[cpp-threads] modes
Alexander Terekhov
alexander.terekhov at gmail.com
Sat May 7 22:26:10 BST 2005
On 5/7/05, Peter Dimov <pdimov at mmltd.net> wrote:
> Doug Lea wrote:
>
> > One question is whether distinguishing CAS.none buys you anything vs
> > CAS.rel wrt performance on current or plausible future platforms.
> > Which I think is mainly another empirical question about PPC-smps. (I
> > don't have one around to check either.)
>
> I think that the answer here is almost certainly "yes" (maybe even with
> capital letters), because PPC CAS.rel is CAS.none prefixed with (lw)sync,
Likewise, under (hypothetical) Sparc RMO, CAS.rel is CAS.none
prefixed with LoadStore + StoreStore (or LoadLoad + StoreLoad (or
LoadStore + StoreLoad (or LoadLoad + StoreStore))). I mean that
"gcc/gas" for Sparc RMO can safely spare that membar prefix
instruction in the case of CAS.none for (hypothetical) Sparc RMO.
regards,
alexander.
More information about the cpp-threads
mailing list