UNTANGLE -- Unlocking Routing and Logic Obfuscation Using Graph Neural Networks-based Link Prediction

Lilas Alrahis, Satwik Patnaik, Muhammad Abdullah Hanif, Muhammad Shafique, and Ozgur Sinanoglu

Contact

About

UNTANGLE is a link prediction-based attack on logic locking. This repository contains the python implementation of UNTANGLE attack in addition to the implementation of random MUX locking and InterLock.

Setup

Step 1: Install the default GNN model (i.e., DGCNN)

$ git clone https://github.com/muhanzhang/pytorch_DGCNN
$ cd pytorch_DGCNN/lib
$ make -j4
$ cd ../..

Step 2: Install Required Packages

Install PyTorch
Install numpy, scipy, networkx, tqdm, sklearn, gensim

Usage

Attacking Random Mux Locking

1) Lock a design

First: Modify line 8 in ./prepare_datasets/perl_scripts/MUX_random_lock.pl and place the full path to theCircuit.pm (This is done only once)
Example, lock the c7552 ISCAS benchmark with key size of 256

$ cd ./prepare_datasets/perl_scripts/
$ perl MUX_random_lock.pl -k 256 -i ../test_c7552/ > log.txt

MUX_random_lock.pl is a Perl script that reads a circuit in Bench format and locks it using 2-input MUXes. It will convert the design into a graph. It assigns unique numerical IDs (0 to N-1) to the nodes (gates). N represents the total number of nodes (gates) in the design.
-k flag specifies the desired key size.
It will generate a directory ../../data/c7552_MUX_K256 which includes:
- The extracted features will be dumped in feat.txt. The ith line in feat.txt represent the feature vector of the node ID = the ith line in count.txt
- The existence of an edge i between two vertices u and v is represented by the entry of ith line in links_train.txt
- The links_test.txt and link_test_n.txt are created to identify the edges exclusive to the testing set. links_test.txt includes all the true MUX connections while link_test_n.txt includes all the false MUX connections
- The cell.txt file includes the mapping between node IDs and gate instances
- The locked_MUX_2_K_256_c7552.bench file represents the locked circuit

2) Train UNTANGLE

$ cd ../../
$ python Main.py --file-name c7552_MUX_K256 --train-name links_train.txt  --test-name links_test.txt --testneg-name link_test_n.txt --hop 2  --save-model > Log_train_c7552_MUX_K256.txt

3) Get the predictions

$ python Main.py  --file-name c7552_MUX_K256 --train-name links_train.txt  --test-name links_test.txt --hop 2  --only-predict > Log_pos_predict_c7552_MUX_K256.txt
$ python Main.py  --file-name c7552_MUX_K256 --train-name links_train.txt  --test-name  link_test_n.txt --hop 2  --only-predict > Log_neg_predict_c7552_MUX_K256.txt

The likelihoods for the links will be dumped in links_test_2__pred.txt and link_test_n_2__pred.txt. Where 2 represents the hop size

4) Parse the predictions

$perl break_MUX.pl c7552_MUX_K256

Citation & Acknowledgement

If you find the code useful, please cite our paper:

ICCAD 2021:

@INPROCEEDINGS{untangle,
  author={Alrahis, Lilas and Patnaik, Satwik and Hanif, Muhammad Abdullah and Shafique, Muhammad and Sinanoglu, Ozgur},
  booktitle={2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD)}, 
  title={UNTANGLE: Unlocking Routing and Logic Obfuscation Using Graph Neural Networks-based Link Prediction}, 
  year={2021},
  volume={},
  number={},
  pages={1-9},
  doi={10.1109/ICCAD51958.2021.9643476}}

We owe many thanks to Muhan Zhang for making his SEAL code available.

lilasrahis/UNTANGLE