From 1241365f1aeb24ef0ffe82970f7c558022ddc85f Mon Sep 17 00:00:00 2001 From: James Morse Date: Mon, 19 Oct 2015 16:31:55 +0100 Subject: openvswitch: Allocate memory for ovs internal device stats. "openvswitch: Remove vport stats" removed the per-vport statistics, in order to use the netdev's statistics fields. "openvswitch: Fix ovs_vport_get_stats()" fixed the export of these stats to user-space, by using the provided netdev_ops to collate them - but ovs internal devices still use an unallocated dev->tstats field to count packets, which are no longer exported by this api. Allocate the dev->tstats field for ovs internal devices, and wire up ndo_get_stats64 with the original implementation of ovs_vport_get_stats(). On its own, "openvswitch: Fix ovs_vport_get_stats()" fixes the OOPs, unmasking a full-on panic on arm64: =============%<============== [] internal_dev_recv+0xa8/0x170 [openvswitch] [] do_output.isra.31+0x60/0x19c [openvswitch] [] do_execute_actions+0x208/0x11c0 [openvswitch] [] ovs_execute_actions+0xc8/0x238 [openvswitch] [] ovs_packet_cmd_execute+0x21c/0x288 [openvswitch] [] genl_family_rcv_msg+0x1b0/0x310 [] genl_rcv_msg+0xa4/0xe4 [] netlink_rcv_skb+0xb0/0xdc [] genl_rcv+0x38/0x50 [] netlink_unicast+0x164/0x210 [] netlink_sendmsg+0x304/0x368 [] sock_sendmsg+0x30/0x4c [SNIP] Kernel panic - not syncing: Fatal exception in interrupt =============%<============== Fixes: 8c876639c985 ("openvswitch: Remove vport stats.") Signed-off-by: James Morse Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- net/openvswitch/vport-internal_dev.c | 46 +++++++++++++++++++++++++++++++++--- 1 file changed, 43 insertions(+), 3 deletions(-) (limited to 'net/openvswitch/vport-internal_dev.c') diff --git a/net/openvswitch/vport-internal_dev.c b/net/openvswitch/vport-internal_dev.c index 388b8a6bf112..b3934126daa8 100644 --- a/net/openvswitch/vport-internal_dev.c +++ b/net/openvswitch/vport-internal_dev.c @@ -106,12 +106,45 @@ static void internal_dev_destructor(struct net_device *dev) free_netdev(dev); } +static struct rtnl_link_stats64 * +internal_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats) +{ + int i; + + memset(stats, 0, sizeof(*stats)); + stats->rx_errors = dev->stats.rx_errors; + stats->tx_errors = dev->stats.tx_errors; + stats->tx_dropped = dev->stats.tx_dropped; + stats->rx_dropped = dev->stats.rx_dropped; + + for_each_possible_cpu(i) { + const struct pcpu_sw_netstats *percpu_stats; + struct pcpu_sw_netstats local_stats; + unsigned int start; + + percpu_stats = per_cpu_ptr(dev->tstats, i); + + do { + start = u64_stats_fetch_begin_irq(&percpu_stats->syncp); + local_stats = *percpu_stats; + } while (u64_stats_fetch_retry_irq(&percpu_stats->syncp, start)); + + stats->rx_bytes += local_stats.rx_bytes; + stats->rx_packets += local_stats.rx_packets; + stats->tx_bytes += local_stats.tx_bytes; + stats->tx_packets += local_stats.tx_packets; + } + + return stats; +} + static const struct net_device_ops internal_dev_netdev_ops = { .ndo_open = internal_dev_open, .ndo_stop = internal_dev_stop, .ndo_start_xmit = internal_dev_xmit, .ndo_set_mac_address = eth_mac_addr, .ndo_change_mtu = internal_dev_change_mtu, + .ndo_get_stats64 = internal_get_stats, }; static struct rtnl_link_ops internal_dev_link_ops __read_mostly = { @@ -161,6 +194,11 @@ static struct vport *internal_dev_create(const struct vport_parms *parms) err = -ENOMEM; goto error_free_vport; } + vport->dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats); + if (!vport->dev->tstats) { + err = -ENOMEM; + goto error_free_netdev; + } dev_net_set(vport->dev, ovs_dp_get_net(vport->dp)); internal_dev = internal_dev_priv(vport->dev); @@ -173,7 +211,7 @@ static struct vport *internal_dev_create(const struct vport_parms *parms) rtnl_lock(); err = register_netdevice(vport->dev); if (err) - goto error_free_netdev; + goto error_unlock; dev_set_promiscuity(vport->dev, 1); rtnl_unlock(); @@ -181,8 +219,10 @@ static struct vport *internal_dev_create(const struct vport_parms *parms) return vport; -error_free_netdev: +error_unlock: rtnl_unlock(); + free_percpu(vport->dev->tstats); +error_free_netdev: free_netdev(vport->dev); error_free_vport: ovs_vport_free(vport); @@ -198,7 +238,7 @@ static void internal_dev_destroy(struct vport *vport) /* unregister_netdevice() waits for an RCU grace period. */ unregister_netdevice(vport->dev); - + free_percpu(vport->dev->tstats); rtnl_unlock(); } -- cgit v1.2.3 From aec15924740edc9886051593bc7769873be9498b Mon Sep 17 00:00:00 2001 From: Pravin B Shelar Date: Tue, 20 Oct 2015 23:00:10 -0700 Subject: openvswitch: Use dev_queue_xmit for vport send. With use of lwtunnel, we can directly call dev_queue_xmit() rather than calling netdev vport send operation. Following change make tunnel vport code bit cleaner. Signed-off-by: Pravin B Shelar Acked-by: Thomas Graf Acked-by: Jiri Benc Signed-off-by: David S. Miller --- net/openvswitch/vport-geneve.c | 2 +- net/openvswitch/vport-gre.c | 2 +- net/openvswitch/vport-internal_dev.c | 8 ++++---- net/openvswitch/vport-netdev.c | 33 +-------------------------------- net/openvswitch/vport-netdev.h | 1 - net/openvswitch/vport-vxlan.c | 2 +- net/openvswitch/vport.c | 30 ++++++++++++++++++++++++++++++ net/openvswitch/vport.h | 7 ++----- 8 files changed, 40 insertions(+), 45 deletions(-) (limited to 'net/openvswitch/vport-internal_dev.c') diff --git a/net/openvswitch/vport-geneve.c b/net/openvswitch/vport-geneve.c index 2735e9c4a3b8..7a568ca8da54 100644 --- a/net/openvswitch/vport-geneve.c +++ b/net/openvswitch/vport-geneve.c @@ -128,7 +128,7 @@ static struct vport_ops ovs_geneve_vport_ops = { .create = geneve_create, .destroy = ovs_netdev_tunnel_destroy, .get_options = geneve_get_options, - .send = ovs_netdev_send, + .send = dev_queue_xmit, .owner = THIS_MODULE, .get_egress_tun_info = geneve_get_egress_tun_info, }; diff --git a/net/openvswitch/vport-gre.c b/net/openvswitch/vport-gre.c index 4d24481669c9..cdb758ab01cf 100644 --- a/net/openvswitch/vport-gre.c +++ b/net/openvswitch/vport-gre.c @@ -94,7 +94,7 @@ static int gre_get_egress_tun_info(struct vport *vport, struct sk_buff *skb, static struct vport_ops ovs_gre_vport_ops = { .type = OVS_VPORT_TYPE_GRE, .create = gre_create, - .send = ovs_netdev_send, + .send = dev_queue_xmit, .get_egress_tun_info = gre_get_egress_tun_info, .destroy = ovs_netdev_tunnel_destroy, .owner = THIS_MODULE, diff --git a/net/openvswitch/vport-internal_dev.c b/net/openvswitch/vport-internal_dev.c index 388b8a6bf112..7f0a8bd08857 100644 --- a/net/openvswitch/vport-internal_dev.c +++ b/net/openvswitch/vport-internal_dev.c @@ -202,22 +202,21 @@ static void internal_dev_destroy(struct vport *vport) rtnl_unlock(); } -static void internal_dev_recv(struct vport *vport, struct sk_buff *skb) +static netdev_tx_t internal_dev_recv(struct sk_buff *skb) { - struct net_device *netdev = vport->dev; + struct net_device *netdev = skb->dev; struct pcpu_sw_netstats *stats; if (unlikely(!(netdev->flags & IFF_UP))) { kfree_skb(skb); netdev->stats.rx_dropped++; - return; + return NETDEV_TX_OK; } skb_dst_drop(skb); nf_reset(skb); secpath_reset(skb); - skb->dev = netdev; skb->pkt_type = PACKET_HOST; skb->protocol = eth_type_trans(skb, netdev); skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN); @@ -229,6 +228,7 @@ static void internal_dev_recv(struct vport *vport, struct sk_buff *skb) u64_stats_update_end(&stats->syncp); netif_rx(skb); + return NETDEV_TX_OK; } static struct vport_ops ovs_internal_vport_ops = { diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c index f7e8dcce7ada..b327368a3848 100644 --- a/net/openvswitch/vport-netdev.c +++ b/net/openvswitch/vport-netdev.c @@ -190,37 +190,6 @@ void ovs_netdev_tunnel_destroy(struct vport *vport) } EXPORT_SYMBOL_GPL(ovs_netdev_tunnel_destroy); -static unsigned int packet_length(const struct sk_buff *skb) -{ - unsigned int length = skb->len - ETH_HLEN; - - if (skb->protocol == htons(ETH_P_8021Q)) - length -= VLAN_HLEN; - - return length; -} - -void ovs_netdev_send(struct vport *vport, struct sk_buff *skb) -{ - int mtu = vport->dev->mtu; - - if (unlikely(packet_length(skb) > mtu && !skb_is_gso(skb))) { - net_warn_ratelimited("%s: dropped over-mtu packet: %d > %d\n", - vport->dev->name, - packet_length(skb), mtu); - vport->dev->stats.tx_errors++; - goto drop; - } - - skb->dev = vport->dev; - dev_queue_xmit(skb); - return; - -drop: - kfree_skb(skb); -} -EXPORT_SYMBOL_GPL(ovs_netdev_send); - /* Returns null if this device is not attached to a datapath. */ struct vport *ovs_netdev_get_vport(struct net_device *dev) { @@ -235,7 +204,7 @@ static struct vport_ops ovs_netdev_vport_ops = { .type = OVS_VPORT_TYPE_NETDEV, .create = netdev_create, .destroy = netdev_destroy, - .send = ovs_netdev_send, + .send = dev_queue_xmit, }; int __init ovs_netdev_init(void) diff --git a/net/openvswitch/vport-netdev.h b/net/openvswitch/vport-netdev.h index bf22fcedbc69..19e29c12adcc 100644 --- a/net/openvswitch/vport-netdev.h +++ b/net/openvswitch/vport-netdev.h @@ -27,7 +27,6 @@ struct vport *ovs_netdev_get_vport(struct net_device *dev); struct vport *ovs_netdev_link(struct vport *vport, const char *name); -void ovs_netdev_send(struct vport *vport, struct sk_buff *skb); void ovs_netdev_detach_dev(struct vport *); int __init ovs_netdev_init(void); diff --git a/net/openvswitch/vport-vxlan.c b/net/openvswitch/vport-vxlan.c index fb3cdb85905d..6f700710d413 100644 --- a/net/openvswitch/vport-vxlan.c +++ b/net/openvswitch/vport-vxlan.c @@ -170,7 +170,7 @@ static struct vport_ops ovs_vxlan_netdev_vport_ops = { .create = vxlan_create, .destroy = ovs_netdev_tunnel_destroy, .get_options = vxlan_get_options, - .send = ovs_netdev_send, + .send = dev_queue_xmit, .get_egress_tun_info = vxlan_get_egress_tun_info, }; diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c index 12a36ac21eda..ef19d0b77d13 100644 --- a/net/openvswitch/vport.c +++ b/net/openvswitch/vport.c @@ -537,3 +537,33 @@ int ovs_vport_get_egress_tun_info(struct vport *vport, struct sk_buff *skb, return vport->ops->get_egress_tun_info(vport, skb, upcall); } + +static unsigned int packet_length(const struct sk_buff *skb) +{ + unsigned int length = skb->len - ETH_HLEN; + + if (skb->protocol == htons(ETH_P_8021Q)) + length -= VLAN_HLEN; + + return length; +} + +void ovs_vport_send(struct vport *vport, struct sk_buff *skb) +{ + int mtu = vport->dev->mtu; + + if (unlikely(packet_length(skb) > mtu && !skb_is_gso(skb))) { + net_warn_ratelimited("%s: dropped over-mtu packet: %d > %d\n", + vport->dev->name, + packet_length(skb), mtu); + vport->dev->stats.tx_errors++; + goto drop; + } + + skb->dev = vport->dev; + vport->ops->send(skb); + return; + +drop: + kfree_skb(skb); +} diff --git a/net/openvswitch/vport.h b/net/openvswitch/vport.h index a413f3ae6a7b..885607f28d56 100644 --- a/net/openvswitch/vport.h +++ b/net/openvswitch/vport.h @@ -153,7 +153,7 @@ struct vport_ops { int (*set_options)(struct vport *, struct nlattr *); int (*get_options)(const struct vport *, struct sk_buff *); - void (*send)(struct vport *, struct sk_buff *); + netdev_tx_t (*send) (struct sk_buff *skb); int (*get_egress_tun_info)(struct vport *, struct sk_buff *, struct dp_upcall_info *upcall); @@ -234,9 +234,6 @@ static inline struct rtable *ovs_tunnel_route_lookup(struct net *net, return rt; } -static inline void ovs_vport_send(struct vport *vport, struct sk_buff *skb) -{ - vport->ops->send(vport, skb); -} +void ovs_vport_send(struct vport *vport, struct sk_buff *skb); #endif /* vport.h */ -- cgit v1.2.3