Re: [PATCH v2] net/9p: add vsock transport

[email protected] Fri, 3 Jul 2026 16:31:25 +0200
Newsgroups dev.linux.lists.v9fs,org.kernel.vger.linux-kernel
Message-ID <[email protected]>
On Fri, Jul 03, 2026 at 09:12 +0000, [email protected] wrote:
> >The privport mount option is not currently supported for vsock and
> >returns -EOPNOTSUPP if specified.
>
> Why not?

The existing p9_bind_privport() implementation is TCP-specific
(uses inet_pton and struct sockaddr_in), so it cannot be reused
for vsock directly. Per Stefan Hajnoczi's suggestion, I opted to
reject the option cleanly for now rather than leave it silently
ignored. Support can be added in a follow-up if needed.

> >-			fd   	  used passed file descriptors for connection
> >+			fd   	  uses passed file descriptors for connection
>
> nit: unrelated change?

I noticed it as I was adding the vsock line below, but you're right,
it's not related, so I'll drop it in v3.

> The rest LGTM, so for vsock point of view:
>
> Acked-by: Stefano Garzarella <[email protected]>

Thank you!

--
Matus