Re: [virtio-dev] Re: [PATCH 0/5] virtio: introduce SUSPEND bit and vq state

[Jason Wang <jasowang@redhat.com>]

On Thu, Sep 21, 2023 at 12:11âPM Parav Pandit <parav@nvidia.com> wrote:
>
>
>
> > From: Jason Wang <jasowang@redhat.com>
> > Sent: Thursday, September 21, 2023 9:32 AM
> >
> > On Thu, Sep 21, 2023 at 11:51âAM Parav Pandit <parav@nvidia.com> wrote:
> > >
> > >
> > >
> > > > From: Jason Wang <jasowang@redhat.com>
> > > > Sent: Thursday, September 21, 2023 8:45 AM The main issue I see so
> > > > far is that you want to couple migration with admin commands but I
> > > > don't see much advantages to doing this.
> > > >
> > > The way I read above comment is, to draw a parallel line: descriptor posting in
> > virtio spec is tied to virtqueues. What is the advantage of it?
> >
> > Are you saying virtio can't live without admin commands? Again, let's not shift
> > concepts.
> >
> No, I did not say that.
> I just donât see how functionalities proposed in [1] can be done without admin commands by the _device_ for member device passthrough requirement.
>
> You made point as "donât see much advantage with migration done using admin commands".

Parav, I think I've clarified several times:

migration using the admin command is probably fine in some use cases.

What's not fine, is:

Mandate the admin command to be the only way for migration.

Are we on the same page for my concern now?

> What is the advantage of descriptor posting using virtqueue. It is the way of virtio spec...
>
> > > Well, it is one way to achieve it.
> > > There may be different way to do all bulk data transfer without admin
> > commands.
> >
> > Why is virtqueue the only way to do bulk data transferring? Can't DMA be
> > initiated by other-way?
> >
>
> Sure, what is the disadvantage of existing mechanism of virtqueue that can do following.
> 1. Ability to do DMA
> 2. agnostic of the DMA who do not want to do DMA

I don't understand this.

> 3. Ability to multiple command executions in parallel

Each device has their self-contained interface, why can't the commands
be executed in parallel.

> 4. Non blocking interface for driver that does not require any kind of polling

Are you saying the interrupt can only work for virtqueue?

>
> Why to invent new DMA scheme which at does all the 4 tasks?

It's simply because admin virtqueue can not work for all the cases. I
think you've agreed on this, no?

> First please list down disadvantages of admin queue + show all 4 things are achieved using new DMA interface.
> That will help to understand why new dma interface is needed.

I can give you a simple example. For example, what happens if we want
to migrate the owner? Having another owner for this owner is not a
good answer.

Thanks


>
>
>
> > Thanks
> >
> > > What it the advantage of it, please list down them in [1] for the command
> > where you can find alternative.
> > >
> > > [1]
> > > https://lore.kernel.org/virtio-comment/20230909142911.524407-7-parav@n
> > > vidia.com/T/#md9fcfa1ba997463de8c7fb8c6d1786b224b0bead
>