[PATCH v3] Avoid memory barrier in read_seqcount() through load acquire

[Peter Zijlstra]

On Wed, Oct 23, 2024 at 04:42:36PM -0700, Christoph Lameter (Ampere) wrote:
> On Wed, 23 Oct 2024, Peter Zijlstra wrote:
> 
> > > I doubt anybody will notice, and smp_load_acquire() is the future. Any
> > > architecture that does badly on it just doesn't matter (and, as
> > > mentioned, I don't think they even exist - "smp_rmb()" is generally at
> > > least as expensive).
> >
> > Do we want to do the complementing patch and make write_seqcount_end()
> > use smp_store_release() ?
> >
> > I think at least ARM (the 32bit thing) has wmb but uses mb for
> > store_release. But I also think I don't really care about that.
> 
> The proper instruction would be something like
> 
> atomic_inc_release(&seqcount)

It would not be, making the increment itself atomic would make the whole
thing far more expensive.