[libvirt] [PATCH 1/1] cpu: Fix one compile error for PPC.

[Jiri Denemark]

On Mon, Jul 29, 2013 at 13:23:23 +0200, Jiri Denemark wrote:
> On Mon, Jul 29, 2013 at 15:41:19 +0800, Li Zhang wrote:
> > From: Li Zhang <zhlcindy at linux.vnet.ibm.com>
> > 
> > CPU data structure is refined, which causes one compile error for PPC.
> > 
> > Signed-off-by: Li Zhang <zhlcindy at linux.vnet.ibm.com>
> > ---
> >  src/cpu/cpu_powerpc.c | 8 ++++----
> >  1 file changed, 4 insertions(+), 4 deletions(-)
> > 
> > diff --git a/src/cpu/cpu_powerpc.c b/src/cpu/cpu_powerpc.c
> > index 62437d3..55a4153 100644
> > --- a/src/cpu/cpu_powerpc.c
> > +++ b/src/cpu/cpu_powerpc.c
> > @@ -354,15 +354,15 @@ ppcDataFree(virCPUDataPtr data)
> >  static virCPUDataPtr
> >  ppcNodeData(void)
> >  {
> > -    virCPUDataPtr data;
> > +    virCPUDataPtr cpuData;
> >  
> > -    if (VIR_ALLOC(data) < 0)
> > +    if (VIR_ALLOC(cpuData) < 0)
> >          return NULL;
> >  
> >      asm("mfpvr %0"
> > -        : "=r" (data->ppc.pvr));
> > +        : "=r" (cpuData->data.ppc.pvr));
> >  
> > -    return data;
> > +    return cpuData;
> >  }
> >  #endif
> 
> I'm fine with this change but what compile error does this fix? The only
> error I could think of is shadowing a global "data" symbol but I doubt
> that's the case as the same structure is called "data" in several other
> functions in cpu_powerpc.c

Oh, I see it now. The bug was that data->ppc.pvr should have been
data->data.ppc.pvr. This makes more sense now... I blame the hot
weather.

Jirka