> BTW, if you're wondering why curculio is still failing the pgbench
> test,
Hmmm, that is interesting! It shows that at least some TAP tests are
useful.
> all is explained here:
>
> https://man.openbsd.org/srandom
>
> Or at least most is explained there.
Yep. They try to be more serious than other systems about PRNG, which is
not bad in itself.
> While curculio is unsurprisingly failing all four seeded_random tests,
> when I try it locally on an OpenBSD 6.4 installation, only the uniform,
> exponential, and gaussian cases reliably "fail". zipfian usually
> doesn't.
> It looks like the zipfian code almost always produces 4000 regardless of
> the seed value, though occasionally it produces 4001. Bad parameters
> for that algorithm, perhaps?
Welcome to the zipfian highly skewed distribution! I'll check the
parameters used in the test, maybe it should use something less extreme.
srandom is only used for initializing the state of various internal rand48
LCG PRNG for pgbench.
Maybe on OpenBSD pg should switch srandom to srandom_deterministic?
--
Fabien.