Re: [PATCH] kvm/ppc/booke64: fix build breakage from Altivec, and disable e6500

[Scott Wood <scottwood@freescale.com>]

On 05/10/2013 01:20:33 PM, Caraman Mihai Claudiu-B02008 wrote:
> 
> 
> > -----Original Message-----
> > From: Wood Scott-B07421
> > Sent: Friday, May 10, 2013 8:42 PM
> > To: Caraman Mihai Claudiu-B02008
> > Cc: Alexander Graf; Wood Scott-B07421; kvm-ppc@vger.kernel.org;
> > kvm@vger.kernel.org
> > Subject: Re: [PATCH] kvm/ppc/booke64: fix build breakage from  
> Altivec,
> > and disable e6500
> >
> > On 05/10/2013 09:11:24 AM, Caraman Mihai Claudiu-B02008 wrote:
> > > > -----Original Message-----
> > > > From: Alexander Graf [mailto:agraf@suse.de]
> > > > Sent: Friday, May 10, 2013 4:16 PM
> > > > To: Caraman Mihai Claudiu-B02008
> > > > Cc: Wood Scott-B07421; kvm-ppc@vger.kernel.org;  
> kvm@vger.kernel.org
> > > > Subject: Re: [PATCH] kvm/ppc/booke64: fix build breakage from
> > > Altivec,
> > > > and disable e6500
> > > >
> > > >
> > > > On 10.05.2013, at 11:40, Caraman Mihai Claudiu-B02008 wrote:
> > > >
> > > > >> -----Original Message-----
> > > > >> From: Wood Scott-B07421
> > > > >> Sent: Friday, May 10, 2013 6:15 AM
> > > > >> To: Alexander Graf
> > > > >> Cc: kvm-ppc@vger.kernel.org; kvm@vger.kernel.org; Wood
> > > Scott-B07421;
> > > > >> Caraman Mihai Claudiu-B02008
> > > > >> Subject: [PATCH] kvm/ppc/booke64: fix build breakage from
> > > Altivec, and
> > > > >> disable e6500
> > > > >>
> > > > >> BookE altivec support brought two new exceptions, but KVM  
> was not
> > > > >> updated, so the build broke for all 64-bit booke with KVM
> > > enabled.
> > > > >
> > > > > We couldn't do that in KVM before having
> > > > BOOKE_INTERRUPT_ALTIVEC_UNAVAIL/
> > > > > BOOKE_INTERRUPT_ALTIVEC_ASSIST. As Tiejun noticed earlier we
> > > should
> > > > have
> > > > > done this in Kumar's tree but we missed that chance. We will  
> face
> > > > similar
> > > > > issues every time an exception handler will be added.
> > > >
> > > > How hard would it be to add? I suppose it's broken in 3.10, so  
> we
> > > need
> > > > something quick before it gets released?
> > >
> > > Not so hard. Yes. I was surprised by this patch given the fact  
> that
> > > we have
> > > planned to send altivec support upstream this days and that we
> > > already have
> > > a similar patch from Tiejun on our list.
> >
> > I didn't see Tiejun's patch...  My goal was just to fix the build  
> break
> > without exposing problems, and to encourage a patch to fix it  
> properly
> > to happen sooner rather than later.  With Tiejun's patch, which is
> > similar to mine except that it doesn't disable e6500 support, a user
> > could BUG() the kernel by forcing an Altivec exception in a guest.   
> I
> > didn't want to go further down the road of adding reflectors for  
> those
> > exceptions, which could make it look like the problem was dealt with
> > even though it's still not done.
> 
> I agree it's quite annoying to hit a build breakage. Reflection is not
> a proper solution for this problem (though we will require it later)
> but program exception injection looks feasible as a simple fix.

Program exception injection still doesn't deal with state corruption.

> I wouldn't want to see e6500 removed for this reason.

It's not being removed, just closed while under construction to prevent  
damage to people passing through. :-)

-Scott