[Qemu-devel] [RFC PATCH 4/4] vhost-user: Add new option to specify vhost-user backend supports

[Tetsuya Mukawa]

This patch adds 'backend_features' option for vhost-user backends.
If this option is specified, QEMU assumes vhost-user backends support
the features specified by user, and QEMU can start without vhost-user
backend.

Here are examples.
* QEMU is configured as vhost-user client.
 -chardev socket,id=chr0,path=/tmp/sock,reconnect=3 \
 -netdev vhost-user,id=net0,chardev=chr0,vhostforce,backend_features=0x68000 \
 -device virtio-net-pci,netdev=net0 \

* QEMU is configured as vhost-user server.
 -chardev socket,id=chr0,path=/tmp/sock,server,nowait \
 -netdev vhost-user,id=net0,chardev=chr0,vhostforce,backend_features=0x68000 \
 -device virtio-net-pci,netdev=net0 \

To know vhost-user backend features that the backend expects, please
specify 0xffffffff as backend_features, then invoke QEMU and check error log
like below.

  Lack of backend features. Expected 0xffffffff, but receives 0x68000

Above log indicates the backend features QEMU should be passed.

Signed-off-by: Tetsuya Mukawa <address@hidden>
---
 hw/net/vhost_net.c             |  9 ++++++++-
 hw/net/virtio-net.c            | 24 ++++++++++++++++++++++++
 hw/scsi/vhost-scsi.c           |  2 +-
 hw/virtio/vhost-user.c         |  9 +++++++++
 hw/virtio/vhost.c              |  8 ++++++--
 include/hw/virtio/vhost.h      |  5 ++++-
 include/hw/virtio/virtio-net.h |  2 ++
 include/net/net.h              |  6 ++++++
 include/net/vhost_net.h        |  2 ++
 net/net.c                      | 18 ++++++++++++++++++
 net/tap.c                      |  5 ++++-
 net/vhost-user.c               | 42 ++++++++++++++++++++++++++++++++++++++++--
 qapi-schema.json               | 10 ++++++++--
 qemu-options.hx                |  3 ++-
 14 files changed, 134 insertions(+), 11 deletions(-)

diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c
index 47f8b89..f974d09 100644
--- a/hw/net/vhost_net.c
+++ b/hw/net/vhost_net.c
@@ -158,8 +158,15 @@ struct vhost_net *vhost_net_init(VhostNetOptions *options)
     net->dev.nvqs = 2;
     net->dev.vqs = net->vqs;
 
+    if (options->has_backend_features) {
+        net->dev.has_backend_features = options->has_backend_features;
+        net->dev.backend_features = options->backend_features;
+    }
+
     r = vhost_dev_init(&net->dev, options->opaque,
-                       options->backend_type, options->force);
+                       options->backend_type, options->force,
+                       options->has_backend_features,
+                       options->backend_features);
     if (r < 0) {
         goto fail;
     }
diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
index 3af6faf..ebe5422 100644
--- a/hw/net/virtio-net.c
+++ b/hw/net/virtio-net.c
@@ -366,6 +366,28 @@ static int peer_has_ufo(VirtIONet *n)
     return n->has_ufo;
 }
 
+static int peer_has_backend_features(VirtIONet *n)
+{
+    if (!peer_has_vnet_hdr(n))
+        return 0;
+
+    n->has_backend_features =
+            qemu_has_backend_features(qemu_get_queue(n->nic)->peer);
+
+    return n->has_backend_features;
+}
+
+static uint64_t peer_backend_features(VirtIONet *n)
+{
+    if (!peer_has_vnet_hdr(n))
+        return 0;
+
+    n->backend_features =
+            qemu_backend_features(qemu_get_queue(n->nic)->peer);
+
+    return n->backend_features;
+}
+
 static void virtio_net_set_mrg_rx_bufs(VirtIONet *n, int mergeable_rx_bufs)
 {
     int i;
@@ -463,6 +485,8 @@ static uint32_t virtio_net_get_features(VirtIODevice *vdev, 
uint32_t features)
     }
 
     if (!get_vhost_net(nc->peer)) {
+        if (peer_has_backend_features(n))
+            features = peer_backend_features(n);
         return features;
     }
     return vhost_net_get_features(get_vhost_net(nc->peer), features);
diff --git a/hw/scsi/vhost-scsi.c b/hw/scsi/vhost-scsi.c
index 335f442..4998a95 100644
--- a/hw/scsi/vhost-scsi.c
+++ b/hw/scsi/vhost-scsi.c
@@ -246,7 +246,7 @@ static void vhost_scsi_realize(DeviceState *dev, Error 
**errp)
     s->dev.backend_features = 0;
 
     ret = vhost_dev_init(&s->dev, (void *)(uintptr_t)vhostfd,
-                         VHOST_BACKEND_TYPE_KERNEL, true);
+                         VHOST_BACKEND_TYPE_KERNEL, true, false, 0);
     if (ret < 0) {
         error_setg(errp, "vhost-scsi: vhost initialization failed: %s",
                    strerror(-ret));
diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
index 4d7e3ba..f0dcb97 100644
--- a/hw/virtio/vhost-user.c
+++ b/hw/virtio/vhost-user.c
@@ -307,6 +307,15 @@ static int vhost_user_call(struct vhost_dev *dev, unsigned 
long int request,
                 error_report("Received bad msg size.");
                 goto close;
             }
+            if (dev->has_backend_features) {
+                if (dev->backend_features != (dev->backend_features & 
msg.u64)) {
+                    error_report("Lack of backend features. "
+                                 "Expected 0x%llx, but receives 0x%lx",
+                                 dev->backend_features, msg.u64);
+                    goto close;
+                }
+            }
+
             *((__u64 *) arg) = msg.u64;
             break;
         case VHOST_USER_GET_VRING_BASE:
diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
index 54851b7..20cb116 100644
--- a/hw/virtio/vhost.c
+++ b/hw/virtio/vhost.c
@@ -811,7 +811,9 @@ static void vhost_virtqueue_cleanup(struct vhost_virtqueue 
*vq)
 }
 
 int vhost_dev_init(struct vhost_dev *hdev, void *opaque,
-                   VhostBackendType backend_type, bool force)
+                   VhostBackendType backend_type, bool force,
+                   bool has_backend_features,
+                   unsigned long long backend_features)
 {
     uint64_t features;
     int i, r;
@@ -833,7 +835,9 @@ int vhost_dev_init(struct vhost_dev *hdev, void *opaque,
 
     r = hdev->vhost_ops->vhost_call(hdev, VHOST_GET_FEATURES, &features);
     if (r < 0) {
-        goto fail;
+        if (!has_backend_features)
+            goto fail;
+        features = backend_features;
     }
 
     for (i = 0; i < hdev->nvqs; ++i) {
diff --git a/include/hw/virtio/vhost.h b/include/hw/virtio/vhost.h
index 8f04888..60306cb 100644
--- a/include/hw/virtio/vhost.h
+++ b/include/hw/virtio/vhost.h
@@ -41,6 +41,7 @@ struct vhost_dev {
     unsigned long long features;
     unsigned long long acked_features;
     unsigned long long backend_features;
+    bool has_backend_features;
     bool started;
     bool log_enabled;
     vhost_log_chunk_t *log;
@@ -55,7 +56,9 @@ struct vhost_dev {
 };
 
 int vhost_dev_init(struct vhost_dev *hdev, void *opaque,
-                   VhostBackendType backend_type, bool force);
+                   VhostBackendType backend_type, bool force,
+                   bool has_backend_features,
+                   unsigned long long backend_features);
 void vhost_dev_cleanup(struct vhost_dev *hdev);
 bool vhost_dev_query(struct vhost_dev *hdev, VirtIODevice *vdev);
 int vhost_dev_start(struct vhost_dev *hdev, VirtIODevice *vdev);
diff --git a/include/hw/virtio/virtio-net.h b/include/hw/virtio/virtio-net.h
index e0dbb41..528da28 100644
--- a/include/hw/virtio/virtio-net.h
+++ b/include/hw/virtio/virtio-net.h
@@ -70,6 +70,8 @@ typedef struct VirtIONet {
     size_t guest_hdr_len;
     uint32_t host_features;
     uint8_t has_ufo;
+    uint8_t has_backend_features;
+    uint64_t backend_features;
     int mergeable_rx_bufs;
     uint8_t promisc;
     uint8_t allmulti;
diff --git a/include/net/net.h b/include/net/net.h
index e66ca03..e2eab3e 100644
--- a/include/net/net.h
+++ b/include/net/net.h
@@ -55,6 +55,8 @@ typedef bool (HasVnetHdrLen)(NetClientState *, int);
 typedef void (UsingVnetHdr)(NetClientState *, bool);
 typedef void (SetOffload)(NetClientState *, int, int, int, int, int);
 typedef void (SetVnetHdrLen)(NetClientState *, int);
+typedef bool (HasBackendFeatures)(NetClientState *);
+typedef unsigned long long (BackendFeatures)(NetClientState *);
 
 typedef struct NetClientInfo {
     NetClientOptionsKind type;
@@ -73,6 +75,8 @@ typedef struct NetClientInfo {
     UsingVnetHdr *using_vnet_hdr;
     SetOffload *set_offload;
     SetVnetHdrLen *set_vnet_hdr_len;
+    HasBackendFeatures *has_backend_features;
+    BackendFeatures *backend_features;
 } NetClientInfo;
 
 struct NetClientState {
@@ -136,6 +140,8 @@ bool qemu_has_ufo(NetClientState *nc);
 bool qemu_has_vnet_hdr(NetClientState *nc);
 bool qemu_has_vnet_hdr_len(NetClientState *nc, int len);
 void qemu_using_vnet_hdr(NetClientState *nc, bool enable);
+bool qemu_has_backend_features(NetClientState *nc);
+unsigned long long qemu_backend_features(NetClientState *nc);
 void qemu_set_offload(NetClientState *nc, int csum, int tso4, int tso6,
                       int ecn, int ufo);
 void qemu_set_vnet_hdr_len(NetClientState *nc, int len);
diff --git a/include/net/vhost_net.h b/include/net/vhost_net.h
index b1c18a3..b4ca8e8 100644
--- a/include/net/vhost_net.h
+++ b/include/net/vhost_net.h
@@ -12,6 +12,8 @@ typedef struct VhostNetOptions {
     NetClientState *net_backend;
     void *opaque;
     bool force;
+    bool has_backend_features;
+    unsigned long long backend_features;
 } VhostNetOptions;
 
 struct vhost_net *vhost_net_init(VhostNetOptions *options);
diff --git a/net/net.c b/net/net.c
index 7427f6a..85e41d0 100644
--- a/net/net.c
+++ b/net/net.c
@@ -459,6 +459,24 @@ void qemu_set_vnet_hdr_len(NetClientState *nc, int len)
     nc->info->set_vnet_hdr_len(nc, len);
 }
 
+bool qemu_has_backend_features(NetClientState *nc)
+{
+    if (!nc || !nc->info->has_backend_features) {
+        return false;
+    }
+
+    return nc->info->has_backend_features(nc);
+}
+
+unsigned long long qemu_backend_features(NetClientState *nc)
+{
+    if (!nc || !nc->info->backend_features) {
+        return false;
+    }
+
+    return nc->info->backend_features(nc);
+}
+
 int qemu_can_send_packet(NetClientState *sender)
 {
     int vm_running = runstate_is_running();
diff --git a/net/tap.c b/net/tap.c
index 968df46..6d25170 100644
--- a/net/tap.c
+++ b/net/tap.c
@@ -636,12 +636,15 @@ static int net_init_tap_one(const NetdevTapOptions *tap, 
NetClientState *peer,
     }
 
     if (tap->has_vhost ? tap->vhost :
-        vhostfdname || (tap->has_vhostforce && tap->vhostforce)) {
+        vhostfdname || tap->has_backend_features ||
+                    (tap->has_vhostforce && tap->vhostforce)) {
         VhostNetOptions options;
 
         options.backend_type = VHOST_BACKEND_TYPE_KERNEL;
         options.net_backend = &s->nc;
         options.force = tap->has_vhostforce && tap->vhostforce;
+        options.has_backend_features = tap->has_backend_features;
+        options.backend_features = tap->backend_features;
 
         if (tap->has_vhostfd || tap->has_vhostfds) {
             vhostfd = monitor_fd_param(cur_mon, vhostfdname, &err);
diff --git a/net/vhost-user.c b/net/vhost-user.c
index d31fc41..2705393 100644
--- a/net/vhost-user.c
+++ b/net/vhost-user.c
@@ -19,6 +19,8 @@ typedef struct VhostUserState {
     NetClientState nc;
     CharDriverState *chr;
     VHostNetState *vhost_net;
+    bool has_backend_features;
+    unsigned long long backend_features;
     int watch;
 } VhostUserState;
 
@@ -54,6 +56,7 @@ static int vhost_user_start(VhostUserState *s)
     options.net_backend = &s->nc;
     options.opaque = s->chr;
     options.force = true;
+    options.backend_features = s->backend_features;
 
     s->vhost_net = vhost_net_init(&options);
 
@@ -91,12 +94,32 @@ static bool vhost_user_has_ufo(NetClientState *nc)
     return true;
 }
 
+static bool vhost_user_has_backend_features(NetClientState *nc)
+{
+    assert(nc->info->type == NET_CLIENT_OPTIONS_KIND_VHOST_USER);
+
+    VhostUserState *s = DO_UPCAST(VhostUserState, nc, nc);
+
+    return s->has_backend_features;
+}
+
+static unsigned long long vhost_user_backend_features(NetClientState *nc)
+{
+    assert(nc->info->type == NET_CLIENT_OPTIONS_KIND_VHOST_USER);
+
+    VhostUserState *s = DO_UPCAST(VhostUserState, nc, nc);
+
+    return s->backend_features;
+}
+
 static NetClientInfo net_vhost_user_info = {
         .type = NET_CLIENT_OPTIONS_KIND_VHOST_USER,
         .size = sizeof(VhostUserState),
         .cleanup = vhost_user_cleanup,
         .has_vnet_hdr = vhost_user_has_vnet_hdr,
         .has_ufo = vhost_user_has_ufo,
+        .has_backend_features = vhost_user_has_backend_features,
+        .backend_features = vhost_user_backend_features,
 };
 
 static void net_vhost_link_down(VhostUserState *s, bool link_down)
@@ -148,7 +171,9 @@ static void net_vhost_user_event(void *opaque, int event)
 }
 
 static int net_vhost_user_init(NetClientState *peer, const char *device,
-                               const char *name, CharDriverState *chr)
+                               const char *name, CharDriverState *chr,
+                               bool has_backend_features,
+                               unsigned long long backend_features)
 {
     NetClientState *nc;
     VhostUserState *s;
@@ -163,6 +188,8 @@ static int net_vhost_user_init(NetClientState *peer, const 
char *device,
     /* We don't provide a receive callback */
     s->nc.receive_disabled = 1;
     s->chr = chr;
+    s->has_backend_features = has_backend_features;
+    s->backend_features = backend_features;
 
     qemu_chr_add_handlers(s->chr, NULL, NULL, net_vhost_user_event, s);
 
@@ -247,6 +274,8 @@ int net_init_vhost_user(const NetClientOptions *opts, const 
char *name,
 {
     const NetdevVhostUserOptions *vhost_user_opts;
     CharDriverState *chr;
+    bool has_backend_features;;
+    unsigned long long backend_features;
 
     assert(opts->kind == NET_CLIENT_OPTIONS_KIND_VHOST_USER);
     vhost_user_opts = opts->vhost_user;
@@ -263,6 +292,15 @@ int net_init_vhost_user(const NetClientOptions *opts, 
const char *name,
         return -1;
     }
 
+    /* backend features */
+    if (vhost_user_opts->has_backend_features) {
+        has_backend_features = true;
+        backend_features = vhost_user_opts->backend_features;
+    } else {
+        has_backend_features = false;
+        backend_features = 0;
+    }
 
-    return net_vhost_user_init(peer, "vhost_user", name, chr);
+    return net_vhost_user_init(peer, "vhost_user", name, chr,
+                                has_backend_features, backend_features);
 }
diff --git a/qapi-schema.json b/qapi-schema.json
index f97ffa1..bab4a74 100644
--- a/qapi-schema.json
+++ b/qapi-schema.json
@@ -2243,6 +2243,8 @@
 #
 # @queues: #optional number of queues to be created for multiqueue capable tap
 #
+# @backend_features: #optional feature flag to support vhost user backend
+#
 # Since 1.2
 ##
 { 'struct': 'NetdevTapOptions',
@@ -2259,7 +2261,8 @@
     '*vhostfd':    'str',
     '*vhostfds':   'str',
     '*vhostforce': 'bool',
-    '*queues':     'uint32'} }
+    '*queues':     'uint32',
+    '*backend_features':'uint64'} }
 
 ##
 # @NetdevSocketOptions
@@ -2444,12 +2447,15 @@
 #
 # @vhostforce: #optional vhost on for non-MSIX virtio guests (default: false).
 #
+# @backend_features: #optional feature flag to support vhost user backend 
(default: 0).
+#
 # Since 2.1
 ##
 { 'struct': 'NetdevVhostUserOptions',
   'data': {
     'chardev':        'str',
-    '*vhostforce':    'bool' } }
+    '*vhostforce':    'bool',
+    '*backend_features':    'uint64' } }
 
 ##
 # @NetClientOptions
diff --git a/qemu-options.hx b/qemu-options.hx
index ec356f6..3ad3486 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -1466,7 +1466,7 @@ DEF("net", HAS_ARG, QEMU_OPTION_net,
     "-net tap[,vlan=n][,name=str],ifname=name\n"
     "                connect the host TAP network interface to VLAN 'n'\n"
 #else
-    "-net 
tap[,vlan=n][,name=str][,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off][,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n]\n"
+    "-net 
tap[,vlan=n][,name=str][,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off][,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n][,backend_features=n]\n"
     "                connect the host TAP network interface to VLAN 'n'\n"
     "                use network scripts 'file' (default=" 
DEFAULT_NETWORK_SCRIPT ")\n"
     "                to configure it and 'dfile' (default=" 
DEFAULT_NETWORK_DOWN_SCRIPT ")\n"
@@ -1486,6 +1486,7 @@ DEF("net", HAS_ARG, QEMU_OPTION_net,
     "                use 'vhostfd=h' to connect to an already opened vhost net 
device\n"
     "                use 'vhostfds=x:y:...:z to connect to multiple already 
opened vhost net devices\n"
     "                use 'queues=n' to specify the number of queues to be 
created for multiqueue TAP\n"
+    "                use 'backend_features=n' to specify the features that 
vhost backend supported\n"
     "-net bridge[,vlan=n][,name=str][,br=bridge][,helper=helper]\n"
     "                connects a host TAP network interface to a host bridge 
device 'br'\n"
     "                (default=" DEFAULT_BRIDGE_INTERFACE ") using the program 
'helper'\n"
-- 
2.1.4