Re: asm-generic: Disallow no-op mb() for SMP systems

From: Peter Zijlstra
Date: Thu Feb 01 2018 - 08:53:48 EST


On Thu, Feb 01, 2018 at 01:32:30PM +0000, Will Deacon wrote:
> On Thu, Feb 01, 2018 at 02:29:09PM +0100, Peter Zijlstra wrote:
> > On Thu, Feb 01, 2018 at 09:27:50PM +0900, Stafford Horne wrote:
> > > I tried to clarify some of this in the spec v1.2 [0] which help formalize some of
> > > the techniques we used for the SMP implementation. Its probably not perfect,
> > > but I added a section "10. Multicore support" and tried to clarify some things
> > > in section 7 on Atomicity. But it seems I dont cover exactly what are are
> > > mentioning here. In general:
> > >
> > > 1 Secondary cores have memory snooping enabled meaning that any write to a
> > > cached address will cause the cache line to be invalidated.
> > > 2 l.swa (store atomic word) implies a store buffer flush.
> >
> > What about l.lwa? Can that observe 'old' values, or rather, miss values
> > stuck in a remote store buffer?
> >
> > This will then cause the first l.swa to fail, which, per the above,
> > would then sync things up? Which means you get that one extra
> > merry-go-round.
>
> That's ok from a correctness perspective, though, as long as store buffers
> are guaranteed to drain.

Depends a bit if you can build control dependencies off of l.swa
succeding or not I think :-) Otherwise you get into that dodgy state you
suffer from where bits can leak right through.

That is, I was thinking what we need for smp_mb__before_atomic.

I could've gotten my brain in a twist or course, which isn't _that_
unusual. I never seem to be able to quite remember the holes you have
with ll/sc on arm64 :-)