Aspect-Oriented Programming with the E Verification Language: A Pragmatic Guide for Testbench Developers

265 Pages·2007·3.27 MB·English

by  David Robinson

Aspect-Oriented
Programming with
the e Verifi cation
Language
A Pragmatic Guide for
Testbench Developers
David Robinson
AMSTERDAM (cid:129) BOSTON (cid:129) HEIDELBERG (cid:129) LONDON
NEW YORK (cid:129) OXFORD (cid:129) PARIS (cid:129) SAN DIEGO
SAN FRANCISCO (cid:129) SINGAPORE (cid:129) SYDNEY (cid:129) TOKYO
Morgan Kaufmann is an imprint of Elsevier
PPrreelliimmss--PP337744221100..iinndddd ii 77//1122//0077 66::0066::3399 PPMM
Publishing Director: Joanne Tracy
Senior Acquisitions Editor: Chuck Glaser
Publishing Services Manager: George Morrison
Project Manager: Mónica González de Mendoza
Assistant Editor: Michele Cronin
Production Assistant: Lianne Hong
Composition: Charon Tec Ltd (A Macmillan Company)
Interior printer: Maple-Vail Book Manufacturing Group
Cover printer: Phoenix Color Corporation
Morgan Kaufmann Publishers is an imprint of Elsevier.
30 Corporate Drive, Suite 400, Burlington, MA 01803, USA
This book is printed on acid-free paper.
© 2007 by Elsevier, Inc. All rights reserved.
Designations used by companies to distinguish their products are often claimed as trademarks or registered
trademarks. In all instances in which Morgan Kaufmann Publishers is aware of a claim, the product names
appear in initial capital or all capital letters. Readers, however, should contact the appropriate companies for
more complete information regarding trademarks and registration.
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or
by any means — electronic, mechanical, photocopying, scanning, or otherwise — without prior written
permission of the publisher.
Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford,
UK: phone: ((cid:2)44) 1865 843830, fax: ((cid:2)44) 1865 853333, E-mail: [email protected].
You may also complete your request online via the Elsevier homepage (http://elsevier.com), by selecting
“Support & Contact” then “Copyright and Permission” and then “Obtaining Permissions.”
Library of Congress Cataloging-in-Publication Data
Robinson, David.
Aspect-oriented programming with the e verifi cation language : a pragmatic guide for testbench developers / David
Robinson.
p. cm.
Includes bibliographical references and index.
ISBN-13: 978-0-12-374210-0 (pbk. : alk. paper) 1. Object-oriented programming (Computer science) I. Title.
QA76.64.R629 2007
005.1(cid:3)17 — dc22
2007019636
ISBN: 978-0-12-374210-0
For information on all Morgan Kaufmann publications, visit
our Web site at www.mkp.com or www.books.elsevier.com
07 08 09 10 11 12 10 9 8 7 6 5 4 3 2 1
Printed in the United States of America
PPrreelliimmss--PP337744221100..iinndddd iiii 77//1122//0077 66::0066::4411 PPMM
Acknowledgments
I’d like to thank Ken Gray for making the innocuous com-
ment that started the whole thing going in the fi rst place.
I’m fairly sure he meant it, but I sometimes can’t help but
feel that I’ve fallen for a large practical joke.
Keeping with the theme of innocuous comments for a
moment, I’d like to thank Tommy Kelly for his very detailed
review comments. During a phone conversation about some
of these he just happened to mention that he had no idea
what I was talking about by the end of the Introduction.
That throw away line started one of the biggest rewrites the
book ever had.
I’d like to thank Jason Sprott for stress testing my OOP
arguments. As something of an OOP, Vera, SystemVerilog,
testbench, and verifi cation guru, having him admit that
there might be a slight possibility that maybe AOP had some
kind of an advantage over OOP on its own was a defi ning
moment. I also have to thank Jason and Tommy for creating
Verilab and inviting me in to play.
That leads me on nicely to thanking all the people (past and
present) in Verilab for making it such a great place to work.
The technical knowledge fl oating around in there is noth-
ing short of amazing, and everyone has contributed to the
making of this book. In particular, I’d like to thank Robert
Fairlie, J.L. Gray, and Gordon McGregor for their detailed
reviews.
Outside of Verilab, Suzanne Peggie, Claudia Blank, Mike
Stellfox, Ernst Zwingenberger, Bernhard Klein, Robert Richter,
PPrreelliimmss--PP337744221100..iinndddd iiiiii 77//1122//0077 66::0066::4411 PPMM
iv Acknowledgments
Christian Beckmann, John MacBeth, Gery Osowiecki, Andrew
Piziali, and Thorsten Lutscher all deserve a special mention
for helping make the book what it is — whether they realized
it or not.
Finally, I’d like to thank Chuck Glaser, Monica Mendoza,
Michele Cronin, and Dennis Schaefer at Morgan Kaufmann,
and all the good folks at Charon Tec Ltd for actually making
the book.
PPrreelliimmss--PP337744221100..iinndddd iivv 77//1122//0077 66::0066::4411 PPMM
Table of Contents
Foreword ix
Preface xi
About Verilab xix
1. Introduction to Aspect Oriented Programming
(AOP) 1
1.1. What are aspects? — Part I 2
1.2. Why do I need aspects? What’s wrong with
crosscutting concerns? 9
1.3. Surely OOP doesn’t have any problems? 11
1.4. Why does AOP help? 18
1.5. Theory vs real life — What else is AOP
good for? 22
1.6. What are aspects? — Part II 27
2. AOP in e 31
2.1. How do I extend a class? 36
2.2. How do I extend a class for multiple values
of a determinant? 41
2.3. How do I extend a type? 42
2.4. How do I introduce a new noncoverage
member to a class? 43
2.5. How do I introduce a coverage group to
a class? 44
2.6. How do I extend a coverage group? 46
2.7. H ow do I change the behavior of a
method? 50
2.8. How do I limit the scope of my extensions? 53
2.9. Using return in method advice 59
2.10. Controlling the order of method
extension calls 65
PPrreelliimmss--PP337744221100..iinndddd vv 77//1122//0077 66::0066::4422 PPMM
vi Table of Contents
3. Using AOP to Organize Your Code 67
3.1. A word about style 69
3.2. What aspects do I want to use? 70
3.3. Mapping aspects to fi les 75
4. Creating Flexible Code 93
5. Creating Pluggable Code 113
5.1. The extendable case statement 120
5.2. The factory pattern 132
6. Improving Your Productivity 145
6.1. Shifting the power 146
6.2. Dealing with broken code 147
6.3. Handling workarounds 149
6.4. Reducing and deferring class complexity 150
6.5. Adding problem-specifi c functionality 153
6.6. Reducing the OOP-induced overhead 155
7. AOP in Action 163
7.1. Creating a class with a selectable
algorithm 164
7.2. Creating a confi guration interface for
an eVC 173
7.3. Using aspects to create a layered
verifi cation environment 180
7.4. Creating reusable layered sequences 188
7.5. Testing your verifi cation environment 196
7.6. Debugging using AOP 204
7.7. Encapsulating tests 210
8. Analysing e Code 215
8.1. The e toolkit 217
8.2. Finding class declarations and extensions 221
8.3. Finding the class inheritance hierarchy 224
8.4. Finding the determinants used by a class 224
PPrreelliimmss--PP337744221100..iinndddd vvii 77//1122//0077 66::0066::4422 PPMM
Table of Contents vii
8.5. Finding method declarations and
extensions 225
8.6. Finding fi eld declarations 226
8.7. Finding event declarations 227
8.8. Finding enumerated type declarations and
extensions 227
8.9. How do I fi nd where a value is added to
a type? 228
8.10. Finding cover group declarations and
extensions 229
8.11. Finding the source of a message in the
log fi le 230
8.12. Finding aspects 231
Bibliography 239
Epilogue 241
Index 243
PPrreelliimmss--PP337744221100..iinndddd vviiii 77//1122//0077 66::0066::4422 PPMM
This page intentionally left blank
Foreword
David Robinson has made a valuable contribution to the
verifi cation community with this book about Aspect-
Oriented Programming (AOP) using e. Coverage-driven veri-
fi cation has entered the mainstream as a critical and central
component of an effective verifi cation methodology, and this
has driven engineers to seek out the best practices for devel-
oping automated verifi cation environments. There are many
compelling reasons why people choose Specman Elite and
the e language to tackle their verifi cation problems — the
declarative nature of the language which makes it easy to
express stimulus constraints, temporal assertions, and func-
tional coverage; the large library of predefi ned routines which
make it easy to create sophisticated checkers and other test-
bench components; the fl exibility provided by the extensibil-
ity of the language; or simply the fact that they fi nd it to be an
incredibly effi cient way to get their job done. I have spent the
last 8 years working with design and verifi cation engineers to
help them create constrained-random, coverage-driven veri-
fi cation environments using the e verifi cation language. They
don’t usually start out saying, “I could really use an AOP lan-
guage to make my verifi cation work a lot easier…” In fact,
most Specman experts and beginners who leverage the power
of AOP in their everyday work to create and use e verifi ca-
tion environments do so without much thought of exactly
what AOP is all about. I, and other verifi cation experts, have
tried for many years to convey in simple terms the power of
e as an AOP language to signifi cantly simplify and accelerate
the development of reusable, automated verifi cation envi-
ronments, but we have not succeeded in a way that matches
PPrreelliimmss--PP337744221100..iinndddd iixx 77//1122//0077 66::0066::4422 PPMM