Fast polynomial multiplications using AVX2. Based on work by Zhenfei Zhang, ported to int32_t by CapacitorSet, GPLv3 licensed.
Usage: make.
It was used to experiment alternative implementations of negacyclic convolution in TFHE; because the experiment proved unfruitful (3x slower than FFT for int32 polynomials with degree 1000), this project will no longer be developed.