linux-kernel - Re: [PATCH] KVM: x86: always allow writing '0' to MSR_KVM_ASYNC_PF

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20200911160455.GB4344@sjchrist-ice>
Date:   Fri, 11 Sep 2020 09:04:56 -0700
From:   Sean Christopherson <sean.j.christopherson@...el.com>
To:     Vitaly Kuznetsov <vkuznets@...hat.com>
Cc:     kvm@...r.kernel.org, Paolo Bonzini <pbonzini@...hat.com>,
        Wanpeng Li <wanpengli@...cent.com>,
        Jim Mattson <jmattson@...gle.com>,
        "Dr . David Alan Gilbert" <dgilbert@...hat.com>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] KVM: x86: always allow writing '0' to MSR_KVM_ASYNC_PF_EN

On Fri, Sep 11, 2020 at 11:31:47AM +0200, Vitaly Kuznetsov wrote:
> Even without in-kernel LAPIC we should allow writing '0' to
> MSR_KVM_ASYNC_PF_EN as we're not enabling the mechanism. In
> particular, QEMU with 'kernel-irqchip=off' fails to start
> a guest with
> 
> qemu-system-x86_64: error: failed to set MSR 0x4b564d02 to 0x0
> 
> Fixes: 9d3c447c72fb2 ("KVM: X86: Fix async pf caused null-ptr-deref")
> Reported-by: Dr. David Alan Gilbert <dgilbert@...hat.com>
> Signed-off-by: Vitaly Kuznetsov <vkuznets@...hat.com>
> ---
>  arch/x86/kvm/x86.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index d39d6cf1d473..44a86f7f2397 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -2730,9 +2730,6 @@ static int kvm_pv_enable_async_pf(struct kvm_vcpu *vcpu, u64 data)
>  	if (data & 0x30)
>  		return 1;
>  
> -	if (!lapic_in_kernel(vcpu))
> -		return 1;
> -
>  	vcpu->arch.apf.msr_en_val = data;
>  
>  	if (!kvm_pv_async_pf_enabled(vcpu)) {
> @@ -2741,6 +2738,9 @@ static int kvm_pv_enable_async_pf(struct kvm_vcpu *vcpu, u64 data)
>  		return 0;
>  	}
>  
> +	if (!lapic_in_kernel(vcpu))

This doesn't actually verify that @data == 0.  kvm_pv_async_pf_enabled()
returns true iff KVM_ASYNC_PF_ENABLED and KVM_ASYNC_PF_DELIVERY_AS_INT are
set, e.g. this would allow setting one and not the other.  This also allows
userspace to set vcpu->arch.apf.msr_en_val to an unsupported value, i.e.
@data has already been propagated to the vcpu and isn't unwound.

Why not just pivot on @data when lapic_in_kernel() is false?  vcpu->arch.apic
is immutable so there's no need to update apf.msr_en_val in either direction.

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 539ea1cd6020..36969d5ec291 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2735,7 +2735,7 @@ static int kvm_pv_enable_async_pf(struct kvm_vcpu *vcpu, u64 data)
                return 1;

        if (!lapic_in_kernel(vcpu))
-               return 1;
+               return data ? 1 : 0;

        vcpu->arch.apf.msr_en_val = data;


> +		return 1;
> +
>  	if (kvm_gfn_to_hva_cache_init(vcpu->kvm, &vcpu->arch.apf.data, gpa,
>  					sizeof(u64)))
>  		return 1;
> -- 
> 2.25.4
>