[lkml]   [2009]   [Feb]   [18]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: Q: smp.c && barriers (Was: [PATCH 1/4] generic-smp: remove single ipi fallback for smp_call_function_many())
On Wed, Feb 18, 2009 at 08:09:21AM -0800, Linus Torvalds wrote:
> On Wed, 18 Feb 2009, Nick Piggin wrote:
> >
> > I agree with you both that we *should* make arch interrupt code
> > do the ordering, but given the subtle lockups on some architectures
> > in this new code, I didn't want to make it significantly weaker...
> >
> > Though perhaps it appears that I have, if I have removed an smp_mb
> > that x86 was relying on to emit an mfence to serialise the apic.
> The thing is, if the architecture doesn't order IPI wrt cache coherency,
> then the "smp_mb()" doesn't really do so _either_.

Oh yes agreed three, which is why I'm saying it is just a hack
and should be removed at some point.

> It might hide some architecture-specific implementation issue, of course,
> so random amounts of "smp_mb()"s sprinkled around might well make some
> architecture "work", but it's in no way guaranteed. A smp_mb() does not
> guarantee that some separate IPI network is ordered - that may well take
> some random machine-specific IO cycle.

Yes, but I didn't want to pull out that smp_mb() at least until
arch maintainers can ack it. Just because there might indeed be
some random issue hidden by it.

 \ /
  Last update: 2009-02-19 01:15    [W:0.210 / U:3.532 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site