Scripts for building and running benchmarks with programs that run on multiple threads and may contain data races.

Usage

1. Run ./setup.sh to build all the benchmark programs.

   • Before running the script, set the CUSTOM_LLVM_BUILD_PATH environment variable to the path to your custom LLVM build.
   • E.g. export CUSTOM_LLVM_BUILD_PATH=/home/daniel/llvm-project/build
   • It should take less than 5 minutes to build everything.

2. Run python3 runner.py to run the test cases. It loads a list of test cases from testcases.yml.

   • You may provide arguments to choose the test set and number of iterations for each test case.
   • Usage: python3 runner.py [test set name] [number of iterations]
   • E.g. python3 runner.py mini 20

3. Results are in report.csv.

4. Modify testcases.yml if you want to add/remove test cases.