Re: [PATCH v4 4/6] vduse-blk: implements vduse-blk export

[Kevin Wolf]

Am 27.04.2022 um 05:11 hat Yongji Xie geschrieben:
> On Wed, Apr 27, 2022 at 1:03 AM Kevin Wolf <kwolf@redhat.com> wrote:
> >
> > Am 06.04.2022 um 09:59 hat Xie Yongji geschrieben:
> > > This implements a VDUSE block backends based on
> > > the libvduse library. We can use it to export the BDSs
> > > for both VM and container (host) usage.
> > >
> > > The new command-line syntax is:
> > >
> > > $ qemu-storage-daemon \
> > >     --blockdev file,node-name=drive0,filename=test.img \
> > >     --export vduse-blk,node-name=drive0,id=vduse-export0,writable=on
> > >
> > > After the qemu-storage-daemon started, we need to use
> > > the "vdpa" command to attach the device to vDPA bus:
> > >
> > > $ vdpa dev add name vduse-export0 mgmtdev vduse
> > >
> > > Also the device must be removed via the "vdpa" command
> > > before we stop the qemu-storage-daemon.
> > >
> > > Signed-off-by: Xie Yongji <xieyongji@bytedance.com>
> >
> > The request handling code is almos the same as for the vhost-user-blk
> > export. I wonder if we could share this code instead of copying.
> >
> 
> I think we can. Will do it v5.
> 
> > The main difference seems to be that you chose not to support discard
> > and write_zeroes yet. I'm curious if there is a reason why the
> > vhost-user-blk code wouldn't work for vdpa there?
> >
> 
> They are different protocols. The data plane is similar, so we can
> share some codes. But the control plane is different, e.g., vhost-user
> can only work for guests but vdpa can work for both guests and hosts.

Yes, sure, but discard/write_zeroes are part of the data plane, no?
You're already sharing (or at the moment copying) the code for the other
request types mostly unchanged, so I wondered what is different about
discard/write_zeroes.

Kevin