Re: [RFC v2 4/9] vhost-vdpa: introduce vhost-vdpa net client

[Cindy Lu <lulu@redhat.com>]

On Sat, May 9, 2020 at 12:42 AM Eric Blake <eblake@redhat.com> wrote:
>
> On 5/8/20 11:32 AM, Cindy Lu wrote:
> > From: Tiwei Bie <tiwei.bie@intel.com>
> >
> > This patch set introduces a new net client type: vhost-vdpa.
> > vhost-vdpa net client will set up a vDPA device which is specified
> > by a "vhostdev" parameter.
> >
> > Co-authored-by: Lingshan Zhu <lingshan.zhu@intel.com>
> > Signed-off-by: Cindy Lu <lulu@redhat.com>
> > ---
>
> Just looking at UI:
>
>
> > +++ b/qapi/net.json
> > @@ -441,6 +441,23 @@
> >       '*queues':        'int' } }
> >
> >   ##
> > +# @NetdevVhostVDPAOptions:
> > +#
> > +# Vhost-vdpa network backend
> > +#
> > +# @vhostdev: name of a vdpa dev path in sysfs
> > +#
> > +# @queues: number of queues to be created for multiqueue vhost-vdpa
> > +#          (default: 1) (Since 5.1)
>
> No need to mark a 'since' tag on a member introduced at the same time as
> the overall struct itself.
>
Will fix this
> > +#
> > +# Since: 5.1
> > +##
> > +{ 'struct': 'NetdevVhostVDPAOptions',
> > +  'data': {
> > +    '*vhostdev':     'str',
>
> What does this default to if omitted?
>
> > +    '*fd':           'str',
>
> Not documented above.
>
> > +    '*queues':       'int' } }
> > +##
>
> Missing blank line separator.
>
Thanks Eric, Will fix  these all
> >   # @NetClientDriver:
> >   #
> >   # Available netdev drivers.
> > @@ -451,7 +468,7 @@
> >   ##
> >   { 'enum': 'NetClientDriver',
> >     'data': [ 'none', 'nic', 'user', 'tap', 'l2tpv3', 'socket', 'vde',
> > -            'bridge', 'hubport', 'netmap', 'vhost-user' ] }
> > +            'bridge', 'hubport', 'netmap', 'vhost-user', 'vhost-vdpa' ] }
>
> Missing a line above that 'vhost-vdpa' is new to 5.1.
>
>
Will fix this
> > @@ -2749,6 +2756,18 @@ qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,sha
> >        -device virtio-net-pci,netdev=net0
> >   @end example
> >
> > +@item -netdev vhost-vdpa,vhostdev=/path/to/dev
> > +Establish a vhost-vdpa netdev, backed by a vhostdev. The chardev should
> > +be a unix domain socket backed one. The vhost-vdpa uses a specifically defined
> > +protocol to pass vhost ioctl replacement messages to an application on the other
> > +end of the socket.
> > +Example:
> > +@example
> > +qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \
> > +     -numa node,memdev=mem \
> > +     -netdev type=vhost-vdpa,id=net0,vhostdev=/path/to/dev \
> > +     -device virtio-net-pci,netdev=net0
> > +@end example
> >   @item -netdev hubport,id=@var{id},hubid=@var{hubid}[,netdev=@var{nd}]
> >
> >   Create a hub port on the emulated hub with ID @var{hubid}.
> >
>
> --
> Eric Blake, Principal Software Engineer
> Red Hat, Inc.           +1-919-301-3226
> Virtualization:  qemu.org | libvirt.org
>