Re: [PATCH net-next v11 05/13] vsock: add netns support to virtio transports

[Stefano Garzarella]

On Fri, Nov 21, 2025 at 11:03:51AM -0800, Bobby Eshleman wrote:
> On Fri, Nov 21, 2025 at 03:39:25PM +0100, Stefano Garzarella wrote:
> > On Thu, Nov 20, 2025 at 09:44:37PM -0800, Bobby Eshleman wrote:
> > > From: Bobby Eshleman <bobbyeshleman@xxxxxxxx>
> > >
> > > Add netns support to loopback and vhost. Keep netns disabled for
> > > virtio-vsock, but add necessary changes to comply with common API
> > > updates.
> > >
> > > This is the patch in the series when vhost-vsock namespaces actually
> > > come online.  Hence, vhost_transport_supports_local_mode() is switched
> > > to return true.
> > >
> > > Signed-off-by: Bobby Eshleman <bobbyeshleman@xxxxxxxx>
> > > ---
> > > Changes in v11:
> > > - reorder with the skb ownership patch for loopback (Stefano)
> > > - toggle vhost_transport_supports_local_mode() to true
> > >
> > > Changes in v10:
> > > - Splitting patches complicates the series with meaningless placeholder
> > >  values that eventually get replaced anyway, so to avoid that this
> > >  patch combines into one. Links to previous patches here:
> > >  - Link: https://lore.kernel.org/all/20251111-vsock-vmtest-v9-3-852787a37bed@xxxxxxxx/
> > >  - Link: https://lore.kernel.org/all/20251111-vsock-vmtest-v9-6-852787a37bed@xxxxxxxx/
> > >  - Link: https://lore.kernel.org/all/20251111-vsock-vmtest-v9-7-852787a37bed@xxxxxxxx/
> > > - remove placeholder values (Stefano)
> > > - update comment describe net/net_mode for
> > >  virtio_transport_reset_no_sock()
> > > ---
> > > drivers/vhost/vsock.c                   | 47 ++++++++++++++++++------
> > > include/linux/virtio_vsock.h            |  8 +++--
> > > net/vmw_vsock/virtio_transport.c        | 10 ++++--
> > > net/vmw_vsock/virtio_transport_common.c | 63 ++++++++++++++++++++++++---------
> > > net/vmw_vsock/vsock_loopback.c          |  8 +++--
> > > 5 files changed, 103 insertions(+), 33 deletions(-)
> >
> > Reviewed-by: Stefano Garzarella <sgarzare@xxxxxxxxxx>
>
> If we move the supports_local_mode() changes into this patch (for virtio
> and loopback, as I bring up in other discussion), should I drop this
> trailer or carry it forward?

I'll take a second look in any case, so maybe better to remove it if the 
patch will change.

Thanks for asking!
Stefano