This repository aims at providing all the necessary stuff for benchmarking RIR and compare it to other (language) implementations.
We use a harness for benchmarking complying with the format of the are-we-fast suite.
To make the benchmarking reproducible we resort on ReBench.
Setup/run.sh rebench.conf Benchmarks path/to/rir path/to/gnur path/to/fastr e:EXPERIMENT -S
Tested experiments are PIR-LLVM, GNU-R and FASTR.