Download Formal Methods and Software Engineering: 17th International by Michael Butler, Sylvain Conchon, Fatiha Zaïdi PDF

By Michael Butler, Sylvain Conchon, Fatiha Zaïdi

This booklet constitutes the refereed court cases of the seventeenth overseas convention on Formal Engineering equipment, ICFEM 2015, held in Paris, France, in November 2015. The 27 revised complete papers provided have been conscientiously reviewed and chosen from eighty two submissions. The papers disguise quite a lot of subject matters within the quarter of formal equipment and software program engineering and are dedicated to advancing the cutting-edge of utilising formal tools in perform. They concentration specifically on combos of conceptual and methodological points with their formal origin and power support.

Show description

Read or Download Formal Methods and Software Engineering: 17th International Conference on Formal Engineering Methods, ICFEM 2015, Paris, France, November 3-5, 2015, Proceedings PDF

Best compilers books

Ant in Action

This moment version of a Manning bestseller has been revised and re-titled to slot the 'In motion' sequence through Steve Loughran, an Ant undertaking committer. "Ant in motion" introduces Ant and the way to take advantage of it for test-driven Java program improvement. Ant itself is relocating to v1. 7, an immense revision, on the finish of 2006 so the timing for the booklet is true.

Modern compiler design

A compiler is likely one of the most important components of a desktops method software program, translating courses written in a high-level language into low-level instructions that the laptop can comprehend and execute. so much compiler layout books concentration in simple terms on strategies for relevant (or procedural) languages like C or Pascal, while sleek Compiler layout additionally covers compilation innovations for object-oriented, sensible, good judgment and allotted languages.

COBOL: From Micro to Mainframe (3rd Edition)

For classes in Cobol Programming. This variation is designed to fulfill your whole COBOL wishes - on a number of structures. The textbook covers all uncomplicated COBOL parts, with extra chapters at the 12 months 2000 challenge, dependent programming and layout, debugging, subprograms, desk processing, sorting, display I/O, sequential dossier upkeep, listed documents, and object-oriented COBOL.

Additional info for Formal Methods and Software Engineering: 17th International Conference on Formal Engineering Methods, ICFEM 2015, Paris, France, November 3-5, 2015, Proceedings

Sample text

00 Proofs of behavior are quite simple as they are calls to the previously studied average1 and average2 functions. The difficult part, as expected, is overflow. 30 S. Boldo It is handled automatically by Gappa, except the proof that x + y does not overflow, provided that |x| < C ≤ 2970 . More precisely, even if y is the biggest floating-point number, if |x| < 2970 , then x ⊕ y will not overflow as it will round to y. 6 Conclusion and Perspectives The initial goal was to prove a program computing the average without overflow.

The fact that average1(x, y) is between min(x, y) and max(x, y) is slightly more difficult as rounding is involved. The implies = 0 implies average1(x, y) = 0 and that 2Ei ≤ x+y facts that x+y 2 2 average1(x, y) = 0 are also quite simple from basic floating-point properties of the rounding. The rounding error here is very small as it is equivalent to only one rounding: average1(x, y) − x+y 1 ≤ ulp 2 2 x+y 2 . Formal Verification of Programs Computing the Floating-Point Average 23 An interesting point is the fact that this algorithm requires x and y to be of different signs in order to not overflow.

It means we want to have the tightest precondition, in order to cover all cases that do not fail. And this requires additional work. Unfortunately, programs are often not carefully written with overflow in mind. There are overflowing examples in an overwhelming proportion of them. Our work is therefore either to give precise conditions for them to work correctly, or to rewrite them. Acknowledgments. The author is indebted to a referee of a previous version of this work, who rightfully pitied the fact that no program existed for a correctly-rounded average and pointed out the previously dismissed average2 function.

Download PDF sample

Rated 4.70 of 5 – based on 17 votes