On 2023/10/30 13:08, Jason Wang wrote:
On Mon, Oct 30, 2023 at 12:03 PM Akihiko Odaki <akihiko.od...@daynix.com> wrote:
On 2023/10/30 12:06, Jason Wang wrote:
在 2023/10/27 15:52, Akihiko Odaki 写道:
On 2023/10/27 15:49, Jason Wang wrote:
On Tue, Oct 17, 2023 at 12:09 PM Akihiko Odaki
<akihiko.od...@daynix.com> wrote:
While netmap implements virtio-net header, it does not implement
receive_raw().
The only user for raw is the announcing. Netmap probably doesn't it
at all.
In my understanding, the announcing *sends* a raw packet.
It's send via NIC and receive by its peer which is the TAP
qemu_send_packet_raw() -> nc -> nc->peer -> peer->receive_raw()?
Anything I miss?
The problem is that the peer can be netmap and netmap also requires a
virtio-net header.
Right, but I don't know whether netmap can migrate.
Thinking of the condition that announcement can happen, I'm not aware
anything that prevents migration with netamp. It also is apparently
possible to make an announcement with HMP/QMP.
In any case, I think it's better to fix qemu_send_packet_raw() for
netmap to prevent potential breakage especially if it costs nothing
(actually it saves some code).
Regards,
Akihiko Odaki
