Adaptec, Inc.

Cypress Semiconductor Corporation

National Semiconductor Corporation

Walter Francis Bridgewater, Jr

Barry Allen Davis

S Babar Raza

Anup Nayak

Dimitris Pantelakis

Mark A Ross

Derwin W Mattos

Patrick P Fasang

Michael H Herschfelt

William C Gintz

A data interface is provided that can de-skew data signals by taking into account different skewing effects on each data signal. The data interface can be used, for example, in a communication system and can be configured to operate in one of three possible modes of operation. In the first mode, de-skewing is fixed prior to the sample logic. In the second mode, de-skewing is periodically changed automatically as the amount of skew changes based on training signals that are periodically sent into the data interface. The combination of the data phase count and the positive and negative clock width pulse counts will then determine where the final transition or edge of each data signal is to be placed within a bit. The third mode of operation involves an override or programmatic modification of the second mode of operation based on values stored in a register.

Data interface that is configurable into separate modes of operation for sub-bit de-skewing of parallel-fed data signals

A clock detection and selection circuit () can include a first counter (-) that generates a first count value CNTaccording to a first clock signal CLKand a second counter (-) that generates a second count value CNTaccording to a second clock signal CLK. First separation-detect logic (-) and second separation-detect logic (-) determine if a pre-specified difference exists between a first count value (CNT/CNT') and second count value (CNT/CNT&#x2032;). According to such determinations, separation information (INFand INF) can be generated indicating which clock signal (CLKor CLK) is faster. Selection logic () can select a faster of the clock signals (CLKor CLK) if the separation information values confirm one another.

Method and device for selecting one of multiple clock signals based on frequency differences of such clock signals

An invention is provided for data bit align. The invention includes a multiplexer that receives a data sample word as data input and also receives a clock sample word as select input. The multiplexer selects a data bit from the data sample word based on the clock sample word. Generally, the multiplexer can select the data bit from the data sample word corresponding to a position of the clock edge in the clock sample word. The invention also includes an output register, which is coupled to the multiplexer. The output register stores the selected data bit from the multiplexer and provides the selected data bit to remaining system components.

Method and apparatus for data bit align

A memory device, apparatus and method are provided for accessing memory cells. The device, apparatus, and method allow two or more electronic subsystems with corresponding external ports to access a single array of memory elements via a single internal port. Address, data, and control signals from each of the external ports are multiplexed onto the word line drivers and de-multiplexed from the sense amplifiers. Multiplexing and de-multiplexing operations are sequenced based on a state machine that receives synchronized signals from the external port. The synchronized signal can be clock signals that are synchronized to a high-speed sampling clock. Synchronization and sequencing functions can occur over a relatively few number of cycles of the high-speed sampling clock to minimize the time for resolving access conflicts, thereby maximizing the number of external ports which can access the internal port of the array.

System and method for staging concurrent accesses to a memory address location via a single port using a high speed sampling clock

An apparatus for correcting duty cycle error is provided which includes circuitry capable of determining existence of a duty cycle error from input data received over data transmissions lines where the circuitry generates duty cycle correction data based on the duty cycle error. The apparatus also includes a digital analog converter (DAC) being coupled to the circuitry where the DAC is capable of receiving a magnitude portion of the duty cycle correction data from the circuitry. The apparatus further includes an adjustable bias driver being coupled to the circuitry, the DAC and the data transmission lines. The adjustable bias driver receives the magnitude portion of the duty cycle correction data from the DAC and receives a polarity portion of the duty cycle correction data from the circuitry where the adjustable bias driver adjusts the polarity of signals applied to the data transmission lines for correcting the duty cycle error.

Methods and apparatus to correct duty cycle

An invention is provided for a deskewer that corrects skew on a data channel. The deskewer includes a delay calculator that calculates deskew data indicating the amount of delay needed to correct skew on a data channel. Coupled to the delay calculator is a deskew circuit that receives the deskew data from the delay calculator and uses the deskew data to delay a bit stream on the data channel.

Method and apparatus for a programmable deskew circuit

A low-voltage differential dual receiver for a SCSI bus uses a symmetrical driver by doing without a termination bias voltage. The data phase and protocol phase of SCSI communication are separated by using two receivers and optimizing each receiver for its particular function. A high-speed receiver is used when transmitting data at high-speeds, and a lower performance, low-speed receiver is used for other SCSI phases. A built-in offset allows the low-speed receiver to operate correctly during bus arbitration. The built-in offset in the low-speed receiver takes the place of the termination bias voltage in a traditional SCSI bus and is implemented in a variety of ways.

Low voltage differential dual receiver

A low-voltage differential dual receiver for a SCSI bus uses a symmetrical driver by doing without a termination bias voltage. The data phase and protocol phase of SCSI communication are separated by using two receivers and optimizing each receiver for its particular function. A high-speed receiver is used when transmitting data at high-speeds, and a lower performance, low-speed receiver is used for other SCSI phases. A built-in offset allows the low-speed receiver to operate correctly during bus arbitration. The built-in offset in the low-speed receiver takes the place of the termination bias voltage in a traditional SCSI bus and is implemented in a variety of ways. In a first example, an N-well generation circuit produces a bulk voltage for one transistor of the differential transistor pair that is different than a supply voltage supplied to the bulk of the other transistor. In a second example, each of the transistors of the pair is implanted with a different dosage to change the threshold voltage for each. In a third example, resistors of different sizes are attached to the source of each transistor in the pair in order to produce a different voltage at each source. In a fourth example, two replica comparators are used to monitor an offset voltage of the differential receiver and to send control signals to an adjustable current source. The current source is adjusted by having an up-down counter switch on and off various legs of the current source.

The first pulse problem for a low-voltage differential SCSI bus driver is remedied by supplying greater power for a first pulse of a bus line after a steady state condition. Activity detection circuitry detects when a signal has remained in a steady state for a number of bus cycles and enables an additional power boosting differential driver to deliver an appropriate amount of power for a limited amount of time in order to produce a quality first pulse while minimizing power output. The extra power needed to remedy the quality of the first pulse is only supplied for the duration of the first pulse so that the output driver strength is minimized and the total power over time that an integrated circuit must dissipate is reduced. In another embodiment, instead of greater than normal power being delivered for a first pulse, the output driver is decreased in its output drive strength while an output remains in a particular state. When the output does finally switch states, it switches at normal strength; the net effect is an increased drive strength from the steady state to the new state. Step down control circuitry is used to determine after a specified number of clock pulses how much power should be stepped down and in how many increments.

Low voltage differential driver with multiple drive strengths

A bias compensator circuit for significantly reducing an offset produced by a termination bias that is associated with a differential pair bus. The differential pair bus is connected to a driver. The bias compensator circuit includes: (a) a signal source connected to a first line of the differential pair; (b) a first switch for switching ON the signal source while the driver is driving; (c) a signal sink connected to a second line of the differential pair; and (d) a second switch for switching ON the signal sink while the driver is driving.

Bias compensator for differential transmission line with voltage bias

In a built-in self test ('BIST') circuit for on-chip testing of an integrated circuit memory, a control logic circuit is responsive to an external signal on a test select pin for controlling the BIST operations of the major circuit groups thereof. The major circuit groups include an address PRPG (pseudo-random pattern generator), which selectively furnishes test addresses or mission addresses to the memory; a data PRPG, which selectively furnishes test data or mission data to the memory; a PSA (parallel signature analyzer) PRPG, which furnishes mission data from the memory in normal mode and determines a signature in test mode; and a decoder, which compares the signature determined by the PSA PRPG with a known correct signature and sets a flag to indicate memory pass/fail. The BIST circuit is modular and extendable for any N-word by M-bit memory. To this end, a counter in the control logic circuit comprises a plurality of counter slices, the address PRPG comprises a plurality of address slices, the data PRPG comprises a plurality of data slices, and the PSA PRPG comprises a plurality of PSA slices. The BIST circuit is formed by replicating slices for the major circuit groups. Only certain feedback paths in the major circuit groups and certain control signal combinatorial circuits in the control logic require customization.

Growing community of inventors

San Jose, CA, United States of America

Walter Francis Bridgewater, Jr