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;
Subject: Re: [PATCH 2/5] KVM: add a check to ensure grow start value is
 nonzero
To:     Zhenzhong Duan <zhenzhong.duan@oracle.com>,
        linux-kernel@vger.kernel.org
Cc:     kvm@vger.kernel.org, mtosatti@redhat.com,
        joao.m.martins@oracle.com, rafael.j.wysocki@intel.com,
        rkrcmar@redhat.com
References: <1572060239-17401-1-git-send-email-zhenzhong.duan@oracle.com>
 <1572060239-17401-3-git-send-email-zhenzhong.duan@oracle.com>
From:   Paolo Bonzini <pbonzini@redhat.com>
Openpgp: preference=signencrypt
Message-ID: <391dd11b-ebbb-28ff-5e57-4a795cd16a1b@redhat.com>
Date:   Mon, 11 Nov 2019 14:49:39 +0100
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101
 Thunderbird/60.8.0
MIME-Version: 1.0
In-Reply-To: <1572060239-17401-3-git-send-email-zhenzhong.duan@oracle.com>
Content-Language: en-US
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 26/10/19 05:23, Zhenzhong Duan wrote:
> vcpu->halt_poll_ns could be zeroed in certain cases (e.g. by
> halt_poll_ns_shrink). If halt_poll_ns_grow_start is zero,
> vcpu->halt_poll_ns will never be larger than zero.
>=20
> Signed-off-by: Zhenzhong Duan <zhenzhong.duan@oracle.com>
> ---
>  virt/kvm/kvm_main.c | 7 +++++++
>  1 file changed, 7 insertions(+)
>=20
> diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> index 2ca2979..1b6fe3b 100644
> --- a/virt/kvm/kvm_main.c
> +++ b/virt/kvm/kvm_main.c
> @@ -2266,6 +2266,13 @@ static void grow_halt_poll_ns(struct kvm_vcpu *vcp=
u)
>  =09=09goto out;
> =20
>  =09val *=3D grow;
> +
> +=09/*
> +=09 * vcpu->halt_poll_ns needs a nonzero start point to grow if it's zer=
o.
> +=09 */
> +=09if (!grow_start)
> +=09=09grow_start =3D 1;
> +
>  =09if (val < grow_start)
>  =09=09val =3D grow_start;
> =20
>=20

Zeroing grow_start will simply disable halt polling.  Is that a problem?

Paolo