Synopsys, Inc.

Kaushik De

Rajarshi Mukherjee

Mahantesh D Narwade

David L Allen

Namit Gupta

Bhaskar Pal

Dipti Ranjan Senapati

Sabyasachi Das

Dhiraj Goswami

Paras Mal Jain

Eduard Rudolf Cerny

Kevin Michael Harer

Sanjay Gulati

Badri P Gopalan

Renu Mehra

Fahim Rahim

Pallab Dasgupta

Partha Pratim Chakrabarti

Suman Nandan

Malitha Kulatunga

Subhamoy Pal

Sachin Bansal

Meirav Nitzan

Stewart Williams

Gaurav Pratap

Godwin R Maben

David Allen

Nishant Patel

Meera Viswanath

In one aspect, a method includes invoking a signoff tool via a first command from an implementation tool running on a register transfer level (RTL) design, and executing a native command of the signoff tool from within the implementation tool. The native command generates a notification. The method also includes determining whether the RTL design passes a low-power signoff check based on the notification and sending the design for final signoff verification based on the determination that the RTL design passes the low-power signoff checks.

Low-power static signoff verification from within an implementation tool

A system enhances a system design to incorporate safety measures. The system receives a system design for processing through various stages of design using design tools, for example electronic design automation tools for introducing safety features in a circuit design. The system receives safety requirements for the system design, the safety requirements specifying safety measures for the system design. The system generates from the safety requirements, a safety specification storing a set of commands. The system generates a system design enhanced with safety measures. The enhanced system design it generated for at least a subset of the plurality of tools. A tool processes the generated safety specification to implement safety measures in the system design according to the received safety requirements.

Implementing and verifying safety measures in a system design based on safety specification generated from safety requirements

Disclosed herein are system, computer-readable storage medium, and method embodiments of machine-learning (ML)-based static verification for derived hardware-design elements. A system including at least one processor may be configured to extract a feature set from a hardware description and evaluate a similarity index of a first hardware element with respect to a second hardware element, using an ML process based on the feature set, wherein the first hardware element is described in the hardware description. The at least one processor may be further configured to update one or more parameters corresponding to a static verification of the hardware description while the static verification is being performed, by providing at least one test attribute, corresponding to the second hardware element, applicable to the first hardware element, in response to determining that the similarity index is within a specified range, and additionally output a first result of the static verification.

Machine learning (ML)-based static verification for derived hardware-design elements

State table complexity reduction in a hierarchical verification flow is provided by identifying peripheral supplies and non-peripheral supplies in a hierarchical group in a hierarchical logical block model of a circuit based on whether logic blocks associated with the power supplies provide outputs to or receive inputs from circuity external to the hierarchical group; merging associated power state tables for the peripheral supplies and the non-peripheral supplies in the hierarchical group to create a merged power state table for the hierarchical group; removing, by a processing device, any power states associated with the non-peripheral supplies from the merged power state table to create a reduced power state table; and modeling a reduced logical block based on the reduced power state table.

State table complexity reduction in a hierarchical verification flow

Apparatuses and methods for performing domain crossing verification of a register transfer level (RTL) representation of an integrated circuit (IC) that includes a memory block are provided. One example method includes receiving an RTL representation of an IC; automatically inferring one or more memory blocks in the RTL representation of the IC; identifying one or more input ports and one or more output ports of the one or more memory blocks; designating the one or more input ports and the one or more output ports as one or more start points and one or more end points; and performing domain crossing analysis on the RTL representation.

Apparatuses and methods for accurate and efficient clock domain and reset domain verification with register transfer level memory inference

A method and EDA software tool for analyzing and verifying that a multi-level power managed system description (IC design) is free of power-state combination conflicts by way of identifying and reconciling voltage level and power-state combination conflicts caused by reused blocks (IP cores). The reconciliation process involves generating Power-State Tables (PSTs) associated with each hierarchical circuit level (e.g., top/system level and lower/block levels) of the IC design using both initial power supply voltage values and reconciled/revised voltage values, which are determined by the main driver voltage levels of each power supply. Initial supply relationships generated using the initial PSTs are then compared with final supply relationships generated using the reconciled PSTs, whereby conflicts are identified when one or more initial supply relationship fails to match a final supply relationship, or when one or more final supply relationship fails to match an initial supply relationship.

Voltage reconciliation in multi-level power managed systems

A method for controlling the functional output of a verification tool upon receipt of a circuit description comprises searching for a predetermined base pattern in the circuit description. The method further comprises searching for predetermined sub-patterns that are assigned to the base pattern, in the circuit description. The method further comprises the validation of each found sub-pattern based on a predetermined rule to minimize the set of reported errors that based on verification of the circuit description.

Verification of circuit structures including sub-structure variants

Systems and techniques for detecting design problems in a circuit design are described. A higher-level abstraction of the circuit design can be synthesized to obtain a lower-level abstraction of the circuit design, and a mapping between signals in the higher-level abstraction and the signals in the lower-level abstraction. A design problem can be detected in the circuit design in response to determining that a possible glitch in a signal in the lower-level abstraction is not blocked when an enable signal is assigned a blocking value. The enable signal and the corresponding blocking value are identified by analyzing the higher-level abstraction.

Accurate glitch detection

A method for functional verification of a circuit description comprises generating a first set of crossover paths based on the circuit description, generating a low power information based on a power design description associated with the circuit description, the low power information determining a set of power state combinations, and generating a second set of crossover paths based on the first set of crossover paths and the low power information, the second set of crossover paths being a subset of the first set of crossover paths. Each of the second set of crossover paths is evaluated to identify circuit description errors, in particular functional circuit description errors.

Minimizing crossover paths for functional verification of a circuit description

Verification-result ranking techniques for root cause analysis are disclosed using violation report analysis and violation weighting. Violation reports are unwieldy and result from a variety of design and process checks. The check coverage can overlap, causing a specific violation to trigger multiple reported violations. High turn around times for violation report analysis increase the risk that selective violation analysis will inadvertently suppress real design bugs. This reduces the odds that static checker reports alone will meet design sign-off criteria. Determining relationships among a plurality of violations for a design permits clustering violations into hot spots. Identification of primary and subsequent contributors to the plurality of violations is based on the relationships among violations. The hot spot with the highest weight is identified, and then subsequent violations are identified to maximize violation coverage. The result is greater efficiency of design violation identification and resolution.

Ranking verification results for root cause analysis

An enhanced RTL simulation including information regarding multi-cycle paths is provided. The multi-cycle path information, which is available in the design constraint file, can be used for timing analysis during RTL simulation. This information can advantageously augment the RTL simulation engine to approximate the cycle delays at the destination registers, thereby providing a more realistic approximation of circuit behavior at the RTL level. Notably, RTL simulation is orders of magnitude faster than gate level simulation. Moreover, design bugs associated with multi-cycle paths are more easily corrected during RTL simulation compared to waiting until the gate level simulation.

Technique for honoring multi-cycle path semantics in RTL simulation

One embodiment of the present invention provides systems and techniques to extract assume properties from a constrained random test-bench. During operation, the system can receive a constrained random test-bench for verifying the design-under-test (DUT), wherein the constrained random test-bench includes a statement which assigns a random value to a random variable according to a constraint. Next, the system can modify the constrained random test-bench by replacing the statement with another statement which assigns a free input variable's value to the random variable. The system can also add a statement to the constrained random test-bench that toggles a marker variable to localize the scope of the statement. The system can then generate an assume property which models the constraint on the free input variable. The assume property can then be used by a formal property verification tool to verify the DUT.

Growing community of inventors

Pleasanton, CA, United States of America

Kaushik De