Software-based self-testing methodology for processor cores

Abstract

At-speed testing of gigahertz processors using external testers may not be technically and economically feasible. Hence, there is an emerging need for low-cost high-quality self-test methodologies that can be used by processors to test themselves at-speed. Currently, built-in self-test (BIST) is the primary self-test methodology available. While memory BIST is commonly used for testing embedded memory cores, complex logic designs such as microprocessors are rarely tested with logic BIST. In this paper, we first analyze the issues associated with current hardware-based logic-BIST methodologies by applying a commercial logic-BIST tool to two processor cores. We then propose a new software-based self-testing methodology for processors, which uses a software tester embedded in the processor memory as a vehicle for applying structural tests. The software tester consists of programs for test generation and test application. Prior to the test, structural tests are prepared for processor components in the form of self-test signatures. During the process of self-test, the test generation program expands the self-test signatures into test sets and the test application program applies the tests to the components under test at the speed of the processor. Application of the novel software-based self-test method demonstrates its significant cost/fault coverage benefits and its ability to apply at-speed test while alleviating the need for high-speed testers.

Structural Software-Based Self-Test of Network-on-Chip

“…For microprocessors, the benefits of structural testing and functional testing are combined by a so-called structural software-based self-test (SBST) [12][13][14][15][16]. Here, ATPG provides deterministic, structural test patterns which are transformed into arguments of a sequence of valid instructions.…”

Infrastructure for Testing Nodes of a Wireless Sensor Network

“…The first includes (Shen & Abraham, 1998;Batcher & Papachristou, 1999) that have a high level of abstraction and are functionally oriented. The second category includes (Kranitis et al, 2005, Chen & Dey, 2001Kranitis, Paschalis, Gizopoulos, & Zorian, 2002), which are structurally oriented and require structural fault-driven test development. From the first category, (Shen & Abraham, 1998) requires a lengthy test set, whereas the approach of (Batcher & Papachristou, 1999) is not purely a software-based method and requires some extra hardware.…”

“…• Testing while chip is running at full functional speed (at-speed testing) (Chen & Dey, 2001) • Generation of deterministic test vectors from SBST program code • Unique signature responses contained within SBST program code and compared by MCU (Chen & Dey, 2001) • No need for expensive automatic test equipment (ATE) • In-field testability • Energy efficiency This self-testing approach is considered an inexpensive way of achieving reliability, availability, and serviceability in a WSN. There are other compelling reasons for the use of SBST, such as certain tests that can determine if the hardware or software of a node has been tampered with by an intruder.…”

“…From the first category, (Shen & Abraham, 1998) requires a lengthy test set, whereas the approach of (Batcher & Papachristou, 1999) is not purely a software-based method and requires some extra hardware. In the second category, a gate level netlist is necessary in (Chen & Dey, 2001), which is difficult information to obtain for COTS chips because of intellectual property (IP) constraints. The SBST methodology in (Kranitis et al, 2002) is based upon the application of deterministic test patterns targeting structural faults of individual processor components.…”

Energy Efficient Software-Based Self-Test for Wireless Sensor Network Nodes

“…The first group includes functional SBST approaches [7], [8] employed at a high level of abstraction. The second category represents a structural SBST [9] [10], [11], which requires a fault-driven test development. Fault coverage of a functional fault model is usually low, even with numeous manual interventions, while the test set can be lengthy [7], [8].…”