We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
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
I'm curious why the following benchmark:
(set-logic QF_ABV) (define-fun s1 () (Array (_ BitVec 1) (_ BitVec 1)) (store (store ((as const (Array (_ BitVec 1) (_ BitVec 1))) #b0) #b0 #b0) #b1 #b1)) (define-fun s3 () (Array (_ BitVec 1) (_ BitVec 1)) (store (store ((as const (Array (_ BitVec 1) (_ BitVec 1))) #b1) #b0 #b0) #b1 #b1)) (assert (distinct s1 s3)) (check-sat) (get-info :reason-unknown)
causes z3 to respond unknown, with the reason:
unknown
unknown (:reason-unknown "smt tactic failed to show goal to be sat/unsat (incomplete (theory array))")
If you change the logic to ALL, it correctly responds unsat.
ALL
unsat
While this isn't necessarily a big deal, I'm curious why QF_ABV logic isn't able to decide this rather simple query.
QF_ABV
(Not that it's relevant; but cvc5 returns unsat with either logic.)
The text was updated successfully, but these errors were encountered:
eb5d036
No branches or pull requests
I'm curious why the following benchmark:
causes z3 to respond
unknown
, with the reason:If you change the logic to
ALL
, it correctly respondsunsat
.While this isn't necessarily a big deal, I'm curious why
QF_ABV
logic isn't able to decide this rather simple query.(Not that it's relevant; but cvc5 returns
unsat
with either logic.)The text was updated successfully, but these errors were encountered: