On Tue, Jul 14, 2015 at 12:38:01PM +0100, Peter Zijlstra wrote: > On Tue, Jul 14, 2015 at 12:31:47PM +0100, Will Deacon wrote: > > #ifndef atomic_add > > #define atomic_add(args...) (void)atomic_add_return_relaxed(args); > > > > It would mean a new architecture only has to define a barrier instruction > > and a handful of relaxed atomics for a bare-minimum atomic.h avoiding > > spinlocks. > > Look at include/asm-generic/atomic.h, all you need is a cmpxchg(). > > We could easily change that to be cmpxchg_relaxed() and a few barriers.
Ok, I'll leave that part for now and implement your original suggestion as a starting point. Will -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
