From mboxrd@z Thu Jan 1 00:00:00 1970 From: Radim =?utf-8?B?S3LEjW3DocWZ?= Subject: Re: [PATCH v1] KVM: x86: fix illegal MP_STATE when in/entering SMM Date: Wed, 22 Mar 2017 17:49:33 +0100 Message-ID: <20170322164932.GA1992@potion> References: <20170322155354.18436-1-david@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: kvm@vger.kernel.org, Paolo Bonzini , Dmitry Vyukov , stable@vger.kernel.org, LKML To: David Hildenbrand Return-path: Content-Disposition: inline In-Reply-To: <20170322155354.18436-1-david@redhat.com> Sender: stable-owner@vger.kernel.org List-Id: kvm.vger.kernel.org 2017-03-22 16:53+0100, David Hildenbrand: > If we already entered/are about to enter SMM, don't allow > switching to INIT/SIPI_RECEIVED, otherwise the next call to > kvm_apic_accept_events() will report a warning. > > Fixes: cd7764fe9f73 ("KVM: x86: latch INITs while in system management mode") > Cc: stable@vger.kernel.org # 4.2+ > Reported-by: Dmitry Vyukov > Signed-off-by: David Hildenbrand > --- > diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c > index 1faf620..7d0ec1b 100644 > --- a/arch/x86/kvm/x86.c > +++ b/arch/x86/kvm/x86.c > @@ -7355,6 +7355,12 @@ int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu, > mp_state->mp_state != KVM_MP_STATE_RUNNABLE) > return -EINVAL; > > + /* INITs are latched while in SMM */ > + if ((is_smm(vcpu) || vcpu->arch.smi_pending) && I think that userspace can still set the mpstate first and then enter SMM with KVM_SET_VCPU_EVENTS, which would result in the same warning ... Don't we need a similar check in kvm_vcpu_ioctl_x86_set_vcpu_events()? Thanks. > + (mp_state->mp_state == KVM_MP_STATE_SIPI_RECEIVED || > + mp_state->mp_state == KVM_MP_STATE_INIT_RECEIVED)) > + return -EINVAL; > + > if (mp_state->mp_state == KVM_MP_STATE_SIPI_RECEIVED) { > vcpu->arch.mp_state = KVM_MP_STATE_INIT_RECEIVED; > set_bit(KVM_APIC_SIPI, &vcpu->arch.apic->pending_events); > -- > 2.9.3 >