[Qemu-devel] [PULL 058/103] Add the vhost-user netdev backend to the command line

["Michael S. Tsirkin" <mst@redhat.com>]

From: Nikolay Nikolaev <n.nikolaev@virtualopensystems.com>

The supplied chardev id will be inspected for supported options. Only
a socket backend, with a set path (i.e. a Unix socket) and optionally
the server parameter set, will be allowed. Other options (nowait, telnet)
will make the chardev unusable and the netdev will not be initialised.

Additional checks for validity:
  - requires `-numa node,memdev=..`
  - requires `-device virtio-net-*`

The `vhostforce` option is used to force vhost-net when we deal with
non-MSIX guests.

Signed-off-by: Antonios Motakis <a.motakis@virtualopensystems.com>
Signed-off-by: Nikolay Nikolaev <n.nikolaev@virtualopensystems.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Luiz Capitulino <lcapitulino@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
---
 qapi-schema.json   |  19 +++++++++-
 hw/net/vhost_net.c |   4 ++
 net/hub.c          |   1 +
 net/net.c          |   3 ++
 net/vhost-user.c   | 109 +++++++++++++++++++++++++++++++++++++++++++++++++++--
 hmp-commands.hx    |   4 +-
 qemu-options.hx    |  18 +++++++++
 7 files changed, 152 insertions(+), 6 deletions(-)

diff --git a/qapi-schema.json b/qapi-schema.json
index 7bc33ea..0837fc5 100644
--- a/qapi-schema.json
+++ b/qapi-schema.json
@@ -3267,6 +3267,22 @@
     '*devname':    'str' } }
 
 ##
+# @NetdevVhostUserOptions
+#
+# Vhost-user network backend
+#
+# @chardev: name of a unix socket chardev
+#
+# @vhostforce: #optional vhost on for non-MSIX virtio guests (default: false).
+#
+# Since 2.1
+##
+{ 'type': 'NetdevVhostUserOptions',
+  'data': {
+    'chardev':        'str',
+    '*vhostforce':    'bool' } }
+
+##
 # @NetClientOptions
 #
 # A discriminated record of network device traits.
@@ -3284,7 +3300,8 @@
     'dump':     'NetdevDumpOptions',
     'bridge':   'NetdevBridgeOptions',
     'hubport':  'NetdevHubPortOptions',
-    'netmap':   'NetdevNetmapOptions' } }
+    'netmap':   'NetdevNetmapOptions',
+    'vhost-user': 'NetdevVhostUserOptions' } }
 
 ##
 # @NetLegacy
diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c
index 5f06736..7ac7c21 100644
--- a/hw/net/vhost_net.c
+++ b/hw/net/vhost_net.c
@@ -15,6 +15,7 @@
 
 #include "net/net.h"
 #include "net/tap.h"
+#include "net/vhost-user.h"
 
 #include "hw/virtio/virtio-net.h"
 #include "net/vhost_net.h"
@@ -360,6 +361,9 @@ VHostNetState *get_vhost_net(NetClientState *nc)
     case NET_CLIENT_OPTIONS_KIND_TAP:
         vhost_net = tap_get_vhost_net(nc);
         break;
+    case NET_CLIENT_OPTIONS_KIND_VHOST_USER:
+        vhost_net = vhost_user_get_vhost_net(nc);
+        break;
     default:
         break;
     }
diff --git a/net/hub.c b/net/hub.c
index 33a99c9..7e0f2d6 100644
--- a/net/hub.c
+++ b/net/hub.c
@@ -322,6 +322,7 @@ void net_hub_check_clients(void)
             case NET_CLIENT_OPTIONS_KIND_TAP:
             case NET_CLIENT_OPTIONS_KIND_SOCKET:
             case NET_CLIENT_OPTIONS_KIND_VDE:
+            case NET_CLIENT_OPTIONS_KIND_VHOST_USER:
                 has_host_dev = 1;
                 break;
             default:
diff --git a/net/net.c b/net/net.c
index 0ff2e40..c91b67b 100644
--- a/net/net.c
+++ b/net/net.c
@@ -786,6 +786,7 @@ static int (* const net_client_init_fun[NET_CLIENT_OPTIONS_KIND_MAX])(
         [NET_CLIENT_OPTIONS_KIND_BRIDGE]    = net_init_bridge,
 #endif
         [NET_CLIENT_OPTIONS_KIND_HUBPORT]   = net_init_hubport,
+        [NET_CLIENT_OPTIONS_KIND_VHOST_USER] = net_init_vhost_user,
 };
 
 
@@ -819,6 +820,7 @@ static int net_client_init1(const void *object, int is_netdev, Error **errp)
         case NET_CLIENT_OPTIONS_KIND_BRIDGE:
 #endif
         case NET_CLIENT_OPTIONS_KIND_HUBPORT:
+        case NET_CLIENT_OPTIONS_KIND_VHOST_USER:
             break;
 
         default:
@@ -907,6 +909,7 @@ static int net_host_check_device(const char *device)
 #ifdef CONFIG_VDE
                                        ,"vde"
 #endif
+                                       ,"vhost-user"
     };
     for (i = 0; i < ARRAY_SIZE(valid_param_list); i++) {
         if (!strncmp(valid_param_list[i], device,
diff --git a/net/vhost-user.c b/net/vhost-user.c
index 4bdd19d..24e050c 100644
--- a/net/vhost-user.c
+++ b/net/vhost-user.c
@@ -12,6 +12,7 @@
 #include "net/vhost_net.h"
 #include "net/vhost-user.h"
 #include "sysemu/char.h"
+#include "qemu/config-file.h"
 #include "qemu/error-report.h"
 
 typedef struct VhostUserState {
@@ -21,9 +22,16 @@ typedef struct VhostUserState {
     VHostNetState *vhost_net;
 } VhostUserState;
 
+typedef struct VhostUserChardevProps {
+    bool is_socket;
+    bool is_unix;
+    bool is_server;
+} VhostUserChardevProps;
+
 VHostNetState *vhost_user_get_vhost_net(NetClientState *nc)
 {
     VhostUserState *s = DO_UPCAST(VhostUserState, nc, nc);
+    assert(nc->info->type == NET_CLIENT_OPTIONS_KIND_VHOST_USER);
     return s->vhost_net;
 }
 
@@ -82,7 +90,7 @@ static bool vhost_user_has_ufo(NetClientState *nc)
 }
 
 static NetClientInfo net_vhost_user_info = {
-        .type = 0,
+        .type = NET_CLIENT_OPTIONS_KIND_VHOST_USER,
         .size = sizeof(VhostUserState),
         .cleanup = vhost_user_cleanup,
         .has_vnet_hdr = vhost_user_has_vnet_hdr,
@@ -148,8 +156,103 @@ static int net_vhost_user_init(NetClientState *peer, const char *device,
     return 0;
 }
 
+static int net_vhost_chardev_opts(const char *name, const char *value,
+                                  void *opaque)
+{
+    VhostUserChardevProps *props = opaque;
+
+    if (strcmp(name, "backend") == 0 && strcmp(value, "socket") == 0) {
+        props->is_socket = true;
+    } else if (strcmp(name, "path") == 0) {
+        props->is_unix = true;
+    } else if (strcmp(name, "server") == 0) {
+        props->is_server = true;
+    } else {
+        error_report("vhost-user does not support a chardev"
+                     " with the following option:\n %s = %s",
+                     name, value);
+        return -1;
+    }
+    return 0;
+}
+
+static CharDriverState *net_vhost_parse_chardev(const NetdevVhostUserOptions *opts)
+{
+    CharDriverState *chr = qemu_chr_find(opts->chardev);
+    VhostUserChardevProps props;
+
+    if (chr == NULL) {
+        error_report("chardev \"%s\" not found", opts->chardev);
+        return NULL;
+    }
+
+    /* inspect chardev opts */
+    memset(&props, 0, sizeof(props));
+    if (qemu_opt_foreach(chr->opts, net_vhost_chardev_opts, &props, true) != 0) {
+        return NULL;
+    }
+
+    if (!props.is_socket || !props.is_unix) {
+        error_report("chardev \"%s\" is not a unix socket",
+                     opts->chardev);
+        return NULL;
+    }
+
+    qemu_chr_fe_claim_no_fail(chr);
+
+    return chr;
+}
+
+static int net_vhost_check_net(QemuOpts *opts, void *opaque)
+{
+    const char *name = opaque;
+    const char *driver, *netdev;
+    const char virtio_name[] = "virtio-net-";
+
+    driver = qemu_opt_get(opts, "driver");
+    netdev = qemu_opt_get(opts, "netdev");
+
+    if (!driver || !netdev) {
+        return 0;
+    }
+
+    if (strcmp(netdev, name) == 0 &&
+        strncmp(driver, virtio_name, strlen(virtio_name)) != 0) {
+        error_report("vhost-user requires frontend driver virtio-net-*");
+        return -1;
+    }
+
+    return 0;
+}
+
 int net_init_vhost_user(const NetClientOptions *opts, const char *name,
-                   NetClientState *peer)
+                        NetClientState *peer)
 {
-    return net_vhost_user_init(peer, "vhost_user", 0, 0, 0);
+    const NetdevVhostUserOptions *vhost_user_opts;
+    CharDriverState *chr;
+    bool vhostforce;
+
+    assert(opts->kind == NET_CLIENT_OPTIONS_KIND_VHOST_USER);
+    vhost_user_opts = opts->vhost_user;
+
+    chr = net_vhost_parse_chardev(vhost_user_opts);
+    if (!chr) {
+        error_report("No suitable chardev found");
+        return -1;
+    }
+
+    /* verify net frontend */
+    if (qemu_opts_foreach(qemu_find_opts("device"), net_vhost_check_net,
+                          (char *)name, true) == -1) {
+        return -1;
+    }
+
+    /* vhostforce for non-MSIX */
+    if (vhost_user_opts->has_vhostforce) {
+        vhostforce = vhost_user_opts->vhostforce;
+    } else {
+        vhostforce = false;
+    }
+
+    return net_vhost_user_init(peer, "vhost_user", name, chr, vhostforce);
 }
diff --git a/hmp-commands.hx b/hmp-commands.hx
index 2e462c0..bc9c032 100644
--- a/hmp-commands.hx
+++ b/hmp-commands.hx
@@ -1206,7 +1206,7 @@ ETEXI
     {
         .name       = "host_net_add",
         .args_type  = "device:s,opts:s?",
-        .params     = "tap|user|socket|vde|netmap|dump [options]",
+        .params     = "tap|user|socket|vde|netmap|vhost-user|dump [options]",
         .help       = "add host VLAN client",
         .mhandler.cmd = net_host_device_add,
     },
@@ -1234,7 +1234,7 @@ ETEXI
     {
         .name       = "netdev_add",
         .args_type  = "netdev:O",
-        .params     = "[user|tap|socket|vde|bridge|hubport|netmap],id=str[,prop=value][,...]",
+        .params     = "[user|tap|socket|vde|bridge|hubport|netmap|vhost-user],id=str[,prop=value][,...]",
         .help       = "add host network device",
         .mhandler.cmd = hmp_netdev_add,
         .command_completion = netdev_add_completion,
diff --git a/qemu-options.hx b/qemu-options.hx
index 5a4eff9..ab06df1 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -1460,6 +1460,7 @@ DEF("netdev", HAS_ARG, QEMU_OPTION_netdev,
 #ifdef CONFIG_NETMAP
     "netmap|"
 #endif
+    "vhost-user|"
     "socket|"
     "hubport],id=str[,option][,option][,...]\n", QEMU_ARCH_ALL)
 STEXI
@@ -1791,6 +1792,23 @@ The hubport netdev lets you connect a NIC to a QEMU "vlan" instead of a single
 netdev.  @code{-net} and @code{-device} with parameter @option{vlan} create the
 required hub automatically.
 
+@item -netdev vhost-user,chardev=@var{id}[,vhostforce=on|off]
+
+Establish a vhost-user netdev, backed by a chardev @var{id}. The chardev should
+be a unix domain socket backed one. The vhost-user uses a specifically defined
+protocol to pass vhost ioctl replacement messages to an application on the other
+end of the socket. On non-MSIX guests, the feature can be forced with
+@var{vhostforce}.
+
+Example:
+@example
+qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \
+     -numa node,memdev=mem \
+     -chardev socket,path=/path/to/socket \
+     -netdev type=vhost-user,id=net0,chardev=chr0 \
+     -device virtio-net-pci,netdev=net0
+@end example
+
 @item -net dump[,vlan=@var{n}][,file=@var{file}][,len=@var{len}]
 Dump network traffic on VLAN @var{n} to file @var{file} (@file{qemu-vlan0.pcap} by default).
 At most @var{len} bytes (64k by default) per packet are stored. The file format is
-- 
MST