[net-next,1/2] virtio_net: Introduce VIRTIO_NET_F_MASTER feature bit

This feature bit can be used by hypervisor to indicate virtio_net device to
act as a master for a directly attached passthru device with the same MAC
address.

Signed-off-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
---
 drivers/net/virtio_net.c        | 3 ++-
 include/uapi/linux/virtio_net.h | 1 +
 2 files changed, 3 insertions(+), 1 deletion(-)

From: Sridhar Samudrala <sridhar.samudrala@intel.com>
Date: Tue,  2 Jan 2018 16:35:37 -0800

> diff --git a/include/uapi/linux/virtio_net.h b/include/uapi/linux/virtio_net.h
> index fc353b518288..a9b4e0836786 100644
> --- a/include/uapi/linux/virtio_net.h
> +++ b/include/uapi/linux/virtio_net.h
> @@ -56,6 +56,7 @@
>  #define VIRTIO_NET_F_MQ	22	/* Device supports Receive Flow
>  					 * Steering */
>  #define VIRTIO_NET_F_CTRL_MAC_ADDR 23	/* Set MAC address */
> +#define VIRTIO_NET_F_MASTER	62	/* act as master for a VF device */

Please respin this series with something in this commit message which
explains why we are using "62" rather than something like "24" here.

Thank you.

On Tue, Jan 02, 2018 at 04:35:37PM -0800, Sridhar Samudrala wrote:
> This feature bit can be used by hypervisor to indicate virtio_net device to
> act as a master for a directly attached passthru device with the same MAC
> address.
> 
> Signed-off-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
> ---
>  drivers/net/virtio_net.c        | 3 ++-
>  include/uapi/linux/virtio_net.h | 1 +
>  2 files changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index 6fb7b658a6cc..46844a1d9a62 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -2796,7 +2796,8 @@ static struct virtio_device_id id_table[] = {
>  	VIRTIO_NET_F_CTRL_RX, VIRTIO_NET_F_CTRL_VLAN, \
>  	VIRTIO_NET_F_GUEST_ANNOUNCE, VIRTIO_NET_F_MQ, \
>  	VIRTIO_NET_F_CTRL_MAC_ADDR, \
> -	VIRTIO_NET_F_MTU, VIRTIO_NET_F_CTRL_GUEST_OFFLOADS
> +	VIRTIO_NET_F_MTU, VIRTIO_NET_F_CTRL_GUEST_OFFLOADS, \
> +	VIRTIO_NET_F_MASTER
>  
>  static unsigned int features[] = {
>  	VIRTNET_FEATURES,
> diff --git a/include/uapi/linux/virtio_net.h b/include/uapi/linux/virtio_net.h
> index fc353b518288..a9b4e0836786 100644
> --- a/include/uapi/linux/virtio_net.h
> +++ b/include/uapi/linux/virtio_net.h
> @@ -56,6 +56,7 @@
>  #define VIRTIO_NET_F_MQ	22	/* Device supports Receive Flow
>  					 * Steering */
>  #define VIRTIO_NET_F_CTRL_MAC_ADDR 23	/* Set MAC address */
> +#define VIRTIO_NET_F_MASTER	62	/* act as master for a VF device */

Well the virtio hardware doesn't really know whether it's a master or
the slave of a bond. In fact quite possibly down the road we'll add a
virtual device on top as others seem to want to do - that other one will
be the master then.

And we do nothing do check it's a VF.

So maybe a better name would be

#define VIRTIO_NET_F_BACKUP	62	/* Act as backup for another device with the same MAC */

>  
>  #ifndef VIRTIO_NET_NO_LEGACY
>  #define VIRTIO_NET_F_GSO	6	/* Host handles pkts w/ any GSO type */
> -- 
> 2.14.3

On 1/3/2018 9:43 AM, Michael S. Tsirkin wrote:
> On Tue, Jan 02, 2018 at 04:35:37PM -0800, Sridhar Samudrala wrote:
>> This feature bit can be used by hypervisor to indicate virtio_net device to
>> act as a master for a directly attached passthru device with the same MAC
>> address.
>>
>> Signed-off-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
>> ---
>>   drivers/net/virtio_net.c        | 3 ++-
>>   include/uapi/linux/virtio_net.h | 1 +
>>   2 files changed, 3 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
>> index 6fb7b658a6cc..46844a1d9a62 100644
>> --- a/drivers/net/virtio_net.c
>> +++ b/drivers/net/virtio_net.c
>> @@ -2796,7 +2796,8 @@ static struct virtio_device_id id_table[] = {
>>   	VIRTIO_NET_F_CTRL_RX, VIRTIO_NET_F_CTRL_VLAN, \
>>   	VIRTIO_NET_F_GUEST_ANNOUNCE, VIRTIO_NET_F_MQ, \
>>   	VIRTIO_NET_F_CTRL_MAC_ADDR, \
>> -	VIRTIO_NET_F_MTU, VIRTIO_NET_F_CTRL_GUEST_OFFLOADS
>> +	VIRTIO_NET_F_MTU, VIRTIO_NET_F_CTRL_GUEST_OFFLOADS, \
>> +	VIRTIO_NET_F_MASTER
>>   
>>   static unsigned int features[] = {
>>   	VIRTNET_FEATURES,
>> diff --git a/include/uapi/linux/virtio_net.h b/include/uapi/linux/virtio_net.h
>> index fc353b518288..a9b4e0836786 100644
>> --- a/include/uapi/linux/virtio_net.h
>> +++ b/include/uapi/linux/virtio_net.h
>> @@ -56,6 +56,7 @@
>>   #define VIRTIO_NET_F_MQ	22	/* Device supports Receive Flow
>>   					 * Steering */
>>   #define VIRTIO_NET_F_CTRL_MAC_ADDR 23	/* Set MAC address */
>> +#define VIRTIO_NET_F_MASTER	62	/* act as master for a VF device */
> Well the virtio hardware doesn't really know whether it's a master or
> the slave of a bond. In fact quite possibly down the road we'll add a
> virtual device on top as others seem to want to do - that other one will
> be the master then.
>
> And we do nothing do check it's a VF.
>
> So maybe a better name would be
>
> #define VIRTIO_NET_F_BACKUP	62	/* Act as backup for another device with the same MAC */
Yes. BACKUP  does make more sense as virtio is only used as a BACKUP 
datapath when the other
device is unplugged.


>
>>   
>>   #ifndef VIRTIO_NET_NO_LEGACY
>>   #define VIRTIO_NET_F_GSO	6	/* Host handles pkts w/ any GSO type */
>> -- 
>> 2.14.3