Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Mon, 4 Mar 2019 20:36:55 +0800
From:   Yang Weijiang <weijiang.yang@intel.com>
To:     Sean Christopherson <sean.j.christopherson@intel.com>
Cc:     pbonzini@redhat.com, rkrcmar@redhat.com, jmattson@google.com,
        linux-kernel@vger.kernel.org, kvm@vger.kernel.org, mst@redhat.com,
        yu-cheng.yu@intel.com, Zhang Yi Z <yi.z.zhang@linux.intel.com>,
        wei.w.wang@intel.com, weijiang.yang@inte.com
Subject: Re: [PATCH v3 6/8] KVM:VMX: Load Guest CET via VMCS when CET is
 enabled in Guest
Message-ID: <20190304123655.GB4185@local-michael-cet-test.sh.intel.com>
References: <20190225132716.6982-1-weijiang.yang@intel.com>
 <20190225132716.6982-7-weijiang.yang@intel.com>
 <20190228161715.GF6166@linux.intel.com>
 <20190228083844.GC12006@local-michael-cet-test.sh.intel.com>
 <20190301145819.GC22584@linux.intel.com>
 <20190303122608.GA32013@local-michael-cet-test.sh.intel.com>
 <20190304184307.GC17120@linux.intel.com>
 <20190304095640.GA3576@local-michael-cet-test.sh.intel.com>
 <20190305031202.GI17120@linux.intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20190305031202.GI17120@linux.intel.com>
User-Agent: Mutt/1.10.1 (2018-07-13)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Mar 04, 2019 at 07:12:02PM -0800, Sean Christopherson wrote:
> On Mon, Mar 04, 2019 at 05:56:40PM +0800, Yang Weijiang wrote:
> > On Mon, Mar 04, 2019 at 10:43:07AM -0800, Sean Christopherson wrote:
> > > On Sun, Mar 03, 2019 at 08:26:08PM +0800, Yang Weijiang wrote:
> > > > On Fri, Mar 01, 2019 at 06:58:19AM -0800, Sean Christopherson wrote:
> > > > > On Thu, Feb 28, 2019 at 04:38:44PM +0800, Yang Weijiang wrote:
> > > > > > On Thu, Feb 28, 2019 at 08:17:15AM -0800, Sean Christopherson wrote:
> > > > > > > On Mon, Feb 25, 2019 at 09:27:14PM +0800, Yang Weijiang wrote:
> > > > > > > > "Load Guest CET state" bit controls whether guest CET states
> > > > > > > > will be loaded at Guest entry. Before doing that, KVM needs
> > > > > > > > to check if CPU CET feature is available.
> > > > > > > > 
> > > > > > > > Signed-off-by: Zhang Yi Z <yi.z.zhang@linux.intel.com>
> > > > > > > > Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
> > > > > > > > ---
> > > > > > > >  arch/x86/kvm/vmx.c | 32 ++++++++++++++++++++++++++++++++
> > > > > > > >  1 file changed, 32 insertions(+)
> > > > > > > > 
> > > > > > > > diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
> > > > > > > > index 89ee086e1729..d32cee9ee079 100644
> > > > > > > > --- a/arch/x86/kvm/vmx.c
> > > > > > > > +++ b/arch/x86/kvm/vmx.c
> > > > > > > > @@ -55,6 +55,7 @@
> > > > > > > >  #include <asm/mmu_context.h>
> > > > > > > >  #include <asm/spec-ctrl.h>
> > > > > > > >  #include <asm/mshyperv.h>
> > > > > > > > +#include <asm/cet.h>
> > > > > > > >  
> > > > > > > >  #include "trace.h"
> > > > > > > >  #include "pmu.h"
> > > > > > > > @@ -4065,6 +4066,20 @@ static inline bool vmx_feature_control_msr_valid(struct kvm_vcpu *vcpu,
> > > > > > > >  	return !(val & ~valid_bits);
> > > > > > > >  }
> > > > > > > >  
> > > > > > > > +static int vmx_guest_cet_cap(struct kvm_vcpu *vcpu)
> > > > > > > > +{
> > > > > > > > +	u32 eax, ebx, ecx, edx;
> > > > > > > > +
> > > > > > > > +	/*
> > > > > > > > +	 * Guest CET can work as long as HW supports the feature, independent
> > > > > > > > +	 * to Host SW enabling status.
> > > > > > > > +	 */
> > > > > > > > +	cpuid_count(7, 0, &eax, &ebx, &ecx, &edx);
> > > > > > > > +
> > > > > > > > +	return ((ecx & bit(X86_FEATURE_SHSTK)) |
> > > > > > > > +		(edx & bit(X86_FEATURE_IBT))) ? 1 : 0;
> > > > > > > 
> > > > > > > Given the holes in the (current) architecture/spec, I think KVM has to
> > > > > > > require both features to be supported in the guest to allow CR4.CET to
> > > > > > > be enabled.
> > > > > > The reason why I use a "OR" here is to keep CET enabling control the
> > > > > > same as that on host, right now on host, users can select to enable  SHSTK or IBT
> > > > > > feature by disabling the unexpected one. It's free to select SHSTK & IBT
> > > > > > or SHSTK | IBT.
> > > > > 
> > > > > Which is not the same as SHSTK != IBT in *hardware*, which is effectively
> > > > > what this is allowing for the guest.  The problem is that the architecture
> > > > > doesn't cleanly separate the two features, i.e. we'd have a virtualization
> > > > > hole where the guest could touch state for a disabled feature.
> > > > > 
> > > > > Regardless, the guest would still be able to selectively enable each CET
> > > > > feature, it would just never see a model where SHSTK != IBT.
> > > > Hi, Sean,
> > > > I'd like to understand your concerns. From my point of view, e.g., 
> > > > when only IBT is enabled, PL3_SSP MSR would be unnecessrily exposed,
> > > > this is the design "limitation", but PL3_SSP keeps 0 if SHSTK is not
> > > > configured. Could you detail your concerns?
> > > 
> > > In your approach, IA32_{S,U}_CET can be written if SHSTK or IBT is exposed
> > > to the guest.  If only SHSTK is exposed, a devious guest can still use IBT
> > > because it can set CR4.CET as well as the enable bits in IA32_{S,U}_CET.
> > > Preventing the guest from using IBT in this scenario is infeasible as it
> > > would require trapping and emulating the XSAVE as well as the relevent CET
> > > MSRs.
> > Cannot agree with you more!
> > This is some design limitation, but from my point of view, once vmm
> > exposes CET capability to guest via CPUID, it grants the guest kernel freedom to choose
> > which features to be enabled, we don't need to add extra constraints on
> > the usage.
> 
> But if KVM allows SHSTK and IBT to be toggled independently then the VMM
> has only exposed SHSTK or IBT, not CET as whole.
> 
> Even if SHSTK and IBT are bundled together the guest still has to opt-in
> to enabling each feature.  I don't see what we gain by pretending that
> SHSTK/IBT can be individually exposed to the guest, and on the flip side
> doing so creates a virtualization hole.
you almost convinced me ;-), maybe I'll make the feature as a bundle in
next release after check with kernel team. BTW, what do you mean by
saying "create a virtualization hole"? Is it what you stated in above
reply?