Modify the Handshaking Protocol (-Xshyper-optimized-handshaking)
- : The default behavior without the option specified. TheautoIntel® oneAPIenables the optimization if it is possible to do so, else it sets to off. Use this value when you want to achieve a higher fDPC++/C++CompilerMAX. When you enable the optimization, theIntel® oneAPIadds pipeline registers to the handshaking paths of the stallable nodes. As a result, you observe higher fDPC++/C++CompilerMAXat the cost of increased area and latency.
- : TheoffIntel® oneAPIattempts to optimize for lower latency at the potential cost of lower fDPC++/C++CompilerMAX. Disabling hyper-optimized handshaking might also decrease area. This is useful for smaller designs where you are willing to give up fMAXfor lower latency and area.
dpcpp -fintelfpga -Xshardware -Xshyper-optimized-handshaking=auto <source_file>.cpp
dpcpp -fintelfpga -Xshardware -Xshyper-optimized-handshaking=off <source_file>.cpp