Re: [RFC] fix asus_hides_smbus_lpc_ich6() for resume

[Jesse Barnes <jbarnes@virtuousgeek.org>]

On Friday, March 14, 2008 7:37 am Rafael J. Wysocki wrote:
> On Friday, 14 of March 2008, Shaohua Li wrote:
> > On Thu, 2008-03-13 at 21:44 -0700, Greg KH wrote:
> > > On Fri, Mar 14, 2008 at 11:49:09AM +0800, Shaohua Li wrote:
> > > > diff --git a/include/linux/pci.h b/include/linux/pci.h
> > > > index 9010f54..821ad02 100644
> > > > --- a/include/linux/pci.h
> > > > +++ b/include/linux/pci.h
> > > > @@ -1016,6 +1016,8 @@ enum pci_fixup_pass {
> > > >  	pci_fixup_final,	/* Final phase of device fixups */
> > > >  	pci_fixup_enable,	/* pci_enable_device() time */
> > > >  	pci_fixup_resume,	/* pci_enable_device() time */
> > > > +	pci_fixup_suspend,
> > > > +	pci_fixup_resume_later,
> > >
> > > Please document when these quirks are run.
> >
> > will do if you think the method is correct.
> >
> > > And why "_later"?  What's wrong with the normal resume time?
> >
> > pci_fixup_resume is called in pci_device_resume_earily(), which is
> > called with interrupt disable, so I added a new one which is called at
> > pci_device_resume(), this routine is called with interrupt enabled.
>
> Is the pci_fixup_resume thing used at all?  If it is, how can I check who
> uses it?

Oh good catch, it looks like it's unused at this point (at least there's no 
DECLARE_PCI_FIXUP_SUSPEND macro in pci.h), it was probably just put there for 
completeness.

Jesse