Agnostic Validation Test Bench For Efuse Connectivity Verification

In semiconductor industry, validation is an important process to discover design bugs and have it fixed before the product is released. Semiconductor integrated circuit is normally refreshed in yearly cadence and it is crucial to have a short design and validation cycle, without compromising the pro...

Full description

Saved in:
Bibliographic Details
Main Author: CHAN , WEI JIAN
Format: Thesis
Language:English
Published: 2017
Subjects:
Online Access:http://eprints.usm.my/38350/1/CHAN_WEI_JIAN_24_Pages.pdf
http://eprints.usm.my/38350/
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Universiti Sains Malaysia
Language: English
Description
Summary:In semiconductor industry, validation is an important process to discover design bugs and have it fixed before the product is released. Semiconductor integrated circuit is normally refreshed in yearly cadence and it is crucial to have a short design and validation cycle, without compromising the product quality. Nowadays, validation process often becomes the bottleneck for product readiness. Integrated circuit validation flow has to be improved in order to keep up with the advancement of integrated circuit design flow. In this work, an improvement method on validation flow is discussed, with particular focus on eFUSE (Electric FUSE) connectivity validation. eFUSE is a feature available in integrated circuit which functions as a central storage for important ‘settings’, and distribute them during system boot up process. eFUSE connectivity validation is needed to ensure each intellectual property is able to retrieve the correct eFUSE value. In this work, the concept of agnostic validation test bench for eFUSE connectivity validation is developed and tested the idea of it is to eliminate manual test development effort, improves validation efficiency and promotes reusability across different projects. By using this methodology, eFUSE connectivity validation time is reduced significantly and recorded an improvement of 28%. There is also an average improvement of 65% in eFUSE coverage percentage. In summary, the eFUSE connectivity validation time frame is shortened, without compromising the test quality.