Tested by various people on various platforms, I'm willing to fix any
breakage this causes.
ok niels@ deraadt@ and mickey@ (after his comments were applied)

make -> ${MAKE}
add phony targets
remove unnecessary subshells
install kernel in a separate target from building.
some SUDO.
okay deraadt@
(checked bootstrap on i386, should be safe elsewhere, and trivial to fix
anyways)

No changes in the generated output, so no regen needed.
ok todd@

crashes sh, EVEN after that possible fix.  It appears to overflow the
ramdisk _before_ it dumps core.  Noone was shown these changes.  They were
just tossed in.  This is not a development process that will work, so I
am just backing out all the changes in the last 24 hours.

order a few places
provide '[ "$DEBUG" ] && set -x' inside R() for debugging help ..

magma and spif were moved into MAKEDEV.mi, remove from here

ok jason@