HAL : hal-00157268, version 1

DOI : 10.1007/s10009-003-0120-y

Fiche détaillée

Récupérer au format BibTeX - EndNote - TEI - RefWorks -

International Journal on Software Tools for Technology Transfer 5, 2-3 (2003) 237-246

Properties of two's complement floating point notations

Sylvie Boldo 1, Marc Daumas 1

(2003)

Few designs, mostly those of Texas Instruments, continue to use tworsquos complement floating point units. Such units are simpler to build and to validate, but they do not comply to the dominant IEEE standard for floating point arithmetic. We compare some properties of the two systems in this text. Some features are lost, but others remain unchanged. One strong example is the case of Sterbenzrsquos theorem and our recent extension. We show in the paper that the theorem and its extension hold for the tworsquos complement architecture. Still, users should ensure that results are large enough on circuits that do not implement gradual underflow. Theorems have been proven and validated using the Coq automatic proof checker.

1 :	Laboratoire de l'Informatique du Parallélisme (LIP)
	Université de Lyon – CNRS : UMR5668 – INRIA – École Normale Supérieure - Lyon – Université Claude Bernard - Lyon I

Domaine

:

Informatique/Arithmétique des ordinateurs

Mots Clés : Digital signal processing - Avionics - Formal proof

hal-00157268, version 1

http://hal.archives-ouvertes.fr/hal-00157268

oai:hal.archives-ouvertes.fr:hal-00157268

Contributeur : Marc Daumas <>

Soumis le : Lundi 25 Juin 2007, 16:33:36

Dernière modification le : Lundi 25 Juin 2007, 16:38:07