-
Notifications
You must be signed in to change notification settings - Fork 274
Enable new SMT backend test suite to be run against cvc5 locally and on CI #6777
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Enable new SMT backend test suite to be run against cvc5 locally and on CI #6777
Conversation
regression/cbmc-incr-smt2/nondeterministic-int-assert/incremental_solver_called.desc
Outdated
Show resolved
Hide resolved
This will need the CVC5 version number being installed in CI to be bumped before the tests are going to pass. The fact that the make build jobs are currently passing suggests that the CVC5 tests are not being run/checked properly on these jobs. |
d708c65
to
c0c15b1
Compare
regression/cbmc-incr-smt2/Makefile
Outdated
@../test.pl -e -p -c "../../../src/cbmc/cbmc --incremental-smt2-solver 'z3 --smt2 -in' --validate-goto-model --validate-ssa-equation --slice-formula" $(exclude_broken_windows_tests) | ||
|
||
test.cvc5: | ||
@../test.pl -e -p -c "../../../src/cbmc/cbmc --incremental-smt2-solver 'cvc5 --lang smt --incremental' --validate-goto-model --validate-ssa-equation --slice-formula" $(exclude_broken_windows_tests) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
--lang=smtlib2.6
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yep, getting on to that.
regression/cbmc-incr-smt2/bitvector-bitwise-operators/shift_right.desc
Outdated
Show resolved
Hide resolved
This allows us to test against more than one solvers in the new backend (currently tested - cvc5, z3). This ensures that we don't invariably issue output that is solver specific, tying the new backend to a single solver by accident.
…ith multiple solvers. Previously the command that was enabling the new SMT backend (and providing the script for invoking the appropriate SMT solver) was passed as part of the options passed in `test.desc`. Since this was removed, and passed as part of the test configuration in the `CMakeLists.txt`, we need to do the same here.
c0c15b1
to
fbd7662
Compare
As of diffblue#6777, the codecov job kept failing for CVC5 wasn't available in this job. The same would have been true of release jobs, but we haven't had one since diffblue#6777 was merged.
As of diffblue#6777, the codecov job kept failing for CVC5 wasn't available in this job. The same would have been true of release jobs, but we haven't had one since diffblue#6777 was merged.
This is the evolution of #6745, given that the bug we were working around has now been fixed by CVC5 developers (here -> cvc5/cvc5#8489).
This allows for a cleaner changeset here, where in we only make changes to tests
and test configuration files to allow to run with multiple solvers (CVC5 and z3).