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;
MIME-Version: 1.0
References: <1567680270-14022-1-git-send-email-wanpengli@tencent.com> <87ftlakhn6.fsf@vitty.brq.redhat.com>
In-Reply-To: <87ftlakhn6.fsf@vitty.brq.redhat.com>
From:   Wanpeng Li <kernellwp@gmail.com>
Date:   Fri, 6 Sep 2019 09:54:23 +0800
Message-ID: <CANRm+CyPxb+ZY2cTdbLL_LBKMJSOaMqnPGKc_ATc6-TMHW-rJw@mail.gmail.com>
Subject: Re: [PATCH] KVM: LAPIC: Fix SynIC Timers inject timer interrupt w/o
 LAPIC present
To:     Vitaly Kuznetsov <vkuznets@redhat.com>
Cc:     LKML <linux-kernel@vger.kernel.org>, kvm <kvm@vger.kernel.org>,
        Paolo Bonzini <pbonzini@redhat.com>,
        =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= <rkrcmar@redhat.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, 5 Sep 2019 at 21:16, Vitaly Kuznetsov <vkuznets@redhat.com> wrote:
>
> Wanpeng Li <kernellwp@gmail.com> writes:
>
> > From: Wanpeng Li <wanpengli@tencent.com>
> >
> > Reported by syzkaller:
> >
> >       kasan: GPF could be caused by NULL-ptr deref or user memory acces=
s
> >       general protection fault: 0000 [#1] PREEMPT SMP KASAN
> >       RIP: 0010:__apic_accept_irq+0x46/0x740 arch/x86/kvm/lapic.c:1029
> >       Call Trace:
> >       kvm_apic_set_irq+0xb4/0x140 arch/x86/kvm/lapic.c:558
> >       stimer_notify_direct arch/x86/kvm/hyperv.c:648 [inline]
> >       stimer_expiration arch/x86/kvm/hyperv.c:659 [inline]
> >       kvm_hv_process_stimers+0x594/0x1650 arch/x86/kvm/hyperv.c:686
> >       vcpu_enter_guest+0x2b2a/0x54b0 arch/x86/kvm/x86.c:7896
> >       vcpu_run+0x393/0xd40 arch/x86/kvm/x86.c:8152
> >       kvm_arch_vcpu_ioctl_run+0x636/0x900 arch/x86/kvm/x86.c:8360
> >       kvm_vcpu_ioctl+0x6cf/0xaf0 arch/x86/kvm/../../../virt/kvm/kvm_mai=
n.c:2765
> >
> > The testcase programs HV_X64_MSR_STIMERn_CONFIG/HV_X64_MSR_STIMERn_COUN=
T,
> > in addition, there is no lapic in the kernel, the counters value are sm=
all
> > enough in order that kvm_hv_process_stimers() inject this already-expir=
ed
> > timer interrupt into the guest through lapic in the kernel which trigge=
rs
> > the NULL deferencing. This patch fixes it by checking lapic_in_kernel,
> > discarding the inject if it is 0.
> >
> > Reported-by: syzbot+dff25ee91f0c7d5c1695@syzkaller.appspotmail.com
> > Cc: Paolo Bonzini <pbonzini@redhat.com>
> > Cc: Radim Kr=C4=8Dm=C3=A1=C5=99 <rkrcmar@redhat.com>
> > Cc: Vitaly Kuznetsov <vkuznets@redhat.com>
> > Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
> > ---
> >  arch/x86/kvm/hyperv.c | 4 +++-
> >  1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
> > index c10a8b1..461fcc5 100644
> > --- a/arch/x86/kvm/hyperv.c
> > +++ b/arch/x86/kvm/hyperv.c
> > @@ -645,7 +645,9 @@ static int stimer_notify_direct(struct kvm_vcpu_hv_=
stimer *stimer)
> >               .vector =3D stimer->config.apic_vector
> >       };
> >
> > -     return !kvm_apic_set_irq(vcpu, &irq, NULL);
> > +     if (lapic_in_kernel(vcpu))
> > +             return !kvm_apic_set_irq(vcpu, &irq, NULL);
> > +     return 0;
> >  }
> >
> >  static void stimer_expiration(struct kvm_vcpu_hv_stimer *stimer)
>
> Hm, but this basically means direct mode synthetic timers won't work
> when LAPIC is not in kernel but the feature will still be advertised to
> the guest, not good. Shall we stop advertizing it? Something like
> (completely untested):
>
> diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
> index 3f5ad84853fb..1dfa594eaab6 100644
> --- a/arch/x86/kvm/hyperv.c
> +++ b/arch/x86/kvm/hyperv.c
> @@ -1856,7 +1856,13 @@ int kvm_vcpu_ioctl_get_hv_cpuid(struct kvm_vcpu *v=
cpu, struct kvm_cpuid2 *cpuid,
>
>                         ent->edx |=3D HV_FEATURE_FREQUENCY_MSRS_AVAILABLE=
;
>                         ent->edx |=3D HV_FEATURE_GUEST_CRASH_MSR_AVAILABL=
E;
> -                       ent->edx |=3D HV_STIMER_DIRECT_MODE_AVAILABLE;
> +
> +                       /*
> +                        * Direct Synthetic timers only make sense with i=
n-kernel
> +                        * LAPIC
> +                        */
> +                       if (lapic_in_kernel(vcpu))
> +                               ent->edx |=3D HV_STIMER_DIRECT_MODE_AVAIL=
ABLE;
>
>                         break;

Thanks, I fold this into v2, syzkaller even didn't check the cpuid, so
I still keep the discard inject part.

                                                                   Wanpeng