diff mbox

[v5,net-next] MPLS: Use mpls_features to activate software MPLS GSO segmentation

Message ID 1401839597-8961-1-git-send-email-horms@verge.net.au
State Accepted, archived
Delegated to: David Miller
Headers show

Commit Message

Simon Horman June 3, 2014, 11:53 p.m. UTC 
If an MPLS packet requires segmentation then use mpls_features
to determine if the software implementation should be used.

As no driver advertises MPLS GSO segmentation this will always be
the case.

I had not noticed that this was necessary before as software MPLS GSO
segmentation was already being used in my test environment. I believe that
the reason for that is the skbs in question always had fragments and the
driver I used does not advertise NETIF_F_FRAGLIST (which seems to be the
case for most drivers). Thus software segmentation was activated by
skb_gso_ok().

This introduces the overhead of an extra call to skb_network_protocol()
in the case where where CONFIG_NET_MPLS_GSO is set and
skb->ip_summed == CHECKSUM_NONE.

Thanks to Jesse Gross for prompting me to investigate this.

Signed-off-by: Simon Horman <horms@verge.net.au>
Acked-by: YAMAMOTO Takashi <yamamoto@valinux.co.jp>

---
v5
* As suggested by Yamamoto-san
  - Remove unused tmp variable
  - Do not use unlikely() for MPLS ethertype check
* Added Ack from Yamamoto-san

v4.2
* As suggested by Eric Dumazet
  - Use htons() instead of cpu_to_be16()
  - Refactor code to pass type to net_mpls_features.

v4.1
* Following fedback from Thomas Graff and Jesse Gross
  - Use ethertype of packet to detect MPLS rather than
    relying on mac_len indicating a gap between the end of L2
    and the beginning of L3. That assumption seems to
    be broken by the GRE GSO code.
  - Move mpls_features handling into harmonize_features()
    This allows an existing call in there to skb_network_protocol()
    to be leveraged.
  - Removed acks as the patch has now changed in a material way

v4
* As suggested by YAMAMOTO Takashi
  - Correct typos in comment
* Added Ack from YAMAMOTO Takashi
v3
* As requested by David Miller
  - Do not mark net_mpls_features as inline
  - Correct alignment of parameters

v2
* Added Ack from Jesse Gross
* Removed duplicate 'Thus' from changelog
---
 net/core/dev.c | 28 +++++++++++++++++++++++++++-
 1 file changed, 27 insertions(+), 1 deletion(-)

Comments

Thomas Graf June 4, 2014, 10:13 a.m. UTC | #1 
On 06/04/14 at 08:53am, Simon Horman wrote:
> If an MPLS packet requires segmentation then use mpls_features
> to determine if the software implementation should be used.
> 
> As no driver advertises MPLS GSO segmentation this will always be
> the case.
> 
> I had not noticed that this was necessary before as software MPLS GSO
> segmentation was already being used in my test environment. I believe that
> the reason for that is the skbs in question always had fragments and the
> driver I used does not advertise NETIF_F_FRAGLIST (which seems to be the
> case for most drivers). Thus software segmentation was activated by
> skb_gso_ok().
> 
> This introduces the overhead of an extra call to skb_network_protocol()
> in the case where where CONFIG_NET_MPLS_GSO is set and
> skb->ip_summed == CHECKSUM_NONE.
> 
> Thanks to Jesse Gross for prompting me to investigate this.
> 
> Signed-off-by: Simon Horman <horms@verge.net.au>
> Acked-by: YAMAMOTO Takashi <yamamoto@valinux.co.jp>

LGTM

Acked-by: Thomas Graf <tgraf@suug.ch>
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
David Miller June 5, 2014, 10:05 p.m. UTC | #2 
From: Simon Horman <horms@verge.net.au>
Date: Wed,  4 Jun 2014 08:53:17 +0900

> If an MPLS packet requires segmentation then use mpls_features
> to determine if the software implementation should be used.
> 
> As no driver advertises MPLS GSO segmentation this will always be
> the case.
> 
> I had not noticed that this was necessary before as software MPLS GSO
> segmentation was already being used in my test environment. I believe that
> the reason for that is the skbs in question always had fragments and the
> driver I used does not advertise NETIF_F_FRAGLIST (which seems to be the
> case for most drivers). Thus software segmentation was activated by
> skb_gso_ok().
> 
> This introduces the overhead of an extra call to skb_network_protocol()
> in the case where where CONFIG_NET_MPLS_GSO is set and
> skb->ip_summed == CHECKSUM_NONE.
> 
> Thanks to Jesse Gross for prompting me to investigate this.
> 
> Signed-off-by: Simon Horman <horms@verge.net.au>
> Acked-by: YAMAMOTO Takashi <yamamoto@valinux.co.jp>

Applied to net-next, thanks Simon.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Simon Horman June 6, 2014, 7:16 a.m. UTC | #3 
On Thu, Jun 05, 2014 at 03:05:42PM -0700, David Miller wrote:
> From: Simon Horman <horms@verge.net.au>
> Date: Wed,  4 Jun 2014 08:53:17 +0900
> 
> > If an MPLS packet requires segmentation then use mpls_features
> > to determine if the software implementation should be used.
> > 
> > As no driver advertises MPLS GSO segmentation this will always be
> > the case.
> > 
> > I had not noticed that this was necessary before as software MPLS GSO
> > segmentation was already being used in my test environment. I believe that
> > the reason for that is the skbs in question always had fragments and the
> > driver I used does not advertise NETIF_F_FRAGLIST (which seems to be the
> > case for most drivers). Thus software segmentation was activated by
> > skb_gso_ok().
> > 
> > This introduces the overhead of an extra call to skb_network_protocol()
> > in the case where where CONFIG_NET_MPLS_GSO is set and
> > skb->ip_summed == CHECKSUM_NONE.
> > 
> > Thanks to Jesse Gross for prompting me to investigate this.
> > 
> > Signed-off-by: Simon Horman <horms@verge.net.au>
> > Acked-by: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
> 
> Applied to net-next, thanks Simon.

Thanks!
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/net/core/dev.c b/net/core/dev.c
index 0355ca5..f726bfb 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2498,13 +2498,39 @@  static int dev_gso_segment(struct sk_buff *skb, netdev_features_t features)
 	return 0;
 }
 
+/* If MPLS offload request, verify we are testing hardware MPLS features
+ * instead of standard features for the netdev.
+ */
+#ifdef CONFIG_NET_MPLS_GSO
+static netdev_features_t net_mpls_features(struct sk_buff *skb,
+					   netdev_features_t features,
+					   __be16 type)
+{
+	if (type == htons(ETH_P_MPLS_UC) || type == htons(ETH_P_MPLS_MC))
+		features &= skb->dev->mpls_features;
+
+	return features;
+}
+#else
+static netdev_features_t net_mpls_features(struct sk_buff *skb,
+					   netdev_features_t features,
+					   __be16 type)
+{
+	return features;
+}
+#endif
+
 static netdev_features_t harmonize_features(struct sk_buff *skb,
 	netdev_features_t features)
 {
 	int tmp;
+	__be16 type;
+
+	type = skb_network_protocol(skb, &tmp);
+	features = net_mpls_features(skb, features, type);
 
 	if (skb->ip_summed != CHECKSUM_NONE &&
-	    !can_checksum_protocol(features, skb_network_protocol(skb, &tmp))) {
+	    !can_checksum_protocol(features, type)) {
 		features &= ~NETIF_F_ALL_CSUM;
 	} else if (illegal_highdma(skb->dev, skb)) {
 		features &= ~NETIF_F_SG;