Re: atomic_ops(3) and sync

[Manuel Bouyer <bouyer%antioche.eu.org@localhost>]

On Wed, Jun 16, 2010 at 12:54:35PM +0300, Antti Kantee wrote:
> On Wed Jun 16 2010 at 11:35:05 +0200, Manuel Bouyer wrote:
> > hi,
> > I just noticed that our atomic_ops(3) don't include a sync instruction.
> > >From what I've seen in the mips documents, neither ll nor sc is a memory
> > barrier, so our atomic_ops are no memory barrier either, and callers
> > have to add a sync instruction before atomic calls.
> > Is this the desired behavior, or should a sync be added in atomic_ops ?
> 
> This behaviour is documented in atomic_ops(3) (your email does not
> provide clear indication as to if you read that part of the manpage,
> so I'll just quote it here):
> 
>     Atomic operations are strongly ordered with respect to each other.
>     The global visibility of other loads and stores before and after an
>     atomic operation is undefined.  Applications that require synchro-
>     nization of loads and stores with respect to an atomic operation must
>     use memory barriers.  See membar_ops(3).
> 
> So if you want to impose ordering on other accesses with respect to the
> atomic op you need to manually add membars.  Atomic itself does not,
> and should not, guarantee it.

OK, thanks. I didn't spot this from the man page. I'm still in
a x86-centric world I guess ...

-- 
Manuel Bouyer <bouyer%antioche.eu.org@localhost>
     NetBSD: 26 ans d'experience feront toujours la difference
--