Frama-C-discuss mailing list archives

This page gathers the archives of the old Frama-C-discuss archives, that was hosted by Inria's gforge before its demise at the end of 2020. To search for mails newer than September 2020, please visit the page of the new mailing list on Renater.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Frama-c-discuss] hint assertions and understanding cooperation between wp and value plugin

Hi,

Le 06/05/2016 11:52, Boris Yakobowski a écrit :
> ACSL is a specification language with a two-valued logic (see Section
> 2.2.2). So predicates can definitely be valid or invalid.

Sure

> Here, Maurice's assert is equivalent to the lemma
> 
> \forall double x; -1 <= x <= 1 ==> x*x - .2 * x + 0.01 >= 0.;

Wrong! because in logic formulas, those + , - , * are interpreted as
operations on real numbers, not floats. Similarly, 0.01 and .2 denote
real numbers and not their approximation as floats.

If one really want to state a lemma equivalently to the code

double t = x*x - .2 * x + 0.01;
//@ assert t >= 0.;

then one must write

//@ lemma l: \forall double x; -1 <= x <= 1 ==>
\round(\nearest_even,
 \round(\nearest_even,
  \round(\nearest_even,x*x) -
  \round(\nearest_even, \round(\nearest_even,.2) * x))
  + \round(\nearest_even,0.01) >= 0.0;

(I'm probably wrong with the syntax, but you got the idea)

I indeed assumed the rounding mode was nearest_even, but of course I
don't really know and if one wants to check this kind of property, it is
important to think about what is the rounding mode

> This lemma is invalid because it is refuted by the double value
> 0x1.9999998p-4. (i.e. 1.5999999940395355*2^-4 or approximately
> 0.0999999996275). See my message on May 1 for more floating-point gory
> details :-)

OK (I did not check but I trust you), a counterexample to the second
variant of the lemma of course, not the first, which is valid.


> On Thu, May 5, 2016 at 3:31 AM, David R. Cok <dcok at grammatech.com
> <mailto:dcok at grammatech.com>> wrote:
> 
>     Sorry: when you say "invalid in ACSL" do you actually mean it is
>     invalid (because it is floating point, not real?) or do you mean it
>     is not provable (and by what) or do you mean it is not a well-formed
>     ACSL statement? ACSL is just a language, not a proof tool. If the
>     first, is it actually the case that there are double precision
>     numbers for which the assertion does not hold?
> 
>     - David
> 
> 
>     On 5/4/2016 2:14 PM, Boris Yakobowski wrote:
>>     My mistake. I coalesced two different things when replying to
>>     Maurice :
>>
>>     - the validity of the lemma/*@ lemma sq_double: \forall real x;
>>     x*x - .2 * x + 0.01 >= 0.; */. It is indeed valid in ACSL, should
>>     be proven by WP whatever the model (since, as Loïc mentioned, it
>>     involves only real) , and (as all lemmas) is completely ignored by
>>     Value;
>>
>>     - the validity status of the assert in double x1 = x*x - .2 * x +
>>     0.01; /*@ assert x1 >= 0; */, which is invalid in ACSL, but can be
>>     proven by the Float model of WP.
>>
>>     On Wed, May 4, 2016 at 2:07 PM, Claude Marché
>>     <Claude.Marche at inria.fr <mailto:Claude.Marche at inria.fr>> wrote:
>>
>>
>>         Sorry to interfere, but I don't understand the meaning of
>>         "wrong in WP"
>>
>>         According to ACSL manual, this lemma is a statement expressed
>>         purely in
>>         mathematical real arithmetic, and as such it is valid. It is
>>         indeed
>>         proved automatically by Z3 4.4.1.
>>
>>         If you want to state a similar property talking about
>>         floating-point
>>         arithmetic, it should be stated differently, typically using a
>>         program
>>
>>         void f(double x) {
>>           double y = x*x - .2 * x + 0.01;
>>           //@ assert y >= 0.0;
>>         }
>>
>>         But I guess it probably wrong because of rounding, even with a
>>         precondition like \abs(x) <= 1.0
>>
>>         My two cents,
>>
>>         - Claude
>>
>>         Le 04/05/2016 13:54, Loïc Correnson a écrit :
>>         >> /*@ lemma sq_double: \forall real x; x*x - .2 * x + 0.01 >=
>>         0.; */
>>         >
>>         > This lemma is definitely wrong in WP with Real model (not
>>         float there).
>>         > At least, it is not provable in the forthcoming release of
>>         Frama-C.
>>         > Is there a bug in some existing release?
>>         >       L.
>>         >
>>         >
>>         > _______________________________________________
>>         > Frama-c-discuss mailing list
>>         > Frama-c-discuss at lists.gforge.inria.fr
>>         <mailto:Frama-c-discuss at lists.gforge.inria.fr>
>>         > http://lists.gforge.inria.fr/mailman/listinfo/frama-c-discuss
>>         >
>>
>>         --
>>         Claude Marché                          | tel: +33 1 69 15 66
>>         08 <tel:%2B33%201%2069%2015%2066%2008>
>>         INRIA Saclay - ÃŽle-de-France           |
>>         Université Paris-sud, Bat. 650         |
>>         <http://www.lri.fr/%7Emarche/%0AF-91405>http://www.lri.fr/~marche/
>>         F-91405 ORSAY Cedex                    |
>>         _______________________________________________
>>         Frama-c-discuss mailing list
>>         Frama-c-discuss at lists.gforge.inria.fr
>>         <mailto:Frama-c-discuss at lists.gforge.inria.fr>
>>         http://lists.gforge.inria.fr/mailman/listinfo/frama-c-discuss
>>
>>
>>
>>
>>     -- 
>>     Boris
>>
>>
>>     _______________________________________________
>>     Frama-c-discuss mailing list
>>     Frama-c-discuss at lists.gforge.inria.fr
>>     <mailto:Frama-c-discuss at lists.gforge.inria.fr>
>>     http://lists.gforge.inria.fr/mailman/listinfo/frama-c-discuss
> 
> 
>     _______________________________________________
>     Frama-c-discuss mailing list
>     Frama-c-discuss at lists.gforge.inria.fr
>     <mailto:Frama-c-discuss at lists.gforge.inria.fr>
>     http://lists.gforge.inria.fr/mailman/listinfo/frama-c-discuss
> 
> 
> 
> 
> -- 
> Boris
> 
> 
> _______________________________________________
> Frama-c-discuss mailing list
> Frama-c-discuss at lists.gforge.inria.fr
> http://lists.gforge.inria.fr/mailman/listinfo/frama-c-discuss
> 

-- 
Claude Marché                          | tel: +33 1 69 15 66 08
INRIA Saclay - ÃŽle-de-France           |
Université Paris-sud, Bat. 650         | http://www.lri.fr/~marche/
F-91405 ORSAY Cedex                    |