Received-SPF: pass (google.com: domain of linux-kernel+bounces-78949-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
Date: Fri, 23 Feb 2024 10:12:53 -0800
In-Reply-To: <34504abb-ff58-4a83-9a63-87f22841adc7@redhat.com>
Precedence: bulk
Mime-Version: 1.0
References: <20240203002343.383056-1-seanjc@google.com> <20240203002343.383056-5-seanjc@google.com>
 <34504abb-ff58-4a83-9a63-87f22841adc7@redhat.com>
Message-ID: <ZdjgJaVqHs7WbFnk@google.com>
Subject: Re: [PATCH v2 4/4] KVM: x86/mmu: Fix a *very* theoretical race in kvm_mmu_track_write()
From: Sean Christopherson <seanjc@google.com>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org, 
	Jim Mattson <jmattson@google.com>, Mingwei Zhang <mizhang@google.com>
Content-Type: text/plain; charset="us-ascii"

On Fri, Feb 23, 2024, Paolo Bonzini wrote:
> On 2/3/24 01:23, Sean Christopherson wrote:
> > Add full memory barriers in kvm_mmu_track_write() and account_shadowed()
> > to plug a (very, very theoretical) race where kvm_mmu_track_write() could
> > miss a 0->1 transition of indirect_shadow_pages and fail to zap relevant,
> > *stale* SPTEs.
> 
> Ok, so we have
> 
> emulator_write_phys
>   overwrite PTE
>   kvm_page_track_write
>     kvm_mmu_track_write
>       // memory barrier missing here
>       if (indirect_shadow_pages)
>         zap();
> 
> and on the other side
> 
>   FNAME(page_fault)
>     FNAME(fetch)
>       kvm_mmu_get_child_sp
>         kvm_mmu_get_shadow_page
>           __kvm_mmu_get_shadow_page
>             kvm_mmu_alloc_shadow_page
>               account_shadowed
>                 indirect shadow pages++
>                 // memory barrier missing here
>       if (FNAME(gpte_changed)) // reads PTE
>         goto out
> 
> If you can weave something like this in the commit message the sequence
> would be a bit clearer.

Roger that.

> > In practice, this bug is likely benign as both the 0=>1 transition and
> > reordering of this scope are extremely rare occurrences.
> 
> I wouldn't call it benign, it's more that it's unobservable in practice but
> the race is real.  However...
> > diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
> > index 3c193b096b45..86b85060534d 100644
> > --- a/arch/x86/kvm/mmu/mmu.c
> > +++ b/arch/x86/kvm/mmu/mmu.c
> > @@ -830,6 +830,14 @@ static void account_shadowed(struct kvm *kvm, struct kvm_mmu_page *sp)
> >   	struct kvm_memory_slot *slot;
> >   	gfn_t gfn;
> > +	/*
> > +	 * Ensure indirect_shadow_pages is elevated prior to re-reading guest
> > +	 * child PTEs in FNAME(gpte_changed), i.e. guarantee either in-flight
> > +	 * emulated writes are visible before re-reading guest PTEs, or that
> > +	 * an emulated write will see the elevated count and acquire mmu_lock
> > +	 * to update SPTEs.  Pairs with the smp_mb() in kvm_mmu_track_write().
> > +	 */
> > +	smp_mb();
> 
> ... this memory barrier needs to be after the increment (the desired
> ordering is store-before-read).

Doh.  I'll post a fixed version as a one-off v3.

Thanks!