We weren't particularly maintaining barrier.h before 9.4, because nothing
was using it in those branches.  Well, nothing until commit 37de8de9 got
back-patched.  That broke 9.2 and 9.3 for some non-mainstream platforms
that we haven't been testing in the buildfarm, including icc on ia64,
HPPA, and Alpha.

This commit effectively back-patches commits e5592c61, 89779bf2,
and 747ca669, though I did it just by copying the file (less copyright
date updates) rather than by cherry-picking those commits.

Per an attempt to run gaur and pademelon over old branches they've
not been run on since ~2013.

x86's memory barrier assembly was marked as clobbering "memory" but
not "cc" even though 'addl' sets various flags. As it turns out gcc on
x86 implicitly assumes "cc" on every inline assembler statement, so
it's not a bug. But as that's poorly documented and might get copied
to architectures or compilers where that's not the case, it seems
better to be precise.

Discussion: 20140919100016.GH4277@alap3.anarazel.de

To keep the code common, backpatch to 9.2 where explicit memory
barriers were introduced.

This was not changed in HEAD, but will be done later as part of a
pgindent run.  Future pgindent runs will also do this.

Report by Tom Lane

Backpatch through all supported branches, but not HEAD

commit-fest.

%esp is no good; must use %rsp there.

This is not actually used anywhere yet, but it gets the basic
infrastructure in place.  It is fairly likely that there are bugs, and
support for some important platforms may be missing, so we'll need to
refine this as we go along.