Re: Emulating device configuration / max_virtqueue_pairs in vhost-vdpa and vhost-user

[Jason Wang <jasowang@redhat.com>]

在 2023/2/1 19:48, Eugenio Perez Martin 写道:
> On Wed, Feb 1, 2023 at 12:20 PM Michael S. Tsirkin <mst@redhat.com> wrote:
>> On Wed, Feb 01, 2023 at 12:14:18PM +0100, Maxime Coquelin wrote:
>>> Thanks Eugenio for working on this.
>>>
>>> On 1/31/23 20:10, Eugenio Perez Martin wrote:
>>>> Hi,
>>>>
>>>> The current approach of offering an emulated CVQ to the guest and map
>>>> the commands to vhost-user is not scaling well:
>>>> * Some devices already offer it, so the transformation is redundant.
>>>> * There is no support for commands with variable length (RSS?)
>>>>
>>>> We can solve both of them by offering it through vhost-user the same
>>>> way as vhost-vdpa do. With this approach qemu needs to track the
>>>> commands, for similar reasons as vhost-vdpa: qemu needs to track the
>>>> device status for live migration. vhost-user should use the same SVQ
>>>> code for this, so we avoid duplications.
>>>>
>>>> One of the challenges here is to know what virtqueue to shadow /
>>>> isolate. The vhost-user device may not have the same queues as the
>>>> device frontend:
>>>> * The first depends on the actual vhost-user device, and qemu fetches
>>>> it with VHOST_USER_GET_QUEUE_NUM at the moment.
>>>> * The qemu device frontend's is set by netdev queues= cmdline parameter in qemu
>>>>
>>>> For the device, the CVQ is the last one it offers, but for the guest
>>>> it is the last one offered in config space.
>>>>
>>>> To create a new vhost-user command to decrease that maximum number of
>>>> queues may be an option. But we can do it without adding more
>>>> commands, remapping the CVQ index at virtqueue setup. I think it
>>>> should be doable using (struct vhost_dev).vq_index and maybe a few
>>>> adjustments here and there.
>>>>
>>>> Thoughts?
>>> I am fine with both proposals.
>>> I think index remapping will require a bit more rework in the DPDK
>>> Vhost-user library, but nothing insurmountable.
>>>
>>> I am currently working on a PoC adding support for VDUSE in the DPDK
>>> Vhost library, and recently added control queue support. We can reuse it
>>> if we want to prototype your proposal.
>>>
>>> Maxime
>>>
>>>> Thanks!
>>>>
>>
>> technically backend knows how many vqs are there, last one is cvq...
>> not sure we need full blown remapping ...
>>
> The number of queues may not be the same between cmdline and the device.
>
> If vhost-user device cmdline wants more queues than offered by the
> device qemu will print an error. But the reverse (to offer the same
> number of queues or less than the device have) is valid at this
> moment.
>
> If we add cvq with this scheme, cvq index will not be the same between
> the guest and the device.
>
> vhost-vdpa totally ignores the queues parameter, so we're losing the
> opportunity to offer a consistent config space in the event of a
> migration. I suggest we should act the same way as I'm proposing here
> with vhost-user, so:
> * QEMU can block the migration in the case the destination cannot
> offer the same number of queues.
> * The guest will not see a change of the config space under its feets.


As we discussed in the past, it would be easier to fail the device 
initialization in this case.

Thanks


>
> Now there are other fields in the config space for sure (mtu, rss
> size, etc), but I think the most complex case is the number of queues
> because cvq.
>
> Is that clearer?
>
> Thanks!
>