Abstract: Techniques for increasing bandwidth in port-per-module memory systems having mismatched memory modules are disclosed. In one exemplary embodiment, the techniques are realized through a memory module for storing data thereon. The memory module comprises a memory component having a first set of interface connections for providing access to a memory core of the memory component and a second set of interface connections for providing access to the memory core of the memory component. The memory module also comprises memory access circuitry for selecting between a first mode wherein a first portion of the memory core is accessible through the first set of interface connections and a second portion of the memory core is accessible through the second set of interface connections, and a second mode wherein both the first portion and the second portion of the memory core are accessible through the first set of interface connections.

Abstract: A processor includes a translation look-aside buffer (TLB) that relates virtual page addresses to both physical page addresses and main-memory addresses. If the processor references a virtual page address in the TLB for which there is no corresponding information in cache, the processor passes the main-memory address directly to main memory, avoiding the latency normally associated with systems that translate a physical page address to a main-memory address before accessing information from main memory.

Abstract: Techniques for increasing bandwidth in port-per-module memory systems having mismatched memory modules are disclosed. In one exemplary embodiment, the techniques are realized through a memory controller for controlling access to a memory module, wherein the memory module has a memory component with a memory core for storing data therein. The memory controller comprises a first set of interface connections for providing access to the memory module, and a second set of interface connections for providing access to the memory module.

Abstract: Various module structures are disclosed which may be used to implement modules having 1 to N channels. Bus systems may be formed by the interconnection of such modules.

Abstract: A memory system having a memory controller and a memory device coupled to the memory controller. The memory controller outputs a write data value to the memory device. The memory device receives the write data value from the memory controller, and compares the write data value with a mask key value. If the write data value matches the mask key value, the memory device does not store the write data value. If the write data value does not match the mask key value, the memory device stores the write data value.

Abstract: A memory system with serialized data transfer. The memory system includes within a memory controller and a plurality of memory devices. The memory controller receives a plurality of write data values from a host and outputs the write data values as respective serial streams of bits. Each of the memory devices receives at least one of the serial streams of bits from the memory controller and converts the serial stream of bits to a set of parallel bits for storage.

Abstract: Various module structures are disclosed which may be used to implement modules having 1 to N channels. Bus systems may be formed by the interconnection of such modules.

Abstract: A variable depth write data buffer is provided in a memory device coupled to a master device by an interconnect structure in an embodiment of the present invention. The variable depth write data buffer reduces a delay, or W-R turnaround bubble, time between a read operation and a write operation of a memory device memory core. The variable depth write buffer is programmable to store 1 to 4 data packets in an embodiment of the present invention. The variable depth write data buffer may also be programmed for multiple memory device configurations. A method preloads write data without address information into a write data buffer and a subsequent WRITE command causes the previously loaded write data to be retrieved from the write data buffer and written to a memory core according to an embodiment of the present invention.

Abstract: Methods and apparatuses for scheduling commands are described. According to various embodiments of the invention delay information, which is associated with a command is issued. The delay information directs the device for which the command is intended to either execute the command immediately or to delay the command for some period of time before execution. Multiple commands can be queued for execution within the device.

Abstract: A method and apparatus for signaling between devices of a memory system is provided. In accordance with an embodiment of the invention, one or more of several capabilities are implemented to provide heretofore unattainable levels of important system metrics, for example, high performance and/or low cost.

Abstract: Described is a memory system in which the memory core organization changes with device width. The number of physical memory banks accessed reduces with device width, resulting in reduced power usage for relatively narrow memory configurations. Increasing the number of logic memory banks for narrow memory widths reduces the likelihood of bank conflicts, and consequently improves speed performance.

Abstract: A method and apparatus for coordinating memory operations among diversely-located memory components is described. In accordance with an embodiment of the invention, wave-pipelining is implemented for an address bus coupled to a plurality of memory components. The plurality of memory components are configured according to coordinates relating to the address bus propagation delay and the data bus propagation delay. A timing signal associated with address and/or control signals which duplicates the propagation delay of these signals is used to coordinate memory operations.

Abstract: A memory device, system and method for allowing an early read operation after one or more write operations is provided according to an embodiment of the present invention. The memory device comprises an interface for providing a first write address, a first write data, and a read address. A memory core is coupled to the interface and includes a first memory section having a first data path and a first address path and a second memory section having a second data path and a second address path. In an embodiment of the present invention, the first data and first address path is independent of the second data and second address path. The first write data is provided on the first data path responsive to the first write address being provided on the first address path while a read data is provided on the second data path responsive to the read address being provided on the second address path.

Abstract: Techniques for increasing bandwidth in port-per-module memory systems having mismatched memory modules are disclosed. In one exemplary embodiment, the techniques are realized through a memory component having a memory core for storing data therein. The memory component comprises a first set of interface connections for providing access to the memory core, and a second set of interface connections for providing access to the memory core. The memory component also comprises memory access circuitry for selecting between a first mode wherein a first portion of the memory core is accessible through the first set of interface connections and a second portion of the memory core is accessible through the second set of interface connections, and a second mode wherein both the first portion and the second portion of the memory core are accessible through the first set of interface connections.

Abstract: A memory system includes a memory controller and a memory component coupled to each other. An interface of the memory component is configured to receive a first signal from the memory controller with read request information, retrieve the read data information from the memory core in response to the request information, and transmit to the memory controller a second signal containing the read data information. The read data information includes read data symbols, where the average duration of the read data symbols, measured at the interface, defines a symbol time interval. A first external access time is measured at the interface between a first read request and read data transmitted by the interface in response to the first read request. A second external access time interval is measured at the interface between a second read request and read data transmitted by the interface in response to the second read request.

Abstract: Apparatus and methods are disclosed for adjusting phase of data signals to compensate for phase-offset variations between devices during normal operation. The phase of data signals are adjusted individually in each transmit data unit and receive data unit across multiple data slices with a common set of phase vector clock signals and a corresponding clock cycle count signal. The transmission of signal information between a first device (such as a memory controller) and a second device (such as a memory component) occurs without errors even when the accumulated delays between the first device and second device change by a half symbol time interval or more during operation of the system. The apparatus reduces the circuitry required, such as phase-lock-loops, for individually adjusting the phase of each transmit data unit and receive data unit across multiple data slices, which in turn results in reduction in complexity and cost of the system.

Abstract: A memory system architecture/interconnect topology that includes at least one point-to-point link between a master, and at least one memory subsystem. The memory subsystem includes a buffer device coupled to a plurality of memory devices. The memory system may be upgraded through dedicated point-to-point links and corresponding memory subsystems.

Abstract: A memory system architecture/interconnect topology that includes at least one point-to-point link between a master, and at least one memory subsystem. The memory subsystem includes a buffer device coupled to a plurality of memory devices. The memory system may be upgraded through dedicated point-to-point links and corresponding memory subsystems. The master communicates to the plurality of memory devices in each memory subsystem through the respective buffer device via each point-to-point link.

Abstract: A method and apparatus for coordinating memory operations among diversely-located memory components is described. In accordance with an embodiment of the invention, wave-pipelining is implemented for an address bus coupled to a plurality of memory components. The plurality of memory components are configured according to coordinates relating to the address bus propagation delay and the data bus propagation delay. A timing signal associated with address and/or control signals which duplicates the propagation delay of these signals is used to coordinate memory operations.

Abstract: A memory architecture includes a memory controller coupled to multiple modules. Each module includes a computing engine coupled to a shared memory. Each computing engine is capable of receiving instructions from the memory controller and processing the received instructions. The shared memory is configured to store main memory data and graphical data. Certain computing engines are capable of processing graphical data. The memory controller may include a graphics controller that provides instructions to the computing engine. An interconnect on each module allows multiple modules to be coupled to the memory controller.