[cpp-threads] Should atomic_add support pointers?

[Andrei Alexandrescu]

I think it should; on some systems it's an intrinsic, on others it can 
be implemented with CAS. It's a useful operation, and good to let the 
library implement it.

Andrei

Peter Dimov wrote:
> I'm trying to nail down the semantics of an atomic ops library and I 
> encountered the following interesting issue:
> 
> Should atomic_add support pointer addition? That is:
> 
> int * p;
> 
> atomic_add( &p, 4 ); // allowed? sensible?
> 
> What do you think?
> 
> -- 
> Peter Dimov
> http://www.pdimov.com
>