summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
...
* new files for fixedpoint engine backendGravatar Unknown2013-05-07
|
* merging changes for fixedpoint engine backendGravatar Ken McMillan2013-05-07
|\
* | Adding fixedpoint engine backendGravatar Ken McMillan2013-05-07
| |
| * havocs and updates to globals added only if they are non-nullGravatar Unknown2013-05-06
| |
| * MergeGravatar Unknown2013-05-06
| |\
| * | fixed bug (reported by Akash) in treatment of linear parameters to callsGravatar Unknown2013-05-06
| | |
| | * AbsHoudini: Each function can specify its own abstract domain. Also addedGravatar akashlal2013-05-05
| |/ | | | | | | typechecking
| * In the typecheck method of HavocCmd, added calls to typecheck the havoced varsGravatar Unknown2013-05-04
| |
| * MergeGravatar Unknown2013-05-04
| |\
| * | fixed bug reported by AkashGravatar Unknown2013-05-04
| | |
| | * Print "\n" after a YieldCmdGravatar akashlal2013-05-03
| | |
| | * Some code refactoringGravatar akashlal2013-05-03
| |/
| * add another test fileGravatar Unknown2013-04-30
|/
* MergeGravatar allydonaldson2013-04-30
|\
* | Staged HoudiniGravatar allydonaldson2013-04-30
| |
| * Added a little bit of virtual-ness to the Inliner class. This is so that I canGravatar akashlal2013-04-28
| | | | | | | | subclass it to implement my own inlining policy.
| * AbsHoudini: Bug fixGravatar akashlal2013-04-28
| |
| * AbsHoudini: Added support for /errorLimit:n, n > 1Gravatar akashlal2013-04-25
| |
| * AbsHoudini: Added predicate-abstraction domain and some examples.Gravatar akashlal2013-04-25
| |
| * AbsHoudini: Added support for DoNotUsLabels, domain for simulating Houdini,Gravatar akashlal2013-04-25
| | | | | | | | few bug fixes, hack to support missing prover declarations.
| * added free ensures to each procedure to compensate for havocing of allocatorGravatar Unknown2013-04-19
| |
| * Test case for OG --> HoudiniGravatar akashlal2013-04-19
| |
| * Made OG-Desugared parsable as a Boogie program.Gravatar akashlal2013-04-19
| |
| * Added a test caseGravatar akashlal2013-04-19
| |
| * MergeGravatar akashlal2013-04-19
| |\
| | * AbsHoudini: Added SCC-based worklist, bug-fix for inline attribute,Gravatar akashlal2013-04-19
| | | | | | | | | | | | intervals domain
| * | Fix mod-set traversal to do visit code inside code expressions.Gravatar Rustan Leino2013-04-18
| |/
| * AbsHoudini: Added support for /inlineDepth, and fixed the regression testsGravatar akashlal2013-04-18
| | | | | | | | (all pass)
| * Nice clean re-implementation of AbstractHoudini. And testsGravatar akashlal2013-04-18
| |
| * refactored og and fixed latest bug reported by chrisGravatar Unknown2013-03-20
| |
| * the allocator corresponding to every linear variableGravatar Unknown2013-03-13
| | | | | | | | | | in the modset of a procedure is now put into the modset itself (in addition to the allocators for the output parameters)
| * Added explicit mod set analysis calls to OG transform and linear transformGravatar Unknown2013-03-13
| |
| * added mod set checking to the linear type checkerGravatar Unknown2013-03-13
| |
| * fixed bugs in linear and ogGravatar Unknown2013-03-11
| | | | | | | | added triggers to the axioms for linear set desugaring
| * fixed bug in model printingGravatar Unknown2013-03-11
| | | | | | | | reverted a previous erroneous fix I had made in model parsing
| * Added another constructor to CallCmdGravatar akashlal2013-03-11
| |
| * Minor changes to AbstractHoudiniGravatar akashlal2013-03-10
| |
| * MergeGravatar Unknown2013-03-09
| |\ | |/ |/|
| * fixed emitter for CallCmd to include asyncGravatar Unknown2013-03-09
| |
* | Refactored MatchCandidateGravatar allydonaldson2013-03-08
| |
* | MergeGravatar allydonaldson2013-03-08
|\ \
* | | MatchCandidate modified to match candidates by variable name, rather than by ↵Gravatar allydonaldson2013-03-08
| | | | | | | | | | | | variable identity. ApplyAssignment method added, which takes a program (not necessarily the same program on which Houdini was invoked, but a program that has the same candidate set), and applies the Houdini assignment to the program.
| | * MergeGravatar Unknown2013-03-07
| | |\ | | |/ | |/|
| | * updated the model parsing (probably caused by some change in Z3's output)Gravatar Unknown2013-03-07
| | |
| | * added support for linear sets without useArrayTheory (but there is some ↵Gravatar Unknown2013-03-07
| | | | | | | | | | | | incompletness)
| * | Fix LeastCommonAncestor method in Graph moduleGravatar Nathan Chong2013-03-07
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Solves a problem with Uniformity Analysis in GPUVerify. "The least common ancestor of two nodes u and v in a rooted tree T is a node w that is an ancestor of both u and v and that has the greatest depth in T" (CLR). This is needed when computing the control dependence relation of a CFG. In this case, we need the least common ancestor of a pair of nodes in the dominator tree so we can reuse the 'intersect' function given by Cooper et al in "A Simple, Fast Dominance Algorithm." The old code picked the ancestor of both u and v that has the *least* depth in T. This caused uniformity analysis, in GPUVerify, to be imprecise (adding blocks as non-uniform when they were infact uniform).
| | * MergeGravatar Unknown2013-03-06
| | |\ | |_|/ |/| |
| | * in the process of adding support for linear sets without /useArrayTheoryGravatar Unknown2013-03-06
| |/
| * removed Test\GPUVerifyGravatar Unknown2013-03-06
| |
| * removing GPUVerify and Dafny.slnGravatar Unknown2013-03-05
| |