By Tony Hoare (auth.), Peter Müller (eds.)

ISBN-10: 3642130097

ISBN-13: 9783642130090

Software defects result in huge, immense bills for the software program and society as a complete. whereas checking out turns out to be useful to discover insects, it's inadequate to teach the absence of sure varieties of error or application satisfies its specification. Such excessive degrees of software program caliber should be completed by means of software program verification, that's, by means of proving the correctness of a application with recognize to its specification. software program verification has obvious super development over the last decade; it remains to be an lively learn subject and is now additionally changing into more and more well known between practitioners. This instructional includes chosen papers from the LASER summer season colleges 2007 and 2008, either one of which serious about correctness - utilized software program Verification in 2007 and Concurrency and Correctness in 2008. subject matters lined contain verification of fine-grain concurrency and transactions, the inside track version for concurrent object-oriented programming, the Spec# programming and verification approach, verification within the prototype verification approach PVS, and multi-core chip design.

Show description

Read or Download Advanced Lectures on Software Engineering: LASER Summer School 2007/2008 PDF

Best computers books

's IN-350 Планирование и внедрение LDAP PDF

Данный курс предназначен для подготовки слушателей к использованию LDAP в качестве сервера идентификации. В курсе рассматриваются вопросы настройки, установки, структурирования и конфигурирования сервера разрешения имен с использованием сервиса LDAP сервера sunlight ONE (Open community setting) в операционной системе Solaris nine.

Additional resources for Advanced Lectures on Software Engineering: LASER Summer School 2007/2008

Example text

Section 5 will show how this can be different. In the following we will introduce an abstraction under the assumption that every processor keeps its call stack lock. In this abstraction we call the request queue lock on a processor p simply the lock on p. As long as the call stack lock on a processor p is in possession of p, a request queue lock on p in possession of a processor q means that processor p will be executing new feature requests in the request queue exclusively on behalf of q. This means that a request queue lock grants exclusive access to all the objects handled by p.

Of course, execution of the region also requires the normal sequential token of the thread that contains it. 3) as a sort of inverse of concurrent composition, with a fan-in at the beginning and a fan-out at the end. For simplicity, we assume that there is only a single shared resource, consisting of everything except the private resources of the currently active individual threads. In most practical applications, many separate resources will need to be declared, but we shall not deal with that complexity here.

In that case, the definition of a nested transaction becomes equivalent to that of an un-nested one. When the whole child transaction has finished, the completions accumulated by the child transactions are triggered. 2 by the transverse arrow from the finally exit of the child transactions T to the new finally entry of the child transactions themselves. It executes the completion actions of all the children, in the same order as the execution of forward actions of the children. 34 T. Hoare Another benefit of the introduction of completion actions is to implement the lazy update design pattern.

Download PDF sample

Advanced Lectures on Software Engineering: LASER Summer School 2007/2008 by Tony Hoare (auth.), Peter Müller (eds.)

by Michael

Rated 4.72 of 5 – based on 15 votes