Re: s_lock.h default definitions are rather confused

I wrote:
> I've launched a run now, expect results from gcc HEAD in an hour and
> a half or so.

... and it's happy.  Thanks!

BTW, the reason I went to the trouble of cranking up the buildfarm scripts
on that machine (and it was painful :-() is that I don't believe any other
buildfarm members are running compilers old enough to complain about some
of the things these will.  In particular:

* I've got gaur configured so it will throw "array subscript of type char"
complaints whenever somebody forgets to cast a <ctype.h> function argument
to unsigned char.

* pademelon will complain about // comments, variable-sized local arrays,
flexible array syntax, non-static function definition after static
declaration, and probably some other C89 violations that I am not
remembering right now.

While I'll not cry too hard when we decide to break C89 compatibility,
I don't want it to happen accidentally; so having a pretty old-school
compiler in the farm seems important to me.
        regards, tom lane