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: <1582022829-27032-1-git-send-email-wanpengli@tencent.com> <87zhdg84n6.fsf@vitty.brq.redhat.com>
In-Reply-To: <87zhdg84n6.fsf@vitty.brq.redhat.com>
From:   Wanpeng Li <kernellwp@gmail.com>
Date:   Wed, 19 Feb 2020 08:47:56 +0800
Message-ID: <CANRm+Cyx+J+YK8FzFBV8LRNPeCaXPc93vjFdpA0D_hA+wrpywQ@mail.gmail.com>
Subject: Re: [PATCH] KVM: LAPIC: Recalculate apic map in batch
To:     Vitaly Kuznetsov <vkuznets@redhat.com>
Cc:     LKML <linux-kernel@vger.kernel.org>, kvm <kvm@vger.kernel.org>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Sean Christopherson <sean.j.christopherson@intel.com>,
        Wanpeng Li <wanpengli@tencent.com>,
        Jim Mattson <jmattson@google.com>,
        Joerg Roedel <joro@8bytes.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, 18 Feb 2020 at 20:24, Vitaly Kuznetsov <vkuznets@redhat.com> wrote:
>
> Wanpeng Li <kernellwp@gmail.com> writes:
>
> > From: Wanpeng Li <wanpengli@tencent.com>
> >
> > In the vCPU reset and set APIC_BASE MSR path, the apic map will be recalculated
> > several times, each time it will consume 10+ us observed by ftrace in my
> > non-overcommit environment since the expensive memory allocate/mutex/rcu etc
> > operations. This patch optimizes it by recaluating apic map in batch, I hope
> > this can benefit the serverless scenario which can frequently create/destroy
> > VMs.
> >
> > Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
>
> An alternative idea: instead of making every caller return bool and
> every call site handle the result (once) just add a
> KVM_REQ_APIC_MAP_RECALC flag or a boolean flag to struct kvm. I
> understand it may not be that easy as it sounds as we may be conunting
> on valid mapping somewhere before we actually get to handiling

Yes.

> KVM_REQ_APIC_MAP_RECALC but we may preserve *some*
> recalculate_apic_map() calls (and make it reset KVM_REQ_APIC_MAP_RECALC).

Paolo, keep the caller return bool or add a booleen flag to struct
kvm, what do you think?

    Wanpeng