[klee-dev] How to show (get) the negative values returned in klee's .pc files?

[Daniel Liew]

Hi,

When sending replies please CC the mailing list.

klee_assume() is a "special function" that adds constraints to the
current path being explored (if the resulting path is possible). See
SpecialFunctionHandler::handleAssume() in SpecialFunctionHandler.cpp

SpecialFunctionHandler::handleAssume() receives std::vector<ref<Expr>
> &arguments as one of its arguments which will be an expression tree
describing the constraint that was passed into klee_assume().

klee_assert() is a macro defined in klee.h that will call the special
function __assert_fail() if the assertion is false. This special
function is handled by SpecialFunctionHandler::handleAssertFail() this
will terminate the path currently being explored.

On 1 May 2013 20:29, General Email <general_mail2011 at yahoo.com> wrote:
> Thanks Daniel,
>
> This is very helpful.
> So, my next question is how does klee generate such a condition based on the
> following set of commands?!
>
> klee_assume(!(0<(x+5)));
> klee_assume(y==x+7);
> klee_assert(y<0);
>
> I would appreciate if you provide me with some guidance of how klee_assume
> and klee_assert work.
> Again thank you so much for your help.
> ________________________________
> From: Daniel Liew <daniel.liew at imperial.ac.uk>
> To: General Email <general_mail2011 at yahoo.com>
> Cc: klee-dev <klee-dev at imperial.ac.uk>
> Sent: Wednesday, May 1, 2013 1:01 PM
> Subject: Re: [klee-dev] How to show (get) the negative values returned in
> klee's .pc files?
>
> The path constraint is in the KQuery language ( see
> http://klee.llvm.org/KQuery.html#ReadLSB_expr ). The stuff in the
> square brackets are the constraints which can be understood to mean
>
> 2880154532 == (array X concatenated together). X was simply an integer
> in your case so this concatenation is just concatenating 4bytes
> together the integer.
>
> Decimal constants (such as 2880154532 ) used in this way as far as I
> know represent the unsigned interpretation of a bitvector.
>
> Assuming that a two-complement representation for signed integers you
> can use gdb to quickly calculate what the decimal representation of
> the signed integer x is.
>
> $ gdb
> (gdb) print /t 2880154532
> $1 = 10101011101010111010101110100100
> # The most significant bit is 1 so we know the number is negative. We
> can invert the bits and add 1 to find the absolute value ( see
> http://en.wikipedia.org/wiki/Two%27s_complement#The_most_negative_number
> )
> (gdb) print ~(2880154532) + 1
> $2 = 1414812764
> (gdb) exit
>
> So this tells you that the decimal value of integer x is -1414812764
> in the constraints you showed.
>
> Hope that helps.
>
> On 1 May 2013 17:16, General Email <general_mail2011 at yahoo.com> wrote:
>> Hi,
>>
>> I need to understand how to use klee_assume and klee_assert.
>> I tried to implement the following assumptions (in the function listed
>> below) which assumed that if a symbolic variable x satisfies the condition
>> !(0<(x+5)) and that if another variable y is set to x+7, I want to check
>> whether y is < 0 or not.
>> ------------------------------------------------------------------
>> void main()
>> {
>>  int x,y;
>>  klee_make_symbolic(&x, sizeof(x), "x");
>>
>>  klee_assume(!(0<(x+5)));
>>  klee_assume(y==x+7);
>>  klee_assert(y<0);
>> }
>> ------------------------------------------------------------------
>> The result from klee showed that the assersion is satisfiable based on the
>> following path constraint which I couldn't understand.
>>
>> array x[4] : w32 -> w8 = symbolic
>> (query [(Eq 2880154532
>>            (ReadLSB w32 0 x))]
>>        false)
>>
>> Also how to get the equivelant negative value of the number 2880154532?
>>
>> Would you please advise?
>> Thanks
>
>