Re: smp.c && barriers (Was: [PATCH 1/4] generic-smp: remove singleipi fallback for smp_call_function_many())

[Ingo Molnar]

* Nick Piggin <npiggin@xxxxxxx> wrote:

> On Thu, Feb 19, 2009 at 01:20:31PM +0100, Ingo Molnar wrote:
> > 
> > * Suresh Siddha <suresh.b.siddha@xxxxxxxxx> wrote:
> > 
> > > On Wed, 2009-02-18 at 11:17 -0800, Ingo Molnar wrote: 
> > > > * Suresh Siddha <suresh.b.siddha@xxxxxxxxx> wrote:
> > > > So what should happen is to move that smp_mb() from the x86 
> > > > generic IPI path to the x86 x2apic IPI path. (and turn it into 
> > > > an smp_wmb() - that should be enough - we dont care about future 
> > > > reads being done sooner than this point.)
> > > 
> > > Ingo, smp_wmb() won't help. x2apic register writes can still 
> > > go ahead of the sfence. According to the SDM, we need a 
> > > serializing instruction or mfence. Our internal experiments 
> > > also proved this.
> > 
> > ah, yes - i got confused about how an x2apic write can pass a 
> > _store_ fence.
> 
> And about how smp_wmb() doesn't emit a store fence ;)

yeah ;-) Only wmb() emits a SFENCE all the time. Writes are 
normally ordered so we map smp_wmb() to barrier().

	Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/