Mentor Graphics Corporation

Other

Siemens Industry Software Gmbh

Mentor Graphics, a Siemens Business

Kun-Han Tsai

Janusz Rajski

Wu-Tung Cheng

Mark A Kassab

Chen Wang

Xijiang Lin

Yu Jane Huang

Shi-Yu Huang

Ruifeng Guo

Dhiraj Goswami

Nagesh Tamarapalli

Gaurav Veda

Mason Chern

Jeo-Yen Lee

Takeo Kobayashi

Naixing Wang

Tzu-Heng Huang

Irith Pomeranz

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 test response patterns in response to the test patterns which are captured by the scan chains. Observed failing bit patterns are determined by comparing the unloaded test response patterns with corresponding good-machine test response patterns. Bit-reduction is performed on the observed failing bit patterns to construct training samples. Using the training samples, machine-learning models for faulty scan cell identification are trained. The bit reduction comprises pattern-based bit compression for good scan chains or cycle-based bit compression for the good scan chains. The bit reduction may further comprise bit-filtering. The bit-filtering may comprises keeping only sensitive bits on faulty scan chains for the training samples construction.

Input data compression for machine learning-based chain diagnosis

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.

Multi-stage machine learning-based chain diagnosis

Systems and methods for a deterministic automatic test generation (ATPG) process including Timing Exception ATPG (TEA). A method includes performing an automated test pattern generation (ATPG) process that uses timing exception information to generate a test pattern for a targeted fault of a circuit design with at least one timing exception path. The method includes testing the targeted fault of the circuit design using the test pattern to produce a test result for the targeted fault.

Deterministic test pattern generation for designs with timing exceptions

Disclosed herein are exemplary methods, apparatus, and systems for performing timing-aware automatic test pattern generation (ATPG) that can be used, for example, to improve the quality of a test set generated for detecting delay defects or holding time defects. In certain embodiments, timing information derived from various sources (e.g. from Standard Delay Format (SDF) files) is integrated into an ATPG tool. The timing information can be used to guide the test generator to detect the faults through certain paths (e.g., paths having a selected length, or range of lengths, such as the longest or shortest paths). To avoid propagating the faults through similar paths repeatedly, a weighted random method can be used to improve the path coverage during test generation. Experimental results show that significant test quality improvement can be achieved when applying embodiments of timing-aware ATPG to industrial designs.

Timing-aware test generation and fault simulation

An integrated circuit for on-chip speed grading comprises test circuitry comprising scan chains and a test controller; and wide-range clock signal generation circuitry comprising phase-locked loop circuitry and frequency divider circuitry. The wide-range clock signal generation circuitry is configured to generate a wide-range test clock signal for the test circuitry to conduct a structural delay test for on-chip speed grading. The wide-range test clock signal is generated based on a test clock signal associated with the test circuitry, a frequency range selection signal and a frequency setting signal.

Wide-range clock signal generation for speed grading of logic cores

Various aspects of the disclose techniques relate to techniques of testing interconnects in stacked designs. A single-pulse signal, generated by a first circuit state element on a first die, is applied to a first end of an interconnect and captured at a second end of the interconnect using a clock port of a second circuit state element on a second die. A faulty interconnect may cause the single-pulse signal too distorted to reach the threshold voltage of the second circuit element.

Method and circuit of pulse-vanishing test

Methods of diagnostic test pattern generation for small delay defects are based on identification and activation of long paths passing through diagnosis suspects. The long paths are determined according to some criteria such as path delay values calculated with SDF (Standard Delay Format) timing information and the number of logic gates on a path. In some embodiments of the invention, the long paths are the longest paths passing through a diagnosis suspect and reaching a corresponding failing observation point selected from the failure log, and N longest paths are identified for each of such pairs.

Diagnostic test pattern generation for small delay defect

Speed-path debug techniques based on at-speed scan test patterns. Potential speed paths are identified based upon detected at-speed scan pattern failures and unknown X-value simulation. When the number of identified speed paths is large, the suspect speed paths are ranked.

Speed-path debug using at-speed scan test patterns

Methods, apparatus, and systems for performing fault diagnosis are disclosed herein. In certain disclosed embodiments, methods for diagnosing faults from compressed test responses are provided. For example, in one exemplary embodiment, a circuit description of an at least partially scan-based circuit-under-test and a compactor for compacting test responses captured in the circuit-under-test is received. A transformation function performed by the compactor to the test responses captured in the circuit-under-test is determined. A diagnostic procedure for evaluating uncompressed test responses is modified into a modified diagnostic procedure that incorporates the transformation function therein. Computer-readable media comprising computer-executable instructions for causing a computer to perform any of the disclosed methods are also provided. Likewise, computer-readable media comprising lists of fault candidates identified by any of the disclosed methods or circuit descriptions created or modified by the disclosed methods are provided.

Compactor independent fault diagnosis

Improved responses can be generated to scan patterns (e.g., test patterns) for an electronic circuit designs having timing exception paths by more accurately determining the unknown values that propagate to observation points in the circuit, where the response is captured. For instance, the responses are determined more accurately by analyzing the effect of sensitizing a timing exception path during each time frame associated with a scan pattern. Path sensitization can be determined based on observing whether values injected at starting points of the timing exception paths due to signal transitions and glitches propagate to their end points. The response can be updated by masking the affected end points and propagating unknown values further in the circuit to determine whether they are captured at observation points of the circuit. For instance, the methods and systems described herein may result in reduced unknowns, improved test coverage and test compression.

Generating responses to patterns stimulating an electronic circuit with timing exception paths

Improved responses can be generated to scan patterns (e.g., test patterns) for an electronic circuit design having timing exception paths by more accurately determining the unknown values that propagate to observation points in the circuit, where the response is captured. For instance, the responses are determined more accurately by analyzing the effect of sensitizing a timing exception path during each time frame associated with a scan pattern. Path sensitization can be determined based on observing whether values injected at starting points of the timing exception paths due to signal transitions and glitches propagate to their end points. The response can be updated by masking the affected end points and propagating unknown values further in the circuit to determine whether they are captured at observation points of the circuit. For instance, the methods and systems described herein may result in reduced unknowns, improved test coverage and test compression.

Growing community of inventors

Lake Oswego, OR, United States of America

Kun-Han Tsai