Skip to content
/ maam Public

A monadic approach to static analysis following the methodology of AAM

License

Notifications You must be signed in to change notification settings

davdar/maam

Repository files navigation

Building and Running

I recommend building with a cabal sandbox. To initialize a cabal sandbox (that will live in the current directory) and install needed dependencies, run:

make sandbox

I have not included dependency bounds in my cabal file. Should you have trouble finding appropriate bounds, here are the versions of ghc and cabal packages that I am using.

base=4.7.0.2
Cabal=1.18.1.5
containers=0.5.5.1
directory=1.2.1.0
ghc=7.8.4
template-haskell=2.9.0.0
text=1.2.0.4

Running

To run the project, displaying an analysis of various lambda-if examples, run:

make run

Example output is included at the end of this readme.

Interactive (GHCI)

To support my custom (well-formatted and colored) pretty printing in ghci, you need to first initialize some ghc flag files:

make init-flags

Then just run:

./ghci.sh

to run Main, or:

./ghci.sh Lang.LamIf.Examples

to run another module, like Lang.LamIf.Examples.

Source Code

All code is in /src.

FP

FP is a core functional programming library which replaces the standard Prelude. FP includes more batteries for things like functors, monads, monad transformers, lenses, pretty printing, parsing, deriving, and more. On the downside, it is non-idiomatic at parts and isn't as mature (i.e. debugged and stable).

MAAM

MAAM is a semantics-independent package for implement path, flow, context and object sensitivity in program analysis. MAAM only contains types and definitions which are analysis specific. Because the monad transformers that capture path and flow sensitivity are fully general purpose, they are defined in FP.Monads, not here. The same goes for lattice structures, which are mostly all defined in FP.Core. If I were to port MAAM to use GHC's Prelude, I would need to rip out maybe 50% of FP to be packaged alongside it.

The only code that ends up being specific to analysis is:

  • Mapping monadic actions to state space transition systems, which is defined in MAAM.MonadStep.
  • Implementations for abstract time to infinite-k (concrete), finite-k and zero-k, which are defined in MAAM.Time.

LamIf

Lang.LamIf implements the following for a small applied lambda calculus with booleans and if-statements:

  • Direct-style syntax (Lang.LamIf.Syntax)
  • Continuation passing style (CPS) syntax (Lang.LamIf.CPS)
  • Parsing (Lang.LamIf.Parser) and pretty printing (Lang.LamIf.Pretty)
  • CPS conversion (Lang.LamIf.Passes)
  • Semantics state-space (Lang.LamIf.StateSpace)
  • Monadic semantics (Lang.LamIf.Semantics)
  • Concrete and abstract value domains (Lang.LamIf.Val)
  • Instantiations of language-independent monads from MAAM (Lang.LamIf.Monads)
  • Orthogonal analysis parameters (Lang.LamIf.Analyses)
  • Example analyses (Lang.LamIf.Examples)

Hask

A semantics for GHC core is implemented in Lang.Hask:

  • CPS syntax and conversion (Lang.Hask.CPS)
  • Pretty printing (Lang.Hask.Pretty)
  • Monadic semantics (Lang.Hask.Semantics)
  • Execution semantics (Lang.Hask.Execution)
  • Instantiations of language-independent monads from MAAM (Lang.Hask.Monads)
  • Concrete value domain (Lang.Hask.ValConcrete)
  • Lifting of an arbitrary value domain to a sum-of-products lattice (Lang.Hask.SumOfProdVal)

While the core semantics for core GHC is implemented, we haven't implemented any GHC primitives yet, but you should be able to get a feel for the semantics without the primitives. (More on this coming soon.)

Example Output

If you execute the project it will compute an abstract interpretation of some very small LamIf programs.

The output includes results for the heap when it reaches any HALT state:

Below is a copy of the (normally ANSI-terminal-color-coded) output of make run. For 0CFA results the state space will contain values and abstract environments mapping variables to themselves, which are degenerate encodings of (unused) call-site sensitivity. The abstract heap will come last, mapping variables to the values they take on. This is also the example from the Galois Transformers paper. More examples programs can be found in /data/lamif-src, with example configurations found in /src/Lang/LamIf/Examples.hs.

Source
let b := 1 - 1 >= 0 in
let v := if b
    then if b then 1 else 2
    else if b then 3 else 4
in
let w := if b then 5 else 6 in <v,w>
Stamped
0:let 0:b := 1:(2:(3:1) - (4:1)) >= (5:0) in
6:let 1:v := 7:if 8:0:b
    then 9:if 10:0:b then 11:1 else 12:2
    else 13:if 14:0:b then 15:3 else 16:4
in
17:let 2:w := 18:if 19:0:b then 20:5 else 21:6 in
22:<23:1:v,24:2:w>
CPS
25:3:a#0 := 2:1 - 1
0:0:b := 1:(3:a#0) >= 0
33:8:k#5 := 32:λ 4:x#1 ->
  6:1:v := 7:4:x#1
  29:7:k#4 := 28:λ 5:x#2 ->
    17:2:w := 18:5:x#2
    26:6:a#3 := 22:<1:v,2:w>
    27:HALT (6:a#3)
  18:if 0:b then 30:(7:k#4) 5 else 31:(7:k#4) 6
7:if 0:b
  then
    9:if 0:b then 34:(8:k#5) 1 else 35:(8:k#5) 2
  else
    13:if 0:b then 36:(8:k#5) 3 else 37:(8:k#5) 4
LT=0 DT=0 V=abstract M=fi G=yes C=link LF=app DF=app
( { ( 27:HALT (6:a#3)
    , ( ∙
      , ∙
      , { 0:b => <x=0:b,lτ=∙,dτ=∙>
        , 1:v => <x=1:v,lτ=∙,dτ=∙>
        , 2:w => <x=2:w,lτ=∙,dτ=∙>
        , 3:a#0 => <x=3:a#0,lτ=∙,dτ=∙>
        , 4:x#1 => <x=4:x#1,lτ=∙,dτ=∙>
        , 5:x#2 => <x=5:x#2,lτ=∙,dτ=∙>
        , 6:a#3 => <x=6:a#3,lτ=∙,dτ=∙>
        }
      )
    )
  }
, { <x=1:v,lτ=∙,dτ=∙> => {1,2,3,4}
  , <x=2:w,lτ=∙,dτ=∙> => {5,6}
  , <x=6:a#3,lτ=∙,dτ=∙> =>
      {<<x=1:v,lτ=∙,dτ=∙>,<x=2:w,lτ=∙,dτ=∙>>}
  }
)
LT=0 DT=0 V=abstract M=fs G=yes C=link LF=app DF=app
{ 27:HALT (6:a#3) =>
    ( { ( ∙
        , ∙
        , { 0:b => <x=0:b,lτ=∙,dτ=∙>
          , 1:v => <x=1:v,lτ=∙,dτ=∙>
          , 2:w => <x=2:w,lτ=∙,dτ=∙>
          , 3:a#0 => <x=3:a#0,lτ=∙,dτ=∙>
          , 4:x#1 => <x=4:x#1,lτ=∙,dτ=∙>
          , 5:x#2 => <x=5:x#2,lτ=∙,dτ=∙>
          , 6:a#3 => <x=6:a#3,lτ=∙,dτ=∙>
          }
        )
      }
    , { <x=1:v,lτ=∙,dτ=∙> => {1,4}
      , <x=2:w,lτ=∙,dτ=∙> => {5,6}
      , <x=6:a#3,lτ=∙,dτ=∙> =>
          {<<x=1:v,lτ=∙,dτ=∙>,<x=2:w,lτ=∙,dτ=∙>>}
      }
    )
}
LT=0 DT=0 V=abstract M=ps G=yes C=link LF=app DF=app
{ ( 27:HALT (6:a#3)
  , ( ( ∙
      , ∙
      , { 0:b => <x=0:b,lτ=∙,dτ=∙>
        , 1:v => <x=1:v,lτ=∙,dτ=∙>
        , 2:w => <x=2:w,lτ=∙,dτ=∙>
        , 3:a#0 => <x=3:a#0,lτ=∙,dτ=∙>
        , 4:x#1 => <x=4:x#1,lτ=∙,dτ=∙>
        , 5:x#2 => <x=5:x#2,lτ=∙,dτ=∙>
        , 6:a#3 => <x=6:a#3,lτ=∙,dτ=∙>
        }
      )
    , { <x=1:v,lτ=∙,dτ=∙> => {1}
      , <x=2:w,lτ=∙,dτ=∙> => {5}
      , <x=6:a#3,lτ=∙,dτ=∙> =>
          {<<x=1:v,lτ=∙,dτ=∙>,<x=2:w,lτ=∙,dτ=∙>>}
      }
    )
  )
, ( 27:HALT (6:a#3)
  , ( ( ∙
      , ∙
      , { 0:b => <x=0:b,lτ=∙,dτ=∙>
        , 1:v => <x=1:v,lτ=∙,dτ=∙>
        , 2:w => <x=2:w,lτ=∙,dτ=∙>
        , 3:a#0 => <x=3:a#0,lτ=∙,dτ=∙>
        , 4:x#1 => <x=4:x#1,lτ=∙,dτ=∙>
        , 5:x#2 => <x=5:x#2,lτ=∙,dτ=∙>
        , 6:a#3 => <x=6:a#3,lτ=∙,dτ=∙>
        }
      )
    , { <x=1:v,lτ=∙,dτ=∙> => {4}
      , <x=2:w,lτ=∙,dτ=∙> => {6}
      , <x=6:a#3,lτ=∙,dτ=∙> =>
          {<<x=1:v,lτ=∙,dτ=∙>,<x=2:w,lτ=∙,dτ=∙>>}
      }
    )
  )
}

About

A monadic approach to static analysis following the methodology of AAM

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •  

Languages