diff --git a/src/plugins/wp/Changelog b/src/plugins/wp/Changelog index 405600b3263fe6ba89283519d2f9b81372e9c413..844b5b50097f8f55d30a604eb2ec14cd13cf1498 100644 --- a/src/plugins/wp/Changelog +++ b/src/plugins/wp/Changelog @@ -20,6 +20,8 @@ # <Prover>: prover ############################################################################### +- Wp [2019/01/28] New floating-point model + ###################### Plugin WP 18.0 (Argon) ###################### diff --git a/src/plugins/wp/doc/manual/wp_intro.tex b/src/plugins/wp/doc/manual/wp_intro.tex index da74357a7b980d3af92104006171c1c51d367aa9..1e3c5e076a67cc3e3d8a823eceda35a8c64bcd7d 100644 --- a/src/plugins/wp/doc/manual/wp_intro.tex +++ b/src/plugins/wp/doc/manual/wp_intro.tex @@ -340,12 +340,15 @@ For tackling this complexity, the \textsf{WP} plug-in relies on several or \texttt{-wp-rte} will generate a warning if some annotation might be not generated. -\item[Float Model:] floating-point operations are \emph{defined} to be - the mathematical ones \emph{with} a rounding operation. This is fully - consistent with the \textsf{IEEE} semantics. Most automated provers - are not able to discharge the generated proof obligations. Special - support for the \textsf{Gappa} theorem prover is available through - \textsf{Why3}. +\item[Float Model:] floating-point values are represent in a special + theory with dedicated operations over \texttt{float} and \texttt{double} + values and conversion from and to their \texttt{real} representation \emph{via} + rounding, as defined by the \textsc{C/ACSL} semantics. + + Although correct with respect to the \textsc{IEEE} specifications, this + model still provides very little support for proving properties with automated + provers. You may add additional properties using \emph{drivers} + as explained later. \item[Real Model:] floating-point operations are \emph{transformed} on reals, with \emph{no} rounding. This is completely unsound with