[cpp-threads] High-level vs. low-level

[Peter Dimov]

Lawrence Crowl wrote:
> On 7/13/06, Peter Dimov <pdimov at mmltd.net> wrote:
>> I'm not sure that a sequentially consistent model can achieve the
>> above "within 5%" performance goal. On x86, for example, all stores
>> would need to be fully locked (lock xchg).
>
> Do you mean just the volatiles?

Yes, I mean the atomic<>s (the C++ equivalent of the Java volatile under the 
current proposal.)