| Age | Commit message (Collapse) | Author |
|---|
|
|
|
This commit adds a new option to produce unsat cores based on our proof infrastructure (whereas previously we could only do so if we were also checking unsat cores) and the corresponding changes to the default settings to account for it. Since now options::unsatCores() and options::produceProofs() are incompatible, several parts of the code where we tested if we were in "old unsat cores mode", by testing the former and the negation of the latter options, are updated accordingly.
This commit also changes how SMT engine sets things by disabling proofs in the theory engine if we are in unsat core mode.
|
|
|
|
When a command is invoked, its result/status may be printed depending on its current verbosity, which (for probably an outdated reason) is stored in SMTEngine. One of my previous PRs modified the SMTEngine to return the verbosity as an sexpr. The modification works correctly when the output language is SMT2, but it breaks down with the AST output language, which prints sexprs in a different way. This PR fixes this bug by making SMTEngine return options as strings instead of sexpr.
|
|
This PR does another round of refactoring of the resource manager and related code.
- it moves the Resource enum out of the ResourceManager class
- it treats the resources in a generic way (storing the statistics in a vector) instead of the manual treatment we had before
- weights no longer live in the options, but in the ResourceManager and are changed accordingly in the ResourceManager constructor
- following the generic treatment of resources, it also removes all the resource-specific options --x-step in favor of a generic --rweight name=weight
- removed several unused methods from the ResourceManager
Note that we handle the Resource enum in a way that allows to easily use other enums as additional resources, for example InferenceId. The general idea is that we will at some point have sensible default weights (so that the cumulative resources somewhat simulate the solver runtime) and users (almost) never need to modify them.
|
|
|
|
This consolidates the interface for asserting input formulas to the PropEngine from SmtSolver.
As a consequence of this PR, this corrects one issue with the justification heuristic where skolem definitions were considered "assertions" by the justification heuristic (e.g. formulas that must be satisfied) instead of just being required for skolems in relevant literals. This was asymmetric from skolem definitions from lemmas, which were not being considered assertions. Now, skolem definitions are never assertions.
I tested this on QF_LIA SMT-LIB with decision=justification with 300 second timeout, essentially no difference in results (+6-5 all close to timeout). Also no difference on QF_S + QF_SLIA.
|
|
|
|
|
|
For some benchmarks, checking MACRO_RESOLUTION can be up to 80% (!!!) of the running time. This commit introduces a new rule that does not perform checking. The old rule and checker are kept for ground truth. Some miscellaneous minor changes are also made in the PR.
|
|
Previously the SMT post-processor would update any assumption as long as it
had a proof for it. This can be a problem when one as assumption introduced in a
scope that should not be expanded. This commit fixes the issue by adding the
option of configuring a proof node updater to track scopes and the assumptions
they introduce, which can be used to determine the prood nodes which should be
updated. It also changes the SMT post-processor to only update assumptions that
have not been introduced in some scope.
This commit fixes an issue found by @Lachnitt during the integration of CVC4 and Isabelle.
|
|
Fixes benchmarks 2 and 3 from #5848.
|
|
Previously, we were traversing proof node as a tree, now we use a dag traversal.
This also makes sure that proofs work when we have a external proof conversion and we are in incremental mode. In such cases, the final proof must be cloned to ensure that we do not overwrite proof nodes, which may be reused across multiple check-sat.
|
|
This is in preparation for refactoring skolem creation throughout the code base to improve proofs and migrate Theory::expandDefinitions to Rewriter::expandDefinitions.
This PR also eliminates some unused code in TheoryArithPrivate.
Followup PRs will start formalizing/eliminating calls to mkDummySkolem.
|
|
Currently, NodeBuilder takes a single template argument: An integer
that determines the expected number of arguments. This argument is used
to determine the size of the d_inlineNvChildSpace array. This array is
used to construct nodes inline. The advantage of this is that we don't
have to allocate a NodeValue on the heap for the node under
construction until we are sure that the node is new. While templating
the array size may save some stack space (or avoid a heap allocation if
we statically know that we a fixed number of children and that number is
greater than 10), it complicates the code and leads to longer compile
times. Thus, this commit removes the template argument and moves some of
the NodeBuilder code to a source file for faster compilation.
CPU build time before change (debug build): 2429.68s
CPU build time after change (debug build): 2228.44s
Signed-off-by: Andres Noetzli noetzli@amazon.com
|
|
|
|
Each theory has its own proof checker, responsible for checking the rules pertaining to that theory. The main proof checker uses these specialized checkers. Previously the main proof checker (of the proof node manager used across the SMT solver) was connected to these theory proof checkers during initialization of the theory. This commit adds an interface to the theories for retrieving its proof checker (analogous to how one retrieves the rewriter of a theory) which is used by a new method in the theory engine to register a theory proof checker to a given proof checker according to a theory id.
This is in preparation for the new unsat cores based on proofs.
|
|
Fixes #4328.
|
|
Adds support for BitVector optimization, which is done via offline binary search. Units tests included.
Also mildly refactors the optimizer architecture.
|
|
This PR does some minor refactorings in preparation for the new statistics: some renamings, removal of now obsolete code and usage of references instead of pointers.
|
|
|
|
|
|
This PR changes ResetCommand to not call SmtEngine::reset but instead reconstruct the api::Solver object. This makes SmtEngine::reset obsolete and removes it, and moves the original options from SmtEngine to api::Solver.
The motivation for this change is twofold:
The ResetCommand should not use SmtEngine directly, but only through the API (i.e. Solver).
As soon as the statistics data lives within the registry, we need to re-register statistics after resetting the SmtEngine. We can now do this within the api::Solver constructor.
|
|
|
|
We require tracking of instantiation lemmas for quantifier elimination. Recently, this feature was removed in favor of reconstructing instantiations via substitutions. This does not quite work if instantiation lemmas have more complex post-processing, e.g. virtual term substitution.
This PR reimplements a much simpler form of instantiation tracking that simply adds instantiation bodies to a vector, per quantified formula. It uses this vector for quantifier elimination.
Fixes #5899.
|
|
This PR adds includes that are missing from source files, but currently provided by other includes.
This mostly concerns <sstream> which is currently included by the statistics, which will change in the future.
|
|
|
|
|
|
This PR does some refactoring on how we handle statistics in the driver, and some minor cleanup along the way.
The SmtEngine now has dedicated statistics for the data collected within the driver and provides utility functions to set them.
The driver pushes the collected statistics to the SmtEngine instead of adding them itself to the statistics registry.
The node manager no longer holds a statistics registry (that nobody used anymore anyway)
The command executor no longer holds a pointer to the SmtEngine itself. The pointer is not necessary and seems to become stale after we call reset on the command executor (which, luckily, we don't seem to do usually)
The main motivation for this change is to make the whole statistics infrastructure private to the library and not exporting it to the outside world.
|
|
This creates a central utility for managing "skolem definitions", e.g. mapping between skolems and the lemmas that define their behavior.
This utility is taken from the satRlv branch. It will also be used for the new implementation of the justification decision heuristic.
Note that this PR takes some helper functions out of term formula removal (e.g. hasSkolems) Prior to this PR, these helper functions were incorrect since term formula removal does not account for all introduced skolems. For instance, Theory::ppRewrite may introduce skolems directly. This PR consolidates these cases into the new class, which is called from PropEngine when lemmas and assertions are added. At the moment, the only use of this method is for CEGQI, which needs to do its own tracking of skolems in certain literals.
It also makes some minor reorganization to prop engine.
|
|
In rare cases, we may reuse skolems for different terms (those that are the same up to purification) due to recent changes in how skolem are generated. This guards for this case in the term formula remover, which avoids assertion failures in cd insert hash map.
Fixes #6132.
|
|
Warning if not a supported version.
|
|
This PR moves the statistics registry from SmtEngine to Env.
|
|
This PR eliminates all remaining uses of SExpr outside of statistics.
|
|
This copies most of the remaining changes to set_defaults.cpp from proof-new.
In particular, it recognizes when proofs must be disabled.
This is required to fix the regressions (locally) and the nightlies.
|
|
|
|
The build system (cmake) will automatically generate an export header
cvc4_export.h, which makes sure that the correct export features are
defined depending on the compiler and target platform. The macro CVC4_EXPORT
replaces CVC4_PUBLIC and its usage is reduced by 2/3.
Co-authored-by: Gereon Kremer <nafur42@gmail.com>
|
|
|
|
This PR uses IntegralHistogramStat instead of HistogramStat when appropriate, that is everywhere.
|
|
Adds a dot printer for proof nodes. Also adds an option to choose the proof format (as a mode).
Signed-off-by: Diego Della Rocca de Camargos <diegodellarocc@gmail.com>
|
|
Towards having proofs working on master.
|
|
This PR does a first round of refactoring on the statistics, in particular the Stat class and derived classes.
It significantly shrinks the class hierarchy, modernizes some code (e.g. use std::chrono instead of clock_gettime), removes unused features (e.g. nesting of statistics) and does some general cleanup and consolidation.
Subsequent PRs are planned to change the ownership model (right now every module owns the Stat object) which makes the whole register / unregister mechanism obsolete.
|
|
|
|
This class contains all globally available utilities for internal code.
|
|
This fixes two issues for preprocessing:
(1) The term preregistration visitor was calling preRegister on terms multiple times in a SAT context, which the linear arithmetic solver is sensitive to.
(2) It was possible for non-preprocessed terms to appear in assertions if they were on the RHS of substitutions learned by non-clausal simplification, and substituted into assertions post-theory-preprocessing.
To fix (1), the SharedTermsVisitor is update to track which theories has preregistered each term, as is done in the PreRegisterVisitor. To fix (2), we no longer apply-subst after theory preprocessing.
These two fixes are required to fix #6071.
Note: we should performance test this on SMT-LIB.
|
|
calculus (#6051)
This makes a simplification to the internal proof calculus. In particular, purification skolems are no longer are special case of witness skolems. They are now independent concepts. The concept of "witness form" is replaced in most places by "original form".
This is required for fixing two issues:
(1) variable shadowing issues in skolemization.
(2) bookkeeping issues for bound variables introduced to construct witness terms. This made the LFSC proof conversion extremely cumbersome for e.g. string reductions.
In this PR, the main changes:
The internals of SkolemManager are changed to use original form vs witness form when necessary. This eliminates the need to do variable renaming in SkolemManager::skolemize.
the rule WITNESS_INTRO is replaced by SKOLEM_INTRO
MACRO_SR_* rules use original form
Proof post processing is simplified
These changes imply that ppRewrite should not return WITNESS terms. Followup PRs will modify arithmetic preprocessing so that its ppRewrite method returns skolems instead.
|
|
This removes use of the logic request utility.
It generally bad practice to change the logic dynamically, e.g. during preprocessing, since it makes it so that CVC4 does not properly initialize. We now insist that logic is changed upfront in set_defaults.
This is in preparation for the smt::Env class, which will change the ownership of logic.
|
|
This is in preparation for enabling CI / proofs on master.
This does not throw an option exception when proofs are enabled, it also makes a fix that was missing on master and needed for regressions to pass on master.
This is partially taken from #5980.
|
|
This PR does some more cleanup of the includes.
|
|
|
