Sun Microsystems, Inc.

Oracle International Corporation

Oracle America, Inc.

Sudhakar Bobba

Pradeep R Trivedi

Tyler J Thorp

Dean Liu

Claude R Gauthier

Gin S Yee

David J Greenhill

Georgios K Konstadinidis

Lynn Ooi

Weiran Kong

Bernard Ho

Disclosed is a digital voltage regulator system and method for mitigating voltage droop in an integrated circuit. If an unacceptable voltage droop is detected, the digital voltage regulator may take action to allow the power supply voltage to recover. A digital voltage regulator in accordance with embodiments discussed herein detects voltage droop by comparing a power supply voltage measurement with a threshold voltage. The threshold voltage may be calibrated based on power supply voltage measurements taken while the integrated circuit is operating.

Microprocessor performance and power optimization through self calibrated inductive voltage droop monitoring and correction

A voltage droop monitoring and correcting circuit for a microprocessor includes: a monitor circuit configured to monitor voltage droops of the microprocessor and perform a temporary clock-skipping technique to compensate for the voltage droops. A method for monitoring and correcting voltage droops of a microprocessor includes: monitoring voltage droops of the microprocessor; and performing a temporary clock-skipping technique to compensate for the voltage droops. A computer system includes memory; a processor operatively connected to the memory; and computer-readable instructions stored in the memory for causing the processor to: monitor voltage droops of the microprocessor; and perform a temporary clock-skipping technique to compensate for the voltage droops.

Microprocessor performance and power optimization through inductive voltage droop monitoring and correction

A technique for actively shielding a signal such that a signal driver of the signal only participates in discharge events is provided. Because the signal driver only participates in discharge events, the signal driver is non-interacting with respect to other driver devices. Shield wires are set such that an active transition on the signal causes a discharge of capacitance between the signal and the shield wires.

Signal shielding technique using active shields for non-interacting driver design

A capacitor structure in a semiconductor device is provided. The capacitor structure includes a first power rail on a topmost level of the semiconductor device, and a second power rail on the topmost level of the semiconductor device. The capacitor structure also includes a dielectric layer disposed over at least a portion of one of the first power rail and the second power rail. The capacitor structure further includes a conductive layer disposed over and between the first power rail and the second power rail where the conductive layer is in electrical contact with the power rail not having the dielectric layer, and the conductive layer is disposed over the dielectric layer.

Thin capacitive structure

A method and apparatus for assigning a set of region-based voltage drop budgets to an integrated circuit is provided. Further, a method for partitioning an integrated circuit into optimal voltage drop regions includes analyzing the integrated circuit for worst-case voltage drop data. The worst-case voltage drop data is used to partition the integrated circuit into a set of voltage drop regions, wherein each voltage drop region is assigned a region-based voltage drop budget. The region-based voltage drop budget assigned to a particular voltage drop region is based on a worst-case voltage drop experienced by that voltage drop region.

Region-based voltage drop budgets for low-power design

A method and apparatus for improving the timing accuracy of an integrated circuit through region-based voltage drop budgets is provided. Further, a method for performing timing analysis on an integrated circuit partitioned into voltage drop regions is provided. During the timing analysis, a set of logic paths segments in each voltage drop region is tested to ensure that the integrated circuit meets a set of predefined timing requirements. Logic path segments that reside in different voltage drop regions are tested using a supply voltage inputted by the respective voltage drop region.

Accuracy of timing analysis using region-based voltage drop budgets

A device that uses an input clock signal to generate an output clock signal with a desired frequency is provided. The device uses a voltage controlled delay element that outputs a reset signal to a flip-flop dependent on a bias signal and the input clock signal. When triggered, the flip-flop outputs a transition on the output clock signal, which, in turn, serves as an input to a duty cycle corrector that generates the bias signal dependent on the configuration of the duty cycle corrector. The duty cycle corrector may be configured to generate the bias signal so as to be able to operatively control the duty cycle of the output clock signal.

Duty cycle corrector

A phase locked loop design uses a diode operatively connected to a loop filter capacitor to control a leakage current of the loop filter capacitor. By positioning a diode in series with the loop filter capacitor, a voltage potential across the loop filter capacitor is reduced, thereby reducing the leakage current of the loop filter capacitor. Moreover, the leakage current of the loop filter capacitor is controlled in that it cannot exceed the current through the diode. Control and reduction of the loop filter capacitor leakage current leads to more reliable and stable phase locked loop behavior.

Phase locked loop design with diode for loop filter capacitance leakage current control

A negative impedance device that accelerates signal transitions on a signal is provided. The negative impedance device is highly responsive to high to low and low to high transitions on the signal, and when one of these types of transitions begins to occur on the signal, the negative impedance device senses the transition and quickly drives the signal to the intended value before a point in time when the signal would have reached the intended value had the negative impedance device not been used. Further, a signal transition accelerator design that reduces signal rise and fall times is provided. Further, a method for accelerating a signal transition is provided.

Transmission gate based signal transition accelerator

A 150 degree bump placement layout for an integrated circuit power grid is provided. This layout improves integrated circuit performance and reliability and gives an integrated circuit designer added flexibility and uniformity in designing the integrated circuit. Further, a patterned bump array for a top metal layer of an integrated circuit having a plurality of 150 degree bump placement structures is provided.

150 degree bump placement layout for an integrated circuit power grid

A modulation circuit arranged to modulate a first voltage from a first power supply grid to produce a desired second voltage not greater than the first voltage on a second power supply grid is provided. A digital register is operatively connected to the modulation circuit to determine the desired second voltage on the second power supply grid. Furthermore, the digital register maintains a value representative of an activity level or an anticipated activity level of a circuit connected to the second power supply grid. The modulation circuit maintains the desired second voltage for the circuit connected to the second power supply grid by transferring charge between capacitances.

Dynamic modulation of on-chip supply voltage for low-power design

A method for preferentially shielding a signal to increase implicit decoupling capacitance is provided. The signal is preferentially shielded by using a probability of the signal being at a specific value to assign a shield potential. Further, an integrated circuit that preferentially shields a signal to increase decoupling capacitance by using a probability of the signal being at a specific value to assign a shield potential is provided. Further, a computer system for preferentially shielding a signal to increase decoupling capacitance by using a probability of the signal being at a specific value to assign a shield potential is provided. Further, a computer readable medium having executable instructions for preferentially shielding a signal to increase implicit decoupling capacitance by using a probability of the signal being at a specific value to assign a shield potential is provided. Further, a method to increase system performance by increasing implicit decoupling capacitance is provided.

Shield assignment using preferential shields

A technique that uses active shields to reduce clock skew is provided. The technique uses a shield wire for shielding the signal wire, a driver stage for driving a leading clock signal on the shielding wire, and a signal wire buffer for driving a lagging clock signal on the signal wire, where the leading clock signal is driven onto the first shield wire a phase difference before the lagging clock signal is driven onto the signal wire.

Clock skew reduction using active shields

A method and apparatus for assigning decoupling capacitors on an integrated circuit such that leakage power is minimized is provided. Particularly, the method and apparatus use an available capacitance area of an integrated circuit, a capacitance requirement of the integrated circuit, an available thin-oxide capacitance amount, and an available thick-oxide capacitance amount to generate an assignment that indicates what percentage of the available capacitance area should be filled with thin-oxide capacitors and what percentage of the available capacitance area should be filled with thick-oxide capacitors in order to meet the capacitance requirement and minimize leakage power attributable to the thin-oxide and thick-oxide capacitors.

Decoupling capacitance assignment technique with minimum leakage power

A method that preferentially shields a signal to increase decoupling capacitance is provided. The signal is preferentially shielded based on a probability of the signal being at a specific value. Because the shield may also be used to form the power and ground grid, a balanced number of power versus ground lines is desired. A method for inverting the signal to balance the number of power versus ground lines is provided. Further, a method to increase system performance by increasing implicit decoupling capacitance is provided.

Growing community of inventors

Sunnyvale, CA, United States of America

Sudhakar Bobba