When functional tests are used for manufacturing testing, their quality for detecting manufacturing defects needs to be evaluated. Evaluating functional tests using a traditional gate-level fault simulation environment has several disadvantages. To alleviate them, we describe a functional level coverage metric for estimating gate-level fault coverage that has a high degree of correlation to gate-level coverage. We borrow concepts from simulation-based design verification by defining fault detection conditions as coverage objects and monitoring their occurrence, also called their hit counts, during RTL simulation. To reduce the simulation overhead, we abstract gate-level fault detection conditions to the architectural level. The resulting hit counts are converted to an estimated fault coverage using a formula. Experimental results are presented on three datapath modules in a high-performance microprocessor considering two applications: identification of functional tests with high gate-level coverage relative to other tests, and identification of modules or module functions that require additional functional tests.
As device geometries scale, product complexity has increased with more and more functionality embedded into integrated chips in recent times. In the processor domain, multiple cores with associated glue logic and cache all on a single die are becoming more and more popular. At the same time, product frequencies have gone up and the need to test for delay defects and marginal circuits is rising continually. This is exacerbated in recent times with the focus on lowpower design. While there has been a lot of progress in scan based delay test, the reliance on functional tests has continued. The biggest concern with scan test effectiveness is related to screening small delay defects. Gross delay defects can be tested using a good set of scan transition fault tests (scan AC tests). However questions remain with respect to the effectiveness of scan tests to screen small delay defects and also to test marginality related failures: for example, speed failures caused by issues like cross-capacitance, power droop etc. Functional tests are being used today for getting the last few DPM to reach quality goals. The reliance on functional tests is higher in products that push the process/design envelope to reach performance/power goals.The continued reliance on functional test for reaching quality goals is driving more efforts to be focused on high level test generation in recent times. As devices have evolved into SOC products with embedded cores, this has opened up new opportunities for core based functional test application. In the microprocessor domain, large caches have been integrated on the same die for some time now. The availability of large onchip cache has been taken advantage of to run functional tests that are stored in the cache. Additionally there have been recent efforts where we have employed the logic of the microprocessor to generate tests on the fly. Essentially we have been able to write intelligent software that resides in the chip, and during test application time, generates and applies interesting functional test sequences, thereby accomplishing a functional BIST technology.As SOC functionality increases, there is a big opportunity to configure the various modules in the die in such a way that the entire device can be tested in functional mode. Such corebased software tests can go a long way to enable at-speed functional test. Surprisingly what is needed to enable this is minimal DFT that configures the peripheral modules on the SOC in special DFT modes. Such a SW based functional test technology is essential when it comes to testing the interfaces between the various modules in SOCs. Even if core level testing is enabled with scan and other direct access test methods, inter-module interactions need functional content application to uncover delay defects and marginal circuit behavior in such interfaces.Finally the effectiveness of the scheme is measured by the fault coverage achieved by such functional tests. While methods to enable random tests to hit targeted module interactions can be devised easily, a...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.