Merge branch 'stacked_vlan_tso'

Toshiaki Makita says: ==================== Stacked vlan TSO On the basis of Netdev 0.1 discussion[1], I made a patch set to enable TSO for packets with multiple vlans. Currently, packets with multiple vlans are always segmented by software, which is caused by that netif_skb_features() drops most feature flags for multiple tagged packets. To allow NICs to segment them, we need to get rid of that check from core. Fortunately, recently introduced ndo_features_check() can be used to move the check to each driver, and this patch set is based on the idea. For the initial patch set, I chose 3 drivers, bonding, team, and igb, as candidates to enable TSO. I tested them and confirmed they works fine with this change. Here are samples of performance test results. As I expected, %sys gets pretty lower than before. * TEST1: vlan (.1Q) on vlan (.1ad) on igb (I350) - before $ netperf -t TCP_STREAM -H 192.168.10.1 -l 60 Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 87380 16384 16384 60.02 933.72 Average: CPU %user %nice %system %iowait %steal %idle Average: all 0.13 0.00 11.28 0.01 0.00 88.58 - after $ netperf -t TCP_STREAM -H 192.168.10.1 -l 60 Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 87380 16384 16384 60.01 936.13 Average: CPU %user %nice %system %iowait %steal %idle Average: all 0.24 0.00 4.17 0.01 0.00 95.58 * TEST2: vlan (.1Q) on bridge (.1ad vlan filtering) on team on igb (I350) - before $ netperf -t TCP_STREAM -H 192.168.10.1 -l 60 Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 87380 16384 16384 60.01 936.28 Average: CPU %user %nice %system %iowait %steal %idle Average: all 0.41 0.00 11.57 0.01 0.00 88.01 - after $ netperf -t TCP_STREAM -H 192.168.10.1 -l 60 Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 87380 16384 16384 60.02 935.72 Average: CPU %user %nice %system %iowait %steal %idle Average: all 0.14 0.00 7.66 0.01 0.00 92.19 In addition to above, I tested these configurations: - vlan (.1Q) on vlan (1.ad) on bonding on igb (I350) - vlan (.1Q) on vlan (1.Q) on igb (I350) - vlan (.1Q) on vlan (1.Q) on team on igb (I350) And didn't find any problem. [1] https://netdev01.org/sessions/18 https://netdev01.org/docs/netdev01_bof_8021ad_makita_150212.pdf ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
author: David S. Miller <davem@davemloft.net> 2015-03-29 16:33:32 -0400
committer: David S. Miller <davem@davemloft.net> 2015-03-29 16:33:32 -0400
commit: afb0bc972b5269a2046a1194c475d929c62fc1a3 (patch)
tree: 1c51d0aad72f03283affc7cc91e8106f3e67d672
parent: 608cd71a9c7c9db76e78a792c5a4101e12fea43f (diff)
parent: 1abbc98a8ac5c6cc016e672e5888b6416ae243f6 (diff)
10 files changed, 96 insertions, 22 deletions
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index c026ce9cd7b6..7b4684ccdb3f 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -4038,6 +4038,7 @@ static const struct net_device_ops bond_netdev_ops = {
        .ndo_fix_features       = bond_fix_features,
        .ndo_bridge_setlink     = ndo_dflt_netdev_switch_port_bridge_setlink,
        .ndo_bridge_dellink     = ndo_dflt_netdev_switch_port_bridge_dellink,
+        .ndo_features_check     = passthru_features_check,
 };
 static const struct device_type bond_type = {
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 9677431c582a..039b0c1f480e 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -12557,6 +12557,7 @@ static netdev_features_t bnx2x_features_check(struct sk_buff *skb,
                                              struct net_device *dev,
                                              netdev_features_t features)
 {
+        features = vlan_features_check(skb, features);
        return vxlan_features_check(skb, features);
 }
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 0e07545ccc97..8457d0306e3a 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -2093,6 +2093,7 @@ static const struct net_device_ops igb_netdev_ops = {
 #endif
        .ndo_fix_features       = igb_fix_features,
        .ndo_set_features       = igb_set_features,
+        .ndo_features_check     = passthru_features_check,
 };
 /**
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index a8339e98ad24..ebc93a101c93 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -2373,6 +2373,7 @@ static netdev_features_t mlx4_en_features_check(struct sk_buff *skb,
                                                struct net_device *dev,
                                                netdev_features_t features)
 {
+        features = vlan_features_check(skb, features);
        return vxlan_features_check(skb, features);
 }
 #endif
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
index a430a34a4434..367f3976df56 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
@@ -507,6 +507,7 @@ static netdev_features_t qlcnic_features_check(struct sk_buff *skb,
                                               struct net_device *dev,
                                               netdev_features_t features)
 {
+        features = vlan_features_check(skb, features);
        return vxlan_features_check(skb, features);
 }
 #endif
diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index a23319fc78ca..6928448f6b7f 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -1979,6 +1979,7 @@ static const struct net_device_ops team_netdev_ops = {
        .ndo_change_carrier     = team_change_carrier,
        .ndo_bridge_setlink     = ndo_dflt_netdev_switch_port_bridge_setlink,
        .ndo_bridge_dellink     = ndo_dflt_netdev_switch_port_bridge_dellink,
+        .ndo_features_check     = passthru_features_check,
 };
 /***********************
diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index b11b28a30b9e..920e4457ce6e 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -561,4 +561,71 @@ static inline void vlan_set_encap_proto(struct sk_buff *skb,
                skb->protocol = htons(ETH_P_802_2);
 }
+/**
+ * skb_vlan_tagged - check if skb is vlan tagged.
+ * @skb: skbuff to query
+ *
+ * Returns true if the skb is tagged, regardless of whether it is hardware
+ * accelerated or not.
+ */
+static inline bool skb_vlan_tagged(const struct sk_buff *skb)
+{
+        if (!skb_vlan_tag_present(skb) &&
+            likely(skb->protocol != htons(ETH_P_8021Q) &&
+                   skb->protocol != htons(ETH_P_8021AD)))
+                return false;
+        return true;
+}
+/**
+ * skb_vlan_tagged_multi - check if skb is vlan tagged with multiple headers.
+ * @skb: skbuff to query
+ *
+ * Returns true if the skb is tagged with multiple vlan headers, regardless
+ * of whether it is hardware accelerated or not.
+ */
+static inline bool skb_vlan_tagged_multi(const struct sk_buff *skb)
+{
+        __be16 protocol = skb->protocol;
+        if (!skb_vlan_tag_present(skb)) {
+                struct vlan_ethhdr *veh;
+                if (likely(protocol != htons(ETH_P_8021Q) &&
+                           protocol != htons(ETH_P_8021AD)))
+                        return false;
+                veh = (struct vlan_ethhdr *)skb->data;
+                protocol = veh->h_vlan_encapsulated_proto;
+        }
+        if (protocol != htons(ETH_P_8021Q) && protocol != htons(ETH_P_8021AD))
+                return false;
+        return true;
+}
+/**
+ * vlan_features_check - drop unsafe features for skb with multiple tags.
+ * @skb: skbuff to query
+ * @features: features to be checked
+ *
+ * Returns features without unsafe ones if the skb has multiple tags.
+ */
+static inline netdev_features_t vlan_features_check(const struct sk_buff *skb,
+                                                    netdev_features_t features)
+{
+        if (skb_vlan_tagged_multi(skb))
+                features = netdev_intersect_features(features,
+                                                     NETIF_F_SG |
+                                                     NETIF_F_HIGHDMA |
+                                                     NETIF_F_FRAGLIST |
+                                                     NETIF_F_GEN_CSUM |
+                                                     NETIF_F_HW_VLAN_CTAG_TX |
+                                                     NETIF_F_HW_VLAN_STAG_TX);
+        return features;
+}
 #endif /* !(_LINUX_IF_VLAN_H_) */
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 08c4ab37189f..967bb4c8caf1 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -3657,6 +3657,9 @@ void netdev_change_features(struct net_device *dev);
 void netif_stacked_transfer_operstate(const struct net_device *rootdev,
                                        struct net_device *dev);
+netdev_features_t passthru_features_check(struct sk_buff *skb,
+                                          struct net_device *dev,
+                                          netdev_features_t features);
 netdev_features_t netif_skb_features(struct sk_buff *skb);
 static inline bool net_gso_ok(netdev_features_t features, int gso_type)
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index f196552ec3c4..8b5ab9033b41 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -554,6 +554,7 @@ static int vlan_dev_init(struct net_device *dev)
        if (dev->features & NETIF_F_VLAN_FEATURES)
                netdev_warn(real_dev, "VLAN features are set incorrectly.  Q-in-Q configurations may not work correctly.\n");
+        dev->vlan_features = real_dev->vlan_features & ~NETIF_F_ALL_FCOE;
        /* ipv6 shared card related stuff */
        dev->dev_id = real_dev->dev_id;
diff --git a/net/core/dev.c b/net/core/dev.c
index a0408d497dae..3a06003ecafd 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2562,12 +2562,26 @@ static netdev_features_t harmonize_features(struct sk_buff *skb,
        return features;
 }
+netdev_features_t passthru_features_check(struct sk_buff *skb,
+                                          struct net_device *dev,
+                                          netdev_features_t features)
+{
+        return features;
+}
+EXPORT_SYMBOL(passthru_features_check);
+static netdev_features_t dflt_features_check(const struct sk_buff *skb,
+                                             struct net_device *dev,
+                                             netdev_features_t features)
+{
+        return vlan_features_check(skb, features);
+}
 netdev_features_t netif_skb_features(struct sk_buff *skb)
 {
        struct net_device *dev = skb->dev;
        netdev_features_t features = dev->features;
        u16 gso_segs = skb_shinfo(skb)->gso_segs;
-        __be16 protocol = skb->protocol;
        if (gso_segs > dev->gso_max_segs || gso_segs < dev->gso_min_segs)
                features &= ~NETIF_F_GSO_MASK;
@@ -2579,34 +2593,17 @@ netdev_features_t netif_skb_features(struct sk_buff *skb)
        if (skb->encapsulation)
                features &= dev->hw_enc_features;
-        if (!skb_vlan_tag_present(skb)) {
+        if (skb_vlan_tagged(skb))
-                if (unlikely(protocol == htons(ETH_P_8021Q) ||
-                             protocol == htons(ETH_P_8021AD))) {
-                        struct vlan_ethhdr *veh = (struct vlan_ethhdr *)skb->data;
-                        protocol = veh->h_vlan_encapsulated_proto;
-                } else {
-                        goto finalize;
-                }
-        }
-        features = netdev_intersect_features(features,
-                                             dev->vlan_features |
-                                             NETIF_F_HW_VLAN_CTAG_TX |
-                                             NETIF_F_HW_VLAN_STAG_TX);
-        if (protocol == htons(ETH_P_8021Q) || protocol == htons(ETH_P_8021AD))
                features = netdev_intersect_features(features,
-                                                     NETIF_F_SG |
+                                                     dev->vlan_features |
-                                                     NETIF_F_HIGHDMA |
-                                                     NETIF_F_FRAGLIST |
-                                                     NETIF_F_GEN_CSUM |
                                                     NETIF_F_HW_VLAN_CTAG_TX |
                                                     NETIF_F_HW_VLAN_STAG_TX);
-finalize:
        if (dev->netdev_ops->ndo_features_check)
                features &= dev->netdev_ops->ndo_features_check(skb, dev,
                                                                features);
+        else
+                features &= dflt_features_check(skb, dev, features);
        return harmonize_features(skb, features);
 }
author	David S. Miller <davem@davemloft.net>	2015-03-29 16:33:32 -0400
committer	David S. Miller <davem@davemloft.net>	2015-03-29 16:33:32 -0400
commit	afb0bc972b5269a2046a1194c475d929c62fc1a3 (patch)
tree	1c51d0aad72f03283affc7cc91e8106f3e67d672
parent	608cd71a9c7c9db76e78a792c5a4101e12fea43f (diff)
parent	1abbc98a8ac5c6cc016e672e5888b6416ae243f6 (diff)