20:18:43phadthaihmm likely a result of improperly optimized ironclad's hashing code... If I used ECL for this I'd likely call openssl or gcrypt directly
20:18:58phadthaibut it's probably a good case to revise/improve ecl's c generation
20:28:40FadeI'm just testing some of my personal systems against various distributions, for portability.
20:28:53Fadebut this is a pretty serious performance regression.
20:29:51Fadethis is at default optimisation settings, so whatever ecl starts up with.