into one if a system has an awesome getentropy().  In that case it
is valid to totally throw away the rsx state in the child.  If the
getentropy() is not very good and has a lazy reseed operation, this
combining is a bad idea, and the reseed should probably continue to
use the "something old, something new" mix.  _rs_allocate() can
accomodate either method, but not on the fly.
ok matthew

1. Use "len" parameter instead of sizeof(*rs).
2. Simplify the atfork handler to be strictly async signal safe by
simply writing to a global volatile sig_atomic_t object, and then
checking for this in _rs_forkdetect().  (Idea from discussions with
Szabolcs Nagy and Rich Felker.)
3. Use memset(rs, 0, sizeof(*rs)) to match OpenBSD's MAP_INHERIT_ZERO
fork semantics to avoid any skew in behavior across platforms.
ok deraadt

we no longer need the spread of dev/log AF_UNIX sockets all over the
various chroot spaces.
ok beck millert aja

/usr/local/share/pkgconfig
ok sthen@

Reminded by Rafael Neves

align with POSIX and other systems.
Pointed out by Elliott Hughes on tech
ok deraadt

making it much easier for libressl -portable to fill in the gaps.
ok bcook beck

and random().  Sigh.

We don't want any of the variables created inside netstart to infect the
rc script.
ok claudio sthen aja

If someone wants to take a shot a modifying this rc script so that it does
not use internal rc.subr functions, be my guest...
spotted by jsg@

address space, and once allocated rs is never deallocated."
document the forkhandler to save reviewers time, with matthew

likely to be more changes here to match the new layout.
ok ingo aja

OK deraadt

OK deraadt

Linux (such as Ubuntu 12.04LTS) that don't have it yet.  Seems the AT_XXX
defines are pulled in by <link.h> now.
ok beck@

in case of catastropy.  But it is so poorly documented that any admin
is more likely to store the labels elsewhere, so let's stop bothering
with providing the directory.  Discussed a bit, no objections.

As discussed a few days ago with aja

them what they are: MUTABLE, as in these are the specific files that
admins do edit in /etc

the existing file can start losing... examples...

there, let's make it operator writeable.  Or let's do a commit changing it
to operator writeable and see the pro and con camps rise up.

files in their installed system.  this extended documentation experience
is available better on the net using a browser installed with pkg_add.
(also note that two of the subsystems involved in this issue are heading
to the bit bucket sometime soon)

ok deraadt@ beck@ kettenis@

seconds values in the announcement timestamps. Damn youngsters.

/etc/services file moved from etc to base so is no longer really
user-editable as changes will be overwritten at update time

ok deraadt@

removed to unconfuse devs
okay aja, theo

explicitely executed by sh(1)).
discussed with deraadt@