You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Idea: provision the timeout for more than 1 unrolling to avoid spending the entire timeout in the first unrolling step. This would (maybe) allow the automatic verification of properties that are true under K-induction, as it would try to unroll even if the solver does not return anything.
The text was updated successfully, but these errors were encountered:
I spent some time looking at how this might be done. Currently, the timeouts are completely external to a solver, and are implemented by a wrapper.
However, the solver does still have access to the timeout, and could implement a two-layered timeout --- one internal to the solver timing out individual unrollings, and the existing external timeout that stops the whole thing eventually. While not the most elegant, it is the least invasive way to achieve this.
(Note that the unrolling cannot be attempted in parallel, as the unrolling templates are stateful.)
Idea: provision the timeout for more than 1 unrolling to avoid spending the entire timeout in the first unrolling step. This would (maybe) allow the automatic verification of properties that are true under K-induction, as it would try to unroll even if the solver does not return anything.
The text was updated successfully, but these errors were encountered: