mirrors/z3
3
0
Fork 0
mirror of https://github.com/Z3Prover/z3 synced 2025-04-24 01:25:31 +00:00
Code Activity
17494 commits 24 branches 48 tags 158 MiB
Commit graph

4037 commits

Author SHA1 Message Date
Nikolaj Bjorner
806a4772bc revert effect of filtering unsupported 
Signed-off-by: Nikolaj Bjorner <nbjorner@microsoft.com>
 2023-01-20 17:28:24 -08:00
Nikolaj Bjorner
4e6d498a60 adding placeholder for refining power of 2 2023-01-20 14:37:05 -08:00
Nikolaj Bjorner
0f4f32c5d0 apply relevancy filtering on unsupported ops, fix term construction bug in bv2fpa_converter fix #6548 2023-01-20 13:05:01 -08:00
Nikolaj Bjorner
f3d6856736 remove msf example, add option to make model converter not reduce models 
Signed-off-by: Nikolaj Bjorner <nbjorner@microsoft.com>
 2023-01-19 20:24:31 -08:00
Nikolaj Bjorner
7368f9f7d3 increase build version, better propagation in euf-egraph, handle assumptions in sat.smt 
- increase build version to 4.12.1. This prepares updated release for MacOs-11 build on x86
- move literal propagation mode in euf-egraph to a callback and traversal of equivalence class. Track antecedent by newest equality instead of root. This makes equality propagation to literals have similar behavior as in legacy solver and appears to result in a speedup (10% fewer conflicts on QF_UF/QG-classification/qg5/iso_icl478.smt2 in preliminary testing)
- fix interaction of pre-processing and assumptions. Pre-processing has to freeze assumption literals so they don't get eliminated. This is similar to dependencies that are already frozen.
 2023-01-17 14:07:07 -08:00
Nikolaj Bjorner
dde5218b29 fix mbqi value caching issue raised by Clemens and Martin 2023-01-15 22:47:34 -05:00
Nikolaj Bjorner
4f7f4376b8 fix bug in new core not detecting conflict, fix #6525, add tactic doc 
Signed-off-by: Nikolaj Bjorner <nbjorner@microsoft.com>
 2023-01-14 17:20:43 -05:00
Nikolaj Bjorner
8970a54eaa expose parameters to control behavior for #5660 2023-01-10 22:06:19 -08:00
Nikolaj Bjorner
d415f07386 memory leak on proof justifications 
Signed-off-by: Nikolaj Bjorner <nbjorner@microsoft.com>
 2023-01-10 18:58:25 -08:00
Nikolaj Bjorner
c3e31149a5 fix #6530 
Signed-off-by: Nikolaj Bjorner <nbjorner@microsoft.com>
 2023-01-10 13:43:17 -08:00
Nikolaj Bjorner
1ddef117a2 several fixes to proof logging in legacy solver 
Signed-off-by: Nikolaj Bjorner <nbjorner@microsoft.com>
 2023-01-08 16:11:31 -08:00
Nikolaj Bjorner
fcea32344e add missing tactic descriptions, add rewrite for tamagochi 
Signed-off-by: Nikolaj Bjorner <nbjorner@microsoft.com>
 2023-01-08 13:32:26 -08:00
Nuno Lopes
d30cb55bae don't flush stream when printing param vals 2023-01-03 09:35:17 +00:00
Nikolaj Bjorner
f6d411d54b experimental feature to access congruence closure of SimpleSolver 
This update includes an experimental feature to access a congruence closure data-structure after search.
It comes with several caveats as pre-processing is free to eliminate terms. It is therefore necessary to use a solver that does not eliminate the terms you want to track for congruence of. This is partially addressed by using SimpleSolver or incremental mode solving.

```python
from z3 import *
s = SimpleSolver()
x, y, z = Ints('x y z')
s.add(x == y)
s.add(y == z)
s.check()
print(s.root(x), s.root(y), s.root(z))
print(s.next(x), s.next(y), s.next(z))
```
 2022-12-30 21:41:27 -08:00
Nikolaj Bjorner
c0f1f33898 dampen second setup of theory_bv 2022-12-30 18:47:32 -08:00
Nuno Lopes
47324af210 be nicer when memout is reached in SMT internalize: return undef rather than crashing 2022-12-29 11:08:57 +00:00
Nikolaj Bjorner
fe8034731d fix #6501 2022-12-19 21:02:55 -08:00
Nuno Lopes
d308b8f555 simplify code + remove unused file 2022-12-11 22:11:19 +00:00
Nikolaj Bjorner
847aec1d30 update dependencies 2022-11-30 22:48:10 -08:00
Nikolaj Bjorner
529f116be0 disable new code until pre-condition gets fixed 
Signed-off-by: Nikolaj Bjorner <nbjorner@microsoft.com>
 2022-11-30 22:29:59 -08:00
Nikolaj Bjorner
85f9c7eefa replace restore_size_trail by more generic restore_vector 
other updates:
- change signature of advance_qhead to simplify call sites
- have model reconstruction replay work on a tail of dependent_expr state, while adding formulas to the tail.
 2022-11-28 11:45:56 +07:00
Nikolaj Bjorner
0a671f2f44 fix #6464 2022-11-23 17:21:51 +07:00
Nikolaj Bjorner
0a28bacd0f remove debug out 2022-11-23 16:42:36 +07:00
Nikolaj Bjorner
6188c536ef add logging of propagations to smt core 
log theory propagations with annotation "smt".
It allows tracking theory propagations (when used in conflicts) in the clause logs similar to the new core.
 2022-11-23 11:37:23 +07:00
Nikolaj Bjorner
5374142e3e continue updates for adding proof-log to smt core 2022-11-23 11:37:23 +07:00
Nikolaj Bjorner
bfae8b2162 set flat_and_or to false in bv rewriter 2022-11-15 05:47:28 -08:00
Nikolaj Bjorner
cbc5b1f4f6 have theory_recfun use recursive function discriminator to control when it is enabled 2022-11-06 12:09:45 -08:00
Nikolaj Bjorner
f004478565 produce tseitin justification for clause proofs when a clause is a "gate". 2022-11-06 12:00:25 -08:00
Nikolaj Bjorner
53b6059276 bypass built-in proof objects for clause trail 
the build-in proof constructors are not flexible when it comes to allowing alternation of justified lemmas and lemmas without justifications.
 2022-11-06 11:59:56 -08:00
Nikolaj Bjorner
84af521514 fixes #6439 #6436 2022-11-04 09:36:06 -07:00
Nikolaj Bjorner
1dca6402fb move model and proof converters to self-contained module 2022-11-03 05:23:01 -07:00
Nikolaj Bjorner
7eee7914bd align format of quantifier instantiation with new core 
So far the format is

(forall ((x Int)) body) (not (body[t/x]))

The alternative could be the clause

(not (forall ((x Int)) body)) body[t/x]

they just better be consistent between engines
 2022-10-21 15:26:00 -07:00
Nikolaj Bjorner
ad5fa9433f add experiment with quot-rem encoding 
experiment seeks to determine whether quot-rem encoding can substitute the division circuit encoding.
A first test suggests it makes no difference.
 2022-10-21 09:25:45 -07:00
Nikolaj Bjorner
842e8057bc log also quantifier generation (besides binding) 
We add also logging for quantifier generation.
It is auxiliary information that is of use for diagnostics (axiom profiler).
 2022-10-20 17:49:15 -07:00
Nikolaj Bjorner
fc30461828 unused variables 
Signed-off-by: Nikolaj Bjorner <nbjorner@microsoft.com>
 2022-10-20 09:09:06 -07:00
Nikolaj Bjorner
07dd1065db added API to monitor clause inferences 
See RELEASE_NOTES for more information
examples pending.
 2022-10-19 08:34:55 -07:00
Nikolaj Bjorner
464d52babe fix #6410 
regression after introducing beta-redex optimization
 2022-10-18 12:34:45 -07:00
Nikolaj Bjorner
2449ba93c5 add (disabled) experiment to use quot-rem instead of division circuit 2022-10-13 15:20:43 +02:00
Nikolaj Bjorner
93e1db0b0b fix #6398 2022-10-13 11:16:14 +02:00
Nikolaj Bjorner
ace727ee0f fix #6391 2022-10-12 09:34:49 +02:00
Nikolaj Bjorner
4623117af8 wip - proof hints 2022-10-08 20:12:57 +02:00
Nikolaj Bjorner
9f78a96c1d wip - trim 2022-10-06 18:19:03 +02:00
Nikolaj Bjorner
107981f099 update proof formats for new core 
- update proof format for quantifier instantiation to track original literals
- update proof replay tools with ability to extract proof object

The formats and features are subject to heavy revisions.

Example
```
(set-option :sat.euf true)
(set-option :sat.smt.proof eufproof.smt2)
(declare-fun f (Int) Int)
(declare-const x Int)
(assert (or (= (f (f (f x))) x) (= (f (f x)) x)))
(assert (not (= (f (f (f (f (f (f x)))))) x)))
(check-sat)
```

eufproof.smt2 is:
```
(declare-fun x () Int)
(declare-fun f (Int) Int)
(define-const $24 Int (f x))
(define-const $25 Int (f $24))
(define-const $26 Int (f $25))
(define-const $27 Bool (= $26 x))
(define-const $28 Bool (= $25 x))
(assume $27 $28)
(define-const $30 Int (f $26))
(define-const $31 Int (f $30))
(define-const $32 Int (f $31))
(define-const $33 Bool (= $32 x))
(assume (not $33))
(declare-fun rup () Proof)
(infer (not $33) rup)
(declare-fun euf (Bool Bool Proof Proof Proof Proof) Proof)
(declare-fun cc (Bool) Proof)
(define-const $42 Bool (= $32 $30))
(define-const $43 Proof (cc $42))
(define-const $40 Bool (= $31 $24))
(define-const $41 Proof (cc $40))
(define-const $38 Bool (= $30 $25))
(define-const $39 Proof (cc $38))
(define-const $36 Bool (= $24 $26))
(define-const $37 Proof (cc $36))
(define-const $34 Bool (not $33))
(define-const $44 Proof (euf $34 $28 $37 $39 $41 $43))
(infer (not $28) $33 $44)
(infer (not $28) rup)
(infer $27 rup)
(declare-fun euf (Bool Bool Proof Proof Proof) Proof)
(define-const $49 Bool (= $32 $26))
(define-const $50 Proof (cc $49))
(define-const $47 Bool (= $31 $25))
(define-const $48 Proof (cc $47))
(define-const $45 Bool (= $24 $30))
(define-const $46 Proof (cc $45))
(define-const $51 Proof (euf $34 $27 $46 $48 $50))
(infer $33 $51)
(infer rup)
```

Example of inspecting proof from Python:

```
from z3 import *

def parse(file):
    s = Solver()
    set_option("solver.proof.save", True)
    set_option("solver.proof.check", False)
    s.from_file(file)
    for step in s.proof().children():
        print(step)

parse("../eufproof.smt2")
```

Proof checking (self-validation) is on by default.
Proof saving is off by default.

You can use the proof logs and the proof terms to retrieve quantifier instantiations from the new core.

The self-checker contains a few built-in tuned checkers but falls back to self-checking inferred clauses using SMT.
 2022-09-28 10:40:43 -07:00
Clemens Eisenhofer
5ca53f37c0
Throw an exception if the variable in decide-callback is already assigned (#6362) 
* Memory leak in .NET user-propagator
The user-propagator object has to be manually disposed (IDisposable), otherwise it stays in memory forever, as it cannot be garbage collected automatically

* Throw an exception if variable passed to decide is already assigned instead of running in an assertion violation
 2022-09-24 09:54:14 -07:00
Nikolaj Bjorner
088898834c filter length limits to be non-skolems and under concat/""/unit 2022-09-15 07:41:13 -07:00
Nikolaj Bjorner
c47ca341b7 fix #6343 
The bug was that axiom generation was not enabled on last_index, so no axioms got created to constrain last-index.
With default settings the solver is now very slow on this example. It is related to that the smallest size of a satisfying assignment is above 24. Pending a good heuristic to find initial seeds and increments for iterative deepening, I am adding another parameter smt.seq.min_unfolding that when set to 30 helps for this example.
 2022-09-14 10:17:25 -07:00
Nuno Lopes
16ef89905d fix infinite loop in internalize 2022-09-14 11:50:53 +01:00
Nikolaj Bjorner
34969b71ee #6340 again - reduce new assertions in fresh iteration 
Signed-off-by: Nikolaj Bjorner <nbjorner@microsoft.com>
 2022-09-13 19:58:32 -07:00
Nikolaj Bjorner
fd5448d26b fix #6340 - again 
Signed-off-by: Nikolaj Bjorner <nbjorner@microsoft.com>
 2022-09-13 17:01:51 -07:00
Nikolaj Bjorner
c30b884247 fix #6340 
Signed-off-by: Nikolaj Bjorner <nbjorner@microsoft.com>
 2022-09-12 11:01:24 -07:00