Software

Here you can find the software our group developed.
Some of them are designed and developed in collaboration with other research groups within and outside Northwestern.

Next is some of the software we are currently maintaining.
The full list can be found here.

Fun fact: my entire website (html files and javascript code) as well as my CV (latex files) are automatically generated by the compiler I developed (at the end of working days) called SIMO (Sleeping In My Office).

NOELLE Offers Empowering LLVM Extensions

NOELLE provides abstractions to accelerate the implementation of advanced code analyses and transformations.

NOELLE
This software is developed for the parallelization, MARC, compiling for sub-cycle microarchitecture activities, and power of modern compilers research directions.
NOELLE is also used in the ATC class.

Introducing the NOELLE's project
Introducing the NOELLE's codebase
Dependences with NOELLE
Profiling with NOELLE
Task with NOELLE
Call graph with NOELLE
Data-Flow analysis with NOELLE
Organizing loops in a forest using NOELLE
A parallelizing compiler built upon NOELLE
NOELLE Gym: where NOELLE-based tools exercise
Would you like to quickly test the impact of your code analysis and/or transformation built upon NOELLE?
NOELLE Gym is designed to help you quickly test your work.

This software is developed for and used in all research directions of our group.
Framework to compile benchmarks into a single bitcode file
NOELLE requires a single bitcode file that includes the whole program.
This is not typically obtained by using well-established benchmark suites like SPEC CPU2017, PARSEC, or MiBench.
So, we developed scripts and patches and we organized them into a framework to compile benchmarks of well-established benchmark suites into a single bitcode file.

This software is developed for and used in all research directions of our group.

Parallelizing compilers

This software is developed for the parallelization research direction.

SCAF
An innovative design to combine alias analyses and (optionally) speculation.
Collaborative Parallelization Framework (CPF)
An innovative design to combine code transformations and parallelization to extract threads automatically.

MARC: Mining Advantages in non-deteRministic Code

This software is developed for the MARC research direction.

CARMOT

VIRGIL: Very sImple thRead pool desiGned with the easy Integration goaL in mind

This is a header-only C++ library that implements a simple thread pool using the modern C++ language. The design of VIRGIL aims for an easy integration with other C++ code.

This software is developed for both the MARC research direction and for the parallelizing compiler one.

LLVM related software

Next there is a set of software packages I built to help LLVM development.

Downloading, installing, and compiling LLVM .
This software is used in almost all research directions our group follows as well as in my Code Analysis and Transformation class.
Template for LLVM middle-end passes .
This software is used in almost all research directions our group follows as well as in my Code Analysis and Transformation class.

My development environment

Next there is the software/configurations I use for my development environment. I hope you will find them as useful as I do.

VIM configuration
This is the VIM configuration I use in a Linux environment.

ZSHRC configuration
This is the ZSHRC configuration I use in a MacOS environment.

Simone Campanoni