Xen devel - Sep 2013 - [PATCH net-next v2 1/2] xen-netback: add a vif-is-connected flag

Having applied my patch to separate vif disconnect and free, I ran into a
BUG when testing resume from S3 with a Windows frontend because the vif task
pointer was not cleared by xenvif_disconnect() and so a double call to this
function tries to stop the thread twice.
Rather than applying a point fix for that issue it seems better to introduce
a boolean to indicate whether the vif is connected or not such that repeated
calls to either xenvif_connect() or xenvif_disconnect() behave appropriately.

Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Cc: David Vrabel <david.vrabel@citrix.com>
Cc: Wei Liu <wei.liu2@citrix.com>
Cc: Ian Campbell <ian.campbell@citrix.com>
---
 drivers/net/xen-netback/common.h    |    1 +
 drivers/net/xen-netback/interface.c |   24 +++++++++++++-----------
 2 files changed, 14 insertions(+), 11 deletions(-)

diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
index 5715318..860d92c 100644
--- a/drivers/net/xen-netback/common.h
+++ b/drivers/net/xen-netback/common.h
@@ -169,6 +169,7 @@ struct xenvif {
 
 	/* Miscellaneous private stuff. */
 	struct net_device *dev;
+	bool connected;
 };
 
 static inline struct xenbus_device *xenvif_to_xenbus_device(struct xenvif *vif)
diff --git a/drivers/net/xen-netback/interface.c
b/drivers/net/xen-netback/interface.c
index 01bb854..94b47f5 100644
--- a/drivers/net/xen-netback/interface.c
+++ b/drivers/net/xen-netback/interface.c
@@ -366,7 +366,7 @@ int xenvif_connect(struct xenvif *vif, unsigned long
tx_ring_ref,
 	int err = -ENOMEM;
 
 	/* Already connected through? */
-	if (vif->tx_irq)
+	if (vif->connected)
 		return 0;
 
 	err = xenvif_map_frontend_rings(vif, tx_ring_ref, rx_ring_ref);
@@ -425,6 +425,7 @@ int xenvif_connect(struct xenvif *vif, unsigned long
tx_ring_ref,
 
 	wake_up_process(vif->task);
 
+	vif->connected = 1;
 	return 0;
 
 err_rx_unbind:
@@ -453,23 +454,24 @@ void xenvif_carrier_off(struct xenvif *vif)
 
 void xenvif_disconnect(struct xenvif *vif)
 {
+	if (!vif->connected)
+		return;
+
 	if (netif_carrier_ok(vif->dev))
 		xenvif_carrier_off(vif);
 
-	if (vif->tx_irq) {
-		if (vif->tx_irq == vif->rx_irq)
-			unbind_from_irqhandler(vif->tx_irq, vif);
-		else {
-			unbind_from_irqhandler(vif->tx_irq, vif);
-			unbind_from_irqhandler(vif->rx_irq, vif);
-		}
-		vif->tx_irq = 0;
+	if (vif->tx_irq == vif->rx_irq)
+		unbind_from_irqhandler(vif->tx_irq, vif);
+	else {
+		unbind_from_irqhandler(vif->tx_irq, vif);
+		unbind_from_irqhandler(vif->rx_irq, vif);
 	}
 
-	if (vif->task)
-		kthread_stop(vif->task);
+	kthread_stop(vif->task);
 
 	xenvif_unmap_frontend_rings(vif);
+
+	vif->connected = 0;
 }
 
 void xenvif_free(struct xenvif *vif)
-- 
1.7.10.4

From: Paul Durrant <paul.durrant@citrix.com>
Date: Fri, 20 Sep 2013 14:57:39 +0100
> @@ -169,6 +169,7 @@ struct xenvif {
>  
>  	/* Miscellaneous private stuff. */
>  	struct net_device *dev;
> +	bool connected;
>  };
>  
>  static inline struct xenbus_device *xenvif_to_xenbus_device(struct xenvif
*vif)
...
> +	vif->connected = 1;
 ...
> +	vif->connected = 0;
Please use ''true'' and ''false'' when assigning
values to something
of "bool" type.

Also these look like bug fixes, and are thus more appropriately
targetted at ''net'' and even potentially -stable as well.

On 2013-9-20 21:57, Paul Durrant wrote:
> Having applied my patch to separate vif disconnect and free, I ran into a
> BUG when testing resume from S3 with a Windows frontend because the vif
task
> pointer was not cleared by xenvif_disconnect() and so a double call to this
> function tries to stop the thread twice.
Or it is better to do more implements in windows netfront? For example, 
when the windows vm hibernates, disconnect the vif as required by 
netback: connect-> closing-> closed.

Thanks
Annie
> Rather than applying a point fix for that issue it seems better to
introduce
> a boolean to indicate whether the vif is connected or not such that
repeated
> calls to either xenvif_connect() or xenvif_disconnect() behave
appropriately.
>
> Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
> Cc: David Vrabel <david.vrabel@citrix.com>
> Cc: Wei Liu <wei.liu2@citrix.com>
> Cc: Ian Campbell <ian.campbell@citrix.com>
> ---
>   drivers/net/xen-netback/common.h    |    1 +
>   drivers/net/xen-netback/interface.c |   24 +++++++++++++-----------
>   2 files changed, 14 insertions(+), 11 deletions(-)
>
> diff --git a/drivers/net/xen-netback/common.h
b/drivers/net/xen-netback/common.h
> index 5715318..860d92c 100644
> --- a/drivers/net/xen-netback/common.h
> +++ b/drivers/net/xen-netback/common.h
> @@ -169,6 +169,7 @@ struct xenvif {
>   
>   	/* Miscellaneous private stuff. */
>   	struct net_device *dev;
> +	bool connected;
>   };
>   
>   static inline struct xenbus_device *xenvif_to_xenbus_device(struct xenvif
*vif)
> diff --git a/drivers/net/xen-netback/interface.c
b/drivers/net/xen-netback/interface.c
> index 01bb854..94b47f5 100644
> --- a/drivers/net/xen-netback/interface.c
> +++ b/drivers/net/xen-netback/interface.c
> @@ -366,7 +366,7 @@ int xenvif_connect(struct xenvif *vif, unsigned long
tx_ring_ref,
>   	int err = -ENOMEM;
>   
>   	/* Already connected through? */
> -	if (vif->tx_irq)
> +	if (vif->connected)
>   		return 0;
>   
>   	err = xenvif_map_frontend_rings(vif, tx_ring_ref, rx_ring_ref);
> @@ -425,6 +425,7 @@ int xenvif_connect(struct xenvif *vif, unsigned long
tx_ring_ref,
>   
>   	wake_up_process(vif->task);
>   
> +	vif->connected = 1;
>   	return 0;
>   
>   err_rx_unbind:
> @@ -453,23 +454,24 @@ void xenvif_carrier_off(struct xenvif *vif)
>   
>   void xenvif_disconnect(struct xenvif *vif)
>   {
> +	if (!vif->connected)
> +		return;
> +
>   	if (netif_carrier_ok(vif->dev))
>   		xenvif_carrier_off(vif);
>   
> -	if (vif->tx_irq) {
> -		if (vif->tx_irq == vif->rx_irq)
> -			unbind_from_irqhandler(vif->tx_irq, vif);
> -		else {
> -			unbind_from_irqhandler(vif->tx_irq, vif);
> -			unbind_from_irqhandler(vif->rx_irq, vif);
> -		}
> -		vif->tx_irq = 0;
> +	if (vif->tx_irq == vif->rx_irq)
> +		unbind_from_irqhandler(vif->tx_irq, vif);
> +	else {
> +		unbind_from_irqhandler(vif->tx_irq, vif);
> +		unbind_from_irqhandler(vif->rx_irq, vif);
>   	}
>   
> -	if (vif->task)
> -		kthread_stop(vif->task);
> +	kthread_stop(vif->task);
>   
>   	xenvif_unmap_frontend_rings(vif);
> +
> +	vif->connected = 0;
>   }
>   
>   void xenvif_free(struct xenvif *vif)

> -----Original Message-----
> From: annie li [mailto:annie.li@oracle.com]
> Sent: 22 September 2013 03:57
> To: Paul Durrant
> Cc: netdev@vger.kernel.org; xen-devel@lists.xen.org; Wei Liu; David Vrabel;
> Ian Campbell
> Subject: Re: [Xen-devel] [PATCH net-next v2 1/2] xen-netback: add a vif-is-
> connected flag
> 
> 
> On 2013-9-20 21:57, Paul Durrant wrote:
> > Having applied my patch to separate vif disconnect and free, I ran
into a
> > BUG when testing resume from S3 with a Windows frontend because the
> vif task
> > pointer was not cleared by xenvif_disconnect() and so a double call to
this
> > function tries to stop the thread twice.
> Or it is better to do more implements in windows netfront? For example,
> when the windows vm hibernates, disconnect the vif as required by
> netback: connect-> closing-> closed.
> 
S3 != hibernation; that is S4. The backend does not go away when the VM goes
into S3 as the domain remains intact. We do go through the correct
closing->closed transition on the way down but, because of the way the
D3->D0 code in the frontend needs to be generalized, we attempt a second
closing->closed transition on the way back up. In the S4 case this ok because
we have a fresh backend, but in the S3 case we don''t and therefore hit
the double-disconnect issue. The fact the backend BUGs in this case clearly
shows a vulnerability in the backend and thus that is where the fix needs to be
made; the frontend is doing nothing wrong.

  Paul