Re: pgbench - add pseudo-random permutation function

Hello Dean,

> OK, attached is an update making this change and simplifying the rotate
> code, which hopefully just leaves the question of what (if anything) to
> do with pg_erand48().

Yep. While looking at it, I have some doubts on this part:

m = (uint64) (pg_erand48(random_state.xseed) * (mask + 1)) | 1;
r = (uint64) (pg_erand48(random_state.xseed) * (mask + 1));
r = (uint64) (pg_erand48(random_state.xseed) * size);

I do not understand why the random values are multiplied by anything in
the first place…

This one looks like a no-op :

r = (uint64) (pg_erand48(random_state.xseed) * size);
v = (v + r) % size;

v = (v + r) % size
= (v + rand * size) % size
=? (v % size + rand * size % size) % size
=? (v % size + 0) % size
= v % size
= v

I'm also skeptical about this one:

r = (uint64) (pg_erand48(random_state.xseed) * (mask + 1));
if (v <= mask)
v = ((v * m) ^ r) & mask;

v = ((v * m) ^ r) & mask
= ((v * m) ^ r) % (mask+1)
= ((v * m) ^ (rand * (mask+1))) % (mask+1)
=? ((v * m) % (mask+1)) ^ (rand * (mask+1) % (mask+1))
=? ((v * m) % (mask+1)) ^ (0)
= (v * m) & mask

Or possibly I'm missing something obvious and I'm wrong with my
arithmetic?

--
Fabien.