Re: [PATCH 01/90] staging: comedi: comedidev.h: introduce comedi_to_pci_dev() helper

[Dan Carpenter <dan.carpenter@oracle.com>]

On Thu, Jul 19, 2012 at 11:20:52AM -0500, H Hartley Sweeten wrote:
> On Thursday, July 19, 2012 2:23 AM, Ian Abbott wrote:
> > On 2012-07-19 02:24, H Hartley Sweeten wrote:
> >> Introduce a wrapper for to_pci_dev() to allow the comedi pci drivers
> >> to store the pci_dev pointer in the comedi_device hw_dev variable and
> >> retrieve it easily.
> >>
> >> Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
> >> Cc: Ian Abbott <abbotti@mev.co.uk>
> >> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> >> ---
> >>   drivers/staging/comedi/comedidev.h | 5 +++++
> >>   1 file changed, 5 insertions(+)
> >>
> >> diff --git a/drivers/staging/comedi/comedidev.h b/drivers/staging/comedi/comedidev.h
> >> index de8c99c..620222d 100644
> >> --- a/drivers/staging/comedi/comedidev.h
> >> +++ b/drivers/staging/comedi/comedidev.h
> >> @@ -446,6 +446,11 @@ static inline void comedi_set_hw_dev(struct comedi_device *dev,
> >>   	}
> >>   }
> >>
> >> +static inline struct pci_dev *comedi_to_pci_dev(struct comedi_device *dev)
> >> +{
> >> +	return to_pci_dev(dev->hw_dev);
> >> +}
> >> +
> >
> > That needs to be something like:
> >
> >	return dev->hw_dev ? to_pci_dev(dev->hw_dev) : NULL;
> 
> Hmm.. I'm not really sure.
> 
> I assumed that the container_of() macro would return NULL if the ptr passed to
> it was NULL. But, I'm not sure how this actually unwinds for that case.
> 
> Greg, do you know if the NULL check is needed? It's is possible that the dev->hw_dev
> pointer could be NULL.
> 

container_of() just does pointer math with the offset.  Since
->hw_dev is not the first member of the pci_dev struct then
to_pci_dev() never returns NULL.  If you give it a NULL pointer it
returns a bogus pointer back.

regards,
dan carpenter