[PATCH 110/112] staging: comedi: vmk80xx: remove subdevice pointer math

[Greg KH]

On Thu, Sep 06, 2012 at 11:16:18AM -0700, Greg KH wrote:
> On Thu, Sep 06, 2012 at 12:37:28PM -0500, H Hartley Sweeten wrote:
> > On Wednesday, September 05, 2012 8:07 PM, Greg KH wrote:
> > > On Wed, Sep 05, 2012 at 06:59:09PM -0700, H Hartley Sweeten wrote:
> > >> Convert the comedi_subdevice access from pointer math to array
> > >> access.
> > >> 
> > >> Signed-off-by: H Hartley Sweeten <hsweeten at visionengravers.com>
> > >
> > > This patch causes a build warning.  Care to send a follow-up patch that
> > > fixes it?
> > 
> > Greg,
> > 
> > What's the build warning you get?
> > 
> > I tried building this again, both with and without the patch, and don't
> > see any warnings for this file.
> 
> I get:
> 
>   CC [M]  drivers/staging/comedi/drivers/vmk80xx.o
> drivers/staging/comedi/drivers/vmk80xx.c: In function ‘vmk80xx_cnt_rinsn’:
> drivers/staging/comedi/drivers/vmk80xx.c:889:6: warning: ‘reg[0]’ may be used uninitialized in this function
> drivers/staging/comedi/drivers/vmk80xx.c: In function ‘vmk80xx_ai_rinsn’:
> drivers/staging/comedi/drivers/vmk80xx.c:531:6: warning: ‘reg[0]’ may be used uninitialized in this function
> 
> But, in looking at your patch, I'm doubting that your patch caused this
> problem, in fact, it was an old version of gcc (4.5.3), after updating
> the system to a modern gcc (4.6.3), it's gone.

Hm, no, a newer gcc doesn't fix this up, but it's still not caused by
your patch, let me go try to resolve it.

greg k-h