Abstract: Training data may be collected based on a set of test-case configurations for each integrated circuit (IC) design in a set of IC designs. The training data may include a set of features extracted from each IC design, and a count of test cycles required for achieving a target test coverage for each test-case configuration. A machine learning (ML) model may be trained using the training data to obtain a trained ML model. The trained ML model may be used to predict a set of ranked test-case configurations for a given IC design based on features extracted from the given IC design.

Abstract: This disclosure describes a system and method of automatically capturing source code and associated artifacts for static analysis. A method includes receiving a current state of a project that includes a set of files in a directory to be captured for analysis and a current capture status of individual files of the set of files, determining a plan including a sequence of actions in response to the current state of the project, and executing the sequence of actions to capture each of the set of files. The sequence of actions includes capturing buildable modules in the set of files with a build-capture based on a default build command and a buildless-capture based on module definition files.

Abstract: A method and system for guided power grid augmentation determines a minimum resistance path for cells within an integrated circuit (IC) design. The minimum resistance path traces a conducting wire connecting a pin of a cell to an IC tap within the IC design. A voltage drop value for each of the cells is determined so as to identify target cells having a voltage drop value that satisfies a voltage drop criteria. Polygons have defined size characteristics are defined around the minimum resistance paths of the target cells, and conductors, such as additional conductors, are generated within the defined polygons.

Abstract: Clock multiplexer circuitry outputs one of a first or second clock signal. First selection circuitry is connected in series with first counter circuitry. The first selection circuitry and the first counter circuitry receive a first clock signal and a first selection signal. A first control signal is generated based on the first clock signal and the first selection signal. Second selection circuitry is connected in series with second counter circuitry. The second selection circuitry and the second counter circuitry receive a second clock signal and a second selection signal. A second control signal is generated based on the second clock signal and the second selection signal. The output circuitry is connected to the first counter circuitry and the second counter circuitry. The output circuitry outputs one of the first clock signal and the second clock signal based on the first control signal and the second control signal.

Abstract: A method comprises receiving an integrated circuit (IC) design file and determining, by one or more processors, dose information from the IC design file. The method further comprises determining, by the one or more processors, a mask vector file from the IC design file, and converting, by the one or more processors, the dose information to a vector file format. Further, the method comprises outputting the dose information in the vector file format and the mask vector file to a mask writer device.

Abstract: Methods and systems are described herein relate to automatic channel identification of high-bandwidth memory channels and subchannel generation. An HBM channel identification system may perform a sequence of operations to identify HBM channels within a netlist of an interposer: channel dimension prediction, channel bounding box prediction, channel orientation derivation, subchannel partition, and subchannel routing region creation. In one example, an HBM channel identification method includes identifying candidate nets within a netlist. A bounding box that includes one or more nets of the candidate nets is determined. Once the bounding box is determined, the orientation of the box is determined and used to determine a pattern of bumps within the bounding box. Finally, a subchannel is generated based on the pattern of bumps.

Abstract: Embodiments relate to a system and method for analyzing strongly connected components (SCCs) in a design of an integrated circuit. In one embodiment, a design of an integrated circuit is received, and a set of loops are identified in the received design. Based on the identified loops, one or more SCCs are determined. Each SCC includes multiple loops having shared paths. For instance, an SCC includes a first loop having a first set of nodes connected via a first set of paths and a second loop having a second set of nodes connected via a second set of paths, such that the first loop and the second loop have at least one path in common. The identified SCCs are then analyzed and presented to the user for consideration when reviewing the design of the integrated circuit.

Abstract: A method for debugging an electronic subsystem is disclosed. The method includes converting a first message in a first protocol format received at a first functional logical block of a plurality of functional logical blocks of an electronic subsystem into a second message in a second protocol format at the first functional logical block, wherein the second message includes a unique identifier (UID), and generating a first trace file corresponding to the first functional logical block, wherein the first trace file includes the UID. The method includes forwarding the second message from the first functional logical block to a second functional logical block. The method includes generating a second trace file corresponding to the second functional logical block, wherein the second trace file includes the UID, and performing an analysis on the first and the second functional logical blocks.

Abstract: An integrated circuit (IC) may include a memory device and a circuit coupled with the memory device. The circuit may precondition the memory device to sustain oscillations, initiate first oscillations in a first loop that includes the memory device, and initiate second oscillations in a second loop that does not include the memory device.

Abstract: A freeform optical surface includes, in part, an off-axis optical surface and a departure optical module. The off-axis optical surface may be an off-axis conic optical surface. The departure optical module may be substantially perpendicular to the off-axis conic optical surface.

Abstract: This disclosure describes a method for finding equivalent classes of hard defects in a stacked MOSFET array. The method includes identifying the stacked MOSFET array in a circuit netlist. The stacked MOSFET array includes standard MOSFETs sharing gate and bulk terminals. The method further includes determining electrical defects for the standard MOSFETs, grouping the electrical defects into at least one intermediate equivalent defect class based on a topological equivalence of the electrical defects, grouping the electrical defects in the at least one intermediate equivalent defect class into at least one final equivalent defect class based on an electrical equivalence of the electrical defects, performing a defect simulation on an electrical defect in the at least one final equivalent defect class, and attributing a result of the defect simulation on the electrical defect to additional electrical defects in the final equivalent defect class.

Abstract: Embodiments relate to a system, program product, and method for integrating compiler-based testing in post-silicon validation. The method includes generating a test program through a device-under-test (DUT). The method also includes generating a plurality of memory intervals and injecting the plurality of memory intervals into the test program. The method further includes injecting a plurality of compiled test snippets into the test program and executing one or more post-silicon validation tests for the DUT with the test program.

Abstract: A circuit to multiplex supply voltages may include a set of chains of transistors. Each chain of transistors may correspond to a voltage supply that is desired to be multiplexed and may include a set of transistors coupled in series. A first end terminal of each chain of transistors may be coupled to a corresponding voltage supply, and a second end terminal of each chain of transistors may be coupled to an output terminal of the circuit. A given supply voltage may be selected by turning on transistors in a chain of transistors that corresponds to the given supply voltage and turning off transistors in other chains of transistors.

Abstract: A method includes: receiving a representation of a mixed-signal integrated circuit design including an analog circuit portion and a digital circuit portion including a plurality of descriptions of a power supply, the descriptions including a power supply network description and a register transfer level (RTL) hardware description language (HDL) description; determining a mismatch between the power supply network description and the HDL description of the power supply; generating a value converter to convert a voltage value associated with the power supply between the power supply network description and the HDL description; and converting, by a processor, between the power supply network description and the HDL description during runtime using the value converter to synchronize the power supply network description and the HDL description of the power supply responsive to the mismatch.

Abstract: Techniques for integrated circuit (IC) design are disclosed. A path margin is determined for an endpoint of a plurality of timing paths for an IC design. This includes identifying a sub-critical path, among the plurality of timing paths, where the sub-critical path has more slack than a critical path relating to the endpoint. The path margin is generated based on a first slack associated with the sub-critical path. A second slack, relating to at least one of the plurality of timing paths, is modified from a first value to a second value, based on the path margin. A design metric relating to the IC design is updated based on the modified second slack. The IC design is configured to be used to fabricate an IC.

Abstract: A method of testing a product using confidence estimates is provided. The method includes identifying a set of candidate tests and estimating a respective confidence score for each candidate test, the confidence scores reflecting a level of confidence that the corresponding candidate tests will pass or fail when being performed on the product, the estimating including determining the respective confidence scores in dependence upon at least one of (i) previously obtained test results, (ii) changes to the product since a previous estimation or regression test has been performed and (iii) information regarding a user. The method includes identifying a candidate test having a confidence score that is below a threshold, in response to the identification of the candidate test, performing the candidate test, and providing, to a user, results of the performing of the candidate test.

Abstract: Disclosed herein are method, system, and storage-medium embodiments for single-pass diagnosis of multiple chain defects in circuit-design testing. Embodiments include processor(s) to select a plurality of a scan chains in a circuit under test and determine presence of at least a first defect in the first scan chain, and a second defect in the first scan chain or in the second scan chain. The plurality of scan chains may include specific scan chains that each have respective pluralities of scan cells. Processor(s) may map the first defect to a first range of first scan cells, and the second defect to a second range of second scan cells. Based at least in part on a failing capture-pattern set, processor(s) may locate the first defect in a first scan cell of the first range, and the second defect in a second scan cell of the first range or the second range.

Abstract: A system and method for placement and simulation of a cell in proximity to a cell with a diffusion break is herein disclosed. According to one embodiment, an integrated circuit is designed to include a first cell that has a first edge and a second edge opposite the first edge. The first cell may also include a diffusion region that extends from the first edge to the second edge with a diffusion break separating the diffusion region. The diffusion break may be spaced away from the second edge by a distance that degrades a metric (e.g., a delay, a slew, dynamic power, or leakage) of a second cell placed next to the second edge beyond a threshold level.

Abstract: A method, a system, and non-transitory computer readable medium for aging analysis are provided. The method includes performing stress simulations for a plurality of process, voltage, temperature (PVT) conditions for a circuit, the circuit including one or more devices, extrapolating device level stresses obtained from the stress simulations into device level parameter degradations to a desired circuit age; and performing degradation simulations for the circuit for the same PVT conditions based on the device level parameter degradations. Each degradation simulation for a PVT condition of the plurality of PVT conditions is performed using the device level parameter degradations associated with the same PVT condition.

Abstract: A method of constructing a waveform from N sampled data captured at N successive points in time, includes, in part, applying the N sampled data, K data at a time, to each of M delayed replicas of a filter that includes K taps so to generate N×M interpolated data. The waveform is then constructed from the N sampled data and the N×M interpolated data.