column redundancy check during a delay to accommodate variations in the timing of the data strobe signal and begins providing the column select signals to the memory cell array after performing the column redundancy check. The memory device may be a double data rate (DDR) synchronous dynamic random access memory (SDRAM), for example.

Abstract: Disclosed herein are exemplary embodiments of an improved write address shift register structure useful for example in a DDR3 DRAM having read/write latency. The disclosed shift register structure propagates write addresses from an address bus outside the device to array decoders to allow latent data to be written into the cells in the memory array at a proper time. The register structure comprises a reduced number of registers (e.g., four) thus eliminating the need for extraneous registers which might otherwise be used to propagate “don't care” addresses. The registers are clocked, and the addresses propagated though the registers, in accordance with a latency bus through which a user defines the desired read/write latency in accordance with user preferences and the desired clock speed of the device. The clock for each register is preferably decoded from the latency bus and hence each register preferably has its own unique clock.

Abstract: An off chip driver impedance adjustment circuit includes a storage circuit adapted to receive and store a drive strength adjustment word. A counter circuit is coupled to the storage circuit to receive the drive strength adjustment word and develops a drive strength count responsive to the drive strength adjustment word. A programmable fuse code to preset the counter. An output driver circuit is coupled to the counter circuit to receive the drive strength count and is adapted to receive a data signal. The output driver circuit develops an output signal on an output responsive to the data signal and adjusts a drive strength as a function of the drive strength count.

Abstract: An apparatus and method for selecting a storage location in a memory device including receiving at least one of a pre-decoded location address signal, a match signal, and a redundant location address enable signal, enabling one of a decoder and a redundant decoder in response to the match signal, wherein the decoder is operable to generate a location select signal for selecting a first location, the decoder being responsive to the pre-decoded location address signal, and wherein the redundant decoder is operable to generate a redundant location select signal for selecting a second location, the redundant decoder being responsive to the redundant location address enable signal, and terminating one of the generation of a location select signal and the generation of a redundant location select signal in response to a precharge signal.

Abstract: A data strobe synchronization circuit includes first and second logic circuits receiving global data strobe pulses and respective enable signal. A control circuit initially applies an enable signal to the first logic circuit so that the first logic circuit generates a first data strobe pulse responsive to each global data strobe pulse. The control circuit receives a write control signal. When the write control signal becomes active, the control circuit terminates the enable signal applied to the first logic circuit and applies an enable signal to the second logic circuit. The second logic circuit then generates a second data strobe pulse responsive to the next global data strobe pulse. The first and second data strobe pulses are used to latch a data signal in respective flip-flops. The data strobe pulses may latch the data signal in pairs of flip-flops on the leading and trailing edges of the data strobe pulses.

Abstract: The present invention provides a clock signal input circuit that is able to provide inverse internal clock signals generated by the same input buffer as the address and data signals which exhibit reduced skew. When a skewed external noninverse clock signal and a corresponding external inverse clock signal are passed through respective reference voltage input buffers there is no reduction in skew between the two internal signals. In a preferred embodiment, the invention provides back to back inverters connected to both lines carrying the noninverted and inverted internal clock signals. The slower internal clock signal has an extra inverter driving it when it switches states and the faster internal clock signal has an extra inverter fighting it when it switches states. The skew of the two signals is reduced, allowing for faster operation of the integrated circuit and a reduction in misread data signals.

Abstract: The present invention provides a clock signal input circuit that is able to provide inverse internal clock signals generated by the same input buffer as the address and data signals which exhibit reduced skew. When a skewed external noninverse clock signal and a corresponding external inverse clock signal are passed through respective reference voltage input buffers there is no reduction in skew between the two internal signals. In a preferred embodiment, the invention provides back to back inverters connected to both lines carrying the noninverted and inverted internal clock signals. The slower internal clock signal has an extra inverter driving it when it switches states and the faster internal clock signal has an extra inverter fighting it when it switches states. The skew of the two signals is reduced, allowing for faster operation of the integrated circuit and a reduction in misread data signals.

Abstract: A calibration circuit for calibrating the input data path of a digital circuit is disclosed. A simple string of a repeating data pattern such as, e.g., “1100,” is sent on the data path. The digital circuit captures the data using a clock signal, examines the data signal for the predetermined pattern and adjusts a delay applied to the data signal until the predetermined pattern is recognized. Then the delay is further adjusted until the predetermined pattern is no longer recognized indicating that an edge of the eye of the data is near a clocking edge of the clocking signal. The delay applied to the data signal is then further adjusted by a predetermined amount to position the clock edge near the center of the data eye.

Abstract: A number of embodiments of memory devices and methods of performing read/write timing calibration of these memory devices using a row or a redundant row. Addressing of the row or redundant row in a memory array of a memory device may be accomplished by using a calibration fuse bank to address a row or a redundant row of the memory array, by using a fuse bank of the memory device to address a redundant row of the memory array, or by storing the row address of a row in a memory controller and providing the row address to the memory device during calibration. A redundant row used for calibration may be a redundant row not utilized by a memory device during repair of its memory array. A row used for calibration may be a row not utilized by a memory device due to the nature of the specific application in which that memory device is being used. A unique data pattern may then be written to and read from the addressed row or redundant row for read/write timing calibration.