linux-kernel - Re: [PATCH RESEND] kvm:x86:Fix error handling in the function kvm_lapic_sync_to

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>] [day] [month] [year] [list]

Date:	Mon, 11 Jan 2016 11:00:42 +0100
From:	Paolo Bonzini <pbonzini@...hat.com>
To:	Nicholas Krause <xerofoify@...il.com>, gleb@...nel.org
Cc:	x86@...nel.org, kvm@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH RESEND] kvm:x86:Fix error handling in the function
 kvm_lapic_sync_to_vapic



On 10/01/2016 06:28, Nicholas Krause wrote:
> This fixes error handling in the function kvm_lapic_sync_to_vapic
> to check the value of the call to kvm_write_guest_cached to see if
> it failed by returning a error code and if so execute WARN_ON to
> allow the user to see in their logs the failure when executing the
> function kvm_lapic_sync_to_vapic.
> 
> Signed-off-by: Nicholas Krause <xerofoify@...il.com>
> ---
>  arch/x86/kvm/lapic.c | 7 ++++---
>  1 file changed, 4 insertions(+), 3 deletions(-)
> 
> diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
> index 2a5ca97..f3946c0 100644
> --- a/arch/x86/kvm/lapic.c
> +++ b/arch/x86/kvm/lapic.c
> @@ -1935,7 +1935,7 @@ static void apic_sync_pv_eoi_to_guest(struct kvm_vcpu *vcpu,
>  void kvm_lapic_sync_to_vapic(struct kvm_vcpu *vcpu)
>  {
>  	u32 data, tpr;
> -	int max_irr, max_isr;
> +	int max_irr, max_isr, ret;
>  	struct kvm_lapic *apic = vcpu->arch.apic;
>  
>  	apic_sync_pv_eoi_to_guest(vcpu, apic);
> @@ -1952,8 +1952,9 @@ void kvm_lapic_sync_to_vapic(struct kvm_vcpu *vcpu)
>  		max_isr = 0;
>  	data = (tpr & 0xff) | ((max_isr & 0xf0) << 8) | (max_irr << 24);
>  
> -	kvm_write_guest_cached(vcpu->kvm, &vcpu->arch.apic->vapic_cache, &data,
> -				sizeof(u32));
> +	ret = kvm_write_guest_cached(vcpu->kvm, &vcpu->arch.apic->vapic_cache, &data,
> +				     sizeof(u32));
> +	WARN_ON(ret);
>  }
>  
>  int kvm_lapic_set_vapic_addr(struct kvm_vcpu *vcpu, gpa_t vapic_addr)
> 

Again... no.  The error can be triggered by the guest.  Incorrect
behavior of the guest should *never* result in a WARN in the host.

Paolo