On Fri, Nov 19, 2021 at 6:20 PM Eugenio Pérez <epere...@redhat.com> wrote: > > Qemu falls back on userland handlers even if vhost-user and vhost-vdpa > cases. These assumes a tap device can handle the packets. > > If a vdpa device fail to start, it can trigger a sigsegv because of > that. Add dummy receivers that return no progress so it can keep > running. > > Fixes: 1e0a84ea49 ("vhost-vdpa: introduce vhost-vdpa net client") > Signed-off-by: Eugenio Pérez <epere...@redhat.com> > --- > net/vhost-vdpa.c | 16 ++++++++++++++++ > 1 file changed, 16 insertions(+) > > diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c > index 2e3c22a8c7..4c75b78304 100644 > --- a/net/vhost-vdpa.c > +++ b/net/vhost-vdpa.c > @@ -170,9 +170,25 @@ static bool vhost_vdpa_check_peer_type(NetClientState > *nc, ObjectClass *oc, > return true; > } > > +/** Dummy receive in case qemu falls back to userland tap networking */ > +static ssize_t vhost_vdpa_receive_iov(NetClientState *nc, > + const struct iovec *iov, int iovcnt) > +{ > + return 0; > +} > + > +/** Dummy receive in case qemu falls back to userland tap networking */ > +static ssize_t vhost_vdpa_receive_raw(NetClientState *nc, const uint8_t *buf, > + size_t size) > +{ > + return 0; > +}
It looks to me the .receive_raw is not need, in nc_sendv_compat() we had: => if (flags & QEMU_NET_PACKET_FLAG_RAW && nc->info->receive_raw) { ret = nc->info->receive_raw(nc, buffer, offset); } else { ret = nc->info->receive(nc, buffer, offset); } Thanks > + > static NetClientInfo net_vhost_vdpa_info = { > .type = NET_CLIENT_DRIVER_VHOST_VDPA, > .size = sizeof(VhostVDPAState), > + .receive_iov = vhost_vdpa_receive_iov, > + .receive_raw = vhost_vdpa_receive_raw, > .cleanup = vhost_vdpa_cleanup, > .has_vnet_hdr = vhost_vdpa_has_vnet_hdr, > .has_ufo = vhost_vdpa_has_ufo, > -- > 2.27.0 >