Git Product home page Git Product logo

symbiotic's Introduction

Symbiotic is an open-source tool for finding bugs in sequential computer programs. It is based on three well-know techniques: instrumentation, static program slicing and symbolic execution. Symbiotic is highly modular and most of its components are self-standing programs of LLVM passes that have their own repositories at https://github.com/staticafi.

Getting started

Downloading Symbiotic

The archive used in SV-COMOP 2018 can be downloaded from https://gitlab.com/sosy-lab/sv-comp/archives/raw/svcomp18/2018/symbiotic.zip

Other tarballs with Symbiotic distribution (not updated reguralry) can be downloaded from https://github.com/staticafi/symbiotic/releases After unpacking, Symbiotic is ready to go.

Docker image

You can use also the docker image:

docker pull mchalupa/symbiotic
docker run -ti mchalupa/symbiotic

Building Symbiotic from Sources

First of all you must clone the repository:

$ git clone https://github.com/staticafi/symbiotic

The build script of Symbiotic uses curl, make, and cmake, so make sure you have them installed (the script will complain otherwise). STP theorem prover further needs bison and flex and minisat needs zlib. These components are needed when building KLEE. However, if you do not want to build Symbiotic with KLEE (and therefore with STP and minisat), then you can comment these components out in the build script (there is no switch for not building these components yet).

If you have all the dependencies, you are ready to run the build.sh script:

$ cd symbiotic
$ ./build.sh -j2

And that should be it! However, if something goes wrong or you need to adust the build process, you can pass different options to the build script. Possible options include:

  • build-type=TYPE (TYPE one of Release, Debug)
  • llvm-version=VERSION (the default VERSION is 3.9.1, other versions are rather experimental)
  • with-llvm=, with-llvm-src=, with-llvm-dir= This set of options orders the script to use already built external LLVM (the build script will build LLVM otherwise if it has not been built already in this folder)
  • with-zlib Build also zlib
  • no-llvm Do not try building LLVM
  • slicer, minisat, stp, klee, scripts, bin Start building Symbiotic from this point

As you can see from the example, you can pass also arguments for make, e.g. -j2, to the build script. If you need to specify paths to header files or libraries, you can do it by passing CFLAGS, CPPFLAGS, LDFLAGS environment variables either by exporting them beforehand, or by passing them as make options (e.g. CFLAGS='-g')

If everything goes well, Symbiotic components are built and also installed to the install/ directory that can be packed or copied wherever you need. This directory is under git control, so you can see the differences between versions or make an archive using git archive command.

There is a known problem that can arise while building KLEE:

llvm-config: error: missing: /home/mchalupa/src/symbiotic/llvm-3.9.1/build/lib/libgtest.a
llvm-config: error: missing: /home/mchalupa/src/symbiotic/llvm-3.9.1/build/lib/libgtest_main.a
CMake Error at cmake/find_llvm.cmake:62 (message):
  Failed running
  /home/mchalupa/src/symbiotic/llvm-3.9.1/build/bin/llvm-config;--system-libs
Call Stack (most recent call first):
  cmake/find_llvm.cmake:163 (_run_llvm_config)
  lib/Basic/CMakeLists.txt:19 (klee_get_llvm_libs)

This is due to b5cd41e2 commit in LLVM. Until we have this fixed, the fastest workaround is to just create empty files build/lib/libgtest.a and build/lib/libgtest_main.a in the LLVM's folder.

When building on mac, you may need to build LLVM with shared libraries (modify the build script) or use with-llvm-* switch with your LLVM build.

Running Symbiotic

You can run Symbiotic directly from the root directory:

scripts/symbiotic <OPTIONS> file.c

Alternatively, you can run Symbiotic also from the install directory:

$ install/bin/symbiotic <OPTIONS> file.c

In the case that something went wrong, try running Symbiotic with --debug=OPT where OPT is one of: compile, instrumentation, slicer, prepare, all. When the source code does not contain everything to compile (i.e. it includes some headers), you can use CFLAGS and CPPFLAGS environment variables to pass additional options to the compiler (clang). Either export them before running Symbiotic, or on one line:

CPPFLAGS='-I /lib/gcc/include' scripts/symbiotic file.c

You can also use --cppflags switch that works exactly the same as environment variables. If the program is split into more files, you can give Symbiotic all the files. At least one of them must contain the 'main' function.

scripts/symbiotic main.c additional_definitions.c lib.c

Use --help switch to see all available options.

Symbiotic Components

Components of Symbiotic can be found at https://github.com/staticafi with the only exception of dg library that is currently at https://github.com/mchalupa/dg. All sowtware used in Symbiotic are open-source projects and are licensed under various open-source licenses (mostly MIT license, Apache-2.0, and University of Illinois Open Source license)

Contact

For more information send an e-mail to [email protected].

symbiotic's People

Contributors

mchalupa avatar xvitovs1 avatar lembergerth avatar strejcek avatar msimacek avatar tomsik68 avatar

Watchers

James Cloos avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.