Abstract: An adaptive test system includes one or more reconfigurable test boards, with each test board including at least one re-configurable test processor. The re-configurable test processors can transmit communicate with one another using an inter-processor communications controller associated with each re-configurable test processor. The communications include configuration information, control information, communication protocols, stimulus data, and responses. Configuration information and stimulus data can also be read from a memory. Configuration information is used to configure one or more re-configurable test processors. Once configured, the re-configurable test processor or processors process the data in order to generate one or more test signals. The one or more test signals are then used to test a DUT.

Abstract: An excitation signal generator (“ESG”) is described. The ESG generates an minimized excitation signal for use in a test system to generate a functional model of a device under test (“DUT”) where extreme values of the minimized excitation signal are increased toward a central value without changing the power spectrum at the DUT.

Abstract: An integrated circuit, including a configurable scan architecture used for an integrated circuit test procedure and quality control. The configurable scan chain architecture has the capability of being reconfigured to one of a variety scan chain architectures based on the constraints of the integrated circuit and the testing device. The present invention minimizes the integrated circuit test time by reconfiguring the scan architecture depending on certain constraints such as the latching frequency, the predetermined I/O frequency, the number of available integrated circuit I/O pins, the number of pins required for a proposed scan architecture, and the number of available pins on the testing device. The configurable scan architecture receives configuration signals which indicate which scan chain architecture should be configured on the integrated circuit that is being tested.

Abstract: A system and method that enables testing of circuitry using an externally generated signature. An external tester is arranged external to a device under test (DUT). Such external tester is operable to input test data to the DUT, receive output data from the DUT, and generate a signature for at least a portion of such received output data. The external tester compares the generated signature with an expected signature to determine whether the DUT is functioning as expected. If the generated signature fails to match an expected signature, then error data can be written to an error map log. Preferably, further interaction with the DUT is not required after detecting that a generated signature fails to match an expected signature in order to perform such error evaluation. Thus, error evaluation can be performed concurrently with testing of the DUT. Mask data may be stored in a compressed form, and decompressed and used for masking certain non-deterministic output bits in generating the signature.

Abstract: The model-based method tests compliance of production devices with the performance specifications of a device design. The production devices are manufactured in accordance with the device design by a manufacturing process. In the method, a simple model form based on the device design and the performance specifications is developed, a stimulus for testing the production devices is specified and each production device is tested. The model form has a basis function and model form parameters for the basis function. The model form parameters are dependent on the manufacturing process and differ in value among the production devices. A production device is tested by measuring the response of the production device to the stimulus; extracting, using the model form, the values of the model form parameters for the production device from the measured response and the stimulus; and checking compliance of the production device with the performance specifications using the extracted values of the model form parameters.

Abstract: A multi-component production devices in accordance with a device design having performance specifications is made by a method comprising receiving respective component behavioral models of the production components constituting the production device; combining the component behavioral models in accordance with the device design to form a device behavioral model; and, prior to assembling the production device, and predicting performance metrics for the production device by performing simulated tests on the device behavioral model.

Abstract: Creating a low-bandwidth channel in a high-bandwidth channel. By taking advantage of extra bandwidth in a high-bandwidth channel, a low-bandwidth channel is created by inserting extra packets. When an inter-packet gap of the proper duration is detected, the extra packet is inserted and any incoming packets on the high-bandwidth channel are stored in an elastic buffer. Observing inter-packet gaps, minimal latency is introduced in the high-bandwidth channel when there is no extra packet in the process of being sent, and the effects of sending a packet on the low-bandwidth channel are absorbed and distributed among other passing traffic.

Abstract: Disclosed are systems and methods which implement multi-site testing in which a test sequence implemented with respect to dice of a set of dice for parallel testing is not identical. Accordingly, heterogeneous test sequences are employed with respect to dice tested in parallel. A sequence of tests for testing individual circuit blocks of dice may be selected for optimizing the time for testing the set of dice for which parallel testing is conducted. Additionally or alternatively, a sequence of tests for testing individual circuit blocks of dice may be selected for managing and/or reducing the resources utilized in testing the dice. Moreover, test sequences may be dynamically determined for the dice of a set of dice being tested in parallel, such as in response to feedback provided by sensors monitoring the dice, to address various side effects.

Abstract: A system and method enable testing of circuitry using an externally generated signature. An external tester is arranged external to a device under test (DUT). Such external tester is operable to input test data to the DUT, receive output data from the DUT, and generate a signature for at least a portion of such received output data. The external tester compares the generated signature with an expected signature to determine whether the DUT is functioning as expected. If the generated signature fails to match an expected signature, then error data can be written to an error map log. Preferably, further interaction with the DUT is not required after detecting that a generated signature fails to match an expected signature in order to perform such error evaluation. Thus, error evaluation can be performed concurrently with testing of the DUT. Mask data may be stored in a compressed form, and decompressed and used for masking certain non-deterministic output bits in generating the signature.

Abstract: A method of compressing a test vector creates a compressed test vector for use in conjunction with automated test equipment (ATE). The method comprises generating a test vector having a sequence of elements, at least one element of which comprises a ‘don't care’ value. A random sequence of elements is produced also. The test vector and the random sequence are segmented. Each segment of the test vector is compared to a corresponding segment of the random sequence to determine whether the corresponding segments match. When a match is found, a first flag value is sequentially inserted into a compression test vector. When a mismatch is found, a second flag value is sequentially inserted into the compression vector as well as the elements of the mismatched test vector segment. The compressed test vector may be decompressed according to the invention directly into a completely specified test vector using the flag values.

Abstract: Disclosed are systems and methods which implement multi-site testing in which a test sequence implemented with respect to dice of a set of dice for parallel testing is not identical. Accordingly, heterogeneous test sequences are employed with respect to dice tested in parallel. A sequence of tests for testing individual circuit blocks of dice may be selected for optimizing the time for testing the set of dice for which parallel testing is conducted. Additionally or alternatively, a sequence of tests for testing individual circuit blocks of dice may be selected for managing and/or reducing the resources utilized in testing the dice. Moreover, test sequences may be dynamically determined for the dice of a set of dice being tested in parallel, such as in response to feedback provided by sensors monitoring the dice, to address various side effects.

Abstract: A system and method are disclosed in which a plurality of dice on a semiconductor wafer are interconnected to enable efficient testing thereof. In certain embodiments a plurality of dice are interconnected in a manner that enables test data to be communicated from a tester system to a plurality of dice for concurrent testing of such plurality of dice. Depending on the amount of interconnection, all or a portion of each of the plurality of dice may be tested concurrently. In certain embodiments, a plurality of dice are interconnected in a manner that enables test data to be communicated from one die to at least one other die. In certain embodiments, a plurality of dice are interconnected in a manner that enables such dice to be tested concurrently while maintaining a repeatable pattern at the reticle level for fabricating such dice.

Abstract: An integrated circuit, including a configurable scan architecture used for an integrated circuit test procedure and quality control. The configurable scan chain architecture has the capability of being reconfigured to one of a variety scan chain architectures based on the constraints of the integrated circuit and the testing device. The present invention minimizes the integrated circuit test time by reconfiguring the scan architecture depending on certain constraints such as the latching frequency, the predetermined I/O frequency, the number of available integrated circuit I/O pins, the number of pins required for a proposed scan architecture, and the number of available pins on the testing device. The configurable scan architecture receives configuration signals which indicate which scan chain architecture should be configured on the integrated circuit that is being tested.

Abstract: A transition splitting apparatus and method reduce a maximum transition rate of a digital signal. The apparatus and method are particularly useful for digital signal processing in communications and for performing digital transition timing testing on a device under test. The apparatus and method split a digital signal into two or more signals, while preserving the relative timing of transitions in the digital signal. A high frequency signal is partitioned by the apparatus and method into a plurality of equivalent lower frequency signals without loss of transition timing information. The apparatus is implemented with readily available components. The maximum transition rate of a digital signal is reduced by a factor of two or more, and is proportional to the number of output signals. A plurality of the apparatuses may be cascaded together into a system to achieve even greater reductions in maximum transition rates.

Abstract: A method of compressing a test vector creates a compressed test vector for use in conjunction with automated test equipment (ATE). The method comprises generating a test vector having a sequence of elements, at least one element of which comprises a ‘don't care’ value. A random sequence of elements is produced also. The test vector and the random sequence are segmented. Each segment of the test vector is compared to a corresponding segment of the random sequence to determine whether the corresponding segments match. When a match is found, a first flag value is sequentially inserted into a compression test vector. When a mismatch is found, a second flag value is sequentially inserted into the compression vector as well as the elements of the mismatched test vector segment. The compressed test vector may be decompressed according to the invention directly into a completely specified test vector using the flag values.

Abstract: A transition splitting apparatus and method reduce a maximum transition rate of a digital signal. The apparatus and method are particularly useful for digital signal processing in communications and for performing digital transition timing testing on a device under test. The apparatus and method split a digital signal into two or more signals, while preserving the relative timing of transitions in the digital signal. A high frequency signal is partitioned by the apparatus and method into a plurality of equivalent lower frequency signals without loss of transition timing information. The apparatus is implemented with readily available components. The maximum transition rate of a digital signal is reduced by a factor of two or more, and is proportional to the number of output signals. A plurality of the apparatuses may be cascaded together into a system to achieve even greater reductions in maximum transition rates.

Abstract: In a multiport memory semiconductor integrated circuit, an efficient method of testing the memory for faults. The method determines a base address in a multiport memory. A plurality of addresses are scanned within the memory which are at a hamming distance of 1 from the base address, such that at least two memory cells in each column of the multiport memory device are accessed in the scan. This allows the detection of cross port faults, address mismatch faults, and bit shorts due to the fact that the faults are exposed when the two memory cells sharing access columns are accessed with test data. The method functions nominally without requiring any detailed information about the placement and routing structure of the multiport memory device.

Abstract: A method of and system for inserting test points within an integrated circuit design. According to the present invention, test points are inserted early in the electronic design process and prior to logic synthesis such that the problem of design constraint violation can be avoided. One embodiment of the present invention includes the computer implemented steps of receiving an unmapped netlist of an integrated circuit design, and receiving from an external source, data that indicates the location and the desired functionality of the test point to be inserted. Thereafter, the present invention inserts a generic test point circuit at the indicated location and generates a modified unmapped netlist. Subsequently, the present invention performs a logic synthesis process on the modified unmapped netlist where the generic test point circuit is degenerated into an actual test point circuit for performing the desired functionality.

Abstract: A circuit for efficiently performing shadow logic testing and memory block testing within a semiconductor integrated circuit. The integrated circuit includes a memory block for storing data. A shadow logic circuit is coupled to the memory block for interfacing the memory block with external circuitry. The shadow logic provides inputs to the memory block and receives data outputs from the memory block. A test collar is coupled between the memory block and the shadow logic. The test collar receives the data inputs from the shadow logic and receives the data outputs from the memory block. The test collar is configured to both provide test inputs to the shadow logic and capture test outputs from the shadow logic independent of the memory block. The test collar is also adapted to both provide tests inputs to the memory block and capture test outputs from the memory block independent of the shadow logic.

Abstract: A circuit for efficiently performing shadow logic testing and memory block testing within a semiconductor integrated circuit. The integrated circuit includes a memory block for storing data. A shadow logic circuit is coupled to the memory block for interfacing the memory block with external circuitry. The shadow logic provides inputs to the memory block and receives data outputs from the memory block. A test collar is coupled between the memory block and the shadow logic. The test collar receives the data inputs from the shadow logic and receives the data outputs from the memory block. The test collar is configured to both provide test inputs to the shadow logic and capture test outputs from the shadow logic independent of the memory block. The test collar is also adapted to both provide tests inputs to the memory block and capture test outputs from the memory block independent of the shadow logic.