Re: memory strangeness (fwd)

On Thu, 4 Jul 2002, Tom Lane wrote:

> Curt Sampson <cjs(at)cynic(dot)net> writes:
> > Shared memory pages, IIRC, are locked, meaning that they cannot be
> > swapped.
>
> Is that really how it works on *BSD? That's great if so --- it's
> exactly what Postgres wants --- but you'll pardon my paranoia about
> assuming that any part of the universe works the way I want it to...

Well, note the IIRC. :-)

But as it happens, I'm wrong. (I was probably remembering the old
behaviour, from before the new VM system in NetBSD.)

Now, under NetBSD, the memory segment is an anonymous memory object
that is indeed pagable, so anybody allocating tons and tons of SysV
shared memory to postgres on a NetBSD box is going very, very wrong
indeed. FreeBSD uses pagable as well, though it appears that you can
set a kern.ipc.shm_use_phys sysctl to a value other than 0 to make it
allocate non-pagable memory to all future requests. (Segments already
created will still be pagable.)

> You don't have to think hard to see that this
> would be a dead loss for Postgres' shared disk buffers.

Quite so.

cjs
--
Curt Sampson <cjs(at)cynic(dot)net> +81 90 7737 2974 http://www.netbsd.org
Don't you know, in this new Dark Age, we're all light. --XTC