Skip to contentSkip to conversation

vhost: disable on tap link down

PATCH Superseded Michael S. Tsirkin 2 participants 4 messages Feb 7, 2011 – Feb 10, 2011

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 vhost: disable on tap link down

qemu makes it possible to disable link at tap
which is not communicated to the guest but
causes all packets to be dropped.

Handle this with vhost simply by moving to the userspace emulation.

Note: it might be a good idea to make peer link status match tap in this case, so the guest gets an event
and updates the carrier state. For now
stay bug for bug compatible with what we used to have.

---

Untested.
Pradeep, mind trying this patch out and reporting? Thanks!

 hw/virtio-net.c |    4 ++++
 net.c           |    7 +++++++
 2 files changed, 11 insertions(+), 0 deletions(-)
hw/virtio-net.c +4 -0
index 671d952..fc2d6f5 100644
@@ -112,6 +112,10 @@ static void virtio_net_vhost_status(VirtIONet *n, uint8_t status)
         return;
     }
 
+    if (n->nic->nc.peer->link_down) {
+        return;
+    }
+
     if (!tap_get_vhost_net(n->nic->nc.peer)) {
         return;
     }
net.c +7 -0
index 9ba5be2..57ee997 100644
@@ -1324,6 +1324,13 @@ done:
     if (vc->info->link_status_changed) {
         vc->info->link_status_changed(vc);
     }
+
+    /* Notify peer. Don't update peer link status: this makes it possible to
+     * disconnect from host network without notifying the guest.
+     * FIXME: is this useful? Could just be an artifact of vlan support. */
+    if (vc->peer && vc->peer->info->link_status_changed) {
+        vc->peer->info->link_status_changed(vc->peer);
+    }
     return 0;
 }
 
Signature · 1 line
1.7.3.2.91.g446ac
#1 lore raw
Quoted 11 lines · On Mon, 7 Feb 2011 15:50:01 +0200 "Michael S. Tsirkin" <[email protected]> wrote:

qemu makes it possible to disable link at tap
which is not communicated to the guest but
causes all packets to be dropped.

Handle this with vhost simply by moving to the userspace emulation.

Note: it might be a good idea to make peer link status match tap in this case, so the guest gets an event
and updates the carrier state. For now
stay bug for bug compatible with what we used to have.

Quoted 6 lines

---

Untested.
Pradeep, mind trying this patch out and reporting?

Hi mst

This patch works. Thanks

--Pradeep

Quoted 41 lines

Thanks!

 hw/virtio-net.c |    4 ++++
 net.c           |    7 +++++++
 2 files changed, 11 insertions(+), 0 deletions(-)
hw/virtio-net.c +4 -0
index 671d952..fc2d6f5 100644
@@ -112,6 +112,10 @@ static void virtio_net_vhost_status(VirtIONet
 *n, uint8_t status) return;
     }
 
+    if (n->nic->nc.peer->link_down) {
+        return;
+    }
+
     if (!tap_get_vhost_net(n->nic->nc.peer)) {
         return;
     }
net.c +2 -0
index 9ba5be2..57ee997 100644
@@ -1324,6 +1324,13 @@ done:
     if (vc->info->link_status_changed) {
         vc->info->link_status_changed(vc);
     }
+
+    /* Notify peer. Don't update peer link status: this makes it

possible to

net.c +2 -0
+     * disconnect from host network without notifying the guest.
+     * FIXME: is this useful? Could just be an artifact of vlan

support. */

net.c +3 -0
+    if (vc->peer && vc->peer->info->link_status_changed) {
+        vc->peer->info->link_status_changed(vc->peer);
+    }
     return 0;
 }
#2 lore raw
Quoted 18 lines · pradeep
Quoted 15 lines · On Mon, 7 Feb 2011 15:50:01 +0200 "Michael S. Tsirkin" <[email protected]> wrote:

qemu makes it possible to disable link at tap
which is not communicated to the guest but
causes all packets to be dropped.

Handle this with vhost simply by moving to the userspace emulation.

Note: it might be a good idea to make peer link status match tap in this case, so the guest gets an event
and updates the carrier state. For now
stay bug for bug compatible with what we used to have.

---

Quoted 6 lines
Quoted 2 lines

Untested.
Pradeep, mind trying this patch out and reporting?

Hi mst

This patch works. Thanks

Strange actually. Did you put the link down before guest booted? I went to test it with set link after guest is up, and it didn't work, I needed this on top - can you ack this in your setup as well pls?

hw/virtio-net.c +2 -5
index fc2d6f5..3e3d73a 100644
@@ -112,14 +112,11 @@ static void virtio_net_vhost_status(VirtIONet *n, uint8_t status)
         return;
     }
 
-    if (n->nic->nc.peer->link_down) {
-        return;
-    }
-
     if (!tap_get_vhost_net(n->nic->nc.peer)) {
         return;
     }
-    if (!!n->vhost_started == virtio_net_started(n, status)) {
+    if (!!n->vhost_started == virtio_net_started(n, status) &&
+        !n->nic->nc.peer->link_down) {
         return;
     }
     if (!n->vhost_started) {
#3 lore raw
Quoted 23 lines · On Tue, 8 Feb 2011 17:41:12 +0200 "Michael S. Tsirkin" <[email protected]> wrote:
Quoted 22 lines · pradeep
Quoted 18 lines · On Mon, 7 Feb 2011 15:50:01 +0200 "Michael S. Tsirkin" <[email protected]> wrote:

qemu makes it possible to disable link at tap
which is not communicated to the guest but
causes all packets to be dropped.

Handle this with vhost simply by moving to the userspace emulation.

Note: it might be a good idea to make peer link status match tap in this case, so the guest gets an event
and updates the carrier state. For now
stay bug for bug compatible with what we used to have.

---

Untested.
Pradeep, mind trying this patch out and reporting?

Quoted 7 lines
Quoted 3 lines

Hi mst

This patch works. Thanks

Strange actually. Did you put the link down before guest booted? I went to test it with set link after guest is up, and it didn't work, I needed this on top - can you ack this in your setup as well pls?

ahhhh. Old one dint. New patch works for me.
Thanks mst

--Pradeep

Quoted 22 lines
hw/virtio-net.c +2 -5
index fc2d6f5..3e3d73a 100644
@@ -112,14 +112,11 @@ static void virtio_net_vhost_status(VirtIONet
 *n, uint8_t status) return;
     }
 
-    if (n->nic->nc.peer->link_down) {
-        return;
-    }
-
     if (!tap_get_vhost_net(n->nic->nc.peer)) {
         return;
     }
-    if (!!n->vhost_started == virtio_net_started(n, status)) {
+    if (!!n->vhost_started == virtio_net_started(n, status) &&
+        !n->nic->nc.peer->link_down) {
         return;
     }
     if (!n->vhost_started) {