Re: [PATCH 17/19] locking/barriers: Kill lockless_dereference

[Ingo Molnar]

* Will Deacon <will.deacon@xxxxxxx> wrote:

> On Tue, Oct 24, 2017 at 11:31:04AM +0200, Ingo Molnar wrote:
> > 
> > * Paul E. McKenney <paulmck@xxxxxxxxxxxxxxxxxx> wrote:
> > 
> > > From: Will Deacon <will.deacon@xxxxxxx>
> > > 
> > > lockless_dereference is a nice idea, but its gained little traction in
> > > kernel code since it's introduction three years ago. This is partly
> > 
> > s/its/it
> > s/it's/its
> 
> Crikey, no idea what happened there!

NP, just habitual drive-by typo fixing from me, would have fixed them silently 
when applying the patches, had it not been for the split-up suggestion :-)

> > > because it's a pain to type, but also because using READ_ONCE instead
> > > will work correctly on all architectures apart from Alpha, which is a
> > > fully supported but somewhat niche architecture these days.
> > > 
> > > This patch moves smp_read_barrier_depends() (a NOP on all architectures
> > > other than Alpha) from lockless_dereference into READ_ONCE, converts
> > > the few actual users over to READ_ONCE and then finally removes
> > > lockless_dereference altogether.
> > 
> > Nit: if we refer to smp_read_barrier_depends() with parentheses (which is the nice 
> > thing to do for function-alike symbols), then we should do the same with 
> > READ_ONCE() and lockless_dereference() as well.
> > 
> > Also, could we please split this into three patches:
> > 
> >  #1: Add smp_read_barrier_depends() to READ_ONCE()
> >  #2: Convert all lockless_dereference() users to READ_ONCE()
> >  #3: Remove the now unused lockless_dereference() API
> > 
> > to make it easier to analyze if bisected to, should any problems arise?
> 
> Sure, I'll do that now.

Thanks!

	Ingo