Benchmark: timing 1000000 iterations of iterative_c, iterative_p, recursive_c, recursive_p...
iterative_c:  1 wallclock secs ( 1.10 usr +  0.00 sys =  1.10 CPU) @ 907441.02/s (n=1000000)
iterative_p: 19 wallclock secs (19.52 usr +  0.00 sys = 19.52 CPU) @ 51234.76/s (n=1000000)
recursive_c:  1 wallclock secs ( 1.16 usr +  0.00 sys =  1.16 CPU) @ 860585.20/s (n=1000000)
recursive_p: 36 wallclock secs (36.26 usr +  0.00 sys = 36.26 CPU) @ 27576.32/s (n=1000000)
                Rate recursive_p iterative_p recursive_c iterative_c
recursive_p  27576/s          --        -46%        -97%        -97%
iterative_p  51235/s         86%          --        -94%        -94%
recursive_c 860585/s       3021%       1580%          --         -5%
iterative_c 907441/s       3191%       1671%          5%          --