Re: [PATCH V3 2/2] mm: Extend memory hotplug API to allow memory hotplug in virtual machines

[Andrew Morton <akpm@linux-foundation.org>]

On Fri, 20 May 2011 01:25:20 +0200 Daniel Kiper <dkiper@net-space.pl> wrote:

> On Thu, May 19, 2011 at 04:01:43PM -0700, Andrew Morton wrote:
> > On Thu, 19 May 2011 22:45:09 +0200
> > Daniel Kiper <dkiper@net-space.pl> wrote:
> >
> > > On Wed, May 18, 2011 at 08:36:02PM -0700, David Rientjes wrote:
> > > > On Tue, 17 May 2011, Daniel Kiper wrote:
> > > >
> > > > > This patch contains online_page_callback and apropriate functions for
> > > > > setting/restoring online page callbacks. It allows to do some machine
> > > > > specific tasks during online page stage which is required to implement
> > > > > memory hotplug in virtual machines. Additionally, __online_page_set_limits(),
> > > > > __online_page_increment_counters() and __online_page_free() function
> > > > > was added to ease generic hotplug operation.
> > > >
> > > > There are several issues with this.
> > > >
> > > > First, this is completely racy and only allows one global callback to be
> > > > in use at a time without looping, which is probably why you had to pass an
> > >
> > > One callback is allowed by design. Currently I do not see
> > > any real usage for more than one callback.
> >
> > I'd suggest that you try using the notifier.h tools here and remove the
> > restriction.  Sure, we may never use the capability but I expect the
> > code will look nice and simple and once it's done, it's done.
> 
> Hmmm... I am a bit confused. Here https://lkml.org/lkml/2011/3/28/510 you
> was against (ab)using notifiers. Here https://lkml.org/lkml/2011/3/29/313
> you proposed currently implemented solution. Maybe I missed something...
> What should I do now ??? I agree that the code should look nice and simple
> and once it's done, it's done.

Oh, OK, the callback's role is to free a page, so there's no sens in
there ever being more than a single registered callback.