From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
To: Laurent Vivier <lvivier@redhat.com>
Cc: qemu-devel@nongnu.org, "Markus Armbruster" <armbru@redhat.com>,
"Eric Blake" <eblake@redhat.com>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Daniel P. Berrangé" <berrange@redhat.com>,
"Jason Wang" <jasowang@redhat.com>,
"Ralph Schmieder" <ralph.schmieder@gmail.com>,
"Stefano Brivio" <sbrivio@redhat.com>
Subject: Re: [RFC PATCH v3 00/11] qapi: net: add unix socket type support to netdev backend
Date: Mon, 20 Jun 2022 19:24:19 +0100 [thread overview]
Message-ID: <YrC7U1XH0pGsn5Df@work-vm> (raw)
In-Reply-To: <20220620101828.518865-1-lvivier@redhat.com>
* Laurent Vivier (lvivier@redhat.com) wrote:
> "-netdev socket" only supports inet sockets.
>
> It's not a complex task to add support for unix sockets, but
> the socket netdev parameters are not defined to manage well unix
> socket parameters.
>
> As discussed in:
>
> "socket.c added support for unix domain socket datagram transport"
> https://lore.kernel.org/qemu-devel/1C0E1BC5-904F-46B0-8044-68E43E67BE60@gmail.com/
>
> This series adds support of unix socket type using SocketAddress QAPI structure.
>
> Two new netdev backends, "stream" and "dgram" are added, that are barely a copy of "socket"
> backend but they use the SocketAddress QAPI to provide socket parameters.
> And then they also implement unix sockets (TCP and UDP).
Had you considered a -netdev chardev?
Dave
> Some examples of CLI syntax:
>
> for TCP:
>
> -netdev stream,id=socket0,addr.type=inet,addr.host=localhost,addr.port=1234
> -netdev stream,id=socket0,server=off,addr.type=inet,addr.host=localhost,addr.port=1234
>
> -netdev dgram,id=socket0,\
> local.type=inet,local.host=localhost,local.port=1234,\
> remote.type=inet,remote.host=localhost,remote.port=1235
>
> for UNIX:
>
> -netdev stream,id=socket0,addr.type=unix,addr.path=/tmp/qemu0
> -netdev stream,id=socket0,server=off,addr.type=unix,addr.path=/tmp/qemu0
>
> -netdev dgram,id=socket0,\
> local.type=unix,local.path=/tmp/qemu0,\
> remote.type=unix,remote.path=/tmp/qemu1
>
> for FD:
>
> -netdev stream,id=socket0,addr.type=fd,addr.str=4
> -netdev stream,id=socket0,server=off,addr.type=fd,addr.str=5
>
> -netdev dgram,id=socket0,local.type=fd,addr.str=4
>
> v3:
> - remove support of "-net" for dgram and stream. They are only
> supported with "-netdev" option.
> - use &error_fatal directly in net_client_inits()
> - update qemu-options.hx
> - move to QIO for stream socket
>
> v2:
> - use "stream" and "dgram" rather than "socket-ng,mode=stream"
> and ""socket-ng,mode=dgram"
> - extract code to bypass qemu_opts_parse_noisily() to
> a new patch
> - do not ignore EINVAL (Stefano)
> - fix "-net" option
>
> CC: Ralph Schmieder <ralph.schmieder@gmail.com>
> CC: Stefano Brivio <sbrivio@redhat.com>
> CC: Daniel P. Berrangé <berrange@redhat.com>
> CC: Markus Armbruster <armbru@redhat.com>
>
> Laurent Vivier (10):
> net: introduce convert_host_port()
> net: remove the @errp argument of net_client_inits()
> qapi: net: introduce a way to bypass qemu_opts_parse_noisily()
> qapi: net: add stream and dgram netdevs
> net: stream: add unix socket
> net: dgram: make dgram_dst generic
> net: dgram: move mcast specific code from net_socket_fd_init_dgram()
> net: dgram: add unix socket
> qemu-sockets: introduce socket_uri()
> net: stream: move to QIO
>
> Stefano Brivio (1):
> net: stream: Don't ignore EINVAL on netdev socket connection
>
> hmp-commands.hx | 2 +-
> include/net/net.h | 3 +-
> include/qemu/sockets.h | 4 +-
> monitor/hmp-cmds.c | 23 +-
> net/clients.h | 6 +
> net/dgram.c | 706 +++++++++++++++++++++++++++++++++++++++++
> net/hub.c | 2 +
> net/meson.build | 2 +
> net/net.c | 149 ++++++---
> net/stream.c | 371 ++++++++++++++++++++++
> qapi/net.json | 40 ++-
> qemu-options.hx | 12 +
> softmmu/vl.c | 5 +-
> util/qemu-sockets.c | 20 ++
> 14 files changed, 1277 insertions(+), 68 deletions(-)
> create mode 100644 net/dgram.c
> create mode 100644 net/stream.c
>
> --
> 2.36.1
>
--
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK
next prev parent reply other threads:[~2022-06-20 19:42 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-06-20 10:18 [RFC PATCH v3 00/11] qapi: net: add unix socket type support to netdev backend Laurent Vivier
2022-06-20 10:18 ` [RFC PATCH v3 01/11] net: introduce convert_host_port() Laurent Vivier
2022-06-20 10:18 ` [RFC PATCH v3 02/11] net: remove the @errp argument of net_client_inits() Laurent Vivier
2022-06-20 11:39 ` Markus Armbruster
2022-06-20 10:18 ` [RFC PATCH v3 03/11] qapi: net: introduce a way to bypass qemu_opts_parse_noisily() Laurent Vivier
2022-06-20 12:43 ` Markus Armbruster
2022-06-20 13:46 ` Laurent Vivier
2022-06-22 8:00 ` Markus Armbruster
2022-06-20 10:18 ` [RFC PATCH v3 04/11] qapi: net: add stream and dgram netdevs Laurent Vivier
2022-06-20 15:21 ` Markus Armbruster
2022-06-20 17:52 ` Laurent Vivier
2022-06-21 8:49 ` Markus Armbruster
2022-06-21 19:27 ` Laurent Vivier
2022-06-22 11:47 ` Markus Armbruster
2022-06-22 16:18 ` Laurent Vivier
2022-06-23 12:49 ` Markus Armbruster
2022-06-20 10:18 ` [RFC PATCH v3 05/11] net: stream: Don't ignore EINVAL on netdev socket connection Laurent Vivier
2022-06-20 10:18 ` [RFC PATCH v3 06/11] net: stream: add unix socket Laurent Vivier
2022-06-20 10:18 ` [RFC PATCH v3 07/11] net: dgram: make dgram_dst generic Laurent Vivier
2022-06-20 10:18 ` [RFC PATCH v3 08/11] net: dgram: move mcast specific code from net_socket_fd_init_dgram() Laurent Vivier
2022-06-20 10:18 ` [RFC PATCH v3 09/11] net: dgram: add unix socket Laurent Vivier
2022-06-20 10:18 ` [RFC PATCH v3 10/11] qemu-sockets: introduce socket_uri() Laurent Vivier
2022-06-20 10:18 ` [RFC PATCH v3 11/11] net: stream: move to QIO Laurent Vivier
2022-06-20 18:24 ` Dr. David Alan Gilbert [this message]
2022-06-21 9:51 ` [RFC PATCH v3 00/11] qapi: net: add unix socket type support to netdev backend Laurent Vivier
2022-06-21 10:31 ` Dr. David Alan Gilbert
2022-06-21 10:54 ` Daniel P. Berrangé
2022-06-21 12:16 ` Dr. David Alan Gilbert
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=YrC7U1XH0pGsn5Df@work-vm \
--to=dgilbert@redhat.com \
--cc=armbru@redhat.com \
--cc=berrange@redhat.com \
--cc=eblake@redhat.com \
--cc=jasowang@redhat.com \
--cc=lvivier@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=ralph.schmieder@gmail.com \
--cc=sbrivio@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).