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: <1557975980-9875-1-git-send-email-wanpengli@tencent.com>
 <1557975980-9875-4-git-send-email-wanpengli@tencent.com> <20190517200509.GJ15006@linux.intel.com>
In-Reply-To: <20190517200509.GJ15006@linux.intel.com>
From:   Wanpeng Li <kernellwp@gmail.com>
Date:   Mon, 20 May 2019 16:43:46 +0800
Message-ID: <CANRm+CxT96pPsqzNXMvJWU-rk3SuZ8yXGBu9BVQdrtyuqAuLdQ@mail.gmail.com>
Subject: Re: [PATCH v3 3/5] KVM: LAPIC: Expose per-vCPU timer_advance_ns to userspace
To:     Sean Christopherson <sean.j.christopherson@intel.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>,
        Liran Alon <liran.alon@oracle.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Sat, 18 May 2019 at 04:05, Sean Christopherson
<sean.j.christopherson@intel.com> wrote:
>
> On Thu, May 16, 2019 at 11:06:18AM +0800, Wanpeng Li wrote:
> > From: Wanpeng Li <wanpengli@tencent.com>
> >
> > Expose per-vCPU timer_advance_ns to userspace, so it is able to
> > query the auto-adjusted value.
> >
> > Cc: Paolo Bonzini <pbonzini@redhat.com>
> > Cc: Radim Kr=C4=8Dm=C3=A1=C5=99 <rkrcmar@redhat.com>
> > Cc: Sean Christopherson <sean.j.christopherson@intel.com>
> > Cc: Liran Alon <liran.alon@oracle.com>
> > Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
> > ---
> >  arch/x86/kvm/debugfs.c | 16 ++++++++++++++++
> >  1 file changed, 16 insertions(+)
> >
> > diff --git a/arch/x86/kvm/debugfs.c b/arch/x86/kvm/debugfs.c
> > index c19c7ed..a6f1f93 100644
> > --- a/arch/x86/kvm/debugfs.c
> > +++ b/arch/x86/kvm/debugfs.c
> > @@ -9,12 +9,22 @@
> >   */
> >  #include <linux/kvm_host.h>
> >  #include <linux/debugfs.h>
> > +#include "lapic.h"
> >
> >  bool kvm_arch_has_vcpu_debugfs(void)
> >  {
> >       return true;
> >  }
> >
> > +static int vcpu_get_timer_advance_ns(void *data, u64 *val)
> > +{
> > +     struct kvm_vcpu *vcpu =3D (struct kvm_vcpu *) data;
> > +     *val =3D vcpu->arch.apic->lapic_timer.timer_advance_ns;
>
> This needs to ensure to check lapic_in_kernel() to ensure apic isn't NULL=
.
> Actually, I think we can skip creation of the parameter entirely if
> lapic_in_kernel() is false.  VMX and SVM both instantiate the lapic
> during kvm_arch_vcpu_create(), which is (obviously) called before
> kvm_arch_create_vcpu_debugfs().

Handle this in v4.

>
> > +     return 0;
> > +}
> > +
> > +DEFINE_SIMPLE_ATTRIBUTE(vcpu_timer_advance_ns_fops, vcpu_get_timer_adv=
ance_ns, NULL, "%llu\n");
> > +
> >  static int vcpu_get_tsc_offset(void *data, u64 *val)
> >  {
> >       struct kvm_vcpu *vcpu =3D (struct kvm_vcpu *) data;
> > @@ -51,6 +61,12 @@ int kvm_arch_create_vcpu_debugfs(struct kvm_vcpu *vc=
pu)
> >       if (!ret)
> >               return -ENOMEM;
> >
> > +     ret =3D debugfs_create_file("lapic_timer_advance_ns", 0444,
> > +                                                     vcpu->debugfs_den=
try,
> > +                                                     vcpu, &vcpu_timer=
_advance_ns_fops);
> > +     if (!ret)
> > +             return -ENOMEM;
> > +
> >       if (kvm_has_tsc_control) {
> >               ret =3D debugfs_create_file("tsc-scaling-ratio", 0444,
> >                                                       vcpu->debugfs_den=
try,
> > --
> > 2.7.4
> >