Silicon Image Gmbh

Other

Agilent Technologies, Inc.

Lattice Semiconductor Corporation

Chinsong Sul

Gijung Ahn

Sungjoon Kim

Heon Kim

Hyukyong Kwon

Andy Ng

Son Truong Nguyen

Hoon Choi

Min-Kyu Kim

Fidel Muradali

Neal C Jaarsma

Garrett O'Brien

The translation based test architecture translates inputs, test control signals, and the chip pin IOs. Translation of test control signals derives dedicated local test control signals for each individual circuit-under-test (CUT) can introduce programmability directly into the test pattern transformation and composition. Using the local test control signals realizes a diversified test functions in each individual CUT without increasing test resource requirement. The translation of IO enable signals of chips can be used to create test scenarios in multi-chip module systems. Transformation of the generated test patterns to derive new test patterns occurs by retranslation of the signals within the generated test patterns. The retranslation can be reassigning test data of the translation layer cells in the generated test patterns and adjustment of corresponding difference in the test patterns. To achieve retranslation, the translation layer cells are identified in the test patterns.

Low cost design for test architecture

The Translation Layer is embedded into each circuit under test (CUT) to modularize test process. The modularized tests are self-contained and performed in isolation. They are composed without consideration of environment constraints. The CUT and its environment constraints can be concurrently be tested in isolation and independently. Interconnections between the CUT and the environment can be tested in the environment constraint test without additional dedicated test logic. The modularized test process allows the test patterns of the environment constraints to be derived from those of the CUT. The resulting test patterns are used to compose the test patterns of a target system. Since the test process is recursive in nature, the modularized test of each constituent subsystem or design core can be performed in isolation in the target system, while the environment constraints and the interconnections are being tested concurrently.

A Design-for-testability method based on composition of test patterns copes with increasing test complexity and cost metric of a large system. System-level structural test patterns from test patterns of constituent subsystems, cores and design IPs are constructed without requiring their design netlists. The delivered test patterns can be utilized 100% in the testing of system. The system-level test pattern is delivered to the device under test, the subsystem test patterns can be scheduled and applied continuously without being interleaved by test deliveries until all of the subsystem test patterns are exercised. Absence of design netlist requirement allows uniform integration of external and internal IPs regardless of availability of test isolation logic or design details. Concurrent test of constituents and their mutual independence in scan operations allows implicit distribution of test protocol signals such as scan enable (SE) and the scan clocks. The method enables at-speed testing of memory shadow logic.

A random number generator and method for testing the same are described. In one embodiment, the random number generator comprises one or more ring oscillator structures, each of the one or more ring oscillator structures having a ring oscillator for use in generating random numbers and having a test structure to reconfigure the ring oscillator into a testable structure.

Test solution for a random number generator

A method and apparatus for a computer memory test structure. An embodiment of a method for testing of a memory board includes testing a memory of the memory board, where testing the memory including use of a built-in self-test structure to provide a first test pattern for the memory. The method further includes testing an IO (input output) interface of the memory with a host, where testing of the IO interface includes use of the built-in self-test structure to provide a second test pattern for the IO interface.

Computer memory test structure

A method and apparatus is disclosed herein for testing of multiple ring oscillators. In one embodiment, the apparatus comprises at least one ring oscillator structure having a ring oscillator having an inverter chain with an odd number of inverters connected back-to-back and operable to produce an oscillatory output, and a test structure coupled to provide either an observability chain input or a test input to the ring oscillator and to receive the oscillatory output as a feedback from the ring oscillator.

Test solution for ring oscillators

A method and apparatus for multi-site testing of computer memory devices. An embodiment of a method of testing computer memory devices includes coupling multiple memory devices, each memory device having a serializer output and a deserializer input, wherein the serializer output of a first memory device is coupled with a deserializer input of one or more of the memory devices of the plurality of memory devices. The method further includes producing test signal patterns using a test generator of each memory device, serializing the test signal pattern at each memory device, and transmitting the serialized test pattern for testing of the memory devices, wherein testing of the memory devices includes a first test mode and a second test mode.

Multi-site testing of computer memory devices and serial IO ports

Embodiments of the invention are generally directed to testing of high-speed input-output devices. An embodiment of a high-speed input-output apparatus includes a transmitter and a receiver, and a loop-back connection from an output of the transmitter to an input of the receiver, the loop-back connection including a first connector and a second connector for transmission of differential signals. The apparatus further includes a first inductor having a first terminal and a second terminal and second inductor having a first terminal and a second terminal, the first terminal of the first inductor being connected to the first connector and the first terminal of the second inductor being connected to the second connector, the second terminal of the first inductor and the second terminal of the second inductor providing a test access port for direct current testing of the apparatus.

Testing of high-speed input-output devices

Embodiments of the invention are generally directed to fault testing for interconnections. An embodiment of a fault analysis apparatus includes a test pattern source to provide a test pattern for an interconnection between a transmitter and a receiver, the interconnection having a transmitter end and a receiver end, the interconnection including a first wire and a second wire, the transmitter transmitting the test pattern on the first wire to the receiver. The apparatus further includes a first switch to open and close a first connection for the first wire, and a second switch to open and close a second connection for the second wire. The first switch and the second switch are to be set according to a configuration to set at least a portion of a test path for the detection of one or more faults in the interconnection.

Fault testing for interconnections

Method, device, and system for testing for errors in high-speed input/output systems. System and device may include a concurrent code checker for checking for errors in encoded data packets through data packets static properties and dynamic properties of the data stream including the packets. Method may involve detecting invalid encoded packets using the data packets static properties and the dynamic properties of the data stream including the packets. Method for optimizing a design of a concurrent code checker logic using don't-care conditions, and concurrent code checker circuit having reduce logic element and semiconductor area requirements.

Concurrent code checker and hardware efficient high-speed I/O having built-in self-test and debug features

Methods and computer readable media for performing scan-based testing of circuits using one or more test clock control structures are disclosed. In one embodiment, a method includes performing an intra-domain test to exercise a first subset of domains of the plurality of circuits implementing dynamic fault detection test patterns. It also includes performing an inter-domain test to exercise a second subset of domains of the plurality of circuits implementing dynamic fault detection test patterns. The dynamic fault detection test patterns can include, for example, last-shift-launch test patterns and broadside test patterns. In various embodiments, the method can include configuring different programmable test clock controllers to test different domains substantially in parallel.

Scan-based testing of devices implementing a test clock control structure ('TCCS')

In some embodiments, a chip includes first and second scan chain segments each including registers and multiplexers to provide to the registers scan input signals during scan input periods and captured output signals during a capture periods. The chip also includes circuitry to provide first and second test clock signals to the registers of the first and second scan chain segments, respectively, wherein the second test clock signal is provided by a different signal path in the circuitry during the scan input periods than during the capture periods, and during the scan input periods the second test clock signal is skewed with respect to the first test clock signal. Other embodiments are described and claimed.

Growing community of inventors

Mountain View, CA, United States of America

Chinsong Sul