-
Notifications
You must be signed in to change notification settings - Fork 1.5k
Bug - Incorrect Optimum Result #2476
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
Comments
There were fixes to PB + SAT integration after the commit mentioned above. With current master Z3 returns 38. |
I see, thanks. |
commit: 9fa9aa0 |
@NikolajBjorner news? |
the example is not small and there have been other bugs to deal with |
I'd like to apologize to you @NikolajBjorner for being disrespectful. Since I re-opened the previous issue instead of filing a new bug, I was unsure whether it had been noticed or not. No pressure at all. |
Not at all, bug reports are always extremely appreciated whether in same or new issue. |
Is this latest version of Z3? Some observations:
|
I used commit 9fa9aa0, which at the time I submitted the formula was the latest one. I did not manually set the random seed. The output i get is attached (out.txt). If there is a way for me to gather info on the random seed used and report it to you, please let me know. I just confirmed the same problem with commit 2e6908b. Indeed |
random seed is something we set from the command-line (or API). For this example, we can set smt.random_seed=1 or smt.random_seed=23 etc. It makes the solver explore different search ordering. |
It reproduces both on Windows and Ubuntu in the last release of Z3. To reproduce the behavior just run "z3 pc_32-4-8-2.new-gen_z3.smt2" without any additional parameters and see that objective is 91. |
Signed-off-by: Nikolaj Bjorner <[email protected]>
I was never able to get 91, however, the assert that Lev pointed to exposes a bug. A fix for this has been pushed. |
Thanks! |
commit: 185b01d
issue: z3 finds 49 as optimal result of
opt.smt2
, instead of 38files: witnesses.zip
notes: The formula
sat.smt2
forces the cost function to be equal 38, and the formulaunsat.smt2
forces the cost function to be better than 38. The solver says that the first formula is satisfiable and the second formula is unsatisfiable.The text was updated successfully, but these errors were encountered: