Re: [PATCH 11/22] KVM: MMU: Add infrastructure for two-level pagewalker

[Avi Kivity]

On 04/28/2010 01:52 PM, Joerg Roedel wrote:
> On Tue, Apr 27, 2010 at 03:34:10PM +0300, Avi Kivity wrote:
>    
> > On 04/27/2010 01:38 PM, Joerg Roedel wrote:
> >      
> > > This patch introduces a mmu-callback to translate gpa
> > > addresses in the walk_addr code. This is later used to
> > > translate l2_gpa addresses into l1_gpa addresses.
> > >
> > > +static inline gfn_t gpa_to_gfn(gpa_t gpa)
> > > +{
> > > +	return (gfn_t)gpa>>   PAGE_SHIFT;
> > > +}
> > > +
> > >        
> > This overflows on 32-bit, since gpa_t is u64 and gfn_t is ulong.
> >      
> Hm, this is a problem outside of this patchset too (for 32bit hosts).
> The best solution is probably to convert gfn_t to u64 too.
>    

If you cast like

   (gfn_t)(gpa >> PAGE_SHIFT)

you avoid the overflow for MAXPHYADDR < 48.  However, I agree that 
converting gfn_t to u64 is best, the minor performance degradation is in 
no way comparable to the corruption that results from a miscast.

--
error compiling committee.c: too many arguments to function

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/