The Hermit Hacker <scrappy@hub.org> writes:
> Two things against it...
> First, its a Linux-ism...he's got it ported to Win and Linux, that's it...
If it depends on mmap(), porting may not be trivial, either.
Still, if it's a drop-in replacement for malloc, I see no reason that
a particular installation couldn't use it with Postgres --- and we could
even recommend that, if it's a big win. The license conflict would
preclude actually including it with Postgres, but pointing to it is
no problem.
Anyone want to try it and see if it provides a noticeable speed
improvement?
regards, tom lane