100 000 cycles:

implementation:    OpenSSL   Naive (4)   With macro (5a)   With T-Box (5b)   Intrinsics (6)
total time:      6.74264ms   23.6295ms          23.494ms         8.70669ms        1.18704ms
avg time:        67.4264ns   236.295ns          234.94ns         87.0669ns        11.8704ns
avg CPU cycles:        255         896               891               330               45



1 000 000 cycles:

implementation:    OpenSSL   Naive (4)   With macro (5a)   With T-Box (5b)   Intrinsics (6)
total time:       65.397ms   23.6295ms          233.56ms         88.0696ms        11.8984ms
avg time:         65.397ns   236.295ns          233.56ns         88.0696ns        11.8984ns
avg CPU cycles:        248         896               885               334               45