Alexandru E. Şuşu. A Vector-Length Agnostic Compiler for the Memory Limited Connex-S Accelerator, ACM Transactions on Embedded Computing Systems (TECS), vol. 19, no. 6, art. 51, pag. 1-30, Oct 2020 (WOS:000595600300011), PDF Download Webpage (paper with small corrections after publication here)
An 8 pages paper: Alexandru E. Şuşu, Compiling Efficiently with Arithmetic Emulation for the Custom-Width Connex Vector Processor, Proceedings of the ACM Workshop on Programming Models for SIMD/Vector Processing (WPMVP), art. 1, pag. 1-8, 2019, Washington, DC, USA, ISBN 978-1-4503-6291-7 (WOS), PDF (local PDF)
Alexandru E. Şuşu, “Vector-Length Agnostic Compilation for the Connex-S Embedded Wide (Scalable) Vector Accelerator", PhD thesis, 2021 - PDF