Methods and systems for performing fault injection testing on an integrated circuit hardware design. The methods include: (a) receiving a raw fault node list identifying one or more fault nodes of the hardware design; (b) receiving information indicating a grouping of the fault nodes in the raw fault node list into a plurality of fault node groups, each fault node group comprising fault nodes that have a same effect on a failure mode of the hardware design; (c) generating a final fault node list based on the fault node groups; (d) selecting a set of fault injection parameters from the final fault node list, the set of fault injection parameters identifying at least one fault node in the final fault node list to fault; (e) performing a fault injection test on the hardware design by causing a fault to be injected into a simulation of the hardware design based on the selected set of fault injection parameters; (f) determining a result of the fault injection test; (g) storing the result of the fault injection test; and repeating (d) to (g) at least once.

A failure diagnostic apparatus includes a path calculation unit which calculates, for each input pattern to a diagnosis target cell, a path affecting an output value of the diagnosis target cell when a failure is assumed as an activation path, a path classification unit which classifies the activation path associated with the input pattern for which the diagnosis target cell has passed a test and the activation path associated with the input pattern for which the diagnosis target cell has failed the test, a path narrowing unit which calculates a first failure candidate path, a second failure candidate path and a normal path of the diagnosis target cell based on classified activation paths, and a result output unit which outputs information on the first failure candidate path, the second failure candidate path and the normal path.

Various aspects of the disclosed technology relate to machine learning-based chain diagnosis. Faults are injected into scan chains in a circuit design. Simulations are performed on the fault-injected circuit design to determine observed failing bit patterns. Bit-reduction is performed on the observed failing bit patterns to construct first training samples. Using the first training samples, first-level machine-learning models are trained. Affine scan cell groups are identified. Second training samples are prepared for each of the affine scan cell groups by performing bit-filtering on a subset of the observed failing bit patterns associated with the faults being injected at scan cells in the each of the affine scan cell groups. Using the second training samples, second-level machine-learning models are trained. The first-level and second-level machine learning models can be applied in a multi-stage machine learning-based chain diagnosis process.

A digital circuit robustness verification method is provided that includes the following steps. An internal storage circuit and an external storage circuit corresponding to a circuit under test are set to store a plurality of random values and a configuration of the circuit under test for performing a predetermined function is set by a processing circuit. A driving signal corresponding to the predetermined function is transmitted to the circuit under test by a previous stage circuit, such that the circuit under test executes the predetermined function to further generate an output signal. The determination as to whether the output signal is correct or not is made by a next stage circuit, and the circuit under test is determined to pass a robustness verification when the output signal is correct.

Embodiments relate to a system for translating design specifications of an electronic circuit. In one embodiment, the design specification is parsed to identify one or more sentences. From the one or more identified sentences, the system extracts semantic concepts. Additionally, for each sentence of the one or more identified sentences, the system determines whether the sentence is translatable. If a target sentence is translatable, the system generates a parse tree for the target sentence, and generates a probabilistic shift-reduce schedule for the generated parse tree. Using the generated probabilistic shift-reduce schedule and optionally the generated parse tree, the system generates code for verifying the condition specified in the target sentence. In some embodiments, to generate the code, the system parses the target sentence using the generated probabilistic shift-reduce schedule.

A method and a system for verifying an integrated circuit stack having a silicon photonic (SIPH) device is introduced. A single first dummy layer is added to at least one terminal of the SIPH device in a first layout of the first integrated circuit, wherein a shape of the single first dummy layer added to the at least one terminal of the SIPH device maps a shape of the at least one terminal of the SIPH device. A first layout versus schematic (LVS) check is performed on the first integrated circuit based on the single first dummy layer added to the at least one terminal of the SIPH device to verify a connection of the SIPH device in the first integrated circuit.

A method, computer program product, and/or system is disclosed for testing integrated circuits, e.g., processors, that includes: generating a software design prototype of the functional behavior of an integrated circuit to be tested; creating a lab All-Events-Trace (AET) normalized model of the integrated circuit, wherein the normalized model captures the functions of the integrated circuit and not the non-functional aspects of the integrated circuit; generating a lab scenario using the software design prototype and the AET normalized model of the integrated circuit for a particular cycle of interest, wherein the lab scenario contains initialization for all signals that have hardware information; and generating a replayed lab normalized AET for the particular cycle of interest.

An electronic assembly has a host wafer having a first circuit including wafer transistors and passive, non-transistor devices. Chiplets have a second circuit including at least one radio frequency (RF) transistor device. Electrical interconnects are between the chiplets and wafer. The electrical interconnects electrically connect the first circuit to the second circuits. Oscillators that have the wafer transistor, the RF transistor and the electrical interconnects produce a signal for built-in self-test circuits for testing an assembly design of the electronic assembly and speeds of the RF chiplet transistors.

A method includes identifying state holding pipeline stages in a pipeline path of a design for test (DFT) of an integrated circuit design, splitting each pattern of a plurality of patterns into a first part and a second part, reformatting the plurality of patterns to generate another plurality of patterns such that the first part and the second part of each pattern of the plurality patterns are included in different patterns of the another plurality of patterns. The length of the first part is a function of a number of the identified pipeline stages.

The disclosure provides using test processors to provide a more flexible solution compared to the existing DFX blocks that are used for controlling test networks in chips. The test processors provide a highly flexible solution since programming of the test processors can be changed at any time; even after manufacturing, and can support practically an unlimited number of core chips in any configuration. The high flexibility provided via the test processors can reduce engineering effort needed in design and verification, accelerate schedules, and may prevent additional tapeouts in case of DFX design bugs. By making debug and diagnosis easier by providing an opportunity to change debug behavior as needed, the time-to-market timeline can be accelerated. Accordingly, the disclosure provides a chip with a test processor, a multi-chip processing system with a test processor, and a method of designing a chip having a test processor.