Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1358246 > unrolled thread

[RFC v2 -next 2/2] virtio_net: Read the advised MTU

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  [RFC v2 -next 2/2] virtio_net: Read the advised MTU Aaron Conole <aconole@redhat.com> - 2016-03-15 22:10 +0100
    Re: [RFC v2 -next 2/2] virtio_net: Read the advised MTU Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> - 2016-03-16 15:20 +0100
      Re: [RFC v2 -next 2/2] virtio_net: Read the advised MTU Aaron Conole <aconole@redhat.com> - 2016-03-17 22:20 +0100
    Re: [RFC v2 -next 2/2] virtio_net: Read the advised MTU "Michael S. Tsirkin" <mst@redhat.com> - 2016-03-16 15:30 +0100
      Re: [RFC v2 -next 2/2] virtio_net: Read the advised MTU Aaron Conole <aconole@redhat.com> - 2016-03-17 22:30 +0100

#1358246 — [RFC v2 -next 2/2] virtio_net: Read the advised MTU

This patch checks the feature bit for the VIRTIO_NET_F_MTU feature. If it
exists, read the advised MTU and use it.

No proper error handling is provided for the case where a user changes the
negotiated MTU. A future commit will add proper error handling. Instead, a
warning is emitted if the guest changes the device MTU after previously
being given advice.

Signed-off-by: Aaron Conole <aconole@bytheb.org>
---
v2:
* Whitespace cleanup in the last hunk
* Code style change around the pr_warn
* Additional test for mtu change before printing warning

 drivers/net/virtio_net.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 767ab11..429fe01 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -146,6 +146,7 @@ struct virtnet_info {
 	virtio_net_ctrl_ack ctrl_status;
 	u8 ctrl_promisc;
 	u8 ctrl_allmulti;
+	bool negotiated_mtu;
 };
 
 struct padded_vnet_hdr {
@@ -1390,8 +1391,11 @@ static const struct ethtool_ops virtnet_ethtool_ops = {
 
 static int virtnet_change_mtu(struct net_device *dev, int new_mtu)
 {
+	struct virtnet_info *vi = netdev_priv(dev);
 	if (new_mtu < MIN_MTU || new_mtu > MAX_MTU)
 		return -EINVAL;
+	if ((vi->negotiated_mtu) && (dev->mtu != new_mtu))
+		pr_warn("changing mtu while the advised mtu bit exists.");
 	dev->mtu = new_mtu;
 	return 0;
 }
@@ -1836,6 +1840,13 @@ static int virtnet_probe(struct virtio_device *vdev)
 	if (virtio_has_feature(vdev, VIRTIO_NET_F_CTRL_VQ))
 		vi->has_cvq = true;
 
+	if (virtio_has_feature(vdev, VIRTIO_NET_F_MTU)) {
+		vi->negotiated_mtu = true;
+		dev->mtu = virtio_cread16(vdev,
+					  offsetof(struct virtio_net_config,
+						   mtu));
+	}
+
 	if (vi->any_header_sg)
 		dev->needed_headroom = vi->hdr_len;
 
@@ -2019,6 +2030,7 @@ static unsigned int features[] = {
 	VIRTIO_NET_F_GUEST_ANNOUNCE, VIRTIO_NET_F_MQ,
 	VIRTIO_NET_F_CTRL_MAC_ADDR,
 	VIRTIO_F_ANY_LAYOUT,
+	VIRTIO_NET_F_MTU,
 };
 
 static struct virtio_driver virtio_net_driver = {
-- 
2.5.0

[toc] | [next] | [standalone]

#1359033

Hello.

On 3/16/2016 12:04 AM, Aaron Conole wrote:

> This patch checks the feature bit for the VIRTIO_NET_F_MTU feature. If it
> exists, read the advised MTU and use it.
>
> No proper error handling is provided for the case where a user changes the
> negotiated MTU. A future commit will add proper error handling. Instead, a
> warning is emitted if the guest changes the device MTU after previously
> being given advice.
>
> Signed-off-by: Aaron Conole <aconole@bytheb.org>
> ---
> v2:
> * Whitespace cleanup in the last hunk
> * Code style change around the pr_warn
> * Additional test for mtu change before printing warning
>
>   drivers/net/virtio_net.c | 12 ++++++++++++
>   1 file changed, 12 insertions(+)
>
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index 767ab11..429fe01 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
[...]
> @@ -1390,8 +1391,11 @@ static const struct ethtool_ops virtnet_ethtool_ops = {
>
>   static int virtnet_change_mtu(struct net_device *dev, int new_mtu)
>   {
> +	struct virtnet_info *vi = netdev_priv(dev);
>   	if (new_mtu < MIN_MTU || new_mtu > MAX_MTU)
>   		return -EINVAL;
> +	if ((vi->negotiated_mtu) && (dev->mtu != new_mtu))

    Inner parens not needed, please be consistent with the code above.

[...]

MBR, Sergei

[toc] | [prev] | [next] | [standalone]

#1360204

Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> writes:

> Hello.
>
> On 3/16/2016 12:04 AM, Aaron Conole wrote:
>
>> This patch checks the feature bit for the VIRTIO_NET_F_MTU feature. If it
>> exists, read the advised MTU and use it.
>>
>> No proper error handling is provided for the case where a user changes the
>> negotiated MTU. A future commit will add proper error handling. Instead, a
>> warning is emitted if the guest changes the device MTU after previously
>> being given advice.
>>
>> Signed-off-by: Aaron Conole <aconole@bytheb.org>
>> ---
>> v2:
>> * Whitespace cleanup in the last hunk
>> * Code style change around the pr_warn
>> * Additional test for mtu change before printing warning
>>
>>   drivers/net/virtio_net.c | 12 ++++++++++++
>>   1 file changed, 12 insertions(+)
>>
>> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
>> index 767ab11..429fe01 100644
>> --- a/drivers/net/virtio_net.c
>> +++ b/drivers/net/virtio_net.c
> [...]
>> @@ -1390,8 +1391,11 @@ static const struct ethtool_ops virtnet_ethtool_ops = {
>>
>>   static int virtnet_change_mtu(struct net_device *dev, int new_mtu)
>>   {
>> +	struct virtnet_info *vi = netdev_priv(dev);
>>   	if (new_mtu < MIN_MTU || new_mtu > MAX_MTU)
>>   		return -EINVAL;
>> +	if ((vi->negotiated_mtu) && (dev->mtu != new_mtu))
>
>    Inner parens not needed, please be consistent with the code above.

Okay, I will do that.

Thanks so much for the review (again), Sergei!

-Aaron

> [...]
>
> MBR, Sergei

[toc] | [prev] | [next] | [standalone]

#1359041

On Tue, Mar 15, 2016 at 05:04:13PM -0400, Aaron Conole wrote:
> This patch checks the feature bit for the VIRTIO_NET_F_MTU feature. If it
> exists, read the advised MTU and use it.
> 
> No proper error handling is provided for the case where a user changes the
> negotiated MTU. A future commit will add proper error handling. Instead, a
> warning is emitted if the guest changes the device MTU after previously
> being given advice.

I don't see this as an error. Device might at best give a hint,
user/network admin always knows best.

> 
> Signed-off-by: Aaron Conole <aconole@bytheb.org>
> ---
> v2:
> * Whitespace cleanup in the last hunk
> * Code style change around the pr_warn
> * Additional test for mtu change before printing warning
> 
>  drivers/net/virtio_net.c | 12 ++++++++++++
>  1 file changed, 12 insertions(+)
> 
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index 767ab11..429fe01 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -146,6 +146,7 @@ struct virtnet_info {
>  	virtio_net_ctrl_ack ctrl_status;
>  	u8 ctrl_promisc;
>  	u8 ctrl_allmulti;
> +	bool negotiated_mtu;
>  };
>  
>  struct padded_vnet_hdr {
> @@ -1390,8 +1391,11 @@ static const struct ethtool_ops virtnet_ethtool_ops = {
>  
>  static int virtnet_change_mtu(struct net_device *dev, int new_mtu)
>  {
> +	struct virtnet_info *vi = netdev_priv(dev);
>  	if (new_mtu < MIN_MTU || new_mtu > MAX_MTU)
>  		return -EINVAL;
> +	if ((vi->negotiated_mtu) && (dev->mtu != new_mtu))
> +		pr_warn("changing mtu while the advised mtu bit exists.");

I don't really see why are we warning here. Just drop this chunk,
as well as the flag in struct virtnet_info.

>  	dev->mtu = new_mtu;
>  	return 0;
>  }
> @@ -1836,6 +1840,13 @@ static int virtnet_probe(struct virtio_device *vdev)
>  	if (virtio_has_feature(vdev, VIRTIO_NET_F_CTRL_VQ))
>  		vi->has_cvq = true;
>  
> +	if (virtio_has_feature(vdev, VIRTIO_NET_F_MTU)) {
> +		vi->negotiated_mtu = true;
> +		dev->mtu = virtio_cread16(vdev,
> +					  offsetof(struct virtio_net_config,
> +						   mtu));
> +	}
> +
>  	if (vi->any_header_sg)
>  		dev->needed_headroom = vi->hdr_len;
>  
> @@ -2019,6 +2030,7 @@ static unsigned int features[] = {
>  	VIRTIO_NET_F_GUEST_ANNOUNCE, VIRTIO_NET_F_MQ,
>  	VIRTIO_NET_F_CTRL_MAC_ADDR,
>  	VIRTIO_F_ANY_LAYOUT,
> +	VIRTIO_NET_F_MTU,
>  };
>  
>  static struct virtio_driver virtio_net_driver = {
> -- 
> 2.5.0

[toc] | [prev] | [next] | [standalone]

#1360207

"Michael S. Tsirkin" <mst@redhat.com> writes:

> On Tue, Mar 15, 2016 at 05:04:13PM -0400, Aaron Conole wrote:
>> This patch checks the feature bit for the VIRTIO_NET_F_MTU feature. If it
>> exists, read the advised MTU and use it.
>> 
>> No proper error handling is provided for the case where a user changes the
>> negotiated MTU. A future commit will add proper error handling. Instead, a
>> warning is emitted if the guest changes the device MTU after previously
>> being given advice.
>
> I don't see this as an error. Device might at best give a hint,
> user/network admin always knows best.
>
>> 
>> Signed-off-by: Aaron Conole <aconole@bytheb.org>
>> ---
>> v2:
>> * Whitespace cleanup in the last hunk
>> * Code style change around the pr_warn
>> * Additional test for mtu change before printing warning
>> 
>>  drivers/net/virtio_net.c | 12 ++++++++++++
>>  1 file changed, 12 insertions(+)
>> 
>> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
>> index 767ab11..429fe01 100644
>> --- a/drivers/net/virtio_net.c
>> +++ b/drivers/net/virtio_net.c
>> @@ -146,6 +146,7 @@ struct virtnet_info {
>>  	virtio_net_ctrl_ack ctrl_status;
>>  	u8 ctrl_promisc;
>>  	u8 ctrl_allmulti;
>> +	bool negotiated_mtu;
>>  };
>>  
>>  struct padded_vnet_hdr {
>> @@ -1390,8 +1391,11 @@ static const struct ethtool_ops virtnet_ethtool_ops = {
>>  
>>  static int virtnet_change_mtu(struct net_device *dev, int new_mtu)
>>  {
>> +	struct virtnet_info *vi = netdev_priv(dev);
>>  	if (new_mtu < MIN_MTU || new_mtu > MAX_MTU)
>>  		return -EINVAL;
>> +	if ((vi->negotiated_mtu) && (dev->mtu != new_mtu))
>> +		pr_warn("changing mtu while the advised mtu bit exists.");
>
> I don't really see why are we warning here. Just drop this chunk,
> as well as the flag in struct virtnet_info.

Okay. I was warning because the user is changing this after telling to
use something different - but if you don't think it's an error, I will
drop it.

Thanks so much for the review, Michael!

-Aaron

>>  	dev->mtu = new_mtu;
>>  	return 0;
>>  }
>> @@ -1836,6 +1840,13 @@ static int virtnet_probe(struct virtio_device *vdev)
>>  	if (virtio_has_feature(vdev, VIRTIO_NET_F_CTRL_VQ))
>>  		vi->has_cvq = true;
>>  
>> +	if (virtio_has_feature(vdev, VIRTIO_NET_F_MTU)) {
>> +		vi->negotiated_mtu = true;
>> +		dev->mtu = virtio_cread16(vdev,
>> +					  offsetof(struct virtio_net_config,
>> +						   mtu));
>> +	}
>> +
>>  	if (vi->any_header_sg)
>>  		dev->needed_headroom = vi->hdr_len;
>>  
>> @@ -2019,6 +2030,7 @@ static unsigned int features[] = {
>>  	VIRTIO_NET_F_GUEST_ANNOUNCE, VIRTIO_NET_F_MQ,
>>  	VIRTIO_NET_F_CTRL_MAC_ADDR,
>>  	VIRTIO_F_ANY_LAYOUT,
>> +	VIRTIO_NET_F_MTU,
>>  };
>>  
>>  static struct virtio_driver virtio_net_driver = {
>> -- 
>> 2.5.0

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web