Abstract: A method and structure for implementing a DRAM memory array as a second level cache memory in a computer system. The computer system includes a central processing unit (CPU), a first level SRAM cache memory, a CPU bus coupled to the CPU, and a second level cache memory which includes a DRAM array coupled to the CPU bus. When accessing the DRAM array, row access and column decoding operations are performed in a self-timed asynchronous manner. Predetermined sequences of column select operations are then performed in a synchronous manner with respect to a clock signal. The DRAM array is operated at a higher frequency than the frequency of the CPU bus clock signal, thereby reducing the access latency of the DRAM array. By operating the DRAM array at a higher frequency than the CPU bus, additional time is provided for precharging the DRAM array. As a result, the precharging of the DRAM array is transparent to the CPU bus.

Abstract: A method and structure for generating a boosted word line voltage for a memory array, such as a DRAM array. To ensure that an adequate voltage is applied to the word line of the memory array during write operations, the word line driver circuit is connected to a boost voltage generator which provides a substantially constant, regulated voltage which is boosted to a level which is approximately equal to the V.sub.cc supply voltage plus the threshold voltage of the memory cell pass transistor. A bias voltage generator provides a negative voltage which is used to bias the substrate of the memory array. The boosted voltage generator and the bias voltage generator can be operated in response to the same clock signal used to operate the memory array. A latch-up prevention circuit is provided to ensure that the word line driver circuit does not latch-up during power-on before an adequate boost voltage has been established.

Abstract: A ring oscillator includes an even-numbered plurality of ring coupled delay stages. Each delay stage includes a differential amplifier, a voltage clamping circuit, and a current source. The differential amplifier receives first and second input signals from a preceding delay stage. The differential amplifier provides a first output signal and a complementary second output signal at first and second nodes, respectively. The voltage clamping circuit is coupled between the first and second nodes to limit a peak-to-peak voltage swing of each of the first and second output signals. The current source is coupled to the differential amplifier and varies a bias current in accordance with a delay bias voltage.

Abstract: A structure and method of implementing a cache memory for a multi-processor system. The cache memory includes a main memory which is coupled to a main memory bus. A plurality of processors can also be coupled to the main memory bus. The main memory includes a plurality of RAM circuit module memory banks. The sense amplifiers of a predetermined number of banks are used as cache memory (i.e., sense amplifier cache lines). The number of banks used with sense amplifiers activated is substantially less than the total number of banks. The banks which are not used as cache memory are kept in a precharged state.

Abstract: A method and structure for implementing pipeline burst read and write operations in a semiconductor memory having a memory cycle time substantially longer than its I/O data cycle-time. The memory system includes a read buffer which stores all data required for a read burst transaction. All read burst data is loaded from the memory to the read buffer at the beginning of each burst read access. The memory is then isolated from the read buffer and prepared to perform the next burst access. During this time, the read data values are provided to the I/O device from the read buffer. A double-buffering technique provides gap-less output data for consecutive pipeline-burst read transactions. The memory system uses a two-entry write buffer in a first in, first out manner for pipeline-burst write operations. Each write buffer entry stores data for an entire burst transaction and a corresponding address.

Abstract: A memory system having several memory devices coupled to a memory controller through an I/O bus, each memory device including multiple memory modules coupled to a chip I/O interface through an internal bus. The system includes a circuit for driving the I/O bus with a reduced CMOS-swing, a circuit for driving the internal bus with a full CMOS-swing in one bus direction and with a reduced CMOS-swing in the other bus direction, a column address generation circuit for allowing sequentially addressed data to be accessed with the decoder delay being eliminated, and a circuit for re-synchronizing data from a source clock to a destination clock with reduced access latency penalty. Simultaneously writing data into multiple circuit modules significantly increases the write bandwidth of the memory. Also included are a dynamic base-address mapping into an address space, a read or write operation across multiple memory modules, a novel I/O bus format, and a protocol and test mode for testing redundant memory sub-arrays.

Abstract: A memory device which utilizes a plurality of memory modules coupled in parallel to a master I/O module through a single directional asymmetrical signal swing (DASS) bus. This structure provides an I/O scheme having symmetrical swing around half the supply voltage, high through-put, high data bandwidth, short access time, low latency and high noise immunity. The memory device utilizes improved column access circuitry including an improved address sequencing circuit and a data amplifier within each memory module. The memory device includes a resynchronization circuit which allows the device to operate either synchronously and asynchronously using the same pins. Each memory module has independent address and command decoders to enable independent operation. Thus, each memory module is activated by commands on the DASS bus only when a memory access operation is performed within the particular memory module. The memory device includes redundant memory modules to replace defective memory modules.

Abstract: A method and structure for controlling the timing of an access to a DRAM array in response to a row access (RAS#) signal and the rising and/or falling edges of a clock signal. Row address decoding and the deactivation of equalization circuits are initiated when the row access signal is received and a first transition of the clock signal is detected. The row address decoding and the deactivation of the equalization circuits are completed before a second transition of the clock signal occurs. The second transition is then used to initiate the turning on of the sense amplifiers of the DRAM array. The sense amplifiers are turned on before a third transition of the clock signal. The third transition of the clock signal is then used to initiate the column address decoding operation of the DRAM array. In an alternative embodiment, the column address decoding is initiated when a column access (CAS#) signal is asserted and the clock signal undergoes the third transition.

Abstract: A method and structure for generating a boosted word line voltage for a memory array, such as a DRAM array. To ensure that an adequate voltage is applied to the word line of the memory array during write operations, the word line driver circuit is connected to a boost voltage generator which provides a substantially constant, regulated voltage which is boosted to a level which is approximately equal to the V.sub.CC supply voltage plus the threshold voltage of the memory cell pass transistor. A bias voltage generator provides a negative voltage which is used to bias the substrate of the memory array. The boosted voltage generator and the bias voltage generator can be operated in response to the same clock signal used to operate the memory array. A latch-up prevention circuit is provided to ensure that the word line driver circuit does not latch-up during power-on before an adequate boost voltage has been established.

Abstract: A fault-tolerant, high-speed wafer scale system comprises a plurality of functional modules, a parallel hierarchical bus which is fault-tolerant to defects in an interconnect network, and one or more bus masters. This bus includes a plurality of bus lines segmented into sections and linked together by programmable bus switches and bus transceivers or repeaters in an interconnect network.

Abstract: A resynchronization circuit for processing a stream of data values read from a memory system, and a method of operating the same. The resynchronization circuit includes a first in, first out (FIFO) memory device, a phase locked loop circuit and a latency control circuit. The FIFO memory device receives a stream of data values and a first clock signal from the memory system. The data values are sequentially read into the FIFO memory device in response to the first clock signal. The phase locked loop circuit receives a second clock signal, and in response generates an output clock signal which leads in phase the second clock signal. The output clock signal is provided to the FIFO memory device to cause the data values to be sequentially read from the FIFO memory device. As a result, a stream of data values is generated which is synchronized with the second clock signal.

Abstract: A method and structure for controlling the timing of an access to a DRAM array in response to a row access (RAS#) signal and the rising and falling edges of a clock signal. Row address decoding and the deactivation of equalization circuits are initiated when the row access signal is received and a rising edge of the clock signal is detected. The row address decoding and the deactivation of the equalization circuits are completed before the falling edge of the clock signal occurs. The falling edge is then used to initiate the turning on of the sense amplifiers of the DRAM array. The sense amplifiers are turned on before the subsequent rising edge of the clock signal. The subsequent rising edge is then used to initiate the column address decoding operation of the DRAM array. A test mode is included which allows the DRAM array to be operated asynchronously for testing purposes.

Abstract: A delay stage for a ring oscillator supplies a first output signal and a second output signal. Each of the first and second output signals has a peak-to-peak voltage swing. The first and second output signals are complementary to each other. The delay stage includes a differential amplifier for generating the first output signal and the second output signal and a voltage clamping circuit for limiting the peak-to-peak voltage swing of the first and second output signals. The voltage clamping circuit is coupled between the first output signal and the second output signal. The differential amplifier includes a first NMOS transistor and a second NMOS transistor for generating the first and second output signals. The differential amplifier also includes a first PMOS transistor and a second PMOS transistor coupled to (1) the first and second NMOS transistors and (2) the voltage clamping circuit for providing bias currents to the first and second NMOS transistors.

Abstract: A system and method for wafer scale integration optimized for medium die size integrated circuits by interconnecting a large number of separate memory (or other circuit) modules on a semiconductor wafer so as to electrically exclude both defective modules and defective interconnect/power segments, and include operative modules and interconnect/power segments. A set of discretionary connections are associated with each of the separate modules and interconnect/power segments and such connections are made (or broken) after a module or interconnect or power segment is tested. A power supply network is set up by combining operative power segments. A bidirectional bus is set up by combining operative interconnect segments to connect to each operative modules. This bidirectional bus consists of one or more hierarchies for speed, power and yield considerations. Each module is assigned an identity code using discretionary connections.

Abstract: A memory system having several memory devices coupled to a memory controller through an I/O bus, each memory device including multiple memory modules coupled to a chip I/O interface through an internal bus. The system includes a circuit for driving the I/O bus with a reduced CMOS-swing, a circuit for driving the internal bus with a full CMOS-swing in one bus direction and with a reduced CMOS-swing in the other bus direction, a column address generation circuit for allowing sequentially addressed data to be accessed with the decoder delay being eliminated, and a circuit for re-synchronizing data from a source clock to a destination clock with reduced access latency penalty. Simultaneously writing data into multiple circuit modules significantly increases the write bandwidth of the memory. Also included are a dynamic base-address mapping into an address space, a read or write operation across multiple memory modules, a novel I/O bus format, and a protocol and test mode for testing redundant memory sub-arrays.

Abstract: Circuitry for performing fine phase adjustment within a phase locked loop is described. The phase selector selects an even phase signal and an odd phase signal from the twelve phase signals output by the VCO. The even and odd phase signals are selected by an even select signal and an odd select signal, respectively. The phase interpolator interpolates between the even phase signal and the odd phase signal to generate an output signal. The affect of the even phase signal and the odd phase signal on the output signal is determined by an even weighting signal and an odd weighting signal, respectively. The weighting signals prevent glitches from appearing on the output signal when either the even phase signal or the odd phase signal is switching. A method of performing fine phase adjustment in a phase locked loop is also described.

Abstract: A bus system is described that minimizes clock-data skew. The bus system includes a data bus, a clockline and synchronization circuitry. The clockline has two clockline segments. Each clockline segment extends the entire length of the data bus and is joined to the other clockline segment by a turnaround at one end of the data bus. The clockline ensures that clock and data signals travel in the same direction. Synchronization circuitry within transmitting devices synchronizes data signals to be coupled onto the data bus with the clock signal used by other devices to receive the data.

Abstract: A high density, static random access memory (SRAM) circuit with single-ended memory cells employs a plurality of (4T-2R) or (6T) type SRAM cells and a regenerative sense amplifier. Each of the SRAM cells employs a single bit-line (BL) and two word lines.

Abstract: Electrical current source circuitry for a bus is described. The circuitry includes transistor circuitry coupled between the bus and ground for controlling bus current, control circuitry coupled to the transistor circuitry, and a controller coupled to the control circuitry for controlling the transistor circuitry. The controller comprises a variable level circuit comprising setting circuitry for setting a desired current for the bus and transistor reference circuitry coupled to the setting circuitry. The variable level circuit provides a first voltage. Voltage reference circuitry provides a reference voltage. Comparison circuitry is coupled to the voltage reference circuitry and to the variable level circuit for comparing the first voltage with the reference voltage. Logic circuitry is responsive to a trigger signal from the comparison circuitry.

Abstract: Briefly, a high density, static, random access memory (SRAM) circuit with ratio independent memory cells employs a number (plurality) of (4T-2R) or (6T) type SRAM cells and a regenerative sense amplifier. Each of the SRAM cells of the present invention differs from corresponding, prior art type SRAM cells in that the SRAM cells of the present invention each include transistors of similar size (channel width).