vhost: restrict postcopy live-migration enablement

Postcopy live-migration feature requires the application to
not populate the guest memory. As the vhost library cannot
prevent the application to that (e.g. preventing the
application to call mlockall()), the feature is disabled by
default.

The application should only enable the feature if it does not
force the guest memory to be populated.

In case the user passes the RTE_VHOST_USER_POSTCOPY_SUPPORT
flag at registration but the feature was not compiled,
registration fails.

For the same reason, postcopy and dequeue zero copy features
are not compatible, so don't advertize postcopy support if
dequeue zero copy is requested.

Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>

diff --git a/doc/guides/prog_guide/vhost_lib.rst b/doc/guides/prog_guide/vhost_lib.rst
index 77af4d7..c77df33 100644 (file)
--- a/doc/guides/prog_guide/vhost_lib.rst
+++ b/doc/guides/prog_guide/vhost_lib.rst
@@ -106,6 +106,14 @@ The following is an overview of some key Vhost API functions:
     Enabling this flag with these Qemu version results in Qemu being blocked
     when multiple queue pairs are declared.
 
+  - ``RTE_VHOST_USER_POSTCOPY_SUPPORT``
+
+    Postcopy live-migration support will be enabled when this flag is set.
+    It is disabled by default.
+
+    Enabling this flag should only be done when the calling application does
+    not pre-fault the guest shared memory, otherwise migration would fail.
+
 * ``rte_vhost_driver_set_features(path, features)``
 
   This function sets the feature bits the vhost-user driver supports. The

diff --git a/lib/librte_vhost/rte_vhost.h b/lib/librte_vhost/rte_vhost.h
index 9292c89..d280ac4 100644 (file)
--- a/lib/librte_vhost/rte_vhost.h
+++ b/lib/librte_vhost/rte_vhost.h
@@ -28,6 +28,7 @@ extern "C" {
 #define RTE_VHOST_USER_NO_RECONNECT    (1ULL << 1)
 #define RTE_VHOST_USER_DEQUEUE_ZERO_COPY       (1ULL << 2)
 #define RTE_VHOST_USER_IOMMU_SUPPORT   (1ULL << 3)
+#define RTE_VHOST_USER_POSTCOPY_SUPPORT                (1ULL << 4)
 
 /** Protocol features. */
 #ifndef VHOST_USER_PROTOCOL_F_MQ

diff --git a/lib/librte_vhost/socket.c b/lib/librte_vhost/socket.c
index c278e9b..01b60ff 100644 (file)
--- a/lib/librte_vhost/socket.c
+++ b/lib/librte_vhost/socket.c
@@ -51,6 +51,8 @@ struct vhost_user_socket {
        uint64_t supported_features;
        uint64_t features;
 
+       uint64_t protocol_features;
+
        /*
         * Device id to identify a specific backend device.
         * It's set to -1 for the default software implementation.
@@ -726,7 +728,7 @@ rte_vhost_driver_get_protocol_features(const char *path,
        did = vsocket->vdpa_dev_id;
        vdpa_dev = rte_vdpa_get_device(did);
        if (!vdpa_dev || !vdpa_dev->ops->get_protocol_features) {
-               *protocol_features = VHOST_USER_PROTOCOL_FEATURES;
+               *protocol_features = vsocket->protocol_features;
                goto unlock_exit;
        }
 
@@ -739,7 +741,7 @@ rte_vhost_driver_get_protocol_features(const char *path,
                goto unlock_exit;
        }
 
-       *protocol_features = VHOST_USER_PROTOCOL_FEATURES
+       *protocol_features = vsocket->protocol_features
                & vdpa_protocol_features;
 
 unlock_exit:
@@ -858,11 +860,21 @@ rte_vhost_driver_register(const char *path, uint64_t flags)
        vsocket->use_builtin_virtio_net = true;
        vsocket->supported_features = VIRTIO_NET_SUPPORTED_FEATURES;
        vsocket->features           = VIRTIO_NET_SUPPORTED_FEATURES;
+       vsocket->protocol_features  = VHOST_USER_PROTOCOL_FEATURES;
 
-       /* Dequeue zero copy can't assure descriptors returned in order */
+       /*
+        * Dequeue zero copy can't assure descriptors returned in order.
+        * Also, it requires that the guest memory is populated, which is
+        * not compatible with postcopy.
+        */
        if (vsocket->dequeue_zero_copy) {
                vsocket->supported_features &= ~(1ULL << VIRTIO_F_IN_ORDER);
                vsocket->features &= ~(1ULL << VIRTIO_F_IN_ORDER);
+
+               RTE_LOG(INFO, VHOST_CONFIG,
+                       "Dequeue zero copy requested, disabling postcopy support\n");
+               vsocket->protocol_features &=
+                       ~(1ULL << VHOST_USER_PROTOCOL_F_PAGEFAULT);
        }
 
        if (!(flags & RTE_VHOST_USER_IOMMU_SUPPORT)) {
@@ -870,6 +882,18 @@ rte_vhost_driver_register(const char *path, uint64_t flags)
                vsocket->features &= ~(1ULL << VIRTIO_F_IOMMU_PLATFORM);
        }
 
+       if (!(flags & RTE_VHOST_USER_POSTCOPY_SUPPORT)) {
+               vsocket->protocol_features &=
+                       ~(1ULL << VHOST_USER_PROTOCOL_F_PAGEFAULT);
+       } else {
+#ifndef RTE_LIBRTE_VHOST_POSTCOPY
+               RTE_LOG(ERR, VHOST_CONFIG,
+                       "Postcopy requested but not compiled\n");
+               ret = -1;
+               goto out_mutex;
+#endif
+       }
+
        if ((flags & RTE_VHOST_USER_CLIENT) != 0) {
                vsocket->reconnect = !(flags & RTE_VHOST_USER_NO_RECONNECT);
                if (vsocket->reconnect && reconn_tid == 0) {

diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
index 0560f2f..508228a 100644 (file)
--- a/lib/librte_vhost/vhost_user.c
+++ b/lib/librte_vhost/vhost_user.c
@@ -1322,7 +1322,11 @@ vhost_user_set_protocol_features(struct virtio_net **pdev,
 {
        struct virtio_net *dev = *pdev;
        uint64_t protocol_features = msg->payload.u64;
-       if (protocol_features & ~VHOST_USER_PROTOCOL_FEATURES) {
+       uint64_t slave_protocol_features = 0;
+
+       rte_vhost_driver_get_protocol_features(dev->ifname,
+                       &slave_protocol_features);
+       if (protocol_features & ~slave_protocol_features) {
                RTE_LOG(ERR, VHOST_CONFIG,
                        "(%d) received invalid protocol features.\n",
                        dev->vid);