[PATCH v5 00/12] qapi: net: add unix socket type support to netdev backend

[PATCH v5 00/12] qapi: net: add unix socket type support to netdev backend

[Laurent Vivier]

"-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).

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

v5:
  - remove RFC prefix
  - put the change of net_client_parse() into its own patch (exit() in the
    function)
  - update comments regarding netdev_is_modern() and netdev_parse_modern()
  - update error case in net_stream_server_init()
  - update qemu-options.hx with unix type

v4:
  - net_client_parse() fails with exit() rather than with return.
  - keep "{ 'name': 'vmnet-host', 'if': 'CONFIG_VMNET' }" on its
    own line in qapi/net.json
  - add a comment in qapi/net.json about parameters usage
  - move netdev_is_modern() check to qemu_init()
  - in netdev_is_modern(), check for JSON and use qemu_opts_do_parse()
    to parse parameters and detect type value.
  - add a blank line after copyright comment

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 (11):
  net: introduce convert_host_port()
  net: remove the @errp argument of net_client_inits()
  net: simplify net_client_parse() error management
  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      |   6 +-
 include/qemu/sockets.h |   4 +-
 monitor/hmp-cmds.c     |  23 +-
 net/clients.h          |   6 +
 net/dgram.c            | 707 +++++++++++++++++++++++++++++++++++++++++
 net/hub.c              |   2 +
 net/meson.build        |   2 +
 net/net.c              | 169 +++++++---
 net/stream.c           | 372 ++++++++++++++++++++++
 qapi/net.json          |  44 ++-
 qemu-options.hx        |  14 +
 softmmu/vl.c           |  16 +-
 util/qemu-sockets.c    |  20 ++
 14 files changed, 1305 insertions(+), 82 deletions(-)
 create mode 100644 net/dgram.c
 create mode 100644 net/stream.c

-- 
2.36.1

[PATCH v5 01/12] net: introduce convert_host_port()

[Laurent Vivier]

Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
---
 include/qemu/sockets.h |  2 ++
 net/net.c              | 62 ++++++++++++++++++++++--------------------
 2 files changed, 34 insertions(+), 30 deletions(-)

diff --git a/include/qemu/sockets.h b/include/qemu/sockets.h
index 038faa157f59..47194b9732f8 100644
--- a/include/qemu/sockets.h
+++ b/include/qemu/sockets.h
@@ -47,6 +47,8 @@ void socket_listen_cleanup(int fd, Error **errp);
 int socket_dgram(SocketAddress *remote, SocketAddress *local, Error **errp);
 
 /* Old, ipv4 only bits.  Don't use for new code. */
+int convert_host_port(struct sockaddr_in *saddr, const char *host,
+                      const char *port, Error **errp);
 int parse_host_port(struct sockaddr_in *saddr, const char *str,
                     Error **errp);
 int socket_init(void);
diff --git a/net/net.c b/net/net.c
index 2db160e0634d..d2288bd3a929 100644
--- a/net/net.c
+++ b/net/net.c
@@ -66,55 +66,57 @@ static QTAILQ_HEAD(, NetClientState) net_clients;
 /***********************************************************/
 /* network device redirectors */
 
-int parse_host_port(struct sockaddr_in *saddr, const char *str,
-                    Error **errp)
+int convert_host_port(struct sockaddr_in *saddr, const char *host,
+                      const char *port, Error **errp)
 {
-    gchar **substrings;
     struct hostent *he;
-    const char *addr, *p, *r;
-    int port, ret = 0;
+    const char *r;
+    long p;
 
     memset(saddr, 0, sizeof(*saddr));
 
-    substrings = g_strsplit(str, ":", 2);
-    if (!substrings || !substrings[0] || !substrings[1]) {
-        error_setg(errp, "host address '%s' doesn't contain ':' "
-                   "separating host from port", str);
-        ret = -1;
-        goto out;
-    }
-
-    addr = substrings[0];
-    p = substrings[1];
-
     saddr->sin_family = AF_INET;
-    if (addr[0] == '\0') {
+    if (host[0] == '\0') {
         saddr->sin_addr.s_addr = 0;
     } else {
-        if (qemu_isdigit(addr[0])) {
-            if (!inet_aton(addr, &saddr->sin_addr)) {
+        if (qemu_isdigit(host[0])) {
+            if (!inet_aton(host, &saddr->sin_addr)) {
                 error_setg(errp, "host address '%s' is not a valid "
-                           "IPv4 address", addr);
-                ret = -1;
-                goto out;
+                           "IPv4 address", host);
+                return -1;
             }
         } else {
-            he = gethostbyname(addr);
+            he = gethostbyname(host);
             if (he == NULL) {
-                error_setg(errp, "can't resolve host address '%s'", addr);
-                ret = -1;
-                goto out;
+                error_setg(errp, "can't resolve host address '%s'", host);
+                return -1;
             }
             saddr->sin_addr = *(struct in_addr *)he->h_addr;
         }
     }
-    port = strtol(p, (char **)&r, 0);
-    if (r == p) {
-        error_setg(errp, "port number '%s' is invalid", p);
+    if (qemu_strtol(port, &r, 0, &p) != 0) {
+        error_setg(errp, "port number '%s' is invalid", port);
+        return -1;
+    }
+    saddr->sin_port = htons(p);
+    return 0;
+}
+
+int parse_host_port(struct sockaddr_in *saddr, const char *str,
+                    Error **errp)
+{
+    gchar **substrings;
+    int ret;
+
+    substrings = g_strsplit(str, ":", 2);
+    if (!substrings || !substrings[0] || !substrings[1]) {
+        error_setg(errp, "host address '%s' doesn't contain ':' "
+                   "separating host from port", str);
         ret = -1;
         goto out;
     }
-    saddr->sin_port = htons(port);
+
+    ret = convert_host_port(saddr, substrings[0], substrings[1], errp);
 
 out:
     g_strfreev(substrings);
-- 
2.36.1

[PATCH v5 02/12] net: remove the @errp argument of net_client_inits()

[Laurent Vivier]

The only caller passes &error_fatal, so use this directly in the function.

It's what we do for -blockdev, -device, and -object.

Suggested-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
---
 include/net/net.h |  2 +-
 net/net.c         | 20 +++++++-------------
 softmmu/vl.c      |  2 +-
 3 files changed, 9 insertions(+), 15 deletions(-)

diff --git a/include/net/net.h b/include/net/net.h
index 523136c7acba..c53c64ac18c4 100644
--- a/include/net/net.h
+++ b/include/net/net.h
@@ -216,7 +216,7 @@ extern const char *host_net_devices[];
 /* from net.c */
 int net_client_parse(QemuOptsList *opts_list, const char *str);
 void show_netdevs(void);
-int net_init_clients(Error **errp);
+void net_init_clients(void);
 void net_check_clients(void);
 void net_cleanup(void);
 void hmp_host_net_add(Monitor *mon, const QDict *qdict);
diff --git a/net/net.c b/net/net.c
index d2288bd3a929..15958f881776 100644
--- a/net/net.c
+++ b/net/net.c
@@ -1562,27 +1562,21 @@ out:
     return ret;
 }
 
-int net_init_clients(Error **errp)
+void net_init_clients(void)
 {
     net_change_state_entry =
         qemu_add_vm_change_state_handler(net_vm_change_state_handler, NULL);
 
     QTAILQ_INIT(&net_clients);
 
-    if (qemu_opts_foreach(qemu_find_opts("netdev"),
-                          net_init_netdev, NULL, errp)) {
-        return -1;
-    }
-
-    if (qemu_opts_foreach(qemu_find_opts("nic"), net_param_nic, NULL, errp)) {
-        return -1;
-    }
+    qemu_opts_foreach(qemu_find_opts("netdev"), net_init_netdev, NULL,
+                      &error_fatal);
 
-    if (qemu_opts_foreach(qemu_find_opts("net"), net_init_client, NULL, errp)) {
-        return -1;
-    }
+    qemu_opts_foreach(qemu_find_opts("nic"), net_param_nic, NULL,
+                      &error_fatal);
 
-    return 0;
+    qemu_opts_foreach(qemu_find_opts("net"), net_init_client, NULL,
+                      &error_fatal);
 }
 
 int net_client_parse(QemuOptsList *opts_list, const char *optarg)
diff --git a/softmmu/vl.c b/softmmu/vl.c
index 54e920ada1a1..c244e8afc4cc 100644
--- a/softmmu/vl.c
+++ b/softmmu/vl.c
@@ -1898,7 +1898,7 @@ static void qemu_create_late_backends(void)
         qtest_server_init(qtest_chrdev, qtest_log, &error_fatal);
     }
 
-    net_init_clients(&error_fatal);
+    net_init_clients();
 
     object_option_foreach_add(object_create_late);
 
-- 
2.36.1

[PATCH v5 03/12] net: simplify net_client_parse() error management

[Laurent Vivier]

All net_client_parse() callers exit in case of error.

Move exit(1) to net_client_parse() and remove error checking from
the callers.

Suggested-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
---
 include/net/net.h |  2 +-
 net/net.c         |  6 ++----
 softmmu/vl.c      | 12 +++---------
 3 files changed, 6 insertions(+), 14 deletions(-)

diff --git a/include/net/net.h b/include/net/net.h
index c53c64ac18c4..e755254443ea 100644
--- a/include/net/net.h
+++ b/include/net/net.h
@@ -214,7 +214,7 @@ extern NICInfo nd_table[MAX_NICS];
 extern const char *host_net_devices[];
 
 /* from net.c */
-int net_client_parse(QemuOptsList *opts_list, const char *str);
+void net_client_parse(QemuOptsList *opts_list, const char *str);
 void show_netdevs(void);
 void net_init_clients(void);
 void net_check_clients(void);
diff --git a/net/net.c b/net/net.c
index 15958f881776..f056e8aebfb2 100644
--- a/net/net.c
+++ b/net/net.c
@@ -1579,13 +1579,11 @@ void net_init_clients(void)
                       &error_fatal);
 }
 
-int net_client_parse(QemuOptsList *opts_list, const char *optarg)
+void net_client_parse(QemuOptsList *opts_list, const char *optarg)
 {
     if (!qemu_opts_parse_noisily(opts_list, optarg, true)) {
-        return -1;
+        exit(1);
     }
-
-    return 0;
 }
 
 /* From FreeBSD */
diff --git a/softmmu/vl.c b/softmmu/vl.c
index c244e8afc4cc..f6461363d608 100644
--- a/softmmu/vl.c
+++ b/softmmu/vl.c
@@ -2794,21 +2794,15 @@ void qemu_init(int argc, char **argv, char **envp)
                 break;
             case QEMU_OPTION_netdev:
                 default_net = 0;
-                if (net_client_parse(qemu_find_opts("netdev"), optarg) == -1) {
-                    exit(1);
-                }
+                net_client_parse(qemu_find_opts("netdev"), optarg);
                 break;
             case QEMU_OPTION_nic:
                 default_net = 0;
-                if (net_client_parse(qemu_find_opts("nic"), optarg) == -1) {
-                    exit(1);
-                }
+                net_client_parse(qemu_find_opts("nic"), optarg);
                 break;
             case QEMU_OPTION_net:
                 default_net = 0;
-                if (net_client_parse(qemu_find_opts("net"), optarg) == -1) {
-                    exit(1);
-                }
+                net_client_parse(qemu_find_opts("net"), optarg);
                 break;
 #ifdef CONFIG_LIBISCSI
             case QEMU_OPTION_iscsi:
-- 
2.36.1

Re: [PATCH v5 03/12] net: simplify net_client_parse() error management

[Markus Armbruster]

Laurent Vivier <lvivier@redhat.com> writes:

> All net_client_parse() callers exit in case of error.
>
> Move exit(1) to net_client_parse() and remove error checking from
> the callers.
>
> Suggested-by: Markus Armbruster <armbru@redhat.com>
> Signed-off-by: Laurent Vivier <lvivier@redhat.com>

Reviewed-by: Markus Armbruster <armbru@redhat.com>

[PATCH v5 04/12] qapi: net: introduce a way to bypass qemu_opts_parse_noisily()

[Laurent Vivier]

As qemu_opts_parse_noisily() flattens the QAPI structures ("type" field
of Netdev structure can collides with "type" field of SocketAddress),
we introduce a way to bypass qemu_opts_parse_noisily() and use directly
visit_type_Netdev() to parse the backend parameters.

More details from Markus:

qemu_init() passes the argument of -netdev, -nic, and -net to
net_client_parse().

net_client_parse() parses with qemu_opts_parse_noisily(), passing
QemuOptsList qemu_netdev_opts for -netdev, qemu_nic_opts for -nic, and
qemu_net_opts for -net.  Their desc[] are all empty, which means any
keys are accepted.  The result of the parse (a QemuOpts) is stored in
the QemuOptsList.

Note that QemuOpts is flat by design.  In some places, we layer non-flat
on top using dotted keys convention, but not here.

net_init_clients() iterates over the stored QemuOpts, and passes them to
net_init_netdev(), net_param_nic(), or net_init_client(), respectively.

These functions pass the QemuOpts to net_client_init().  They also do
other things with the QemuOpts, which we can ignore here.

net_client_init() uses the opts visitor to convert the (flat) QemOpts to
a (non-flat) QAPI object Netdev.  Netdev is also the argument of QMP
command netdev_add.

The opts visitor was an early attempt to support QAPI in
(QemuOpts-based) CLI.  It restricts QAPI types to a certain shape; see
commit eb7ee2cbeb "qapi: introduce OptsVisitor".

A more modern way to support QAPI is qobject_input_visitor_new_str().
It uses keyval_parse() instead of QemuOpts for KEY=VALUE,... syntax, and
it also supports JSON syntax.  The former isn't quite as expressive as
JSON, but it's a lot closer than QemuOpts + opts visitor.

This commit paves the way to use of the modern way instead.

Signed-off-by: Laurent Vivier <lvivier@redhat.com>
---
 include/net/net.h |  2 ++
 net/net.c         | 57 +++++++++++++++++++++++++++++++++++++++++++++++
 softmmu/vl.c      |  6 ++++-
 3 files changed, 64 insertions(+), 1 deletion(-)

diff --git a/include/net/net.h b/include/net/net.h
index e755254443ea..826e14a78734 100644
--- a/include/net/net.h
+++ b/include/net/net.h
@@ -214,6 +214,8 @@ extern NICInfo nd_table[MAX_NICS];
 extern const char *host_net_devices[];
 
 /* from net.c */
+bool netdev_is_modern(const char *optarg);
+void netdev_parse_modern(const char *optarg);
 void net_client_parse(QemuOptsList *opts_list, const char *str);
 void show_netdevs(void);
 void net_init_clients(void);
diff --git a/net/net.c b/net/net.c
index f056e8aebfb2..83300ea1b816 100644
--- a/net/net.c
+++ b/net/net.c
@@ -54,6 +54,7 @@
 #include "net/colo-compare.h"
 #include "net/filter.h"
 #include "qapi/string-output-visitor.h"
+#include "qapi/qobject-input-visitor.h"
 
 /* Net bridge is currently not supported for W32. */
 #if !defined(_WIN32)
@@ -63,6 +64,16 @@
 static VMChangeStateEntry *net_change_state_entry;
 static QTAILQ_HEAD(, NetClientState) net_clients;
 
+typedef struct NetdevQueueEntry {
+    Netdev *nd;
+    Location loc;
+    QSIMPLEQ_ENTRY(NetdevQueueEntry) entry;
+} NetdevQueueEntry;
+
+typedef QSIMPLEQ_HEAD(, NetdevQueueEntry) NetdevQueue;
+
+static NetdevQueue nd_queue = QSIMPLEQ_HEAD_INITIALIZER(nd_queue);
+
 /***********************************************************/
 /* network device redirectors */
 
@@ -1562,6 +1573,20 @@ out:
     return ret;
 }
 
+static void netdev_init_modern(void)
+{
+    while (!QSIMPLEQ_EMPTY(&nd_queue)) {
+        NetdevQueueEntry *nd = QSIMPLEQ_FIRST(&nd_queue);
+
+        QSIMPLEQ_REMOVE_HEAD(&nd_queue, entry);
+        loc_push_restore(&nd->loc);
+        net_client_init1(nd->nd, true, &error_fatal);
+        loc_pop(&nd->loc);
+        qapi_free_Netdev(nd->nd);
+        g_free(nd);
+    }
+}
+
 void net_init_clients(void)
 {
     net_change_state_entry =
@@ -1569,6 +1594,8 @@ void net_init_clients(void)
 
     QTAILQ_INIT(&net_clients);
 
+    netdev_init_modern();
+
     qemu_opts_foreach(qemu_find_opts("netdev"), net_init_netdev, NULL,
                       &error_fatal);
 
@@ -1579,6 +1606,36 @@ void net_init_clients(void)
                       &error_fatal);
 }
 
+/*
+ * Does this -netdev argument use modern rather than traditional syntax?
+ * Modern syntax is to be parsed with netdev_parse_modern().
+ * Traditional syntax is to be parsed with net_client_parse().
+ */
+bool netdev_is_modern(const char *optarg)
+{
+    return false;
+}
+
+/*
+ * netdev_parse_modern() uses modern, more expressive syntax than
+ * net_client_parse(), but supports only the netdev option.
+ * netdev_parse_modern() appends to @nd_queue, whereas net_client_parse()
+ * appends to @qemu_netdev_opts.
+ */
+void netdev_parse_modern(const char *optarg)
+{
+    Visitor *v;
+    NetdevQueueEntry *nd;
+
+    v = qobject_input_visitor_new_str(optarg, "type", &error_fatal);
+    nd = g_new(NetdevQueueEntry, 1);
+    visit_type_Netdev(v, NULL, &nd->nd, &error_fatal);
+    visit_free(v);
+    loc_save(&nd->loc);
+
+    QSIMPLEQ_INSERT_TAIL(&nd_queue, nd, entry);
+}
+
 void net_client_parse(QemuOptsList *opts_list, const char *optarg)
 {
     if (!qemu_opts_parse_noisily(opts_list, optarg, true)) {
diff --git a/softmmu/vl.c b/softmmu/vl.c
index f6461363d608..f557e2d8cb8a 100644
--- a/softmmu/vl.c
+++ b/softmmu/vl.c
@@ -2794,7 +2794,11 @@ void qemu_init(int argc, char **argv, char **envp)
                 break;
             case QEMU_OPTION_netdev:
                 default_net = 0;
-                net_client_parse(qemu_find_opts("netdev"), optarg);
+                if (netdev_is_modern(optarg)) {
+                    netdev_parse_modern(optarg);
+                } else {
+                    net_client_parse(qemu_find_opts("netdev"), optarg);
+                }
                 break;
             case QEMU_OPTION_nic:
                 default_net = 0;
-- 
2.36.1

Re: [PATCH v5 04/12] qapi: net: introduce a way to bypass qemu_opts_parse_noisily()

[Markus Armbruster]

Laurent Vivier <lvivier@redhat.com> writes:

> As qemu_opts_parse_noisily() flattens the QAPI structures ("type" field
> of Netdev structure can collides with "type" field of SocketAddress),
> we introduce a way to bypass qemu_opts_parse_noisily() and use directly
> visit_type_Netdev() to parse the backend parameters.
>
> More details from Markus:
>
> qemu_init() passes the argument of -netdev, -nic, and -net to
> net_client_parse().
>
> net_client_parse() parses with qemu_opts_parse_noisily(), passing
> QemuOptsList qemu_netdev_opts for -netdev, qemu_nic_opts for -nic, and
> qemu_net_opts for -net.  Their desc[] are all empty, which means any
> keys are accepted.  The result of the parse (a QemuOpts) is stored in
> the QemuOptsList.
>
> Note that QemuOpts is flat by design.  In some places, we layer non-flat
> on top using dotted keys convention, but not here.
>
> net_init_clients() iterates over the stored QemuOpts, and passes them to
> net_init_netdev(), net_param_nic(), or net_init_client(), respectively.
>
> These functions pass the QemuOpts to net_client_init().  They also do
> other things with the QemuOpts, which we can ignore here.
>
> net_client_init() uses the opts visitor to convert the (flat) QemOpts to
> a (non-flat) QAPI object Netdev.  Netdev is also the argument of QMP
> command netdev_add.
>
> The opts visitor was an early attempt to support QAPI in
> (QemuOpts-based) CLI.  It restricts QAPI types to a certain shape; see
> commit eb7ee2cbeb "qapi: introduce OptsVisitor".
>
> A more modern way to support QAPI is qobject_input_visitor_new_str().
> It uses keyval_parse() instead of QemuOpts for KEY=VALUE,... syntax, and
> it also supports JSON syntax.  The former isn't quite as expressive as
> JSON, but it's a lot closer than QemuOpts + opts visitor.
>
> This commit paves the way to use of the modern way instead.
>
> Signed-off-by: Laurent Vivier <lvivier@redhat.com>
> ---
>  include/net/net.h |  2 ++
>  net/net.c         | 57 +++++++++++++++++++++++++++++++++++++++++++++++
>  softmmu/vl.c      |  6 ++++-
>  3 files changed, 64 insertions(+), 1 deletion(-)
>
> diff --git a/include/net/net.h b/include/net/net.h
> index e755254443ea..826e14a78734 100644
> --- a/include/net/net.h
> +++ b/include/net/net.h
> @@ -214,6 +214,8 @@ extern NICInfo nd_table[MAX_NICS];
>  extern const char *host_net_devices[];
>  
>  /* from net.c */
> +bool netdev_is_modern(const char *optarg);
> +void netdev_parse_modern(const char *optarg);
>  void net_client_parse(QemuOptsList *opts_list, const char *str);
>  void show_netdevs(void);
>  void net_init_clients(void);
> diff --git a/net/net.c b/net/net.c
> index f056e8aebfb2..83300ea1b816 100644
> --- a/net/net.c
> +++ b/net/net.c
> @@ -54,6 +54,7 @@
>  #include "net/colo-compare.h"
>  #include "net/filter.h"
>  #include "qapi/string-output-visitor.h"
> +#include "qapi/qobject-input-visitor.h"
>  
>  /* Net bridge is currently not supported for W32. */
>  #if !defined(_WIN32)
> @@ -63,6 +64,16 @@
>  static VMChangeStateEntry *net_change_state_entry;
>  static QTAILQ_HEAD(, NetClientState) net_clients;
>  
> +typedef struct NetdevQueueEntry {
> +    Netdev *nd;
> +    Location loc;
> +    QSIMPLEQ_ENTRY(NetdevQueueEntry) entry;
> +} NetdevQueueEntry;
> +
> +typedef QSIMPLEQ_HEAD(, NetdevQueueEntry) NetdevQueue;
> +
> +static NetdevQueue nd_queue = QSIMPLEQ_HEAD_INITIALIZER(nd_queue);
> +
>  /***********************************************************/
>  /* network device redirectors */
>  
> @@ -1562,6 +1573,20 @@ out:
>      return ret;
>  }
>  
> +static void netdev_init_modern(void)
> +{
> +    while (!QSIMPLEQ_EMPTY(&nd_queue)) {
> +        NetdevQueueEntry *nd = QSIMPLEQ_FIRST(&nd_queue);
> +
> +        QSIMPLEQ_REMOVE_HEAD(&nd_queue, entry);
> +        loc_push_restore(&nd->loc);
> +        net_client_init1(nd->nd, true, &error_fatal);
> +        loc_pop(&nd->loc);
> +        qapi_free_Netdev(nd->nd);
> +        g_free(nd);
> +    }
> +}
> +
>  void net_init_clients(void)
>  {
>      net_change_state_entry =
> @@ -1569,6 +1594,8 @@ void net_init_clients(void)
>  
>      QTAILQ_INIT(&net_clients);
>  
> +    netdev_init_modern();
> +
>      qemu_opts_foreach(qemu_find_opts("netdev"), net_init_netdev, NULL,
>                        &error_fatal);
>  
> @@ -1579,6 +1606,36 @@ void net_init_clients(void)
>                        &error_fatal);
>  }
>  
> +/*
> + * Does this -netdev argument use modern rather than traditional syntax?
> + * Modern syntax is to be parsed with netdev_parse_modern().
> + * Traditional syntax is to be parsed with net_client_parse().
> + */
> +bool netdev_is_modern(const char *optarg)
> +{
> +    return false;
> +}
> +
> +/*
> + * netdev_parse_modern() uses modern, more expressive syntax than
> + * net_client_parse(), but supports only the netdev option.

Suggest "the -netdev option".

> + * netdev_parse_modern() appends to @nd_queue, whereas net_client_parse()
> + * appends to @qemu_netdev_opts.
> + */
> +void netdev_parse_modern(const char *optarg)
> +{
> +    Visitor *v;
> +    NetdevQueueEntry *nd;
> +
> +    v = qobject_input_visitor_new_str(optarg, "type", &error_fatal);
> +    nd = g_new(NetdevQueueEntry, 1);
> +    visit_type_Netdev(v, NULL, &nd->nd, &error_fatal);
> +    visit_free(v);
> +    loc_save(&nd->loc);
> +
> +    QSIMPLEQ_INSERT_TAIL(&nd_queue, nd, entry);
> +}
> +
>  void net_client_parse(QemuOptsList *opts_list, const char *optarg)
>  {
>      if (!qemu_opts_parse_noisily(opts_list, optarg, true)) {
> diff --git a/softmmu/vl.c b/softmmu/vl.c
> index f6461363d608..f557e2d8cb8a 100644
> --- a/softmmu/vl.c
> +++ b/softmmu/vl.c
> @@ -2794,7 +2794,11 @@ void qemu_init(int argc, char **argv, char **envp)
>                  break;
>              case QEMU_OPTION_netdev:
>                  default_net = 0;
> -                net_client_parse(qemu_find_opts("netdev"), optarg);
> +                if (netdev_is_modern(optarg)) {
> +                    netdev_parse_modern(optarg);
> +                } else {
> +                    net_client_parse(qemu_find_opts("netdev"), optarg);
> +                }
>                  break;
>              case QEMU_OPTION_nic:
>                  default_net = 0;

Reviewed-by: Markus Armbruster <armbru@redhat.com>

[PATCH v5 05/12] qapi: net: add stream and dgram netdevs

[Laurent Vivier]

Copied from socket netdev file and modified to use SocketAddress
to be able to introduce new features like unix socket.

"udp" and "mcast" are squashed into dgram netdev, multicast is detected
according to the IP address type.
"listen" and "connect" modes are managed by stream netdev. An optional
parameter "server" defines the mode (server by default)

Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
---
 hmp-commands.hx |   2 +-
 net/clients.h   |   6 +
 net/dgram.c     | 631 ++++++++++++++++++++++++++++++++++++++++++++++++
 net/hub.c       |   2 +
 net/meson.build |   2 +
 net/net.c       |  30 ++-
 net/stream.c    | 423 ++++++++++++++++++++++++++++++++
 qapi/net.json   |  44 +++-
 qemu-options.hx |  12 +
 9 files changed, 1148 insertions(+), 4 deletions(-)
 create mode 100644 net/dgram.c
 create mode 100644 net/stream.c

diff --git a/hmp-commands.hx b/hmp-commands.hx
index c9d465735ac0..99c2bc37cc10 100644
--- a/hmp-commands.hx
+++ b/hmp-commands.hx
@@ -1276,7 +1276,7 @@ ERST
     {
         .name       = "netdev_add",
         .args_type  = "netdev:O",
-        .params     = "[user|tap|socket|vde|bridge|hubport|netmap|vhost-user"
+        .params     = "[user|tap|socket|stream|dgram|vde|bridge|hubport|netmap|vhost-user"
 #ifdef CONFIG_VMNET
                       "|vmnet-host|vmnet-shared|vmnet-bridged"
 #endif
diff --git a/net/clients.h b/net/clients.h
index c9157789f2ce..ed8bdfff1e7c 100644
--- a/net/clients.h
+++ b/net/clients.h
@@ -40,6 +40,12 @@ int net_init_hubport(const Netdev *netdev, const char *name,
 int net_init_socket(const Netdev *netdev, const char *name,
                     NetClientState *peer, Error **errp);
 
+int net_init_stream(const Netdev *netdev, const char *name,
+                    NetClientState *peer, Error **errp);
+
+int net_init_dgram(const Netdev *netdev, const char *name,
+                   NetClientState *peer, Error **errp);
+
 int net_init_tap(const Netdev *netdev, const char *name,
                  NetClientState *peer, Error **errp);
 
diff --git a/net/dgram.c b/net/dgram.c
new file mode 100644
index 000000000000..dbe65102d174
--- /dev/null
+++ b/net/dgram.c
@@ -0,0 +1,631 @@
+/*
+ * QEMU System Emulator
+ *
+ * Copyright (c) 2003-2008 Fabrice Bellard
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+#include "qemu/osdep.h"
+
+#include "net/net.h"
+#include "clients.h"
+#include "monitor/monitor.h"
+#include "qapi/error.h"
+#include "qemu/error-report.h"
+#include "qemu/option.h"
+#include "qemu/sockets.h"
+#include "qemu/iov.h"
+#include "qemu/main-loop.h"
+#include "qemu/cutils.h"
+
+typedef struct NetDgramState {
+    NetClientState nc;
+    int listen_fd;
+    int fd;
+    SocketReadState rs;
+  /* contains inet host and port destination iff connectionless (SOCK_DGRAM) */
+    struct sockaddr_in dgram_dst;
+    IOHandler *send_fn;           /* differs between SOCK_STREAM/SOCK_DGRAM */
+    bool read_poll;               /* waiting to receive data? */
+    bool write_poll;              /* waiting to transmit data? */
+} NetDgramState;
+
+static void net_dgram_accept(void *opaque);
+static void net_dgram_writable(void *opaque);
+
+static void net_dgram_update_fd_handler(NetDgramState *s)
+{
+    qemu_set_fd_handler(s->fd,
+                        s->read_poll ? s->send_fn : NULL,
+                        s->write_poll ? net_dgram_writable : NULL,
+                        s);
+}
+
+static void net_dgram_read_poll(NetDgramState *s, bool enable)
+{
+    s->read_poll = enable;
+    net_dgram_update_fd_handler(s);
+}
+
+static void net_dgram_write_poll(NetDgramState *s, bool enable)
+{
+    s->write_poll = enable;
+    net_dgram_update_fd_handler(s);
+}
+
+static void net_dgram_writable(void *opaque)
+{
+    NetDgramState *s = opaque;
+
+    net_dgram_write_poll(s, false);
+
+    qemu_flush_queued_packets(&s->nc);
+}
+
+static ssize_t net_dgram_receive_dgram(NetClientState *nc,
+                                       const uint8_t *buf, size_t size)
+{
+    NetDgramState *s = DO_UPCAST(NetDgramState, nc, nc);
+    ssize_t ret;
+
+    do {
+        if (s->dgram_dst.sin_family != AF_UNIX) {
+            ret = sendto(s->fd, buf, size, 0,
+                         (struct sockaddr *)&s->dgram_dst,
+                         sizeof(s->dgram_dst));
+        } else {
+            ret = send(s->fd, buf, size, 0);
+        }
+    } while (ret == -1 && errno == EINTR);
+
+    if (ret == -1 && errno == EAGAIN) {
+        net_dgram_write_poll(s, true);
+        return 0;
+    }
+    return ret;
+}
+
+static void net_dgram_send_completed(NetClientState *nc, ssize_t len)
+{
+    NetDgramState *s = DO_UPCAST(NetDgramState, nc, nc);
+
+    if (!s->read_poll) {
+        net_dgram_read_poll(s, true);
+    }
+}
+
+static void net_dgram_rs_finalize(SocketReadState *rs)
+{
+    NetDgramState *s = container_of(rs, NetDgramState, rs);
+
+    if (qemu_send_packet_async(&s->nc, rs->buf,
+                               rs->packet_len,
+                               net_dgram_send_completed) == 0) {
+        net_dgram_read_poll(s, false);
+    }
+}
+
+static void net_dgram_send(void *opaque)
+{
+    NetDgramState *s = opaque;
+    int size;
+    int ret;
+    uint8_t buf1[NET_BUFSIZE];
+    const uint8_t *buf;
+
+    size = recv(s->fd, buf1, sizeof(buf1), 0);
+    if (size < 0) {
+        if (errno != EWOULDBLOCK) {
+            goto eoc;
+        }
+    } else if (size == 0) {
+        /* end of connection */
+    eoc:
+        net_dgram_read_poll(s, false);
+        net_dgram_write_poll(s, false);
+        if (s->listen_fd != -1) {
+            qemu_set_fd_handler(s->listen_fd, net_dgram_accept, NULL, s);
+        }
+        closesocket(s->fd);
+
+        s->fd = -1;
+        net_socket_rs_init(&s->rs, net_dgram_rs_finalize, false);
+        s->nc.link_down = true;
+        memset(s->nc.info_str, 0, sizeof(s->nc.info_str));
+
+        return;
+    }
+    buf = buf1;
+
+    ret = net_fill_rstate(&s->rs, buf, size);
+
+    if (ret == -1) {
+        goto eoc;
+    }
+}
+
+static void net_dgram_send_dgram(void *opaque)
+{
+    NetDgramState *s = opaque;
+    int size;
+
+    size = recv(s->fd, s->rs.buf, sizeof(s->rs.buf), 0);
+    if (size < 0) {
+        return;
+    }
+    if (size == 0) {
+        /* end of connection */
+        net_dgram_read_poll(s, false);
+        net_dgram_write_poll(s, false);
+        return;
+    }
+    if (qemu_send_packet_async(&s->nc, s->rs.buf, size,
+                               net_dgram_send_completed) == 0) {
+        net_dgram_read_poll(s, false);
+    }
+}
+
+static int net_dgram_mcast_create(struct sockaddr_in *mcastaddr,
+                                  struct in_addr *localaddr,
+                                  Error **errp)
+{
+    struct ip_mreq imr;
+    int fd;
+    int val, ret;
+#ifdef __OpenBSD__
+    unsigned char loop;
+#else
+    int loop;
+#endif
+
+    if (!IN_MULTICAST(ntohl(mcastaddr->sin_addr.s_addr))) {
+        error_setg(errp, "specified mcastaddr %s (0x%08x) "
+                   "does not contain a multicast address",
+                   inet_ntoa(mcastaddr->sin_addr),
+                   (int)ntohl(mcastaddr->sin_addr.s_addr));
+        return -1;
+    }
+
+    fd = qemu_socket(PF_INET, SOCK_DGRAM, 0);
+    if (fd < 0) {
+        error_setg_errno(errp, errno, "can't create datagram socket");
+        return -1;
+    }
+
+    /*
+     * Allow multiple sockets to bind the same multicast ip and port by setting
+     * SO_REUSEADDR. This is the only situation where SO_REUSEADDR should be set
+     * on windows. Use socket_set_fast_reuse otherwise as it sets SO_REUSEADDR
+     * only on posix systems.
+     */
+    val = 1;
+    ret = setsockopt(fd, SOL_SOCKET, SO_REUSEADDR, &val, sizeof(val));
+    if (ret < 0) {
+        error_setg_errno(errp, errno, "can't set socket option SO_REUSEADDR");
+        goto fail;
+    }
+
+    ret = bind(fd, (struct sockaddr *)mcastaddr, sizeof(*mcastaddr));
+    if (ret < 0) {
+        error_setg_errno(errp, errno, "can't bind ip=%s to socket",
+                         inet_ntoa(mcastaddr->sin_addr));
+        goto fail;
+    }
+
+    /* Add host to multicast group */
+    imr.imr_multiaddr = mcastaddr->sin_addr;
+    if (localaddr) {
+        imr.imr_interface = *localaddr;
+    } else {
+        imr.imr_interface.s_addr = htonl(INADDR_ANY);
+    }
+
+    ret = setsockopt(fd, IPPROTO_IP, IP_ADD_MEMBERSHIP,
+                     &imr, sizeof(struct ip_mreq));
+    if (ret < 0) {
+        error_setg_errno(errp, errno,
+                         "can't add socket to multicast group %s",
+                         inet_ntoa(imr.imr_multiaddr));
+        goto fail;
+    }
+
+    /* Force mcast msgs to loopback (eg. several QEMUs in same host */
+    loop = 1;
+    ret = setsockopt(fd, IPPROTO_IP, IP_MULTICAST_LOOP,
+                     &loop, sizeof(loop));
+    if (ret < 0) {
+        error_setg_errno(errp, errno,
+                         "can't force multicast message to loopback");
+        goto fail;
+    }
+
+    /* If a bind address is given, only send packets from that address */
+    if (localaddr != NULL) {
+        ret = setsockopt(fd, IPPROTO_IP, IP_MULTICAST_IF,
+                         localaddr, sizeof(*localaddr));
+        if (ret < 0) {
+            error_setg_errno(errp, errno,
+                             "can't set the default network send interface");
+            goto fail;
+        }
+    }
+
+    qemu_socket_set_nonblock(fd);
+    return fd;
+fail:
+    if (fd >= 0) {
+        closesocket(fd);
+    }
+    return -1;
+}
+
+static void net_dgram_cleanup(NetClientState *nc)
+{
+    NetDgramState *s = DO_UPCAST(NetDgramState, nc, nc);
+    if (s->fd != -1) {
+        net_dgram_read_poll(s, false);
+        net_dgram_write_poll(s, false);
+        close(s->fd);
+        s->fd = -1;
+    }
+    if (s->listen_fd != -1) {
+        qemu_set_fd_handler(s->listen_fd, NULL, NULL, NULL);
+        closesocket(s->listen_fd);
+        s->listen_fd = -1;
+    }
+}
+
+static NetClientInfo net_dgram_socket_info = {
+    .type = NET_CLIENT_DRIVER_DGRAM,
+    .size = sizeof(NetDgramState),
+    .receive = net_dgram_receive_dgram,
+    .cleanup = net_dgram_cleanup,
+};
+
+static NetDgramState *net_dgram_fd_init_dgram(NetClientState *peer,
+                                              const char *model,
+                                              const char *name,
+                                              int fd, int is_fd,
+                                              SocketAddress *mcast,
+                                              Error **errp)
+{
+    struct sockaddr_in saddr;
+    int newfd;
+    NetClientState *nc;
+    NetDgramState *s;
+    SocketAddress *sa;
+    SocketAddressType sa_type;
+
+    sa = socket_local_address(fd, errp);
+    if (!sa) {
+        return NULL;
+    }
+    sa_type = sa->type;
+    qapi_free_SocketAddress(sa);
+
+    /*
+     * fd passed: multicast: "learn" dgram_dst address from bound address and
+     * save it. Because this may be "shared" socket from a "master" process,
+     * datagrams would be recv() by ONLY ONE process: we must "clone" this
+     * dgram socket --jjo
+     */
+
+    if (is_fd && mcast != NULL) {
+            if (convert_host_port(&saddr, mcast->u.inet.host,
+                                  mcast->u.inet.port, errp) < 0) {
+                goto err;
+            }
+            /* must be bound */
+            if (saddr.sin_addr.s_addr == 0) {
+                error_setg(errp, "can't setup multicast destination address");
+                goto err;
+            }
+            /* clone dgram socket */
+            newfd = net_dgram_mcast_create(&saddr, NULL, errp);
+            if (newfd < 0) {
+                goto err;
+            }
+            /* clone newfd to fd, close newfd */
+            dup2(newfd, fd);
+            close(newfd);
+
+    }
+
+    nc = qemu_new_net_client(&net_dgram_socket_info, peer, model, name);
+
+    s = DO_UPCAST(NetDgramState, nc, nc);
+
+    s->fd = fd;
+    s->listen_fd = -1;
+    s->send_fn = net_dgram_send_dgram;
+    net_socket_rs_init(&s->rs, net_dgram_rs_finalize, false);
+    net_dgram_read_poll(s, true);
+
+    /* mcast: save bound address as dst */
+    if (is_fd && mcast != NULL) {
+        s->dgram_dst = saddr;
+        snprintf(nc->info_str, sizeof(nc->info_str),
+                 "fd=%d (cloned mcast=%s:%d)",
+                 fd, inet_ntoa(saddr.sin_addr), ntohs(saddr.sin_port));
+    } else {
+        if (sa_type == SOCKET_ADDRESS_TYPE_UNIX) {
+            s->dgram_dst.sin_family = AF_UNIX;
+        }
+
+        snprintf(nc->info_str, sizeof(nc->info_str), "fd=%d %s", fd,
+                 SocketAddressType_str(sa_type));
+    }
+
+    return s;
+
+err:
+    closesocket(fd);
+    return NULL;
+}
+
+static void net_dgram_connect(void *opaque)
+{
+    NetDgramState *s = opaque;
+    s->send_fn = net_dgram_send;
+    net_dgram_read_poll(s, true);
+}
+
+static void net_dgram_accept(void *opaque)
+{
+    NetDgramState *s = opaque;
+    struct sockaddr_in saddr;
+    socklen_t len;
+    int fd;
+
+    for (;;) {
+        len = sizeof(saddr);
+        fd = qemu_accept(s->listen_fd, (struct sockaddr *)&saddr, &len);
+        if (fd < 0 && errno != EINTR) {
+            return;
+        } else if (fd >= 0) {
+            qemu_set_fd_handler(s->listen_fd, NULL, NULL, NULL);
+            break;
+        }
+    }
+
+    s->fd = fd;
+    s->nc.link_down = false;
+    net_dgram_connect(s);
+    snprintf(s->nc.info_str, sizeof(s->nc.info_str), "connection from %s:%d",
+             inet_ntoa(saddr.sin_addr), ntohs(saddr.sin_port));
+}
+
+static int net_dgram_mcast_init(NetClientState *peer,
+                                const char *model,
+                                const char *name,
+                                SocketAddress *remote,
+                                SocketAddress *local,
+                                Error **errp)
+{
+    NetDgramState *s;
+    int fd, ret;
+    struct sockaddr_in saddr;
+
+    if (remote->type != SOCKET_ADDRESS_TYPE_INET) {
+        error_setg(errp, "multicast only support inet type");
+        return -1;
+    }
+
+    if (convert_host_port(&saddr, remote->u.inet.host, remote->u.inet.port,
+                          errp) < 0) {
+        return -1;
+    }
+
+    if (!local) {
+        fd = net_dgram_mcast_create(&saddr, NULL, errp);
+        if (fd < 0) {
+            return -1;
+        }
+    } else {
+        switch (local->type) {
+        case SOCKET_ADDRESS_TYPE_INET: {
+            struct in_addr localaddr;
+
+            if (inet_aton(local->u.inet.host, &localaddr) == 0) {
+                error_setg(errp, "localaddr '%s' is not a valid IPv4 address",
+                           local->u.inet.host);
+                return -1;
+            }
+
+            fd = net_dgram_mcast_create(&saddr, &localaddr, errp);
+            if (fd < 0) {
+                return -1;
+            }
+            break;
+        }
+        case SOCKET_ADDRESS_TYPE_FD:
+            fd = monitor_fd_param(monitor_cur(), local->u.fd.str, errp);
+            if (fd == -1) {
+                return -1;
+            }
+            ret = qemu_socket_try_set_nonblock(fd);
+            if (ret < 0) {
+                error_setg_errno(errp, -ret, "%s: Can't use file descriptor %d",
+                                 name, fd);
+                return -1;
+            }
+            break;
+        default:
+            error_setg(errp, "only support inet or fd type for local");
+            return -1;
+        }
+    }
+
+    s = net_dgram_fd_init_dgram(peer, model, name, fd,
+                                 local->type == SOCKET_ADDRESS_TYPE_FD,
+                                 remote, errp);
+    if (!s) {
+        return -1;
+    }
+
+    s->dgram_dst = saddr;
+
+    snprintf(s->nc.info_str, sizeof(s->nc.info_str), "mcast=%s:%d",
+             inet_ntoa(saddr.sin_addr), ntohs(saddr.sin_port));
+    return 0;
+
+}
+
+static int net_dgram_udp_init(NetClientState *peer,
+                              const char *model,
+                              const char *name,
+                              SocketAddress *remote,
+                              SocketAddress *local,
+                              Error **errp)
+{
+    NetDgramState *s;
+    int fd, ret;
+    struct sockaddr_in raddr_in;
+    gchar *info_str;
+
+    if (remote) {
+        if (local->type == SOCKET_ADDRESS_TYPE_FD) {
+            error_setg(errp, "don't set remote with local.fd");
+            return -1;
+        }
+        if (remote->type != local->type) {
+            error_setg(errp, "remote and local types must be the same");
+            return -1;
+        }
+    } else {
+        if (local->type != SOCKET_ADDRESS_TYPE_FD) {
+            error_setg(errp, "type=inet requires remote parameter");
+            return -1;
+        }
+    }
+
+    switch (local->type) {
+    case SOCKET_ADDRESS_TYPE_INET: {
+        struct sockaddr_in laddr_in;
+
+        if (convert_host_port(&laddr_in, local->u.inet.host, local->u.inet.port,
+                              errp) < 0) {
+            return -1;
+        }
+
+        if (convert_host_port(&raddr_in, remote->u.inet.host,
+                              remote->u.inet.port, errp) < 0) {
+            return -1;
+        }
+
+        fd = qemu_socket(PF_INET, SOCK_DGRAM, 0);
+        if (fd < 0) {
+            error_setg_errno(errp, errno, "can't create datagram socket");
+            return -1;
+        }
+
+        ret = socket_set_fast_reuse(fd);
+        if (ret < 0) {
+            error_setg_errno(errp, errno,
+                             "can't set socket option SO_REUSEADDR");
+            closesocket(fd);
+            return -1;
+        }
+        ret = bind(fd, (struct sockaddr *)&laddr_in, sizeof(laddr_in));
+        if (ret < 0) {
+            error_setg_errno(errp, errno, "can't bind ip=%s to socket",
+                             inet_ntoa(laddr_in.sin_addr));
+            closesocket(fd);
+            return -1;
+        }
+        qemu_socket_set_nonblock(fd);
+
+        info_str = g_strdup_printf("udp=%s:%d/%s:%d",
+                 inet_ntoa(laddr_in.sin_addr), ntohs(laddr_in.sin_port),
+                 inet_ntoa(raddr_in.sin_addr), ntohs(raddr_in.sin_port));
+
+        break;
+    }
+    case SOCKET_ADDRESS_TYPE_FD:
+        fd = monitor_fd_param(monitor_cur(), local->u.fd.str, errp);
+        if (fd == -1) {
+            return -1;
+        }
+        ret = qemu_socket_try_set_nonblock(fd);
+        if (ret < 0) {
+            error_setg_errno(errp, -ret, "%s: Can't use file descriptor %d",
+                             name, fd);
+            return -1;
+        }
+        break;
+    default:
+        error_setg(errp, "only support inet or fd type for local");
+        return -1;
+    }
+
+    s = net_dgram_fd_init_dgram(peer, model, name, fd, 0, NULL, errp);
+    if (!s) {
+        return -1;
+    }
+
+    if (remote) {
+        s->dgram_dst = raddr_in;
+
+        pstrcpy(s->nc.info_str, sizeof(s->nc.info_str), info_str);
+        g_free(info_str);
+    }
+    return 0;
+}
+
+static int net_dgram_init(NetClientState *peer,
+                          const char *model,
+                          const char *name,
+                          SocketAddress *remote,
+                          SocketAddress *local,
+                          Error **errp)
+{
+    /* detect multicast address */
+    if (remote && remote->type == SOCKET_ADDRESS_TYPE_INET) {
+        struct sockaddr_in mcastaddr;
+
+        if (convert_host_port(&mcastaddr, remote->u.inet.host,
+                              remote->u.inet.port, errp) < 0) {
+            return -1;
+        }
+
+        if (IN_MULTICAST(ntohl(mcastaddr.sin_addr.s_addr))) {
+            return net_dgram_mcast_init(peer, model, name, remote, local,
+                                           errp);
+        }
+    }
+    /* unicast address */
+    if (!local) {
+        error_setg(errp, "dgram requires local= parameter");
+        return -1;
+    }
+    return net_dgram_udp_init(peer, model, name, remote, local, errp);
+}
+
+int net_init_dgram(const Netdev *netdev, const char *name,
+                   NetClientState *peer, Error **errp)
+{
+    const NetdevDgramOptions *sock;
+
+    assert(netdev->type == NET_CLIENT_DRIVER_DGRAM);
+    sock = &netdev->u.dgram;
+
+    return net_dgram_init(peer, "dgram", name, sock->remote, sock->local,
+                          errp);
+}
diff --git a/net/hub.c b/net/hub.c
index 1375738bf121..67ca53485638 100644
--- a/net/hub.c
+++ b/net/hub.c
@@ -313,6 +313,8 @@ void net_hub_check_clients(void)
             case NET_CLIENT_DRIVER_USER:
             case NET_CLIENT_DRIVER_TAP:
             case NET_CLIENT_DRIVER_SOCKET:
+            case NET_CLIENT_DRIVER_STREAM:
+            case NET_CLIENT_DRIVER_DGRAM:
             case NET_CLIENT_DRIVER_VDE:
             case NET_CLIENT_DRIVER_VHOST_USER:
                 has_host_dev = 1;
diff --git a/net/meson.build b/net/meson.build
index 754e2d1d405c..896d86d43914 100644
--- a/net/meson.build
+++ b/net/meson.build
@@ -13,6 +13,8 @@ softmmu_ss.add(files(
   'net.c',
   'queue.c',
   'socket.c',
+  'stream.c',
+  'dgram.c',
   'util.c',
 ))
 
diff --git a/net/net.c b/net/net.c
index 83300ea1b816..41e38148bc5e 100644
--- a/net/net.c
+++ b/net/net.c
@@ -48,6 +48,7 @@
 #include "qemu/qemu-print.h"
 #include "qemu/main-loop.h"
 #include "qemu/option.h"
+#include "qemu/keyval.h"
 #include "qapi/error.h"
 #include "qapi/opts-visitor.h"
 #include "sysemu/runstate.h"
@@ -1014,6 +1015,8 @@ static int (* const net_client_init_fun[NET_CLIENT_DRIVER__MAX])(
 #endif
         [NET_CLIENT_DRIVER_TAP]       = net_init_tap,
         [NET_CLIENT_DRIVER_SOCKET]    = net_init_socket,
+        [NET_CLIENT_DRIVER_STREAM]    = net_init_stream,
+        [NET_CLIENT_DRIVER_DGRAM]     = net_init_dgram,
 #ifdef CONFIG_VDE
         [NET_CLIENT_DRIVER_VDE]       = net_init_vde,
 #endif
@@ -1101,6 +1104,8 @@ void show_netdevs(void)
     int idx;
     const char *available_netdevs[] = {
         "socket",
+        "stream",
+        "dgram",
         "hubport",
         "tap",
 #ifdef CONFIG_SLIRP
@@ -1613,7 +1618,30 @@ void net_init_clients(void)
  */
 bool netdev_is_modern(const char *optarg)
 {
-    return false;
+    QemuOpts *opts;
+    bool is_modern;
+    const char *type;
+    static QemuOptsList dummy_opts = {
+        .name = "netdev",
+        .implied_opt_name = "type",
+        .head = QTAILQ_HEAD_INITIALIZER(dummy_opts.head),
+        .desc = { { } },
+    };
+
+    if (optarg[0] == '{') {
+        /* This is JSON, which means it's modern syntax */
+        return true;
+    }
+
+    opts = qemu_opts_create(&dummy_opts, NULL, false, &error_abort);
+    qemu_opts_do_parse(opts, optarg, dummy_opts.implied_opt_name,
+                       &error_abort);
+    type = qemu_opt_get(opts, "type");
+    is_modern = !g_strcmp0(type, "stream") || !g_strcmp0(type, "dgram");
+
+    qemu_opts_reset(&dummy_opts);
+
+    return is_modern;
 }
 
 /*
diff --git a/net/stream.c b/net/stream.c
new file mode 100644
index 000000000000..0851e90becca
--- /dev/null
+++ b/net/stream.c
@@ -0,0 +1,423 @@
+/*
+ * QEMU System Emulator
+ *
+ * Copyright (c) 2003-2008 Fabrice Bellard
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+#include "qemu/osdep.h"
+
+#include "net/net.h"
+#include "clients.h"
+#include "monitor/monitor.h"
+#include "qapi/error.h"
+#include "qemu/error-report.h"
+#include "qemu/option.h"
+#include "qemu/sockets.h"
+#include "qemu/iov.h"
+#include "qemu/main-loop.h"
+#include "qemu/cutils.h"
+
+typedef struct NetStreamState {
+    NetClientState nc;
+    int listen_fd;
+    int fd;
+    SocketReadState rs;
+    unsigned int send_index;      /* number of bytes sent*/
+    IOHandler *send_fn;
+    bool read_poll;               /* waiting to receive data? */
+    bool write_poll;              /* waiting to transmit data? */
+} NetStreamState;
+
+static void net_stream_accept(void *opaque);
+static void net_stream_writable(void *opaque);
+
+static void net_stream_update_fd_handler(NetStreamState *s)
+{
+    qemu_set_fd_handler(s->fd,
+                        s->read_poll ? s->send_fn : NULL,
+                        s->write_poll ? net_stream_writable : NULL,
+                        s);
+}
+
+static void net_stream_read_poll(NetStreamState *s, bool enable)
+{
+    s->read_poll = enable;
+    net_stream_update_fd_handler(s);
+}
+
+static void net_stream_write_poll(NetStreamState *s, bool enable)
+{
+    s->write_poll = enable;
+    net_stream_update_fd_handler(s);
+}
+
+static void net_stream_writable(void *opaque)
+{
+    NetStreamState *s = opaque;
+
+    net_stream_write_poll(s, false);
+
+    qemu_flush_queued_packets(&s->nc);
+}
+
+static ssize_t net_stream_receive(NetClientState *nc, const uint8_t *buf,
+                                  size_t size)
+{
+    NetStreamState *s = DO_UPCAST(NetStreamState, nc, nc);
+    uint32_t len = htonl(size);
+    struct iovec iov[] = {
+        {
+            .iov_base = &len,
+            .iov_len  = sizeof(len),
+        }, {
+            .iov_base = (void *)buf,
+            .iov_len  = size,
+        },
+    };
+    size_t remaining;
+    ssize_t ret;
+
+    remaining = iov_size(iov, 2) - s->send_index;
+    ret = iov_send(s->fd, iov, 2, s->send_index, remaining);
+
+    if (ret == -1 && errno == EAGAIN) {
+        ret = 0; /* handled further down */
+    }
+    if (ret == -1) {
+        s->send_index = 0;
+        return -errno;
+    }
+    if (ret < (ssize_t)remaining) {
+        s->send_index += ret;
+        net_stream_write_poll(s, true);
+        return 0;
+    }
+    s->send_index = 0;
+    return size;
+}
+
+static void net_stream_send_completed(NetClientState *nc, ssize_t len)
+{
+    NetStreamState *s = DO_UPCAST(NetStreamState, nc, nc);
+
+    if (!s->read_poll) {
+        net_stream_read_poll(s, true);
+    }
+}
+
+static void net_stream_rs_finalize(SocketReadState *rs)
+{
+    NetStreamState *s = container_of(rs, NetStreamState, rs);
+
+    if (qemu_send_packet_async(&s->nc, rs->buf,
+                               rs->packet_len,
+                               net_stream_send_completed) == 0) {
+        net_stream_read_poll(s, false);
+    }
+}
+
+static void net_stream_send(void *opaque)
+{
+    NetStreamState *s = opaque;
+    int size;
+    int ret;
+    uint8_t buf1[NET_BUFSIZE];
+    const uint8_t *buf;
+
+    size = recv(s->fd, buf1, sizeof(buf1), 0);
+    if (size < 0) {
+        if (errno != EWOULDBLOCK) {
+            goto eoc;
+        }
+    } else if (size == 0) {
+        /* end of connection */
+    eoc:
+        net_stream_read_poll(s, false);
+        net_stream_write_poll(s, false);
+        if (s->listen_fd != -1) {
+            qemu_set_fd_handler(s->listen_fd, net_stream_accept, NULL, s);
+        }
+        closesocket(s->fd);
+
+        s->fd = -1;
+        net_socket_rs_init(&s->rs, net_stream_rs_finalize, false);
+        s->nc.link_down = true;
+        memset(s->nc.info_str, 0, sizeof(s->nc.info_str));
+
+        return;
+    }
+    buf = buf1;
+
+    ret = net_fill_rstate(&s->rs, buf, size);
+
+    if (ret == -1) {
+        goto eoc;
+    }
+}
+
+static void net_stream_cleanup(NetClientState *nc)
+{
+    NetStreamState *s = DO_UPCAST(NetStreamState, nc, nc);
+    if (s->fd != -1) {
+        net_stream_read_poll(s, false);
+        net_stream_write_poll(s, false);
+        close(s->fd);
+        s->fd = -1;
+    }
+    if (s->listen_fd != -1) {
+        qemu_set_fd_handler(s->listen_fd, NULL, NULL, NULL);
+        closesocket(s->listen_fd);
+        s->listen_fd = -1;
+    }
+}
+
+static void net_stream_connect(void *opaque)
+{
+    NetStreamState *s = opaque;
+    s->send_fn = net_stream_send;
+    net_stream_read_poll(s, true);
+}
+
+static NetClientInfo net_stream_info = {
+    .type = NET_CLIENT_DRIVER_STREAM,
+    .size = sizeof(NetStreamState),
+    .receive = net_stream_receive,
+    .cleanup = net_stream_cleanup,
+};
+
+static NetStreamState *net_stream_fd_init_stream(NetClientState *peer,
+                                                 const char *model,
+                                                 const char *name,
+                                                 int fd, int is_connected)
+{
+    NetClientState *nc;
+    NetStreamState *s;
+
+    nc = qemu_new_net_client(&net_stream_info, peer, model, name);
+
+    snprintf(nc->info_str, sizeof(nc->info_str), "fd=%d", fd);
+
+    s = DO_UPCAST(NetStreamState, nc, nc);
+
+    s->fd = fd;
+    s->listen_fd = -1;
+    net_socket_rs_init(&s->rs, net_stream_rs_finalize, false);
+
+    /* Disable Nagle algorithm on TCP sockets to reduce latency */
+    socket_set_nodelay(fd);
+
+    if (is_connected) {
+        net_stream_connect(s);
+    } else {
+        qemu_set_fd_handler(s->fd, NULL, net_stream_connect, s);
+    }
+    return s;
+}
+
+static void net_stream_accept(void *opaque)
+{
+    NetStreamState *s = opaque;
+    struct sockaddr_in saddr;
+    socklen_t len;
+    int fd;
+
+    for (;;) {
+        len = sizeof(saddr);
+        fd = qemu_accept(s->listen_fd, (struct sockaddr *)&saddr, &len);
+        if (fd < 0 && errno != EINTR) {
+            return;
+        } else if (fd >= 0) {
+            qemu_set_fd_handler(s->listen_fd, NULL, NULL, NULL);
+            break;
+        }
+    }
+
+    s->fd = fd;
+    s->nc.link_down = false;
+    net_stream_connect(s);
+    snprintf(s->nc.info_str, sizeof(s->nc.info_str),
+             "connection from %s:%d",
+             inet_ntoa(saddr.sin_addr), ntohs(saddr.sin_port));
+}
+
+static int net_stream_server_init(NetClientState *peer,
+                                  const char *model,
+                                  const char *name,
+                                  SocketAddress *addr,
+                                  Error **errp)
+{
+    NetClientState *nc;
+    NetStreamState *s;
+    int fd, ret;
+
+    switch (addr->type) {
+    case SOCKET_ADDRESS_TYPE_INET: {
+        struct sockaddr_in saddr_in;
+
+        if (convert_host_port(&saddr_in, addr->u.inet.host, addr->u.inet.port,
+                              errp) < 0) {
+            return -1;
+        }
+
+        fd = qemu_socket(PF_INET, SOCK_STREAM, 0);
+        if (fd < 0) {
+            error_setg_errno(errp, errno, "can't create stream socket");
+            return -1;
+        }
+        qemu_socket_set_nonblock(fd);
+
+        socket_set_fast_reuse(fd);
+
+        ret = bind(fd, (struct sockaddr *)&saddr_in, sizeof(saddr_in));
+        if (ret < 0) {
+            error_setg_errno(errp, errno, "can't bind ip=%s to socket",
+                             inet_ntoa(saddr_in.sin_addr));
+            closesocket(fd);
+            return -1;
+        }
+        break;
+    }
+    case SOCKET_ADDRESS_TYPE_FD:
+        fd = monitor_fd_param(monitor_cur(), addr->u.fd.str, errp);
+        if (fd == -1) {
+            return -1;
+        }
+        ret = qemu_socket_try_set_nonblock(fd);
+        if (ret < 0) {
+            error_setg_errno(errp, -ret, "%s: Can't use file descriptor %d",
+                             name, fd);
+            return -1;
+        }
+        break;
+    default:
+        error_setg(errp, "only support inet or fd type");
+        return -1;
+    }
+
+    ret = listen(fd, 0);
+    if (ret < 0) {
+        error_setg_errno(errp, errno, "can't listen on socket");
+        closesocket(fd);
+        return -1;
+    }
+
+    nc = qemu_new_net_client(&net_stream_info, peer, model, name);
+    s = DO_UPCAST(NetStreamState, nc, nc);
+    s->fd = -1;
+    s->listen_fd = fd;
+    s->nc.link_down = true;
+    net_socket_rs_init(&s->rs, net_stream_rs_finalize, false);
+
+    qemu_set_fd_handler(s->listen_fd, net_stream_accept, NULL, s);
+    return 0;
+}
+
+static int net_stream_client_init(NetClientState *peer,
+                                  const char *model,
+                                  const char *name,
+                                  SocketAddress *addr,
+                                  Error **errp)
+{
+    NetStreamState *s;
+    int fd, connected, ret;
+    gchar *info_str;
+
+    switch (addr->type) {
+    case SOCKET_ADDRESS_TYPE_INET: {
+        struct sockaddr_in saddr_in;
+
+        if (convert_host_port(&saddr_in, addr->u.inet.host, addr->u.inet.port,
+                              errp) < 0) {
+            return -1;
+        }
+
+        fd = qemu_socket(PF_INET, SOCK_STREAM, 0);
+        if (fd < 0) {
+            error_setg_errno(errp, errno, "can't create stream socket");
+            return -1;
+        }
+        qemu_socket_set_nonblock(fd);
+
+        connected = 0;
+        for (;;) {
+            ret = connect(fd, (struct sockaddr *)&saddr_in, sizeof(saddr_in));
+            if (ret < 0) {
+                if (errno == EINTR || errno == EWOULDBLOCK) {
+                    /* continue */
+                } else if (errno == EINPROGRESS ||
+                           errno == EALREADY ||
+                           errno == EINVAL) {
+                    break;
+                } else {
+                    error_setg_errno(errp, errno, "can't connect socket");
+                    closesocket(fd);
+                    return -1;
+                }
+            } else {
+                connected = 1;
+                break;
+            }
+        }
+        info_str = g_strdup_printf("connect to %s:%d",
+                                   inet_ntoa(saddr_in.sin_addr),
+                                   ntohs(saddr_in.sin_port));
+        break;
+    }
+    case SOCKET_ADDRESS_TYPE_FD:
+        fd = monitor_fd_param(monitor_cur(), addr->u.fd.str, errp);
+        if (fd == -1) {
+            return -1;
+        }
+        ret = qemu_socket_try_set_nonblock(fd);
+        if (ret < 0) {
+            error_setg_errno(errp, -ret, "%s: Can't use file descriptor %d",
+                             name, fd);
+            return -1;
+        }
+        connected = 1;
+        info_str = g_strdup_printf("connect to fd %d", fd);
+        break;
+    default:
+        error_setg(errp, "only support inet or fd type");
+        return -1;
+    }
+
+    s = net_stream_fd_init_stream(peer, model, name, fd, connected);
+
+    pstrcpy(s->nc.info_str, sizeof(s->nc.info_str), info_str);
+    g_free(info_str);
+
+    return 0;
+}
+
+int net_init_stream(const Netdev *netdev, const char *name,
+                    NetClientState *peer, Error **errp)
+{
+    const NetdevStreamOptions *sock;
+
+    assert(netdev->type == NET_CLIENT_DRIVER_STREAM);
+    sock = &netdev->u.stream;
+
+    if (!sock->has_server || sock->server) {
+        return net_stream_server_init(peer, "stream", name, sock->addr, errp);
+    }
+    return net_stream_client_init(peer, "stream", name, sock->addr, errp);
+}
diff --git a/qapi/net.json b/qapi/net.json
index d6f7cfd4d656..32a9b1a5ac6c 100644
--- a/qapi/net.json
+++ b/qapi/net.json
@@ -7,6 +7,7 @@
 ##
 
 { 'include': 'common.json' }
+{ 'include': 'sockets.json' }
 
 ##
 # @set_link:
@@ -566,6 +567,42 @@
     '*isolated':  'bool' },
   'if': 'CONFIG_VMNET' }
 
+##
+# @NetdevStreamOptions:
+#
+# Configuration info for stream socket netdev
+#
+# @addr: socket address to listen on (server=true)
+#        or connect to (server=false)
+# @server: create server socket (default: true)
+#
+# Since: 7.1
+##
+{ 'struct': 'NetdevStreamOptions',
+  'data': {
+    'addr':   'SocketAddress',
+    '*server': 'bool' } }
+
+##
+# @NetdevDgramOptions:
+#
+# Configuration info for datagram socket netdev.
+#
+# @remote: remote address
+# @local: local address
+#
+# The code checks there is at least one of these options and reports an error
+# if not. If remote address is present and it's a multicast address, local
+# address is optional. Otherwise local address is required and remote address
+# is optional.
+#
+# Since: 7.1
+##
+{ 'struct': 'NetdevDgramOptions',
+  'data': {
+    '*local':  'SocketAddress',
+    '*remote': 'SocketAddress' } }
+
 ##
 # @NetClientDriver:
 #
@@ -579,8 +616,9 @@
 #        @vmnet-bridged since 7.1
 ##
 { 'enum': 'NetClientDriver',
-  'data': [ 'none', 'nic', 'user', 'tap', 'l2tpv3', 'socket', 'vde',
-            'bridge', 'hubport', 'netmap', 'vhost-user', 'vhost-vdpa',
+  'data': [ 'none', 'nic', 'user', 'tap', 'l2tpv3', 'socket', 'stream',
+            'dgram', 'vde', 'bridge', 'hubport', 'netmap', 'vhost-user',
+            'vhost-vdpa',
             { 'name': 'vmnet-host', 'if': 'CONFIG_VMNET' },
             { 'name': 'vmnet-shared', 'if': 'CONFIG_VMNET' },
             { 'name': 'vmnet-bridged', 'if': 'CONFIG_VMNET' }] }
@@ -610,6 +648,8 @@
     'tap':      'NetdevTapOptions',
     'l2tpv3':   'NetdevL2TPv3Options',
     'socket':   'NetdevSocketOptions',
+    'stream':   'NetdevStreamOptions',
+    'dgram':    'NetdevDgramOptions',
     'vde':      'NetdevVdeOptions',
     'bridge':   'NetdevBridgeOptions',
     'hubport':  'NetdevHubPortOptions',
diff --git a/qemu-options.hx b/qemu-options.hx
index 377d22fbd82f..2be0cbd2549b 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -2722,6 +2722,18 @@ DEF("netdev", HAS_ARG, QEMU_OPTION_netdev,
     "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n"
     "                configure a network backend to connect to another network\n"
     "                using an UDP tunnel\n"
+    "-netdev stream,id=str[,server=on|off],addr.type=inet,addr.host=host,addr.port=port\n"
+    "-netdev stream,id=str[,server=on|off],addr.type=fd,addr.str=h\n"
+    "                configure a network backend to connect to another network\n"
+    "                using a socket connection in stream mode.\n"
+    "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=inet,local.host=addr]\n"
+    "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=fd,local.str=h]\n"
+    "                configure a network backend to connect to a multicast maddr and port\n"
+    "                use 'local.host=addr' to specify the host address to send packets from\n"
+    "-netdev dgram,id=str,local.type=inet,local.host=addr,local.port=port[,remote.type=inet,remote.host=addr,remote.port=port]\n"
+    "-netdev dgram,id=str,local.type=fd,local.str=h\n"
+    "                configure a network backend to connect to another network\n"
+    "                using an UDP tunnel\n"
 #ifdef CONFIG_VDE
     "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n"
     "                configure a network backend to connect to port 'n' of a vde switch\n"
-- 
2.36.1

Re: [PATCH v5 05/12] qapi: net: add stream and dgram netdevs

[Markus Armbruster]

Laurent Vivier <lvivier@redhat.com> writes:

> Copied from socket netdev file and modified to use SocketAddress
> to be able to introduce new features like unix socket.
>
> "udp" and "mcast" are squashed into dgram netdev, multicast is detected
> according to the IP address type.
> "listen" and "connect" modes are managed by stream netdev. An optional
> parameter "server" defines the mode (server by default)
>
> Signed-off-by: Laurent Vivier <lvivier@redhat.com>
> Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
> ---

[...]

> diff --git a/qapi/net.json b/qapi/net.json
> index d6f7cfd4d656..32a9b1a5ac6c 100644
> --- a/qapi/net.json
> +++ b/qapi/net.json
> @@ -7,6 +7,7 @@
>  ##
>  
>  { 'include': 'common.json' }
> +{ 'include': 'sockets.json' }
>  
>  ##
>  # @set_link:
> @@ -566,6 +567,42 @@
>      '*isolated':  'bool' },
>    'if': 'CONFIG_VMNET' }
>  
> +##
> +# @NetdevStreamOptions:
> +#
> +# Configuration info for stream socket netdev
> +#
> +# @addr: socket address to listen on (server=true)
> +#        or connect to (server=false)

Only some address types are valid, as discussed in review of v4.
Shouldn't we document that here?

> +# @server: create server socket (default: true)
> +#
> +# Since: 7.1
> +##
> +{ 'struct': 'NetdevStreamOptions',
> +  'data': {
> +    'addr':   'SocketAddress',
> +    '*server': 'bool' } }
> +
> +##
> +# @NetdevDgramOptions:
> +#
> +# Configuration info for datagram socket netdev.
> +#
> +# @remote: remote address
> +# @local: local address
> +#
> +# The code checks there is at least one of these options and reports an error
> +# if not. If remote address is present and it's a multicast address, local
> +# address is optional. Otherwise local address is required and remote address
> +# is optional.
> +#
> +# Since: 7.1
> +##
> +{ 'struct': 'NetdevDgramOptions',
> +  'data': {
> +    '*local':  'SocketAddress',
> +    '*remote': 'SocketAddress' } }

Hard to see, but the space in "} }" is funny: it's U+00A0
(NO-BREAK-SPACE) encoded in UTF-8.  Make it a plain old space, please.

Only some combinations of address types are valid, as discussed in
review of v4.  Shouldn't we document that here?

> +
>  ##
>  # @NetClientDriver:
>  #
> @@ -579,8 +616,9 @@
>  #        @vmnet-bridged since 7.1
>  ##
>  { 'enum': 'NetClientDriver',
> -  'data': [ 'none', 'nic', 'user', 'tap', 'l2tpv3', 'socket', 'vde',
> -            'bridge', 'hubport', 'netmap', 'vhost-user', 'vhost-vdpa',
> +  'data': [ 'none', 'nic', 'user', 'tap', 'l2tpv3', 'socket', 'stream',
> +            'dgram', 'vde', 'bridge', 'hubport', 'netmap', 'vhost-user',
> +            'vhost-vdpa',
>              { 'name': 'vmnet-host', 'if': 'CONFIG_VMNET' },
>              { 'name': 'vmnet-shared', 'if': 'CONFIG_VMNET' },
>              { 'name': 'vmnet-bridged', 'if': 'CONFIG_VMNET' }] }
> @@ -610,6 +648,8 @@
>      'tap':      'NetdevTapOptions',
>      'l2tpv3':   'NetdevL2TPv3Options',
>      'socket':   'NetdevSocketOptions',
> +    'stream':   'NetdevStreamOptions',
> +    'dgram':    'NetdevDgramOptions',
>      'vde':      'NetdevVdeOptions',
>      'bridge':   'NetdevBridgeOptions',
>      'hubport':  'NetdevHubPortOptions',
> diff --git a/qemu-options.hx b/qemu-options.hx
> index 377d22fbd82f..2be0cbd2549b 100644
> --- a/qemu-options.hx
> +++ b/qemu-options.hx
> @@ -2722,6 +2722,18 @@ DEF("netdev", HAS_ARG, QEMU_OPTION_netdev,
>      "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n"
>      "                configure a network backend to connect to another network\n"
>      "                using an UDP tunnel\n"
> +    "-netdev stream,id=str[,server=on|off],addr.type=inet,addr.host=host,addr.port=port\n"
> +    "-netdev stream,id=str[,server=on|off],addr.type=fd,addr.str=h\n"
> +    "                configure a network backend to connect to another network\n"
> +    "                using a socket connection in stream mode.\n"
> +    "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=inet,local.host=addr]\n"
> +    "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=fd,local.str=h]\n"
> +    "                configure a network backend to connect to a multicast maddr and port\n"
> +    "                use 'local.host=addr' to specify the host address to send packets from\n"

I think we use ``local.host=addr`` markup.

> +    "-netdev dgram,id=str,local.type=inet,local.host=addr,local.port=port[,remote.type=inet,remote.host=addr,remote.port=port]\n"
> +    "-netdev dgram,id=str,local.type=fd,local.str=h\n"
> +    "                configure a network backend to connect to another network\n"
> +    "                using an UDP tunnel\n"
>  #ifdef CONFIG_VDE
>      "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n"
>      "                configure a network backend to connect to port 'n' of a vde switch\n"

I intend to look at this again after we decide what to do about the doc
comments in net.json.

Re: [PATCH v5 05/12] qapi: net: add stream and dgram netdevs

[Laurent Vivier]

On 29/06/2022 13:20, Markus Armbruster wrote:
> Laurent Vivier <lvivier@redhat.com> writes:
> 
>> Copied from socket netdev file and modified to use SocketAddress
>> to be able to introduce new features like unix socket.
>>
>> "udp" and "mcast" are squashed into dgram netdev, multicast is detected
>> according to the IP address type.
>> "listen" and "connect" modes are managed by stream netdev. An optional
>> parameter "server" defines the mode (server by default)
>>
>> Signed-off-by: Laurent Vivier <lvivier@redhat.com>
>> Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
>> ---
> 
> [...]
...
>> +# @server: create server socket (default: true)
>> +#
>> +# Since: 7.1
>> +##
>> +{ 'struct': 'NetdevStreamOptions',
>> +  'data': {
>> +    'addr':   'SocketAddress',
>> +    '*server': 'bool' } }
>> +
>> +##
>> +# @NetdevDgramOptions:
>> +#
>> +# Configuration info for datagram socket netdev.
>> +#
>> +# @remote: remote address
>> +# @local: local address
>> +#
>> +# The code checks there is at least one of these options and reports an error
>> +# if not. If remote address is present and it's a multicast address, local
>> +# address is optional. Otherwise local address is required and remote address
>> +# is optional.
>> +#
>> +# Since: 7.1
>> +##
>> +{ 'struct': 'NetdevDgramOptions',
>> +  'data': {
>> +    '*local':  'SocketAddress',
>> +    '*remote': 'SocketAddress' } }
> 
> Hard to see, but the space in "} }" is funny: it's U+00A0
> (NO-BREAK-SPACE) encoded in UTF-8.  Make it a plain old space, please.
> 

I'm sorry, this happens sometime because I use a french macintosh keyboard, and to do '}' 
I have to press alt+) and if I don't release fast enough 'alt' I produce an alt+space that 
seems to generate this (invisible) character code...
(I've the same problem with '|' = alt+shift+l, but bash doesn't like it)

I'm updating the series with all your comments.

Thank you for the reviews.

Laurent

Re: [PATCH v5 05/12] qapi: net: add stream and dgram netdevs

[Markus Armbruster]

Laurent Vivier <lvivier@redhat.com> writes:

> Copied from socket netdev file and modified to use SocketAddress
> to be able to introduce new features like unix socket.
>
> "udp" and "mcast" are squashed into dgram netdev, multicast is detected
> according to the IP address type.
> "listen" and "connect" modes are managed by stream netdev. An optional
> parameter "server" defines the mode (server by default)
>
> Signed-off-by: Laurent Vivier <lvivier@redhat.com>
> Reviewed-by: Stefano Brivio <sbrivio@redhat.com>

Suggest to add a summary of what we learned during review.  Let me try.

  The two new types need to parsed the modern way with -netdev, because
  <explanation goes here>.

  The previous commit paved the way for parsing the modern way, but
  omitted one detail: how to pick modern vs. traditional, in
  netdev_is_modern().

  We want to pick based on the value of parameter "type".  But how to
  extract it from the option argument?

  Parsing the option argument, either the modern or the traditional way,
  extracts it for us, but only if parsing succeeds.

  If parsing fails, there is no good option.  No matter which parser we
  pick, it'll be the wrong one for some arguments, and the error
  reporting will be confusing.

  Fortunately, the traditional parser accepts *anything* when called in
  a certain way.  This maximizes our chance to extract the value of
  "type", and in turn minimizes the risk of confusing error reporting.

How do you like it?

[PATCH v5 06/12] net: stream: Don't ignore EINVAL on netdev socket connection

[Laurent Vivier]

From: Stefano Brivio <sbrivio@redhat.com>

Other errors are treated as failure by net_stream_client_init(),
but if connect() returns EINVAL, we'll fail silently. Remove the
related exception.

Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
[lvivier: applied to net/stream.c]
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
---
 net/stream.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/net/stream.c b/net/stream.c
index 0851e90becca..e8afbaca50b6 100644
--- a/net/stream.c
+++ b/net/stream.c
@@ -363,8 +363,7 @@ static int net_stream_client_init(NetClientState *peer,
                 if (errno == EINTR || errno == EWOULDBLOCK) {
                     /* continue */
                 } else if (errno == EINPROGRESS ||
-                           errno == EALREADY ||
-                           errno == EINVAL) {
+                           errno == EALREADY) {
                     break;
                 } else {
                     error_setg_errno(errp, errno, "can't connect socket");
-- 
2.36.1

[PATCH v5 07/12] net: stream: add unix socket

[Laurent Vivier]

Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
---
 net/stream.c    | 106 +++++++++++++++++++++++++++++++++++++++++++++---
 qemu-options.hx |   1 +
 2 files changed, 102 insertions(+), 5 deletions(-)

diff --git a/net/stream.c b/net/stream.c
index e8afbaca50b6..1a2bc237707e 100644
--- a/net/stream.c
+++ b/net/stream.c
@@ -235,7 +235,7 @@ static NetStreamState *net_stream_fd_init_stream(NetClientState *peer,
 static void net_stream_accept(void *opaque)
 {
     NetStreamState *s = opaque;
-    struct sockaddr_in saddr;
+    struct sockaddr_storage saddr;
     socklen_t len;
     int fd;
 
@@ -253,9 +253,27 @@ static void net_stream_accept(void *opaque)
     s->fd = fd;
     s->nc.link_down = false;
     net_stream_connect(s);
-    snprintf(s->nc.info_str, sizeof(s->nc.info_str),
-             "connection from %s:%d",
-             inet_ntoa(saddr.sin_addr), ntohs(saddr.sin_port));
+    switch (saddr.ss_family) {
+    case AF_INET: {
+        struct sockaddr_in *saddr_in = (struct sockaddr_in *)&saddr;
+
+        snprintf(s->nc.info_str, sizeof(s->nc.info_str),
+                 "connection from %s:%d",
+                 inet_ntoa(saddr_in->sin_addr), ntohs(saddr_in->sin_port));
+        break;
+    }
+    case AF_UNIX: {
+        struct sockaddr_un saddr_un;
+
+        len = sizeof(saddr_un);
+        getsockname(s->listen_fd, (struct sockaddr *)&saddr_un, &len);
+        snprintf(s->nc.info_str, sizeof(s->nc.info_str),
+                 "connect from %s", saddr_un.sun_path);
+        break;
+    }
+    default:
+        g_assert_not_reached();
+    }
 }
 
 static int net_stream_server_init(NetClientState *peer,
@@ -295,6 +313,42 @@ static int net_stream_server_init(NetClientState *peer,
         }
         break;
     }
+    case SOCKET_ADDRESS_TYPE_UNIX: {
+        struct sockaddr_un saddr_un;
+
+        ret = unlink(addr->u.q_unix.path);
+        if (ret < 0 && errno != ENOENT) {
+            error_setg_errno(errp, errno, "failed to unlink socket %s",
+                             addr->u.q_unix.path);
+            return -1;
+        }
+
+        saddr_un.sun_family = PF_UNIX;
+        ret = snprintf(saddr_un.sun_path, sizeof(saddr_un.sun_path), "%s",
+                       addr->u.q_unix.path);
+        if (ret < 0 || ret >= sizeof(saddr_un.sun_path)) {
+            error_setg(errp, "UNIX socket path '%s' is too long",
+                       addr->u.q_unix.path);
+            error_append_hint(errp, "Path must be less than %zu bytes\n",
+                              sizeof(saddr_un.sun_path));
+        }
+
+        fd = qemu_socket(PF_UNIX, SOCK_STREAM, 0);
+        if (fd < 0) {
+            error_setg_errno(errp, errno, "can't create stream socket");
+            return -1;
+        }
+        qemu_socket_set_nonblock(fd);
+
+        ret = bind(fd, (struct sockaddr *)&saddr_un, sizeof(saddr_un));
+        if (ret < 0) {
+            error_setg_errno(errp, errno, "can't create socket with path: %s",
+                             saddr_un.sun_path);
+            closesocket(fd);
+            return -1;
+        }
+        break;
+    }
     case SOCKET_ADDRESS_TYPE_FD:
         fd = monitor_fd_param(monitor_cur(), addr->u.fd.str, errp);
         if (fd == -1) {
@@ -380,6 +434,48 @@ static int net_stream_client_init(NetClientState *peer,
                                    ntohs(saddr_in.sin_port));
         break;
     }
+    case SOCKET_ADDRESS_TYPE_UNIX: {
+        struct sockaddr_un saddr_un;
+
+        saddr_un.sun_family = PF_UNIX;
+        ret = snprintf(saddr_un.sun_path, sizeof(saddr_un.sun_path), "%s",
+                       addr->u.q_unix.path);
+        if (ret < 0 || ret >= sizeof(saddr_un.sun_path)) {
+            error_setg(errp, "UNIX socket path '%s' is too long",
+                       addr->u.q_unix.path);
+            error_append_hint(errp, "Path must be less than %zu bytes\n",
+                              sizeof(saddr_un.sun_path));
+        }
+
+        fd = qemu_socket(PF_UNIX, SOCK_STREAM, 0);
+        if (fd < 0) {
+            error_setg_errno(errp, errno, "can't create stream socket");
+            return -1;
+        }
+        qemu_socket_set_nonblock(fd);
+
+        connected = 0;
+        for (;;) {
+            ret = connect(fd, (struct sockaddr *)&saddr_un, sizeof(saddr_un));
+            if (ret < 0) {
+                if (errno == EINTR || errno == EWOULDBLOCK) {
+                    /* continue */
+                } else if (errno == EAGAIN ||
+                           errno == EALREADY) {
+                    break;
+                } else {
+                    error_setg_errno(errp, errno, "can't connect socket");
+                    closesocket(fd);
+                    return -1;
+                }
+            } else {
+                connected = 1;
+                break;
+            }
+        }
+        info_str = g_strdup_printf(" connect to %s", saddr_un.sun_path);
+        break;
+    }
     case SOCKET_ADDRESS_TYPE_FD:
         fd = monitor_fd_param(monitor_cur(), addr->u.fd.str, errp);
         if (fd == -1) {
@@ -395,7 +491,7 @@ static int net_stream_client_init(NetClientState *peer,
         info_str = g_strdup_printf("connect to fd %d", fd);
         break;
     default:
-        error_setg(errp, "only support inet or fd type");
+        error_setg(errp, "only support inet, unix or fd type");
         return -1;
     }
 
diff --git a/qemu-options.hx b/qemu-options.hx
index 2be0cbd2549b..23ff12c1b183 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -2723,6 +2723,7 @@ DEF("netdev", HAS_ARG, QEMU_OPTION_netdev,
     "                configure a network backend to connect to another network\n"
     "                using an UDP tunnel\n"
     "-netdev stream,id=str[,server=on|off],addr.type=inet,addr.host=host,addr.port=port\n"
+    "-netdev stream,id=str[,server=on|off],addr.type=unix,addr.path=path\n"
     "-netdev stream,id=str[,server=on|off],addr.type=fd,addr.str=h\n"
     "                configure a network backend to connect to another network\n"
     "                using a socket connection in stream mode.\n"
-- 
2.36.1

Re: [PATCH v5 07/12] net: stream: add unix socket

[Dr. David Alan Gilbert]

* Laurent Vivier (lvivier@redhat.com) wrote:
> Signed-off-by: Laurent Vivier <lvivier@redhat.com>
> Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
> ---
>  net/stream.c    | 106 +++++++++++++++++++++++++++++++++++++++++++++---
>  qemu-options.hx |   1 +
>  2 files changed, 102 insertions(+), 5 deletions(-)
> 
> diff --git a/net/stream.c b/net/stream.c
> index e8afbaca50b6..1a2bc237707e 100644
> --- a/net/stream.c
> +++ b/net/stream.c
> @@ -235,7 +235,7 @@ static NetStreamState *net_stream_fd_init_stream(NetClientState *peer,
>  static void net_stream_accept(void *opaque)
>  {
>      NetStreamState *s = opaque;
> -    struct sockaddr_in saddr;
> +    struct sockaddr_storage saddr;
>      socklen_t len;
>      int fd;
>  
> @@ -253,9 +253,27 @@ static void net_stream_accept(void *opaque)
>      s->fd = fd;
>      s->nc.link_down = false;
>      net_stream_connect(s);
> -    snprintf(s->nc.info_str, sizeof(s->nc.info_str),
> -             "connection from %s:%d",
> -             inet_ntoa(saddr.sin_addr), ntohs(saddr.sin_port));
> +    switch (saddr.ss_family) {
> +    case AF_INET: {
> +        struct sockaddr_in *saddr_in = (struct sockaddr_in *)&saddr;
> +
> +        snprintf(s->nc.info_str, sizeof(s->nc.info_str),
> +                 "connection from %s:%d",
> +                 inet_ntoa(saddr_in->sin_addr), ntohs(saddr_in->sin_port));
> +        break;
> +    }
> +    case AF_UNIX: {
> +        struct sockaddr_un saddr_un;
> +
> +        len = sizeof(saddr_un);
> +        getsockname(s->listen_fd, (struct sockaddr *)&saddr_un, &len);
> +        snprintf(s->nc.info_str, sizeof(s->nc.info_str),
> +                 "connect from %s", saddr_un.sun_path);
> +        break;
> +    }
> +    default:
> +        g_assert_not_reached();
> +    }
>  }
>  
>  static int net_stream_server_init(NetClientState *peer,
> @@ -295,6 +313,42 @@ static int net_stream_server_init(NetClientState *peer,
>          }
>          break;
>      }
> +    case SOCKET_ADDRESS_TYPE_UNIX: {
> +        struct sockaddr_un saddr_un;
> +
> +        ret = unlink(addr->u.q_unix.path);
> +        if (ret < 0 && errno != ENOENT) {
> +            error_setg_errno(errp, errno, "failed to unlink socket %s",
> +                             addr->u.q_unix.path);
> +            return -1;
> +        }
> +
> +        saddr_un.sun_family = PF_UNIX;
> +        ret = snprintf(saddr_un.sun_path, sizeof(saddr_un.sun_path), "%s",
> +                       addr->u.q_unix.path);
> +        if (ret < 0 || ret >= sizeof(saddr_un.sun_path)) {
> +            error_setg(errp, "UNIX socket path '%s' is too long",
> +                       addr->u.q_unix.path);
> +            error_append_hint(errp, "Path must be less than %zu bytes\n",
> +                              sizeof(saddr_un.sun_path));

return here?
> +
> +        fd = qemu_socket(PF_UNIX, SOCK_STREAM, 0);
> +        if (fd < 0) {
> +            error_setg_errno(errp, errno, "can't create stream socket");
> +            return -1;
> +        }
> +        qemu_socket_set_nonblock(fd);
> +
> +        ret = bind(fd, (struct sockaddr *)&saddr_un, sizeof(saddr_un));
> +        if (ret < 0) {
> +            error_setg_errno(errp, errno, "can't create socket with path: %s",
> +                             saddr_un.sun_path);
> +            closesocket(fd);
> +            return -1;
> +        }
> +        break;
> +    }
>      case SOCKET_ADDRESS_TYPE_FD:
>          fd = monitor_fd_param(monitor_cur(), addr->u.fd.str, errp);
>          if (fd == -1) {
> @@ -380,6 +434,48 @@ static int net_stream_client_init(NetClientState *peer,
>                                     ntohs(saddr_in.sin_port));
>          break;
>      }
> +    case SOCKET_ADDRESS_TYPE_UNIX: {
> +        struct sockaddr_un saddr_un;
> +
> +        saddr_un.sun_family = PF_UNIX;
> +        ret = snprintf(saddr_un.sun_path, sizeof(saddr_un.sun_path), "%s",
> +                       addr->u.q_unix.path);
> +        if (ret < 0 || ret >= sizeof(saddr_un.sun_path)) {
> +            error_setg(errp, "UNIX socket path '%s' is too long",
> +                       addr->u.q_unix.path);
> +            error_append_hint(errp, "Path must be less than %zu bytes\n",
> +                              sizeof(saddr_un.sun_path));

and here

(Can you not do a lot of this just by calling util/qemu-sockets.c
unix_connect?)

> +        }
> +
> +        fd = qemu_socket(PF_UNIX, SOCK_STREAM, 0);
> +        if (fd < 0) {
> +            error_setg_errno(errp, errno, "can't create stream socket");
> +            return -1;
> +        }
> +        qemu_socket_set_nonblock(fd);
> +
> +        connected = 0;
> +        for (;;) {
> +            ret = connect(fd, (struct sockaddr *)&saddr_un, sizeof(saddr_un));
> +            if (ret < 0) {
> +                if (errno == EINTR || errno == EWOULDBLOCK) {
> +                    /* continue */
> +                } else if (errno == EAGAIN ||
> +                           errno == EALREADY) {
> +                    break;
> +                } else {
> +                    error_setg_errno(errp, errno, "can't connect socket");
> +                    closesocket(fd);
> +                    return -1;
> +                }
> +            } else {
> +                connected = 1;
> +                break;
> +            }
> +        }
> +        info_str = g_strdup_printf(" connect to %s", saddr_un.sun_path);
> +        break;
> +    }
>      case SOCKET_ADDRESS_TYPE_FD:
>          fd = monitor_fd_param(monitor_cur(), addr->u.fd.str, errp);
>          if (fd == -1) {
> @@ -395,7 +491,7 @@ static int net_stream_client_init(NetClientState *peer,
>          info_str = g_strdup_printf("connect to fd %d", fd);
>          break;
>      default:
> -        error_setg(errp, "only support inet or fd type");
> +        error_setg(errp, "only support inet, unix or fd type");
>          return -1;
>      }
>  
> diff --git a/qemu-options.hx b/qemu-options.hx
> index 2be0cbd2549b..23ff12c1b183 100644
> --- a/qemu-options.hx
> +++ b/qemu-options.hx
> @@ -2723,6 +2723,7 @@ DEF("netdev", HAS_ARG, QEMU_OPTION_netdev,
>      "                configure a network backend to connect to another network\n"
>      "                using an UDP tunnel\n"
>      "-netdev stream,id=str[,server=on|off],addr.type=inet,addr.host=host,addr.port=port\n"
> +    "-netdev stream,id=str[,server=on|off],addr.type=unix,addr.path=path\n"
>      "-netdev stream,id=str[,server=on|off],addr.type=fd,addr.str=h\n"
>      "                configure a network backend to connect to another network\n"
>      "                using a socket connection in stream mode.\n"
> -- 
> 2.36.1
> 
-- 
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK

Re: [PATCH v5 07/12] net: stream: add unix socket

[Laurent Vivier]

On 30/06/2022 11:28, Dr. David Alan Gilbert wrote:
> * Laurent Vivier (lvivier@redhat.com) wrote:
>> Signed-off-by: Laurent Vivier <lvivier@redhat.com>
>> Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
>> ---
>>   net/stream.c    | 106 +++++++++++++++++++++++++++++++++++++++++++++---
>>   qemu-options.hx |   1 +
>>   2 files changed, 102 insertions(+), 5 deletions(-)
>>
>> diff --git a/net/stream.c b/net/stream.c
>> index e8afbaca50b6..1a2bc237707e 100644
>> --- a/net/stream.c
>> +++ b/net/stream.c
>> @@ -235,7 +235,7 @@ static NetStreamState *net_stream_fd_init_stream(NetClientState *peer,
>>   static void net_stream_accept(void *opaque)
>>   {
>>       NetStreamState *s = opaque;
>> -    struct sockaddr_in saddr;
>> +    struct sockaddr_storage saddr;
>>       socklen_t len;
>>       int fd;
>>   
>> @@ -253,9 +253,27 @@ static void net_stream_accept(void *opaque)
>>       s->fd = fd;
>>       s->nc.link_down = false;
>>       net_stream_connect(s);
>> -    snprintf(s->nc.info_str, sizeof(s->nc.info_str),
>> -             "connection from %s:%d",
>> -             inet_ntoa(saddr.sin_addr), ntohs(saddr.sin_port));
>> +    switch (saddr.ss_family) {
>> +    case AF_INET: {
>> +        struct sockaddr_in *saddr_in = (struct sockaddr_in *)&saddr;
>> +
>> +        snprintf(s->nc.info_str, sizeof(s->nc.info_str),
>> +                 "connection from %s:%d",
>> +                 inet_ntoa(saddr_in->sin_addr), ntohs(saddr_in->sin_port));
>> +        break;
>> +    }
>> +    case AF_UNIX: {
>> +        struct sockaddr_un saddr_un;
>> +
>> +        len = sizeof(saddr_un);
>> +        getsockname(s->listen_fd, (struct sockaddr *)&saddr_un, &len);
>> +        snprintf(s->nc.info_str, sizeof(s->nc.info_str),
>> +                 "connect from %s", saddr_un.sun_path);
>> +        break;
>> +    }
>> +    default:
>> +        g_assert_not_reached();
>> +    }
>>   }
>>   
>>   static int net_stream_server_init(NetClientState *peer,
>> @@ -295,6 +313,42 @@ static int net_stream_server_init(NetClientState *peer,
>>           }
>>           break;
>>       }
>> +    case SOCKET_ADDRESS_TYPE_UNIX: {
>> +        struct sockaddr_un saddr_un;
>> +
>> +        ret = unlink(addr->u.q_unix.path);
>> +        if (ret < 0 && errno != ENOENT) {
>> +            error_setg_errno(errp, errno, "failed to unlink socket %s",
>> +                             addr->u.q_unix.path);
>> +            return -1;
>> +        }
>> +
>> +        saddr_un.sun_family = PF_UNIX;
>> +        ret = snprintf(saddr_un.sun_path, sizeof(saddr_un.sun_path), "%s",
>> +                       addr->u.q_unix.path);
>> +        if (ret < 0 || ret >= sizeof(saddr_un.sun_path)) {
>> +            error_setg(errp, "UNIX socket path '%s' is too long",
>> +                       addr->u.q_unix.path);
>> +            error_append_hint(errp, "Path must be less than %zu bytes\n",
>> +                              sizeof(saddr_un.sun_path));
> 
> return here?
>> +
>> +        fd = qemu_socket(PF_UNIX, SOCK_STREAM, 0);
>> +        if (fd < 0) {
>> +            error_setg_errno(errp, errno, "can't create stream socket");
>> +            return -1;
>> +        }
>> +        qemu_socket_set_nonblock(fd);
>> +
>> +        ret = bind(fd, (struct sockaddr *)&saddr_un, sizeof(saddr_un));
>> +        if (ret < 0) {
>> +            error_setg_errno(errp, errno, "can't create socket with path: %s",
>> +                             saddr_un.sun_path);
>> +            closesocket(fd);
>> +            return -1;
>> +        }
>> +        break;
>> +    }
>>       case SOCKET_ADDRESS_TYPE_FD:
>>           fd = monitor_fd_param(monitor_cur(), addr->u.fd.str, errp);
>>           if (fd == -1) {
>> @@ -380,6 +434,48 @@ static int net_stream_client_init(NetClientState *peer,
>>                                      ntohs(saddr_in.sin_port));
>>           break;
>>       }
>> +    case SOCKET_ADDRESS_TYPE_UNIX: {
>> +        struct sockaddr_un saddr_un;
>> +
>> +        saddr_un.sun_family = PF_UNIX;
>> +        ret = snprintf(saddr_un.sun_path, sizeof(saddr_un.sun_path), "%s",
>> +                       addr->u.q_unix.path);
>> +        if (ret < 0 || ret >= sizeof(saddr_un.sun_path)) {
>> +            error_setg(errp, "UNIX socket path '%s' is too long",
>> +                       addr->u.q_unix.path);
>> +            error_append_hint(errp, "Path must be less than %zu bytes\n",
>> +                              sizeof(saddr_un.sun_path));
> 
> and here
> 
> (Can you not do a lot of this just by calling util/qemu-sockets.c
> unix_connect?)

Thank you for your comments.

In fact, in the last patch, I replace all this code by QIO functions.

I keep all the intermediate patches to be sure I don't break anything from the original 
behaviour of -netdev socket and to ease debugging by allowing to bisect between the 
different steps in the change.

Thanks,
Laurent

Re: [PATCH v5 07/12] net: stream: add unix socket

[Markus Armbruster]

Laurent Vivier <lvivier@redhat.com> writes:

> On 30/06/2022 11:28, Dr. David Alan Gilbert wrote:
>> * Laurent Vivier (lvivier@redhat.com) wrote:
>>> Signed-off-by: Laurent Vivier <lvivier@redhat.com>
>>> Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
>>> ---
>>>   net/stream.c    | 106 +++++++++++++++++++++++++++++++++++++++++++++---
>>>   qemu-options.hx |   1 +
>>>   2 files changed, 102 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/net/stream.c b/net/stream.c
>>> index e8afbaca50b6..1a2bc237707e 100644
>>> --- a/net/stream.c
>>> +++ b/net/stream.c
>>> @@ -235,7 +235,7 @@ static NetStreamState *net_stream_fd_init_stream(NetClientState *peer,
>>>   static void net_stream_accept(void *opaque)
>>>   {
>>>       NetStreamState *s = opaque;
>>> -    struct sockaddr_in saddr;
>>> +    struct sockaddr_storage saddr;
>>>       socklen_t len;
>>>       int fd;
>>>   
>>> @@ -253,9 +253,27 @@ static void net_stream_accept(void *opaque)
>>>       s->fd = fd;
>>>       s->nc.link_down = false;
>>>       net_stream_connect(s);
>>> -    snprintf(s->nc.info_str, sizeof(s->nc.info_str),
>>> -             "connection from %s:%d",
>>> -             inet_ntoa(saddr.sin_addr), ntohs(saddr.sin_port));
>>> +    switch (saddr.ss_family) {
>>> +    case AF_INET: {
>>> +        struct sockaddr_in *saddr_in = (struct sockaddr_in *)&saddr;
>>> +
>>> +        snprintf(s->nc.info_str, sizeof(s->nc.info_str),
>>> +                 "connection from %s:%d",
>>> +                 inet_ntoa(saddr_in->sin_addr), ntohs(saddr_in->sin_port));
>>> +        break;
>>> +    }
>>> +    case AF_UNIX: {
>>> +        struct sockaddr_un saddr_un;
>>> +
>>> +        len = sizeof(saddr_un);
>>> +        getsockname(s->listen_fd, (struct sockaddr *)&saddr_un, &len);
>>> +        snprintf(s->nc.info_str, sizeof(s->nc.info_str),
>>> +                 "connect from %s", saddr_un.sun_path);
>>> +        break;
>>> +    }
>>> +    default:
>>> +        g_assert_not_reached();
>>> +    }
>>>   }
>>>   
>>>   static int net_stream_server_init(NetClientState *peer,
>>> @@ -295,6 +313,42 @@ static int net_stream_server_init(NetClientState *peer,
>>>           }
>>>           break;
>>>       }
>>> +    case SOCKET_ADDRESS_TYPE_UNIX: {
>>> +        struct sockaddr_un saddr_un;
>>> +
>>> +        ret = unlink(addr->u.q_unix.path);
>>> +        if (ret < 0 && errno != ENOENT) {
>>> +            error_setg_errno(errp, errno, "failed to unlink socket %s",
>>> +                             addr->u.q_unix.path);
>>> +            return -1;
>>> +        }
>>> +
>>> +        saddr_un.sun_family = PF_UNIX;
>>> +        ret = snprintf(saddr_un.sun_path, sizeof(saddr_un.sun_path), "%s",
>>> +                       addr->u.q_unix.path);
>>> +        if (ret < 0 || ret >= sizeof(saddr_un.sun_path)) {
>>> +            error_setg(errp, "UNIX socket path '%s' is too long",
>>> +                       addr->u.q_unix.path);
>>> +            error_append_hint(errp, "Path must be less than %zu bytes\n",
>>> +                              sizeof(saddr_un.sun_path));
>> 
>> return here?

If you don't, and ...

>>> +
>>> +        fd = qemu_socket(PF_UNIX, SOCK_STREAM, 0);
>>> +        if (fd < 0) {

... we then get here, we trip error_setv()'s assertion "no error has
been set".

>>> +            error_setg_errno(errp, errno, "can't create stream socket");
>>> +            return -1;
>>> +        }
>>> +        qemu_socket_set_nonblock(fd);
>>> +
>>> +        ret = bind(fd, (struct sockaddr *)&saddr_un, sizeof(saddr_un));
>>> +        if (ret < 0) {
>>> +            error_setg_errno(errp, errno, "can't create socket with path: %s",
>>> +                             saddr_un.sun_path);
>>> +            closesocket(fd);
>>> +            return -1;
>>> +        }
>>> +        break;
>>> +    }
>>>       case SOCKET_ADDRESS_TYPE_FD:
>>>           fd = monitor_fd_param(monitor_cur(), addr->u.fd.str, errp);
>>>           if (fd == -1) {
>>> @@ -380,6 +434,48 @@ static int net_stream_client_init(NetClientState *peer,
>>>                                      ntohs(saddr_in.sin_port));
>>>           break;
>>>       }
>>> +    case SOCKET_ADDRESS_TYPE_UNIX: {
>>> +        struct sockaddr_un saddr_un;
>>> +
>>> +        saddr_un.sun_family = PF_UNIX;
>>> +        ret = snprintf(saddr_un.sun_path, sizeof(saddr_un.sun_path), "%s",
>>> +                       addr->u.q_unix.path);
>>> +        if (ret < 0 || ret >= sizeof(saddr_un.sun_path)) {
>>> +            error_setg(errp, "UNIX socket path '%s' is too long",
>>> +                       addr->u.q_unix.path);
>>> +            error_append_hint(errp, "Path must be less than %zu bytes\n",
>>> +                              sizeof(saddr_un.sun_path));
>> 
>> and here

Likewise.

[...]

[PATCH v5 08/12] net: dgram: make dgram_dst generic

[Laurent Vivier]

dgram_dst is a sockaddr_in structure. To be able to use it with
unix socket, use a pointer to a generic sockaddr structure.

Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
---
 net/dgram.c | 76 +++++++++++++++++++++++++++++++----------------------
 1 file changed, 45 insertions(+), 31 deletions(-)

diff --git a/net/dgram.c b/net/dgram.c
index dbe65102d174..dcc2205305c5 100644
--- a/net/dgram.c
+++ b/net/dgram.c
@@ -40,9 +40,8 @@ typedef struct NetDgramState {
     int listen_fd;
     int fd;
     SocketReadState rs;
-  /* contains inet host and port destination iff connectionless (SOCK_DGRAM) */
-    struct sockaddr_in dgram_dst;
-    IOHandler *send_fn;           /* differs between SOCK_STREAM/SOCK_DGRAM */
+    struct sockaddr *dgram_dst; /* contains destination iff connectionless */
+    IOHandler *send_fn;
     bool read_poll;               /* waiting to receive data? */
     bool write_poll;              /* waiting to transmit data? */
 } NetDgramState;
@@ -86,10 +85,9 @@ static ssize_t net_dgram_receive_dgram(NetClientState *nc,
     ssize_t ret;
 
     do {
-        if (s->dgram_dst.sin_family != AF_UNIX) {
-            ret = sendto(s->fd, buf, size, 0,
-                         (struct sockaddr *)&s->dgram_dst,
-                         sizeof(s->dgram_dst));
+        if (s->dgram_dst) {
+            ret = sendto(s->fd, buf, size, 0, s->dgram_dst,
+                         sizeof(struct sockaddr_in));
         } else {
             ret = send(s->fd, buf, size, 0);
         }
@@ -290,6 +288,8 @@ static void net_dgram_cleanup(NetClientState *nc)
         closesocket(s->listen_fd);
         s->listen_fd = -1;
     }
+    g_free(s->dgram_dst);
+    s->dgram_dst = NULL;
 }
 
 static NetClientInfo net_dgram_socket_info = {
@@ -306,7 +306,7 @@ static NetDgramState *net_dgram_fd_init_dgram(NetClientState *peer,
                                               SocketAddress *mcast,
                                               Error **errp)
 {
-    struct sockaddr_in saddr;
+    struct sockaddr_in *saddr = NULL;
     int newfd;
     NetClientState *nc;
     NetDgramState *s;
@@ -328,24 +328,25 @@ static NetDgramState *net_dgram_fd_init_dgram(NetClientState *peer,
      */
 
     if (is_fd && mcast != NULL) {
-            if (convert_host_port(&saddr, mcast->u.inet.host,
-                                  mcast->u.inet.port, errp) < 0) {
+            saddr = g_new(struct sockaddr_in, 1);
+
+            if (convert_host_port(saddr, mcast->u.inet.host, mcast->u.inet.port,
+                                  errp) < 0) {
                 goto err;
             }
             /* must be bound */
-            if (saddr.sin_addr.s_addr == 0) {
+            if (saddr->sin_addr.s_addr == 0) {
                 error_setg(errp, "can't setup multicast destination address");
                 goto err;
             }
             /* clone dgram socket */
-            newfd = net_dgram_mcast_create(&saddr, NULL, errp);
+            newfd = net_dgram_mcast_create(saddr, NULL, errp);
             if (newfd < 0) {
                 goto err;
             }
             /* clone newfd to fd, close newfd */
             dup2(newfd, fd);
             close(newfd);
-
     }
 
     nc = qemu_new_net_client(&net_dgram_socket_info, peer, model, name);
@@ -359,16 +360,13 @@ static NetDgramState *net_dgram_fd_init_dgram(NetClientState *peer,
     net_dgram_read_poll(s, true);
 
     /* mcast: save bound address as dst */
-    if (is_fd && mcast != NULL) {
-        s->dgram_dst = saddr;
+    if (saddr) {
+        g_assert(s->dgram_dst == NULL);
+        s->dgram_dst = (struct sockaddr *)saddr;
         snprintf(nc->info_str, sizeof(nc->info_str),
                  "fd=%d (cloned mcast=%s:%d)",
-                 fd, inet_ntoa(saddr.sin_addr), ntohs(saddr.sin_port));
+                 fd, inet_ntoa(saddr->sin_addr), ntohs(saddr->sin_port));
     } else {
-        if (sa_type == SOCKET_ADDRESS_TYPE_UNIX) {
-            s->dgram_dst.sin_family = AF_UNIX;
-        }
-
         snprintf(nc->info_str, sizeof(nc->info_str), "fd=%d %s", fd,
                  SocketAddressType_str(sa_type));
     }
@@ -376,6 +374,7 @@ static NetDgramState *net_dgram_fd_init_dgram(NetClientState *peer,
     return s;
 
 err:
+    g_free(saddr);
     closesocket(fd);
     return NULL;
 }
@@ -421,21 +420,24 @@ static int net_dgram_mcast_init(NetClientState *peer,
 {
     NetDgramState *s;
     int fd, ret;
-    struct sockaddr_in saddr;
+    struct sockaddr_in *saddr;
 
     if (remote->type != SOCKET_ADDRESS_TYPE_INET) {
         error_setg(errp, "multicast only support inet type");
         return -1;
     }
 
-    if (convert_host_port(&saddr, remote->u.inet.host, remote->u.inet.port,
+    saddr = g_new(struct sockaddr_in, 1);
+    if (convert_host_port(saddr, remote->u.inet.host, remote->u.inet.port,
                           errp) < 0) {
+        g_free(saddr);
         return -1;
     }
 
     if (!local) {
-        fd = net_dgram_mcast_create(&saddr, NULL, errp);
+        fd = net_dgram_mcast_create(saddr, NULL, errp);
         if (fd < 0) {
+            g_free(saddr);
             return -1;
         }
     } else {
@@ -444,13 +446,15 @@ static int net_dgram_mcast_init(NetClientState *peer,
             struct in_addr localaddr;
 
             if (inet_aton(local->u.inet.host, &localaddr) == 0) {
+                g_free(saddr);
                 error_setg(errp, "localaddr '%s' is not a valid IPv4 address",
                            local->u.inet.host);
                 return -1;
             }
 
-            fd = net_dgram_mcast_create(&saddr, &localaddr, errp);
+            fd = net_dgram_mcast_create(saddr, &localaddr, errp);
             if (fd < 0) {
+                g_free(saddr);
                 return -1;
             }
             break;
@@ -458,16 +462,19 @@ static int net_dgram_mcast_init(NetClientState *peer,
         case SOCKET_ADDRESS_TYPE_FD:
             fd = monitor_fd_param(monitor_cur(), local->u.fd.str, errp);
             if (fd == -1) {
+                g_free(saddr);
                 return -1;
             }
             ret = qemu_socket_try_set_nonblock(fd);
             if (ret < 0) {
+                g_free(saddr);
                 error_setg_errno(errp, -ret, "%s: Can't use file descriptor %d",
                                  name, fd);
                 return -1;
             }
             break;
         default:
+            g_free(saddr);
             error_setg(errp, "only support inet or fd type for local");
             return -1;
         }
@@ -477,13 +484,16 @@ static int net_dgram_mcast_init(NetClientState *peer,
                                  local->type == SOCKET_ADDRESS_TYPE_FD,
                                  remote, errp);
     if (!s) {
+        g_free(saddr);
         return -1;
     }
 
-    s->dgram_dst = saddr;
+    g_assert(s->dgram_dst == NULL);
+    s->dgram_dst = (struct sockaddr *)saddr;
 
     snprintf(s->nc.info_str, sizeof(s->nc.info_str), "mcast=%s:%d",
-             inet_ntoa(saddr.sin_addr), ntohs(saddr.sin_port));
+             inet_ntoa(saddr->sin_addr), ntohs(saddr->sin_port));
+
     return 0;
 
 }
@@ -497,8 +507,8 @@ static int net_dgram_udp_init(NetClientState *peer,
 {
     NetDgramState *s;
     int fd, ret;
-    struct sockaddr_in raddr_in;
     gchar *info_str;
+    struct sockaddr *dgram_dst;
 
     if (remote) {
         if (local->type == SOCKET_ADDRESS_TYPE_FD) {
@@ -518,7 +528,7 @@ static int net_dgram_udp_init(NetClientState *peer,
 
     switch (local->type) {
     case SOCKET_ADDRESS_TYPE_INET: {
-        struct sockaddr_in laddr_in;
+        struct sockaddr_in laddr_in, raddr_in;
 
         if (convert_host_port(&laddr_in, local->u.inet.host, local->u.inet.port,
                               errp) < 0) {
@@ -552,9 +562,12 @@ static int net_dgram_udp_init(NetClientState *peer,
         }
         qemu_socket_set_nonblock(fd);
 
+        dgram_dst = g_malloc(sizeof(raddr_in));
+        memcpy(dgram_dst, &raddr_in, sizeof(raddr_in));
+
         info_str = g_strdup_printf("udp=%s:%d/%s:%d",
-                 inet_ntoa(laddr_in.sin_addr), ntohs(laddr_in.sin_port),
-                 inet_ntoa(raddr_in.sin_addr), ntohs(raddr_in.sin_port));
+                        inet_ntoa(laddr_in.sin_addr), ntohs(laddr_in.sin_port),
+                        inet_ntoa(raddr_in.sin_addr), ntohs(raddr_in.sin_port));
 
         break;
     }
@@ -581,7 +594,8 @@ static int net_dgram_udp_init(NetClientState *peer,
     }
 
     if (remote) {
-        s->dgram_dst = raddr_in;
+        g_assert(s->dgram_dst == NULL);
+        s->dgram_dst = dgram_dst;
 
         pstrcpy(s->nc.info_str, sizeof(s->nc.info_str), info_str);
         g_free(info_str);
-- 
2.36.1

[PATCH v5 09/12] net: dgram: move mcast specific code from net_socket_fd_init_dgram()

[Laurent Vivier]

It is less complex to manage special cases directly in
net_dgram_mcast_init() and net_dgram_udp_init().

Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
---
 net/dgram.c | 143 +++++++++++++++++++++++++++-------------------------
 1 file changed, 73 insertions(+), 70 deletions(-)

diff --git a/net/dgram.c b/net/dgram.c
index dcc2205305c5..16e2d909755c 100644
--- a/net/dgram.c
+++ b/net/dgram.c
@@ -302,52 +302,11 @@ static NetClientInfo net_dgram_socket_info = {
 static NetDgramState *net_dgram_fd_init_dgram(NetClientState *peer,
                                               const char *model,
                                               const char *name,
-                                              int fd, int is_fd,
-                                              SocketAddress *mcast,
+                                              int fd,
                                               Error **errp)
 {
-    struct sockaddr_in *saddr = NULL;
-    int newfd;
     NetClientState *nc;
     NetDgramState *s;
-    SocketAddress *sa;
-    SocketAddressType sa_type;
-
-    sa = socket_local_address(fd, errp);
-    if (!sa) {
-        return NULL;
-    }
-    sa_type = sa->type;
-    qapi_free_SocketAddress(sa);
-
-    /*
-     * fd passed: multicast: "learn" dgram_dst address from bound address and
-     * save it. Because this may be "shared" socket from a "master" process,
-     * datagrams would be recv() by ONLY ONE process: we must "clone" this
-     * dgram socket --jjo
-     */
-
-    if (is_fd && mcast != NULL) {
-            saddr = g_new(struct sockaddr_in, 1);
-
-            if (convert_host_port(saddr, mcast->u.inet.host, mcast->u.inet.port,
-                                  errp) < 0) {
-                goto err;
-            }
-            /* must be bound */
-            if (saddr->sin_addr.s_addr == 0) {
-                error_setg(errp, "can't setup multicast destination address");
-                goto err;
-            }
-            /* clone dgram socket */
-            newfd = net_dgram_mcast_create(saddr, NULL, errp);
-            if (newfd < 0) {
-                goto err;
-            }
-            /* clone newfd to fd, close newfd */
-            dup2(newfd, fd);
-            close(newfd);
-    }
 
     nc = qemu_new_net_client(&net_dgram_socket_info, peer, model, name);
 
@@ -359,24 +318,7 @@ static NetDgramState *net_dgram_fd_init_dgram(NetClientState *peer,
     net_socket_rs_init(&s->rs, net_dgram_rs_finalize, false);
     net_dgram_read_poll(s, true);
 
-    /* mcast: save bound address as dst */
-    if (saddr) {
-        g_assert(s->dgram_dst == NULL);
-        s->dgram_dst = (struct sockaddr *)saddr;
-        snprintf(nc->info_str, sizeof(nc->info_str),
-                 "fd=%d (cloned mcast=%s:%d)",
-                 fd, inet_ntoa(saddr->sin_addr), ntohs(saddr->sin_port));
-    } else {
-        snprintf(nc->info_str, sizeof(nc->info_str), "fd=%d %s", fd,
-                 SocketAddressType_str(sa_type));
-    }
-
     return s;
-
-err:
-    g_free(saddr);
-    closesocket(fd);
-    return NULL;
 }
 
 static void net_dgram_connect(void *opaque)
@@ -421,6 +363,7 @@ static int net_dgram_mcast_init(NetClientState *peer,
     NetDgramState *s;
     int fd, ret;
     struct sockaddr_in *saddr;
+    gchar *info_str;
 
     if (remote->type != SOCKET_ADDRESS_TYPE_INET) {
         error_setg(errp, "multicast only support inet type");
@@ -440,6 +383,9 @@ static int net_dgram_mcast_init(NetClientState *peer,
             g_free(saddr);
             return -1;
         }
+        info_str = g_strdup_printf("mcast=%s:%d",
+                                   inet_ntoa(saddr->sin_addr),
+                                   ntohs(saddr->sin_port));
     } else {
         switch (local->type) {
         case SOCKET_ADDRESS_TYPE_INET: {
@@ -457,9 +403,14 @@ static int net_dgram_mcast_init(NetClientState *peer,
                 g_free(saddr);
                 return -1;
             }
+            info_str = g_strdup_printf("mcast=%s:%d",
+                                       inet_ntoa(saddr->sin_addr),
+                                       ntohs(saddr->sin_port));
             break;
         }
-        case SOCKET_ADDRESS_TYPE_FD:
+        case SOCKET_ADDRESS_TYPE_FD: {
+            int newfd;
+
             fd = monitor_fd_param(monitor_cur(), local->u.fd.str, errp);
             if (fd == -1) {
                 g_free(saddr);
@@ -472,7 +423,46 @@ static int net_dgram_mcast_init(NetClientState *peer,
                                  name, fd);
                 return -1;
             }
+
+            /*
+             * fd passed: multicast: "learn" dgram_dst address from bound
+             * address and save it. Because this may be "shared" socket from a
+             * "master" process, datagrams would be recv() by ONLY ONE process:
+             * we must "clone" this dgram socket --jjo
+             */
+
+            saddr = g_new(struct sockaddr_in, 1);
+
+            if (convert_host_port(saddr, local->u.inet.host, local->u.inet.port,
+                                  errp) < 0) {
+                g_free(saddr);
+                closesocket(fd);
+                return -1;
+            }
+
+            /* must be bound */
+            if (saddr->sin_addr.s_addr == 0) {
+                error_setg(errp, "can't setup multicast destination address");
+                g_free(saddr);
+                closesocket(fd);
+                return -1;
+            }
+            /* clone dgram socket */
+            newfd = net_dgram_mcast_create(saddr, NULL, errp);
+            if (newfd < 0) {
+                g_free(saddr);
+                closesocket(fd);
+                return -1;
+            }
+            /* clone newfd to fd, close newfd */
+            dup2(newfd, fd);
+            close(newfd);
+
+            info_str = g_strdup_printf("fd=%d (cloned mcast=%s:%d)",
+                                       fd, inet_ntoa(saddr->sin_addr),
+                                       ntohs(saddr->sin_port));
             break;
+        }
         default:
             g_free(saddr);
             error_setg(errp, "only support inet or fd type for local");
@@ -480,9 +470,7 @@ static int net_dgram_mcast_init(NetClientState *peer,
         }
     }
 
-    s = net_dgram_fd_init_dgram(peer, model, name, fd,
-                                 local->type == SOCKET_ADDRESS_TYPE_FD,
-                                 remote, errp);
+    s = net_dgram_fd_init_dgram(peer, model, name, fd, errp);
     if (!s) {
         g_free(saddr);
         return -1;
@@ -491,8 +479,8 @@ static int net_dgram_mcast_init(NetClientState *peer,
     g_assert(s->dgram_dst == NULL);
     s->dgram_dst = (struct sockaddr *)saddr;
 
-    snprintf(s->nc.info_str, sizeof(s->nc.info_str), "mcast=%s:%d",
-             inet_ntoa(saddr->sin_addr), ntohs(saddr->sin_port));
+    pstrcpy(s->nc.info_str, sizeof(s->nc.info_str), info_str);
+    g_free(info_str);
 
     return 0;
 
@@ -571,7 +559,10 @@ static int net_dgram_udp_init(NetClientState *peer,
 
         break;
     }
-    case SOCKET_ADDRESS_TYPE_FD:
+    case SOCKET_ADDRESS_TYPE_FD: {
+        SocketAddress *sa;
+        SocketAddressType sa_type;
+
         fd = monitor_fd_param(monitor_cur(), local->u.fd.str, errp);
         if (fd == -1) {
             return -1;
@@ -582,23 +573,35 @@ static int net_dgram_udp_init(NetClientState *peer,
                              name, fd);
             return -1;
         }
+
+        sa = socket_local_address(fd, errp);
+        if (sa) {
+            sa_type = sa->type;
+            qapi_free_SocketAddress(sa);
+
+            info_str = g_strdup_printf("fd=%d %s", fd,
+                                       SocketAddressType_str(sa_type));
+        } else {
+            info_str = g_strdup_printf("fd=%d", fd);
+        }
         break;
+    }
     default:
         error_setg(errp, "only support inet or fd type for local");
         return -1;
     }
 
-    s = net_dgram_fd_init_dgram(peer, model, name, fd, 0, NULL, errp);
+    s = net_dgram_fd_init_dgram(peer, model, name, fd, errp);
     if (!s) {
         return -1;
     }
 
+    pstrcpy(s->nc.info_str, sizeof(s->nc.info_str), info_str);
+    g_free(info_str);
+
     if (remote) {
         g_assert(s->dgram_dst == NULL);
         s->dgram_dst = dgram_dst;
-
-        pstrcpy(s->nc.info_str, sizeof(s->nc.info_str), info_str);
-        g_free(info_str);
     }
     return 0;
 }
-- 
2.36.1

[PATCH v5 10/12] net: dgram: add unix socket

[Laurent Vivier]

Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
---
 net/dgram.c     | 65 ++++++++++++++++++++++++++++++++++++++++++++++---
 qemu-options.hx |  1 +
 2 files changed, 63 insertions(+), 3 deletions(-)

diff --git a/net/dgram.c b/net/dgram.c
index 16e2d909755c..9f3eafee3b67 100644
--- a/net/dgram.c
+++ b/net/dgram.c
@@ -86,8 +86,15 @@ static ssize_t net_dgram_receive_dgram(NetClientState *nc,
 
     do {
         if (s->dgram_dst) {
-            ret = sendto(s->fd, buf, size, 0, s->dgram_dst,
-                         sizeof(struct sockaddr_in));
+            socklen_t len;
+
+            if (s->dgram_dst->sa_family == AF_INET) {
+                len = sizeof(struct sockaddr_in);
+            } else {
+                len = sizeof(struct sockaddr_un);
+            }
+
+            ret = sendto(s->fd, buf, size, 0, s->dgram_dst, len);
         } else {
             ret = send(s->fd, buf, size, 0);
         }
@@ -509,7 +516,7 @@ static int net_dgram_udp_init(NetClientState *peer,
         }
     } else {
         if (local->type != SOCKET_ADDRESS_TYPE_FD) {
-            error_setg(errp, "type=inet requires remote parameter");
+            error_setg(errp, "type=inet or unix require remote parameter");
             return -1;
         }
     }
@@ -559,6 +566,58 @@ static int net_dgram_udp_init(NetClientState *peer,
 
         break;
     }
+    case SOCKET_ADDRESS_TYPE_UNIX: {
+        struct sockaddr_un laddr_un, raddr_un;
+
+        ret = unlink(local->u.q_unix.path);
+        if (ret < 0 && errno != ENOENT) {
+            error_setg_errno(errp, errno, "failed to unlink socket %s",
+                             local->u.q_unix.path);
+            return -1;
+        }
+
+        laddr_un.sun_family = PF_UNIX;
+        ret = snprintf(laddr_un.sun_path, sizeof(laddr_un.sun_path), "%s",
+                       local->u.q_unix.path);
+        if (ret < 0 || ret >= sizeof(laddr_un.sun_path)) {
+            error_setg(errp, "UNIX socket path '%s' is too long",
+                       local->u.q_unix.path);
+            error_append_hint(errp, "Path must be less than %zu bytes\n",
+                              sizeof(laddr_un.sun_path));
+        }
+
+        raddr_un.sun_family = PF_UNIX;
+        ret = snprintf(raddr_un.sun_path, sizeof(raddr_un.sun_path), "%s",
+                       remote->u.q_unix.path);
+        if (ret < 0 || ret >= sizeof(raddr_un.sun_path)) {
+            error_setg(errp, "UNIX socket path '%s' is too long",
+                       remote->u.q_unix.path);
+            error_append_hint(errp, "Path must be less than %zu bytes\n",
+                              sizeof(raddr_un.sun_path));
+        }
+
+        fd = qemu_socket(PF_UNIX, SOCK_DGRAM, 0);
+        if (fd < 0) {
+            error_setg_errno(errp, errno, "can't create datagram socket");
+            return -1;
+        }
+
+        ret = bind(fd, (struct sockaddr *)&laddr_un, sizeof(laddr_un));
+        if (ret < 0) {
+            error_setg_errno(errp, errno, "can't bind unix=%s to socket",
+                             laddr_un.sun_path);
+            closesocket(fd);
+            return -1;
+        }
+        qemu_socket_set_nonblock(fd);
+
+        dgram_dst = g_malloc(sizeof(raddr_un));
+        memcpy(dgram_dst, &raddr_un, sizeof(raddr_un));
+
+        info_str = g_strdup_printf("udp=%s:%s",
+                                   laddr_un.sun_path, raddr_un.sun_path);
+        break;
+    }
     case SOCKET_ADDRESS_TYPE_FD: {
         SocketAddress *sa;
         SocketAddressType sa_type;
diff --git a/qemu-options.hx b/qemu-options.hx
index 23ff12c1b183..950e932f528f 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -2732,6 +2732,7 @@ DEF("netdev", HAS_ARG, QEMU_OPTION_netdev,
     "                configure a network backend to connect to a multicast maddr and port\n"
     "                use 'local.host=addr' to specify the host address to send packets from\n"
     "-netdev dgram,id=str,local.type=inet,local.host=addr,local.port=port[,remote.type=inet,remote.host=addr,remote.port=port]\n"
+    "-netdev dgram,id=str,local.type=unix,local.path=path[,remote.type=unix,remote.path=path]\n"
     "-netdev dgram,id=str,local.type=fd,local.str=h\n"
     "                configure a network backend to connect to another network\n"
     "                using an UDP tunnel\n"
-- 
2.36.1

[PATCH v5 11/12] qemu-sockets: introduce socket_uri()

[Laurent Vivier]

Format a string URI from a SocketAddress.

Original code from hmp-cmds.c:SocketAddress_to_str()

Replace 'tcp:' by 'inet:' (because 'inet' can be also 'udp').
Replace 'tcp:' by 'vsock:' with vsock socket type.

Signed-off-by: Laurent Vivier <lvivier@redhat.com>
---
 include/qemu/sockets.h |  2 +-
 monitor/hmp-cmds.c     | 23 +----------------------
 util/qemu-sockets.c    | 20 ++++++++++++++++++++
 3 files changed, 22 insertions(+), 23 deletions(-)

diff --git a/include/qemu/sockets.h b/include/qemu/sockets.h
index 47194b9732f8..3e2ae7e21705 100644
--- a/include/qemu/sockets.h
+++ b/include/qemu/sockets.h
@@ -41,6 +41,7 @@ int unix_listen(const char *path, Error **errp);
 int unix_connect(const char *path, Error **errp);
 
 SocketAddress *socket_parse(const char *str, Error **errp);
+char *socket_uri(SocketAddress *addr);
 int socket_connect(SocketAddress *addr, Error **errp);
 int socket_listen(SocketAddress *addr, int num, Error **errp);
 void socket_listen_cleanup(int fd, Error **errp);
@@ -123,5 +124,4 @@ SocketAddress *socket_address_flatten(SocketAddressLegacy *addr);
  * Return 0 on success.
  */
 int socket_address_parse_named_fd(SocketAddress *addr, Error **errp);
-
 #endif /* QEMU_SOCKETS_H */
diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
index ca98df04952b..8ebb437d1b6a 100644
--- a/monitor/hmp-cmds.c
+++ b/monitor/hmp-cmds.c
@@ -197,27 +197,6 @@ void hmp_info_mice(Monitor *mon, const QDict *qdict)
     qapi_free_MouseInfoList(mice_list);
 }
 
-static char *SocketAddress_to_str(SocketAddress *addr)
-{
-    switch (addr->type) {
-    case SOCKET_ADDRESS_TYPE_INET:
-        return g_strdup_printf("tcp:%s:%s",
-                               addr->u.inet.host,
-                               addr->u.inet.port);
-    case SOCKET_ADDRESS_TYPE_UNIX:
-        return g_strdup_printf("unix:%s",
-                               addr->u.q_unix.path);
-    case SOCKET_ADDRESS_TYPE_FD:
-        return g_strdup_printf("fd:%s", addr->u.fd.str);
-    case SOCKET_ADDRESS_TYPE_VSOCK:
-        return g_strdup_printf("tcp:%s:%s",
-                               addr->u.vsock.cid,
-                               addr->u.vsock.port);
-    default:
-        return g_strdup("unknown address type");
-    }
-}
-
 void hmp_info_migrate(Monitor *mon, const QDict *qdict)
 {
     MigrationInfo *info;
@@ -375,7 +354,7 @@ void hmp_info_migrate(Monitor *mon, const QDict *qdict)
         monitor_printf(mon, "socket address: [\n");
 
         for (addr = info->socket_address; addr; addr = addr->next) {
-            char *s = SocketAddress_to_str(addr->value);
+            char *s = socket_uri(addr->value);
             monitor_printf(mon, "\t%s\n", s);
             g_free(s);
         }
diff --git a/util/qemu-sockets.c b/util/qemu-sockets.c
index 13b5b197f9ea..4efc2ce8b074 100644
--- a/util/qemu-sockets.c
+++ b/util/qemu-sockets.c
@@ -1098,6 +1098,26 @@ int unix_connect(const char *path, Error **errp)
     return sock;
 }
 
+char *socket_uri(SocketAddress *addr)
+{
+    switch (addr->type) {
+    case SOCKET_ADDRESS_TYPE_INET:
+        return g_strdup_printf("inet:%s:%s",
+                               addr->u.inet.host,
+                               addr->u.inet.port);
+    case SOCKET_ADDRESS_TYPE_UNIX:
+        return g_strdup_printf("unix:%s",
+                               addr->u.q_unix.path);
+    case SOCKET_ADDRESS_TYPE_FD:
+        return g_strdup_printf("fd:%s", addr->u.fd.str);
+    case SOCKET_ADDRESS_TYPE_VSOCK:
+        return g_strdup_printf("vsock:%s:%s",
+                               addr->u.vsock.cid,
+                               addr->u.vsock.port);
+    default:
+        return g_strdup("unknown address type");
+    }
+}
 
 SocketAddress *socket_parse(const char *str, Error **errp)
 {
-- 
2.36.1

Re: [PATCH v5 11/12] qemu-sockets: introduce socket_uri()

[Dr. David Alan Gilbert]

* Laurent Vivier (lvivier@redhat.com) wrote:
> Format a string URI from a SocketAddress.
> 
> Original code from hmp-cmds.c:SocketAddress_to_str()
> 
> Replace 'tcp:' by 'inet:' (because 'inet' can be also 'udp').

I think that's OK, it'll look a little odd in migration where
the syntax for a migration URI is tcp: (and is a tcp stream)
and then when you 'info migrate', which is where this is origianlly
used, it turns into inet.

> Replace 'tcp:' by 'vsock:' with vsock socket type.

> Signed-off-by: Laurent Vivier <lvivier@redhat.com>

Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

> ---
>  include/qemu/sockets.h |  2 +-
>  monitor/hmp-cmds.c     | 23 +----------------------
>  util/qemu-sockets.c    | 20 ++++++++++++++++++++
>  3 files changed, 22 insertions(+), 23 deletions(-)
> 
> diff --git a/include/qemu/sockets.h b/include/qemu/sockets.h
> index 47194b9732f8..3e2ae7e21705 100644
> --- a/include/qemu/sockets.h
> +++ b/include/qemu/sockets.h
> @@ -41,6 +41,7 @@ int unix_listen(const char *path, Error **errp);
>  int unix_connect(const char *path, Error **errp);
>  
>  SocketAddress *socket_parse(const char *str, Error **errp);
> +char *socket_uri(SocketAddress *addr);
>  int socket_connect(SocketAddress *addr, Error **errp);
>  int socket_listen(SocketAddress *addr, int num, Error **errp);
>  void socket_listen_cleanup(int fd, Error **errp);
> @@ -123,5 +124,4 @@ SocketAddress *socket_address_flatten(SocketAddressLegacy *addr);
>   * Return 0 on success.
>   */
>  int socket_address_parse_named_fd(SocketAddress *addr, Error **errp);
> -
>  #endif /* QEMU_SOCKETS_H */
> diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
> index ca98df04952b..8ebb437d1b6a 100644
> --- a/monitor/hmp-cmds.c
> +++ b/monitor/hmp-cmds.c
> @@ -197,27 +197,6 @@ void hmp_info_mice(Monitor *mon, const QDict *qdict)
>      qapi_free_MouseInfoList(mice_list);
>  }
>  
> -static char *SocketAddress_to_str(SocketAddress *addr)
> -{
> -    switch (addr->type) {
> -    case SOCKET_ADDRESS_TYPE_INET:
> -        return g_strdup_printf("tcp:%s:%s",
> -                               addr->u.inet.host,
> -                               addr->u.inet.port);
> -    case SOCKET_ADDRESS_TYPE_UNIX:
> -        return g_strdup_printf("unix:%s",
> -                               addr->u.q_unix.path);
> -    case SOCKET_ADDRESS_TYPE_FD:
> -        return g_strdup_printf("fd:%s", addr->u.fd.str);
> -    case SOCKET_ADDRESS_TYPE_VSOCK:
> -        return g_strdup_printf("tcp:%s:%s",
> -                               addr->u.vsock.cid,
> -                               addr->u.vsock.port);
> -    default:
> -        return g_strdup("unknown address type");
> -    }
> -}
> -
>  void hmp_info_migrate(Monitor *mon, const QDict *qdict)
>  {
>      MigrationInfo *info;
> @@ -375,7 +354,7 @@ void hmp_info_migrate(Monitor *mon, const QDict *qdict)
>          monitor_printf(mon, "socket address: [\n");
>  
>          for (addr = info->socket_address; addr; addr = addr->next) {
> -            char *s = SocketAddress_to_str(addr->value);
> +            char *s = socket_uri(addr->value);
>              monitor_printf(mon, "\t%s\n", s);
>              g_free(s);
>          }
> diff --git a/util/qemu-sockets.c b/util/qemu-sockets.c
> index 13b5b197f9ea..4efc2ce8b074 100644
> --- a/util/qemu-sockets.c
> +++ b/util/qemu-sockets.c
> @@ -1098,6 +1098,26 @@ int unix_connect(const char *path, Error **errp)
>      return sock;
>  }
>  
> +char *socket_uri(SocketAddress *addr)
> +{
> +    switch (addr->type) {
> +    case SOCKET_ADDRESS_TYPE_INET:
> +        return g_strdup_printf("inet:%s:%s",
> +                               addr->u.inet.host,
> +                               addr->u.inet.port);
> +    case SOCKET_ADDRESS_TYPE_UNIX:
> +        return g_strdup_printf("unix:%s",
> +                               addr->u.q_unix.path);
> +    case SOCKET_ADDRESS_TYPE_FD:
> +        return g_strdup_printf("fd:%s", addr->u.fd.str);
> +    case SOCKET_ADDRESS_TYPE_VSOCK:
> +        return g_strdup_printf("vsock:%s:%s",
> +                               addr->u.vsock.cid,
> +                               addr->u.vsock.port);
> +    default:
> +        return g_strdup("unknown address type");
> +    }
> +}
>  
>  SocketAddress *socket_parse(const char *str, Error **errp)
>  {
> -- 
> 2.36.1
> 
-- 
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK

[PATCH v5 12/12] net: stream: move to QIO

[Laurent Vivier]

Use QIOChannel, QIOChannelSocket and QIONetListener.

Signed-off-by: Laurent Vivier <lvivier@redhat.com>
---
 net/stream.c | 476 ++++++++++++++++++---------------------------------
 1 file changed, 165 insertions(+), 311 deletions(-)

diff --git a/net/stream.c b/net/stream.c
index 1a2bc237707e..6e9952c7c3ef 100644
--- a/net/stream.c
+++ b/net/stream.c
@@ -34,48 +34,36 @@
 #include "qemu/iov.h"
 #include "qemu/main-loop.h"
 #include "qemu/cutils.h"
+#include "io/channel.h"
+#include "io/channel-socket.h"
+#include "io/net-listener.h"
 
 typedef struct NetStreamState {
     NetClientState nc;
-    int listen_fd;
-    int fd;
+    QIOChannel *listen_ioc;
+    QIONetListener *listener;
+    QIOChannel *ioc;
+    guint ioc_read_tag;
+    guint ioc_write_tag;
     SocketReadState rs;
     unsigned int send_index;      /* number of bytes sent*/
-    IOHandler *send_fn;
-    bool read_poll;               /* waiting to receive data? */
-    bool write_poll;              /* waiting to transmit data? */
 } NetStreamState;
 
-static void net_stream_accept(void *opaque);
-static void net_stream_writable(void *opaque);
+static void net_stream_listen(QIONetListener *listener,
+                              QIOChannelSocket *cioc,
+                              void *opaque);
 
-static void net_stream_update_fd_handler(NetStreamState *s)
+static gboolean net_stream_writable(QIOChannel *ioc,
+                                    GIOCondition condition,
+                                    gpointer data)
 {
-    qemu_set_fd_handler(s->fd,
-                        s->read_poll ? s->send_fn : NULL,
-                        s->write_poll ? net_stream_writable : NULL,
-                        s);
-}
-
-static void net_stream_read_poll(NetStreamState *s, bool enable)
-{
-    s->read_poll = enable;
-    net_stream_update_fd_handler(s);
-}
-
-static void net_stream_write_poll(NetStreamState *s, bool enable)
-{
-    s->write_poll = enable;
-    net_stream_update_fd_handler(s);
-}
-
-static void net_stream_writable(void *opaque)
-{
-    NetStreamState *s = opaque;
+    NetStreamState *s = data;
 
-    net_stream_write_poll(s, false);
+    s->ioc_write_tag = 0;
 
     qemu_flush_queued_packets(&s->nc);
+
+    return G_SOURCE_REMOVE;
 }
 
 static ssize_t net_stream_receive(NetClientState *nc, const uint8_t *buf,
@@ -92,12 +80,15 @@ static ssize_t net_stream_receive(NetClientState *nc, const uint8_t *buf,
             .iov_len  = size,
         },
     };
+    struct iovec local_iov[2];
+    unsigned int nlocal_iov;
     size_t remaining;
     ssize_t ret;
 
-    remaining = iov_size(iov, 2) - s->send_index;
-    ret = iov_send(s->fd, iov, 2, s->send_index, remaining);
 
+    remaining = iov_size(iov, 2) - s->send_index;
+    nlocal_iov = iov_copy(local_iov, 2, iov, 2, s->send_index, remaining);
+    ret = qio_channel_writev(s->ioc, local_iov, nlocal_iov, NULL);
     if (ret == -1 && errno == EAGAIN) {
         ret = 0; /* handled further down */
     }
@@ -107,19 +98,25 @@ static ssize_t net_stream_receive(NetClientState *nc, const uint8_t *buf,
     }
     if (ret < (ssize_t)remaining) {
         s->send_index += ret;
-        net_stream_write_poll(s, true);
+        s->ioc_write_tag = qio_channel_add_watch(s->ioc, G_IO_OUT,
+                                                 net_stream_writable, s, NULL);
         return 0;
     }
     s->send_index = 0;
     return size;
 }
 
+static gboolean net_stream_send(QIOChannel *ioc,
+                                GIOCondition condition,
+                                gpointer data);
+
 static void net_stream_send_completed(NetClientState *nc, ssize_t len)
 {
     NetStreamState *s = DO_UPCAST(NetStreamState, nc, nc);
 
-    if (!s->read_poll) {
-        net_stream_read_poll(s, true);
+    if (!s->ioc_read_tag) {
+        s->ioc_read_tag = qio_channel_add_watch(s->ioc, G_IO_IN,
+                                                net_stream_send, s, NULL);
     }
 }
 
@@ -130,19 +127,24 @@ static void net_stream_rs_finalize(SocketReadState *rs)
     if (qemu_send_packet_async(&s->nc, rs->buf,
                                rs->packet_len,
                                net_stream_send_completed) == 0) {
-        net_stream_read_poll(s, false);
+        if (s->ioc_read_tag) {
+            g_source_remove(s->ioc_read_tag);
+            s->ioc_read_tag = 0;
+        }
     }
 }
 
-static void net_stream_send(void *opaque)
+static gboolean net_stream_send(QIOChannel *ioc,
+                                GIOCondition condition,
+                                gpointer data)
 {
-    NetStreamState *s = opaque;
+    NetStreamState *s = data;
     int size;
     int ret;
-    uint8_t buf1[NET_BUFSIZE];
-    const uint8_t *buf;
+    char buf1[NET_BUFSIZE];
+    const char *buf;
 
-    size = recv(s->fd, buf1, sizeof(buf1), 0);
+    size = qio_channel_read(s->ioc, buf1, sizeof(buf1), NULL);
     if (size < 0) {
         if (errno != EWOULDBLOCK) {
             goto eoc;
@@ -150,52 +152,63 @@ static void net_stream_send(void *opaque)
     } else if (size == 0) {
         /* end of connection */
     eoc:
-        net_stream_read_poll(s, false);
-        net_stream_write_poll(s, false);
-        if (s->listen_fd != -1) {
-            qemu_set_fd_handler(s->listen_fd, net_stream_accept, NULL, s);
+        s->ioc_read_tag = 0;
+        if (s->ioc_write_tag) {
+            g_source_remove(s->ioc_write_tag);
+            s->ioc_write_tag = 0;
+        }
+        if (s->listener) {
+            qio_net_listener_set_client_func(s->listener, net_stream_listen,
+                                             s, NULL);
         }
-        closesocket(s->fd);
+        object_unref(OBJECT(s->ioc));
+        s->ioc = NULL;
 
-        s->fd = -1;
         net_socket_rs_init(&s->rs, net_stream_rs_finalize, false);
         s->nc.link_down = true;
         memset(s->nc.info_str, 0, sizeof(s->nc.info_str));
 
-        return;
+        return G_SOURCE_REMOVE;
     }
     buf = buf1;
 
-    ret = net_fill_rstate(&s->rs, buf, size);
+    ret = net_fill_rstate(&s->rs, (const uint8_t *)buf, size);
 
     if (ret == -1) {
         goto eoc;
     }
+
+    return G_SOURCE_CONTINUE;
 }
 
 static void net_stream_cleanup(NetClientState *nc)
 {
     NetStreamState *s = DO_UPCAST(NetStreamState, nc, nc);
-    if (s->fd != -1) {
-        net_stream_read_poll(s, false);
-        net_stream_write_poll(s, false);
-        close(s->fd);
-        s->fd = -1;
+    if (s->ioc) {
+        if (QIO_CHANNEL_SOCKET(s->ioc)->fd != -1) {
+            if (s->ioc_read_tag) {
+                g_source_remove(s->ioc_read_tag);
+                s->ioc_read_tag = 0;
+            }
+            if (s->ioc_write_tag) {
+                g_source_remove(s->ioc_write_tag);
+                s->ioc_write_tag = 0;
+            }
+        }
+        object_unref(OBJECT(s->ioc));
+        s->ioc = NULL;
     }
-    if (s->listen_fd != -1) {
-        qemu_set_fd_handler(s->listen_fd, NULL, NULL, NULL);
-        closesocket(s->listen_fd);
-        s->listen_fd = -1;
+    if (s->listen_ioc) {
+        if (s->listener) {
+            qio_net_listener_disconnect(s->listener);
+            object_unref(OBJECT(s->listener));
+            s->listener = NULL;
+        }
+        object_unref(OBJECT(s->listen_ioc));
+        s->listen_ioc = NULL;
     }
 }
 
-static void net_stream_connect(void *opaque)
-{
-    NetStreamState *s = opaque;
-    s->send_fn = net_stream_send;
-    net_stream_read_poll(s, true);
-}
-
 static NetClientInfo net_stream_info = {
     .type = NET_CLIENT_DRIVER_STREAM,
     .size = sizeof(NetStreamState),
@@ -203,77 +216,64 @@ static NetClientInfo net_stream_info = {
     .cleanup = net_stream_cleanup,
 };
 
-static NetStreamState *net_stream_fd_init_stream(NetClientState *peer,
-                                                 const char *model,
-                                                 const char *name,
-                                                 int fd, int is_connected)
+static void net_stream_listen(QIONetListener *listener,
+                              QIOChannelSocket *cioc,
+                              void *opaque)
 {
-    NetClientState *nc;
-    NetStreamState *s;
+    NetStreamState *s = opaque;
+    SocketAddress *addr;
+    char *uri;
 
-    nc = qemu_new_net_client(&net_stream_info, peer, model, name);
+    object_ref(OBJECT(cioc));
 
-    snprintf(nc->info_str, sizeof(nc->info_str), "fd=%d", fd);
+    qio_net_listener_set_client_func(s->listener, NULL, s, NULL);
 
-    s = DO_UPCAST(NetStreamState, nc, nc);
+    s->ioc = QIO_CHANNEL(cioc);
+    qio_channel_set_name(s->ioc, "stream-server");
+    s->nc.link_down = false;
 
-    s->fd = fd;
-    s->listen_fd = -1;
-    net_socket_rs_init(&s->rs, net_stream_rs_finalize, false);
+    s->ioc_read_tag = qio_channel_add_watch(s->ioc, G_IO_IN, net_stream_send,
+                                            s, NULL);
 
-    /* Disable Nagle algorithm on TCP sockets to reduce latency */
-    socket_set_nodelay(fd);
+    addr = qio_channel_socket_get_remote_address(cioc, NULL);
+    g_assert(addr != NULL);
+    uri = socket_uri(addr);
+    pstrcpy(s->nc.info_str, sizeof(s->nc.info_str), uri);
+    g_free(uri);
+    qapi_free_SocketAddress(addr);
 
-    if (is_connected) {
-        net_stream_connect(s);
-    } else {
-        qemu_set_fd_handler(s->fd, NULL, net_stream_connect, s);
-    }
-    return s;
 }
 
-static void net_stream_accept(void *opaque)
+static void net_stream_server_listening(QIOTask *task, gpointer opaque)
 {
     NetStreamState *s = opaque;
-    struct sockaddr_storage saddr;
-    socklen_t len;
-    int fd;
-
-    for (;;) {
-        len = sizeof(saddr);
-        fd = qemu_accept(s->listen_fd, (struct sockaddr *)&saddr, &len);
-        if (fd < 0 && errno != EINTR) {
-            return;
-        } else if (fd >= 0) {
-            qemu_set_fd_handler(s->listen_fd, NULL, NULL, NULL);
-            break;
-        }
-    }
-
-    s->fd = fd;
-    s->nc.link_down = false;
-    net_stream_connect(s);
-    switch (saddr.ss_family) {
-    case AF_INET: {
-        struct sockaddr_in *saddr_in = (struct sockaddr_in *)&saddr;
+    QIOChannelSocket *listen_sioc = QIO_CHANNEL_SOCKET(s->listen_ioc);
+    SocketAddress *addr;
+    int ret;
 
-        snprintf(s->nc.info_str, sizeof(s->nc.info_str),
-                 "connection from %s:%d",
-                 inet_ntoa(saddr_in->sin_addr), ntohs(saddr_in->sin_port));
-        break;
+    if (listen_sioc->fd < 0) {
+        snprintf(s->nc.info_str, sizeof(s->nc.info_str), "connection error");
+        return;
     }
-    case AF_UNIX: {
-        struct sockaddr_un saddr_un;
 
-        len = sizeof(saddr_un);
-        getsockname(s->listen_fd, (struct sockaddr *)&saddr_un, &len);
+    addr = qio_channel_socket_get_local_address(listen_sioc, NULL);
+    g_assert(addr != NULL);
+    ret = qemu_socket_try_set_nonblock(listen_sioc->fd);
+    if (addr->type == SOCKET_ADDRESS_TYPE_FD && ret < 0) {
         snprintf(s->nc.info_str, sizeof(s->nc.info_str),
-                 "connect from %s", saddr_un.sun_path);
-        break;
-    }
-    default:
-        g_assert_not_reached();
+                 "can't use file descriptor %s (errno %d)",
+                 addr->u.fd.str, -ret);
+        return;
     }
+    g_assert(ret == 0);
+    qapi_free_SocketAddress(addr);
+
+    s->nc.link_down = true;
+    s->listener = qio_net_listener_new();
+
+    net_socket_rs_init(&s->rs, net_stream_rs_finalize, false);
+    qio_net_listener_set_client_func(s->listener, net_stream_listen, s, NULL);
+    qio_net_listener_add(s->listener, listen_sioc);
 }
 
 static int net_stream_server_init(NetClientState *peer,
@@ -284,104 +284,55 @@ static int net_stream_server_init(NetClientState *peer,
 {
     NetClientState *nc;
     NetStreamState *s;
-    int fd, ret;
+    QIOChannelSocket *listen_sioc = qio_channel_socket_new();
 
-    switch (addr->type) {
-    case SOCKET_ADDRESS_TYPE_INET: {
-        struct sockaddr_in saddr_in;
+    nc = qemu_new_net_client(&net_stream_info, peer, model, name);
+    s = DO_UPCAST(NetStreamState, nc, nc);
 
-        if (convert_host_port(&saddr_in, addr->u.inet.host, addr->u.inet.port,
-                              errp) < 0) {
-            return -1;
-        }
+    s->listen_ioc = QIO_CHANNEL(listen_sioc);
+    qio_channel_socket_listen_async(listen_sioc, addr, 0,
+                                    net_stream_server_listening, s,
+                                    NULL, NULL);
 
-        fd = qemu_socket(PF_INET, SOCK_STREAM, 0);
-        if (fd < 0) {
-            error_setg_errno(errp, errno, "can't create stream socket");
-            return -1;
-        }
-        qemu_socket_set_nonblock(fd);
+    return 0;
+}
 
-        socket_set_fast_reuse(fd);
+static void net_stream_client_connected(QIOTask *task, gpointer opaque)
+{
+    NetStreamState *s = opaque;
+    QIOChannelSocket *sioc = QIO_CHANNEL_SOCKET(s->ioc);
+    SocketAddress *addr;
+    gchar *uri;
+    int ret;
 
-        ret = bind(fd, (struct sockaddr *)&saddr_in, sizeof(saddr_in));
-        if (ret < 0) {
-            error_setg_errno(errp, errno, "can't bind ip=%s to socket",
-                             inet_ntoa(saddr_in.sin_addr));
-            closesocket(fd);
-            return -1;
-        }
-        break;
+    if (sioc->fd < 0) {
+        snprintf(s->nc.info_str, sizeof(s->nc.info_str), "connection error");
+        return;
     }
-    case SOCKET_ADDRESS_TYPE_UNIX: {
-        struct sockaddr_un saddr_un;
-
-        ret = unlink(addr->u.q_unix.path);
-        if (ret < 0 && errno != ENOENT) {
-            error_setg_errno(errp, errno, "failed to unlink socket %s",
-                             addr->u.q_unix.path);
-            return -1;
-        }
-
-        saddr_un.sun_family = PF_UNIX;
-        ret = snprintf(saddr_un.sun_path, sizeof(saddr_un.sun_path), "%s",
-                       addr->u.q_unix.path);
-        if (ret < 0 || ret >= sizeof(saddr_un.sun_path)) {
-            error_setg(errp, "UNIX socket path '%s' is too long",
-                       addr->u.q_unix.path);
-            error_append_hint(errp, "Path must be less than %zu bytes\n",
-                              sizeof(saddr_un.sun_path));
-        }
 
-        fd = qemu_socket(PF_UNIX, SOCK_STREAM, 0);
-        if (fd < 0) {
-            error_setg_errno(errp, errno, "can't create stream socket");
-            return -1;
-        }
-        qemu_socket_set_nonblock(fd);
-
-        ret = bind(fd, (struct sockaddr *)&saddr_un, sizeof(saddr_un));
-        if (ret < 0) {
-            error_setg_errno(errp, errno, "can't create socket with path: %s",
-                             saddr_un.sun_path);
-            closesocket(fd);
-            return -1;
-        }
-        break;
-    }
-    case SOCKET_ADDRESS_TYPE_FD:
-        fd = monitor_fd_param(monitor_cur(), addr->u.fd.str, errp);
-        if (fd == -1) {
-            return -1;
-        }
-        ret = qemu_socket_try_set_nonblock(fd);
-        if (ret < 0) {
-            error_setg_errno(errp, -ret, "%s: Can't use file descriptor %d",
-                             name, fd);
-            return -1;
-        }
-        break;
-    default:
-        error_setg(errp, "only support inet or fd type");
-        return -1;
-    }
+    addr = qio_channel_socket_get_remote_address(sioc, NULL);
+    g_assert(addr != NULL);
+    uri = socket_uri(addr);
+    pstrcpy(s->nc.info_str, sizeof(s->nc.info_str), uri);
+    g_free(uri);
 
-    ret = listen(fd, 0);
-    if (ret < 0) {
-        error_setg_errno(errp, errno, "can't listen on socket");
-        closesocket(fd);
-        return -1;
+    ret = qemu_socket_try_set_nonblock(sioc->fd);
+    if (addr->type == SOCKET_ADDRESS_TYPE_FD && ret < 0) {
+        snprintf(s->nc.info_str, sizeof(s->nc.info_str),
+                 "can't use file descriptor %s (errno %d)",
+                 addr->u.fd.str, -ret);
+        return;
     }
+    g_assert(ret == 0);
+    qapi_free_SocketAddress(addr);
 
-    nc = qemu_new_net_client(&net_stream_info, peer, model, name);
-    s = DO_UPCAST(NetStreamState, nc, nc);
-    s->fd = -1;
-    s->listen_fd = fd;
-    s->nc.link_down = true;
     net_socket_rs_init(&s->rs, net_stream_rs_finalize, false);
 
-    qemu_set_fd_handler(s->listen_fd, net_stream_accept, NULL, s);
-    return 0;
+    /* Disable Nagle algorithm on TCP sockets to reduce latency */
+    qio_channel_set_delay(s->ioc, false);
+
+    s->ioc_read_tag = qio_channel_add_watch(s->ioc, G_IO_IN, net_stream_send,
+                                            s, NULL);
 }
 
 static int net_stream_client_init(NetClientState *peer,
@@ -391,114 +342,17 @@ static int net_stream_client_init(NetClientState *peer,
                                   Error **errp)
 {
     NetStreamState *s;
-    int fd, connected, ret;
-    gchar *info_str;
-
-    switch (addr->type) {
-    case SOCKET_ADDRESS_TYPE_INET: {
-        struct sockaddr_in saddr_in;
-
-        if (convert_host_port(&saddr_in, addr->u.inet.host, addr->u.inet.port,
-                              errp) < 0) {
-            return -1;
-        }
-
-        fd = qemu_socket(PF_INET, SOCK_STREAM, 0);
-        if (fd < 0) {
-            error_setg_errno(errp, errno, "can't create stream socket");
-            return -1;
-        }
-        qemu_socket_set_nonblock(fd);
-
-        connected = 0;
-        for (;;) {
-            ret = connect(fd, (struct sockaddr *)&saddr_in, sizeof(saddr_in));
-            if (ret < 0) {
-                if (errno == EINTR || errno == EWOULDBLOCK) {
-                    /* continue */
-                } else if (errno == EINPROGRESS ||
-                           errno == EALREADY) {
-                    break;
-                } else {
-                    error_setg_errno(errp, errno, "can't connect socket");
-                    closesocket(fd);
-                    return -1;
-                }
-            } else {
-                connected = 1;
-                break;
-            }
-        }
-        info_str = g_strdup_printf("connect to %s:%d",
-                                   inet_ntoa(saddr_in.sin_addr),
-                                   ntohs(saddr_in.sin_port));
-        break;
-    }
-    case SOCKET_ADDRESS_TYPE_UNIX: {
-        struct sockaddr_un saddr_un;
-
-        saddr_un.sun_family = PF_UNIX;
-        ret = snprintf(saddr_un.sun_path, sizeof(saddr_un.sun_path), "%s",
-                       addr->u.q_unix.path);
-        if (ret < 0 || ret >= sizeof(saddr_un.sun_path)) {
-            error_setg(errp, "UNIX socket path '%s' is too long",
-                       addr->u.q_unix.path);
-            error_append_hint(errp, "Path must be less than %zu bytes\n",
-                              sizeof(saddr_un.sun_path));
-        }
+    NetClientState *nc;
+    QIOChannelSocket *sioc = qio_channel_socket_new();
 
-        fd = qemu_socket(PF_UNIX, SOCK_STREAM, 0);
-        if (fd < 0) {
-            error_setg_errno(errp, errno, "can't create stream socket");
-            return -1;
-        }
-        qemu_socket_set_nonblock(fd);
-
-        connected = 0;
-        for (;;) {
-            ret = connect(fd, (struct sockaddr *)&saddr_un, sizeof(saddr_un));
-            if (ret < 0) {
-                if (errno == EINTR || errno == EWOULDBLOCK) {
-                    /* continue */
-                } else if (errno == EAGAIN ||
-                           errno == EALREADY) {
-                    break;
-                } else {
-                    error_setg_errno(errp, errno, "can't connect socket");
-                    closesocket(fd);
-                    return -1;
-                }
-            } else {
-                connected = 1;
-                break;
-            }
-        }
-        info_str = g_strdup_printf(" connect to %s", saddr_un.sun_path);
-        break;
-    }
-    case SOCKET_ADDRESS_TYPE_FD:
-        fd = monitor_fd_param(monitor_cur(), addr->u.fd.str, errp);
-        if (fd == -1) {
-            return -1;
-        }
-        ret = qemu_socket_try_set_nonblock(fd);
-        if (ret < 0) {
-            error_setg_errno(errp, -ret, "%s: Can't use file descriptor %d",
-                             name, fd);
-            return -1;
-        }
-        connected = 1;
-        info_str = g_strdup_printf("connect to fd %d", fd);
-        break;
-    default:
-        error_setg(errp, "only support inet, unix or fd type");
-        return -1;
-    }
+    nc = qemu_new_net_client(&net_stream_info, peer, model, name);
+    s = DO_UPCAST(NetStreamState, nc, nc);
 
-    s = net_stream_fd_init_stream(peer, model, name, fd, connected);
+    s->ioc = QIO_CHANNEL(sioc);
 
-    pstrcpy(s->nc.info_str, sizeof(s->nc.info_str), info_str);
-    g_free(info_str);
+    qio_channel_socket_connect_async(sioc, addr,
+                                     net_stream_client_connected, s,
+                                     NULL, NULL);
 
     return 0;
 }
-- 
2.36.1