[osg-submissions] Use atomic reference counts

[Robert Osfield]

Hi Mathias,

On Wed, Jun 4, 2008 at 10:35 AM, Mathias Fröhlich
<M.Froehlich at science-computing.de> wrote:
> I would normally agree here.
> But the basic trick here is that this *is* *inline*. That means that we do not
> have a function call overhead in some of the implementations. That is the
> gcc/icc one and the sgi one. The sun implementation and the win32
> implementatinon will call library functions anyway. But we can avoid one call
> here.

I was not thinking about moving the implementing into the library,
rather the use of Atomic might eventually be utilised by the library
itself - I'm in agreement w.r.t importance of inlining.

> I have this kind of change, to move that matrix selection into a header in
> preperation also.
> IMO it is *way* better to have a installed library shared object match the
> headers without any commandline define that is required to make it even link
> with the library...
>
> So all in all I would vote for such a Configure file.

I'm open to moving optional configurations used into a single header
file per library i.e. include/osg/Config, and having this written to
from CMake.  The Config.in shouldn't be in the include/* directories
though as this can lead to confusing, I'd suggest perhaps the the src
directory as at least there is one src directory per include.

Thoughts?
Robert.