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;
From:   Vitaly Kuznetsov <vkuznets@redhat.com>
To:     Tianyu Lan <Tianyu.Lan@microsoft.com>
Cc:     "pbonzini\@redhat.com" <pbonzini@redhat.com>,
        "rkrcmar\@redhat.com" <rkrcmar@redhat.com>,
        "tglx\@linutronix.de" <tglx@linutronix.de>,
        "mingo\@redhat.com" <mingo@redhat.com>,
        "hpa\@zytor.com" <hpa@zytor.com>,
        "x86\@kernel.org" <x86@kernel.org>,
        "kvm\@vger.kernel.org" <kvm@vger.kernel.org>,
        "linux-kernel\@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        KY Srinivasan <kys@microsoft.com>
Subject: Re: [RFC Patch 3/3] KVM/x86: Add tlb_remote_flush callback support for vmcs
References: <20180604090749.489-1-Tianyu.Lan@microsoft.com>
        <20180604090749.489-4-Tianyu.Lan@microsoft.com>
        <87h8m8qbso.fsf@vitty.brq.redhat.com>
        <f868a339-03fc-4ba6-4882-4bb6232f599d@microsoft.com>
Date:   Thu, 14 Jun 2018 12:33:53 +0200
In-Reply-To: <f868a339-03fc-4ba6-4882-4bb6232f599d@microsoft.com> (Tianyu
        Lan's message of "Thu, 14 Jun 2018 10:13:46 +0000")
Message-ID: <87d0wtpsha.fsf@vitty.brq.redhat.com>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/25.3 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Tianyu Lan <Tianyu.Lan@microsoft.com> writes:

> On 6/12/2018 11:12 PM, Vitaly Kuznetsov wrote:
>> Tianyu Lan <Tianyu.Lan@microsoft.com> writes:
>> 
>>>
>>> +static int vmx_remote_flush_tlb(struct kvm *kvm)
>>> +{
>>> +	struct kvm_vcpu *vcpu = kvm_get_vcpu(kvm, 0);
>>> +
>>> +	if (!VALID_PAGE(vcpu->arch.mmu.root_hpa))
>>> +		return -1;
>> 
>> Why vcpu0? Can arch.mmu.root_hpa-s differ across vCPUs? What happens if
>> they do?
>
> Yes, it may take place that arch.mmu.root_hpa is differ across vCPUs.
> We may check all vcpu root_hpa and use the hypercall when there is only 
> one validated ept table. If not, go back to current way.
>

I'd suggest an optimization: keep track of wether root_hpas are equal
across all vcpus (check this on change).

-- 
  Vitaly