Coder Social home page Coder Social logo

quazirafi / ir2vec Goto Github PK

View Code? Open in Web Editor NEW

This project forked from iith-compilers/ir2vec

0.0 0.0 0.0 371.74 MB

Implementation of IR2Vec, published in ACM TACO

License: Other

Shell 0.03% C++ 1.75% Python 0.44% C 0.11% CMake 0.02% LLVM 89.30% Jupyter Notebook 8.34%

ir2vec's Introduction

IR2Vec

IR2Vec is a LLVM IR based framework to generate distributed representations for the source code in an unsupervised manner, which can be used to represent programs as input to solve machine learning tasks that take programs as inputs.

This repo contains the source code and relevant information described in the paper (arXiv). Please see here for more details.

IR2Vec: LLVM IR Based Scalable Program Embeddings, S. VenkataKeerthy, Rohit Aggarwal, Shalini Jain, Maunendra Sankar Desarkar, Ramakrishna Upadrasta, and Y. N. Srikant

LLVM Tests Publish pre-commit checks

Image

Table Of Contents

Requirements

(Experiments are done on an Ubuntu 18.04 machine)

Binaries and Libraries - Artifacts

Binaries and Libraries (.a and .so) are autogenerated for every relevant checkin using GitHub Actions. Such generated artifacts are tagged along with the successful runs of Publish workflow and can be found here.

Building from source

  1. mkdir build && cd build
  2. IR2Vec uses Eigen library. If your system already have Eigen (3.3.7) setup, you can skip this step.
    1. Download and extract the released version.
      • wget https://gitlab.com/libeigen/eigen/-/archive/3.3.7/eigen-3.3.7.tar.gz
      • tar -xvzf eigen-3.3.7.tar.gz
    2. mkdir eigen-build && cd eigen-build
    3. cmake ../eigen-3.3.7 && make
    4. cd ../
  3. cmake -DLT_LLVM_INSTALL_DIR=<path_to_LLVM_build_dir> -DEigen3_DIR=<path_to_eigen_build_dir> [-DCMAKE_INSTALL_PREFIX=<install_dir>] ../src
  4. make [&& make install]

This process would generate ir2vec binary under build/bin directory, libIR2Vec.a and libIR2Vec.so under build/lib directory.

To ensure the correctness, run make verify-all

Generating program representations

IR2Vec can be used either as a stand alone tool using binary, or can be integrated with any third party tools using libraries. Please see below for the usage instructions.

Using Binary

ir2vec -<mode> -vocab <seedEmbedding-file-path> -o <output-file> -level <p|f> -class <class-number> <input-ll-file>

Command-Line options

  • mode - can be one of sym/fa
    • sym denotes Symbolic representation
    • fa denotes Flow-Aware representation
  • vocab - the path to the seed embeddings file
  • o - file in which the embeddings are to be appended; (Note : If file doesn’t exist, new file would be created, else embeddings would be appended)
  • level - can be one of chars p/f.
    • p denotes program level encoding
    • f denotes function level encoding
  • class - only non-mandatory argument. Used for the purpose of mentioning class labels for classification tasks (To be used with the level p). Defaults to -1. When, not equal to -1, the pass prints class-number followed by the corresponding embeddings

Please use --help for further details.

Format of the output embeddings in output_file

  • If the level is p:
<class-number> <Embeddings>

class-number would be printed only if it is not -1

  • If the level is f
<function-name> = <Embeddings>

Flow-Aware Embeddings

  • ir2vec -fa -vocab vocabulary/seedEmbeddingVocab-300-llvm12.txt -o <output_file> -level <p|f> -class <class-number> <input_ll_file>

Symbolic Embeddings

  • ir2vec -sym -vocab vocabulary/seedEmbeddingVocab-300-llvm12.txt -o <output_file> -level <p|f> -class <class-number> <input_ll_file>

Using Libraries

The libraries can be installed by passing the installation location to the CMAKE_INSTALL_PREFIX flag during cmake followed by make install. The interfaces are available in IR2Vec.h. External projects that would like to use IR2Vec can access the functionality using these exposed interfaces on including IR2Vec.h from the installed location after linking statically or dynamically.

  • If the project does not use LLVM, LLVM dependencies have to be linked and included separately.
  • Please ensure that the IR2Vec libraries are compiled with compatible LLVM.
    • If you are getting errors, please recompile IR2Vec by passing the current LLVM install directory path to LT_LLVM_INSTALL_DIR during cmake.

The following template can be used to link IR2vec libraries on a cmake based project.

set(IR2VEC_INSTALL_DIR "" CACHE PATH "IR2Vec installation directory")
include_directories("${IR2VEC_INSTALL_DIR}/include")
target_link_libraries(<your_executable_or_library> PUBLIC ${IR2VEC_INSTALL_DIR}/lib/<libIR2Vec.a or libIR2Vec.so>)

And then pass the location of IR2Vec's install prefix to DIR2VEC_INSTALL_DIR during cmake.

The following example snippet shows how to query the exposed vector representations.

#include "IR2Vec.h"

// Creating object to generate FlowAware representation
auto ir2vec =
      IR2Vec::Embeddings(<LLVM Module>, IR2Vec::IR2VecMode::FlowAware,
                         "./vocabulary/seedEmbeddingVocab-300-llvm12.txt");

// Getting Instruction vectors corresponding to the instructions in <LLVM Module>
auto instVecMap = ir2vec.getInstVecMap();
// Access the generated vectors
for (auto instVec : instVecMap) {
    outs() << "Instruction : ";
    instVec.first->print(outs());
    outs() << ": ";

    for (auto val : instVec.second)
      outs() << val << "\t";
}

// Getting vectors corresponding to the functions in <LLVM Module>
auto funcVecMap = ir2vec.getFunctionVecMap();
// Access the generated vectors
for (auto funcVec : funcVecMap) {
    outs() << "Function : " << funcVec.first->getName() << "\n";
    for (auto val : funcVec.second)
      outs() << val << "\t";
  }

// Getting the program vector
auto pgmVec = ir2vec.getProgramVector();
// Access the generated vector
for (auto val : pgmVec)
    outs() << val << "\t";

Experiments

Note

The results mentioned in the experiment's scripts/the published version are not updated for this branch. The experimental results for this branch would be different when compared to the published version. For comparison, use the release corresponding to v0.1.0.

Citation

@article{VenkataKeerthy-2020-IR2Vec,
author = {VenkataKeerthy, S. and Aggarwal, Rohit and Jain, Shalini and Desarkar, Maunendra Sankar and Upadrasta, Ramakrishna and Srikant, Y. N.},
title = {{IR2Vec: LLVM IR Based Scalable Program Embeddings}},
year = {2020},
issue_date = {December 2020},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
volume = {17},
number = {4},
issn = {1544-3566},
url = {https://doi.org/10.1145/3418463},
doi = {10.1145/3418463},
journal = {ACM Trans. Archit. Code Optim.},
month = dec,
articleno = {32},
numpages = {27},
keywords = {heterogeneous systems, representation learning, compiler optimizations, LLVM, intermediate representations}
}

Contributions

Please feel free to raise issues to file a bug, to pose a question, or to initiate any related discussions. Pull requests are welcome :)

License

IR2Vec is released under a BSD 4-Clause License. See the LICENSE file for more details.

ir2vec's People

Contributors

svkeerthy avatar anilavakundu avatar rohitaggarwal007 avatar sid18996 avatar chriscummins avatar actions-user avatar shikharjain97 avatar

Recommend Projects

  • React photo React

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

  • Vue.js photo Vue.js

    🖖 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. 📊📈🎉

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 ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.