conv -- 1D convolution
   -1 -- float, support=full
   -2 -- float, support=same
   -3 -- float, support=min
   -4 -- complex<float>, support=full
   -5 -- complex<float>, support=same
   -6 -- complex<float>, support=min

 Parameters:
  -param N      -- size of coefficient vector (default 16)
  -start N      -- starting problem size 2^N (default 4 or 16 points)
  -loop_start N -- initial number of calibration loops (default 5000)