Questions about the performance reported in the paper (DAC'18)

[Dai-dirk]

Hi,

I'm interested in the performance reported in Tables 2 and 3, but I don't know how it was obtained. I've read the source codes of PULP_HDC/src/main.c, I guess the performance is the #cycle when the outermost loop iterates once (i.e., ix + 1), am I right? Looking for your guideness~

Best regards,

Yuan