A user's reflections on the art of high level synthesis

Yu Rong Pan, Santhosh Kumar Pilakkat, Kay-Chuan Benny Tan, Wai-Meng Mok · 2014

“It worked perfectly on the demo programs, but just doesn't work for slightly complicated code”. One of our researchers commented this on his previous experience with High Level Synthesis (HLS). On the other hand, when I first preached to one of our hardware engineers to use HLS, he was immediately sold after I told him that HLS can perform automatic pipelining. After decades of academic and industrial research and development, and several waves of commercial attempts, robust HLS tool flows that are able to generate high performance hardware blocks, perform rapid design space exploration and ease design verifications are finally upon us. While universities and major industrial companies begin to adopt and incorporate HLS into their standard design repertoire [1], many more have not been aware of or experienced this paradigm shift. In this talk, we argue that it is time to start HLS, and share our experience and reflections from our recent first hand close contact with state-of-the-art commercial HLS tools in our product development cycle.

Read the paper · More papers on PaperTik