Google Inc.

Synopsys, Inc.

Magma Design Automation, Inc.

Esilicon Corporation

Elastix Corporation

Emre Tuncer

Jeffrey Adgate Dean

Chian-Min Richard Ho

Azalia Mirhoseini

Anna Darling Goldie

Alessandra Nardi

Mustafa Nazim Yazgan

Anand Babu

Ya Wang

William Hang

Anirudh Devgan

Tao Lin

Hui Zheng

Jordi Cortadella

Alfred Wong

Vivek Raghavan

Pramod Thazhathethil

Amir Ajami

Luciano Lavagno

Vigyan Singhal

Huachang Xu

Ramprasad Raghavan

Fanny Gur

Manish Harnur

Hamid Savoj

Premal Buch

Srinath R Naidu

Aliaksandr Antonau

Kaushik Balamukundhan

Yiran Li

Detecting voltage-based attacks on an integrated circuit (IC) is difficult in the presence of clock jitter. Propagating signals can exhibit a total delay that is due to a delay component resulting from a voltage-based attack and a delay characteristic resulting from clock fluctuation. Voltage-variation detection circuitry includes first and second voltage-dependent circuits and a voltage analysis circuit. The voltage-dependent circuits produce first and second signals that are indicative of a voltage level responsive to a clock signal and based on different first and second voltage sensitivities. The voltage analysis circuit generates a voltage alert signal based on the first and second signals. A combined signal neutralizes the delay characteristic in the first and second signals, but the delay component due to the voltage variation can be at least partially maintained. Thus, a voltage-based attack is detectable in the presence of clock fluctuation by using two voltage-dependent circuits.

Voltage-variation detection under clock fluctuations

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating a computer chip floorplan. One of the methods includes obtaining netlist data for a computer chip; and generating a computer chip floorplan, comprising placing a respective node at each time step in a sequence comprising a plurality of time steps, the placing comprising, for each time step: generating an input representation for the time step; processing the input representation using a node placement neural network having a plurality of network parameters, wherein the node placement neural network is configured to process the input representation in accordance with current values of the network parameters to generate a score distribution over a plurality of positions on the surface of the computer chip; and assigning the node to be placed at the time step to a position from the plurality of positions using the score distribution.

Generating integrated circuit floorplans using neural networks

This document describes techniques and systems for leakage screening based on power prediction. In particular, the described systems and techniques estimate, during a silicon manufacturing process, use-case power (e.g., low power, ambient power, high power, gaming power) to apply leakage screening for apart (e.g., a chip package). In some aspects, measurable silicon parameters (e.g., leakage values, bin values, processor sensor values) may be used for use-case power prediction. Using the described techniques, a maximum allowable predicted use-case power can be determined and used for leakage screening regardless of an individual rail leakage or voltage bin assignment.

Leakage screening based on use-case power prediction

A method for performing leakage analysis includes receiving information specifying an integrated circuit. A neighborhood of shapes associated with the integrated circuit is then determined. Leakage information associated with the integrated circuit is generated based on the neighborhood of shapes. The neighborhood of shapes may be determined by determining a first set of spacings to a boundary of a first cell from an internal shape. A second set of spacings may be determined from the boundary of the first cell to a shape of a second cell. A lithography process may be characterized using the first and second set of spacings.

Lithography aware leakage analysis

A system for automatically transforming a given synchronous circuit description into an equivalent and provably correct desynchronized circuit description. Included in the automated transformation are techniques for synthesizing a variability-aware controller using a two-phase protocol, techniques for synthesizing a variability-aware controller using gated clocks and testability circuits, techniques for synthesizing a variability-aware controller optimized for performance, techniques for initializing the synthesized controller, techniques for dynamically minimizing power requirements, and techniques for interfacing the desynchronized circuit with external synchronous circuits. Also disclosed are techniques for implementing a system for automatically transforming a synchronous circuit description into an equivalent and provably correct desynchronized circuit description within the context of an electronic design automation design flow. Exemplary circuits used in the application of the aforementioned techniques are provided. Application of mathematical models and techniques used for proving equivalence between the input description and the resulting desynchronized circuit are presented and explained.

Variability-aware scheme for high-performance asynchronous circuit voltage regulation

A method for performing timing analysis includes receiving information specifying an integrated circuit. A neighborhood of shapes associated with the integrated circuit is then determined. Delay information associated with the integrated circuit is generated based on the neighborhood of shapes. The neighborhood of shapes may be determined by determining a first set of spacings to a boundary of a first cell from an internal shape. A second set of spacings may be determined from the boundary of the first cell to a shape of a second cell. A lithography process may be characterized using the first and second set of spacings.

Lithography aware timing analysis

A circuit interconnection structure for synchronizing a network of oscillators placed on a semiconductor substrate. One such structure comprises a first synchronizing circuit electrically coupled to a second synchronizing circuit through tunable delay circuits. Also disclosed are methods to tune oscillators placed in different regions of a circuit having multiple clock domains by estimating the relative slack of a first group of signals within the circuit with regard to the period of a first clock domain, and estimating the relative slack of the second group of signals within the circuit with regard to the period of second clock domain, wherein the estimating is performed at process and operational corners that cover the variability of the circuit at different speed conditions, then calculating tuning values for the oscillator delays for each region such that the oscillator delay slack matches the worst relative slack of the signals of the same region.

Network of tightly coupled performance monitors for determining the maximum frequency of operation of a semiconductor IC

Variability-aware scheme for asynchronous circuit initialization

A system and a method are disclosed for circuit analysis. A circuit modeling system calculates sensitivities of gates for statistical static timing analysis of a circuit. Timing distribution sensitivities of gates and correlations between the sensitivities are determined. A Monte Carlo simulation is run using the sensitivities to determine timing distribution of paths and determine probabilities of paths being the critical path. Aggregate sensitivities for cells are also determined.

Aggregate sensitivity for statistical static timing analysis

A process for reducing cross-talk noise in a VLSI circuit is disclosed. The process identifies a victim net in an integrated circuit and calculates a change in ground capacitance for the victim net to identify a noise amplitude less than or equal to a maximum allowable noise height. The process selects from a library one cell or a grouping of cells having an input capacitance for the victim net closest to the change in ground capacitance. The selected cell or grouping of cells is coupled to the victim net so that its change in ground capacitance provides a noise amplitude less than (or less than or equal to) an allowable maximum noise height that may be a predetermined value. A system for reducing cross-talk noise in a VLSI circuit is also disclosed.

Reduction of cross-talk noise in VLSI circuits

Growing community of inventors

Santa Cruz, CA, United States of America

Emre Tuncer