digital logic - How to write a for-generate statement to generate multiple instances of this parallel sorting network circuit?

I'm trying to implement a parallel sorting network. I'm curious to how one would write a for-generate statement to generate 128 instances of this sorting circuit?

It's supposed to sort 256 8-bit numbers from least to greatest without a clock using only combinational logic.

This is a link to the paper on the sorting network I'm trying to use:

https://www.researchgate.net/publication/261152327_Zero-delay_FPGA-based_odd-even_sorting_network?enrichId=rgreq-27e12d3d7cce338ad2cbbdffa750edb4-XXX&enrichSource=Y292ZXJQYWdlOzI2MTE1MjMyNztBUzoyMzg2MDc5Njk1NTAzNDNAMTQzMzg5OTk4MTg4Nw%3D%3D&el=1_x_3&_esc=publicationCoverPdf

Links to the circuit and sorting network diagram are below.