Re: [PATCH 4/4] x86 gart: rename symbols only used for the GART implementation

[Andi Kleen <ak@suse.de>]

On Tuesday 23 October 2007 20:04:36 H. Peter Anvin wrote:
> Dave Jones wrote:
> > On Tue, Oct 23, 2007 at 07:47:11PM +0200, Joerg Roedel wrote:
> >  > On Tue, Oct 23, 2007 at 07:43:47PM +0200, Andi Kleen wrote:
> >  > > On Tuesday 23 October 2007 19:41:33 Joerg Roedel wrote:
> >  > > > This patch renames the 4 symbols iommu_hole_init(), iommu_aperture,
> >  > > > iommu_aperture_allowed, iommu_aperture_disabled.
> >  > > > 
> >  > > > It replaces the iommu_ with gart_ in the symbol name. All these symbols are
> >  > > > only used for the GART implementation of IOMMUs.
> >  > > 
> >  > > That makes it still potentially conflicting with the AGP GART code.
> >  > 
> >  > Maybe yes. But the AGP GART driver conflicts with the GART IOMMU config
> >  > option. So I don't see a problem here.
> >  
> > It shouldn't.  It's perfectly feasible to use both IOMMU and GART for AGP at
> > the same time.
> > 
> 
> And even if did, it's a big difference between being not able to compile 
> in two drivers in the same kernel and not being able to use them at the 
> same time.

You can already compile them in all. The functions are not really
conflicting -- otherwise allyesconfig would not work.

I think his goal was to get an prefix that describes the module uniquely.
gart_* clearly does not fulfill that criteria.

So basically he's replacing an ambigious-with-other-IOMMU-implementations
prefix with an ambigious-with-AGP prefix. Seems like a rather pointless
change.

-And