Re: [PATCH] PCI Hotplug: fakephp: fix deadlock... again

[Jesse Barnes <jbarnes@virtuousgeek.org>]

On Monday, September 01, 2008 5:19 pm Alex Chiang wrote:
> * Matthew Wilcox <matthew@wil.cx>:
> > On Mon, Sep 01, 2008 at 12:40:18PM -0600, Alex Chiang wrote:
> > > We get the "slot already scheduled for removal" because that
> > > particular device has 2 functions, and we're creating slots
> > > on a per-slot basis now, not a per-function basis.
> > >
> > > Although, I wonder, Willy -- is that really the right thing
> > > to do? Seems like fakephp would be more useful if we did
> > > operate on a per-function basis, and not per-slot. Especially
> > > given Yu's work with SR-IOV, where we can apparently have
> > > lots of functions per a physical device.
> >
> > I suspect it depends on what you believe the point of fakephp
> > is.
>
> Ok, this was all developed before I started working in this area,
> or Linux, even. ;)
>
> > My assumption was that it was a way to fake what would happen
> > if you had a hotplug controller for a particular slot.  In that
> > context, the change I made was clearly correct.  If you want to
> > use it for hot-removing individual functions from a Linux guest
> > running under a hypervisor (for example), that's much less
> > useful.
>
> Sure, that sounds reasonable.
>
> In that case, my patch should fix the stupid regression I
> introduced, but if others think that fakephp would be more useful
> for hot-removing functions, I wouldn't object to reverting the
> original patch.

I don't have a preference here; whatever is most useful for people is probably 
what we should go for.  Zhao?

Thanks,
Jesse