Abstract:
A signal on a transmitter tracks noise on a ground node in a manner decoupled from a positive node of a power supply. The signal is transmitted from the transmitter to the receiver. A reference voltage is generated on the receiver to track noise on a ground node in the receiver. Consequently, the received signal and the reference voltage have substantially the same noise characteristics, which become common mode noise that can be cancelled out when these two signals are compared against each other. In a further embodiment, the reference voltage is compared against a predetermined calibration pattern. An error signal is generated based on a difference between the sampler output and the predetermined calibration pattern. The error signal is then used to adjust the reference voltage so that the DC level of the reference voltage is positioned substantially in the middle of the received signal.
Abstract:
The embodiments described herein describe technologies for using the memory modules in different modes of operation, such as in a standard multi-drop mode or as in a dynamic point-to-point (DPP) mode (also referred to herein as an enhanced mode). The memory modules can also be inserted in the sockets of the memory system in different configurations.
Abstract:
A system that calibrates timing relationships between signals involved in performing write operations is described. This system includes a memory controller which is coupled to a set of memory chips, wherein each memory chip includes a phase detector configured to calibrate a phase relationship between a data-strobe signal and a clock signal received at the memory chip from the memory controller during a write operation. Furthermore, the memory controller is configured to perform one or more write-read-validate operations to calibrate a clock-cycle relationship between the data-strobe signal and the clock signal, wherein the write-read-validate operations involve varying a delay on the data-strobe signal relative to the clock signal by a multiple of a clock period.
Abstract:
The disclosed embodiments relate to components of a memory system that support error detection and correction by means of storage and retrieval of error correcting codes. In specific embodiments, this memory system includes a memory device, which further contains a memory bank. During operation, the memory device receives a request to concurrently access a data word at a first row in a first storage region of the memory bank and error information associated with the data at a second row in a second storage region of the memory bank. Moreover, the memory request includes a first row address identifying the first row and a second row address identifying the second row. Next, the memory device routes the first row address and the second row address to a first row decoder and a second row decoder in the memory bank, respectively. Finally, the memory device uses the first row decoder to decode the first row address to access the first row and concurrently uses the second row decoder to decode the second row address to access the second row.
Abstract:
Embodiments of an integrated circuit (IC) comprising a delay-locked loop (DLL) are described. Some embodiments include first circuitry to generate a first clock signal by delaying an input clock signal by a first delay, second circuitry to determine a code based on the input clock signal and the first clock signal, and third circuitry to produce an output clock signal based on the input clock signal and the code. In some embodiments, the power consumption of the DLL circuitry is reduced by powering down at least some parts of the DLL circuitry for most of the time. In some embodiments, the clock signal that is used to clock the command-and-address circuitry of a memory device is used to clock the on-die-termination latency counter circuitry.
Abstract:
A memory storage scheme specially adapted for wear leveling (or other reorganization of logical memory space). Memory space includes a logical memory space of M addressable blocks of data, stored as rows or pages, and N substitute rows or pages. Data is periodically shuffled by copying data from one of the M addressable blocks to a substitute row, with the donating row then becoming part of substitute memory space, available for ensuing wear leveling operations, using a stride address. The disclosed techniques enable equation-based address translation, obviating need for an address translation table. An embodiment performs address translation entirely in hardware, for example, integrated with a memory device to perform wear leveling or data scrambling, in a manner entirely transparent to a memory controller. In addition, the stride address can represent an offset greater than one (e.g., greater than one row) and can be dynamically varied.
Abstract:
A multi-chip package with signal line compression for testing of the multi-chip package. The multi-chip package includes an interposer and two or more integrated circuits attached to the interposer. The interposer includes multiple data signal lines for data communications between the two integrated circuits. The data signal lines are also coupled to one or more test contacts through an interface circuit. The number of test contacts is smaller than the number of signal lines, which allows a large number of signal lines to be tested with a smaller number of test contacts.
Abstract:
A transmitter for providing channel equalization that includes a first driver and second driver having a high pass filter. The first driver generates a first output signal representing a digital input signal. The second driver generates a second output signal representing a high pass filtered version of the digital input signal. The first and second output signals are summed to provide a third output signal that is channel equalized for transmission over a channel.
Abstract:
A memory system that supports error detection and correction (EDC) coverage. The memory system includes a memory module with at least two groups of memory devices that store data and another memory device that stores error checking information (e.g., Error Correcting Code) for both groups of memory devices. The memory module also includes a memory buffer that determines an address for accessing the error checking information based on whether data is transferred with the first group of memory devices or the second group of memory devices. Alternatively, the memory controller may determine the address for accessing the error checking information to reduce or eliminate the need for a memory buffer.
Abstract:
A memory device comprising a programmable command-and-address (CA) interface and/or a programmable data interface is described. In an operational mode, two or more CA interfaces may be active. In another operational mode, at least one, but not all, CA interfaces may be active. In an operational mode, all of the data interfaces may be active. In another operational mode, at least one, but not all, data interfaces may be active. The memory device can include circuitry to select: an operational mode; a sub-mode within an operational mode; one or more CA interfaces as the active CA interface(s); a main CA interface from multiple active CA interfaces; and/or one or more data interfaces as the active data interfaces. The circuitry may perform these selection(s) based on one or more bits in one or more registers and/or one or more signals received on one or more pins.