- Add a Rational module that implements a field over rationals.
- Add rationals to Datatype
- Move field, nat, finite and linear into libraries/arithmetic
- Update linear_filter_test to use the Rational module
- Update oracles

Maxime Jacquemin authored 1 year ago and David Bühler committed 1 year ago

We can now represents filters that are not centered around zero.
The invariant computation now computes bounds for each state dimension.
The module is documented.

Maxime Jacquemin authored 1 year ago and David Bühler committed 1 year ago

Avoiding confusing between physical and structural equality.

Maxime Jacquemin authored 1 year ago and David Bühler committed 1 year ago

- Change Finite.for_each parameters order for a more natural one
- Used Format boxes for some pretty_printers
- Used Finite.for_each instead of a nasty for loop in vector product
- Utilitary function for index computation in matrices
- Simpler computation of the spectral radius

Maxime Jacquemin authored 1 year ago and David Bühler committed 1 year ago

- The functions `Nat.of_int`, `Nat.of_strictly_positive_int` and
  `Finite.of_int` now returns an option instead of forcing valid
  bounds on their inputs.

- The spectral exponent search is simpler and returns the maximal
  exponent in the given limit that returns a spectral radius lower
  than one.

- A bug has been fixed in the invariant computation. The sum over
  the inputs missed an element.

Maxime Jacquemin authored 1 year ago and David Bühler committed 1 year ago

Under the hood, it's just int. But only valid nat and finite can be
built through the given constructors.

(blind make headers from specifications)

- Fix Value/Value#5
- Fix Value/Value#14

Some case studies (e.g. dyad) use some ugly casts from fd_set_t which lead
to the analysis stopping too early.
Changing the representation of fd_set_t should also help it better conform to
the standard (since a fd_set_t should be able to hold FD_SETSIZE elements).

[Ptests] preserve LOG after STDOPT directive

See merge request frama-c/frama-c!2073

This is absolutely not a sneaky attempt to relaunch a build (now that OCI seems
in better shape) pushing a nearly empty commit.