Re: [PATCH] ssb: fix pcicore build breakage

["Michael Büsch" <m@bues.ch>]

On Fri, 2011-05-13 at 15:43 +0200, Michael Büsch wrote: 
> On Fri, 2011-05-13 at 09:26 -0400, John W. Linville wrote: 
> > drivers/ssb/main.c:1336: error: 'SSB_PCICORE_BCAST_ADDR' undeclared (first use in this function)
> > drivers/ssb/main.c:1337: error: 'SSB_PCICORE_BCAST_DATA' undeclared (first use in this function)
> > drivers/ssb/main.c:1349: error: 'struct ssb_pcicore' has no member named 'dev'
> > 
> > Reported-by: Randy Dunlap <randy.dunlap@oracle.com>
> > Signed-off-by: John W. Linville <linville@tuxdriver.com>
> > ---
> >  drivers/ssb/main.c |    2 ++
> >  1 files changed, 2 insertions(+), 0 deletions(-)
> > 
> > diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c
> > index ee2937c..a53fc4d 100644
> > --- a/drivers/ssb/main.c
> > +++ b/drivers/ssb/main.c
> > @@ -1332,6 +1332,7 @@ EXPORT_SYMBOL(ssb_bus_powerup);
> >  static void ssb_broadcast_value(struct ssb_device *dev,
> >  				u32 address, u32 data)
> >  {
> > +#ifdef CONFIG_SSB_DRIVER_PCICORE
> >  	/* This is used for both, PCI and ChipCommon core, so be careful. */
> >  	BUILD_BUG_ON(SSB_PCICORE_BCAST_ADDR != SSB_CHIPCO_BCAST_ADDR);
> >  	BUILD_BUG_ON(SSB_PCICORE_BCAST_DATA != SSB_CHIPCO_BCAST_DATA);
> > @@ -1340,6 +1341,7 @@ static void ssb_broadcast_value(struct ssb_device *dev,
> >  	ssb_read32(dev, SSB_PCICORE_BCAST_ADDR); /* flush */
> >  	ssb_write32(dev, SSB_PCICORE_BCAST_DATA, data);
> >  	ssb_read32(dev, SSB_PCICORE_BCAST_DATA); /* flush */
> > +#endif
> >  }
> >  
> >  void ssb_commit_settings(struct ssb_bus *bus)
> 
> This clearly is incorrect. Read the comment, please.

I think we should probably have two ssb_broadcast_value functions. One
for PCI core and one for chipcommon. That seems the best solution, even
if the constants happen to be the same. It's only a few lines of code.
(If the compiler is clever enough, it might probably be able to optimize
it out anyway).

-- 
Greetings Michael.