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 RFC 00/15] KVM: Dirty ring interface
To:     Peter Xu <peterx@redhat.com>, Paolo Bonzini <pbonzini@redhat.com>
Cc:     linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
        Sean Christopherson <sean.j.christopherson@intel.com>,
        "Dr . David Alan Gilbert" <dgilbert@redhat.com>,
        Vitaly Kuznetsov <vkuznets@redhat.com>,
        "Michael S. Tsirkin" <mst@redhat.com>
References: <20191129213505.18472-1-peterx@redhat.com>
 <776732ca-06c8-c529-0899-9d2ffacf7789@redhat.com>
 <20191204193357.GE19939@xz-x1>
From:   Jason Wang <jasowang@redhat.com>
Message-ID: <d24d7100-0100-b74f-0761-37196ce16f7f@redhat.com>
Date:   Thu, 5 Dec 2019 14:49:30 +0800
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: <20191204193357.GE19939@xz-x1>
Content-Language: en-US
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On 2019/12/5 =E4=B8=8A=E5=8D=883:33, Peter Xu wrote:
> On Wed, Dec 04, 2019 at 06:39:48PM +0800, Jason Wang wrote:
>> On 2019/11/30 =E4=B8=8A=E5=8D=885:34, Peter Xu wrote:
>>> Branch is here:https://github.com/xzpeter/linux/tree/kvm-dirty-ring
>>>
>>> Overview
>>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>>>
>>> This is a continued work from Lei Cao<lei.cao@stratus.com>  and Paolo
>>> on the KVM dirty ring interface.  To make it simple, I'll still start
>>> with version 1 as RFC.
>>>
>>> The new dirty ring interface is another way to collect dirty pages for
>>> the virtual machine, but it is different from the existing dirty
>>> logging interface in a few ways, majorly:
>>>
>>>     - Data format: The dirty data was in a ring format rather than a
>>>       bitmap format, so the size of data to sync for dirty logging does
>>>       not depend on the size of guest memory any more, but speed of
>>>       dirtying.  Also, the dirty ring is per-vcpu (currently plus
>>>       another per-vm ring, so total ring number is N+1), while the dirt=
y
>>>       bitmap is per-vm.
>>>
>>>     - Data copy: The sync of dirty pages does not need data copy any mo=
re,
>>>       but instead the ring is shared between the userspace and kernel b=
y
>>>       page sharings (mmap() on either the vm fd or vcpu fd)
>>>
>>>     - Interface: Instead of using the old KVM_GET_DIRTY_LOG,
>>>       KVM_CLEAR_DIRTY_LOG interfaces, the new ring uses a new interface
>>>       called KVM_RESET_DIRTY_RINGS when we want to reset the collected
>>>       dirty pages to protected mode again (works like
>>>       KVM_CLEAR_DIRTY_LOG, but ring based)
>>>
>>> And more.
>>
>> Looks really interesting, I wonder if we can make this as a library then=
 we
>> can reuse it for vhost.
> So iiuc this ring will majorly for (1) data exchange between kernel
> and user, and (2) shared memory.  I think from that pov yeh it should
> work even for vhost.
>
> It shouldn't be hard to refactor the interfaces to avoid kvm elements,
> however I'm not sure how to do that best.  Maybe like irqbypass and
> put it into virt/lib/ as a standlone module?  Would it worth it?


Maybe, and it looks to me some dirty pages reporting API for VFIO is=20
proposed in the same time. It will be helpful to unify them (or at least=20
leave a chance for other users).

Thanks


>
> Paolo, what's your take?
>